15 years ago2007-05-06 Erik Bunce <kde@bunce.us>
bdash [Mon, 7 May 2007 02:36:38 +0000 (02:36 +0000)]
2007-05-06  Erik Bunce  <kde@bunce.us>

        Reviewed by Mark Rowe.

        * WebKit.pri:
        Removed unnecessary dependency on MacPorts (/opt/local)

2007-05-06  Erik Bunce  <kde@bunce.us>

        Reviewed by Mark Rowe.

        * WebCore.pro:
        Removed unnecessary dependency on MacPorts (/opt/local)

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21277 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago2007-05-06 Mitz Pettel <mitz@webkit.org>
ap [Sun, 6 May 2007 16:42:10 +0000 (16:42 +0000)]
2007-05-06  Mitz Pettel  <mitz@webkit.org>

        Reviewed by Hyatt.

        - fix http://bugs.webkit.org/show_bug.cgi?id=13557
          Crash when dragging selection over absolutely positioned generated content on block element

        Test: fast/css-generated-content/positioned-background-hit-test-crash.html

        * rendering/RenderLayer.cpp:
        (WebCore::RenderLayer::enclosingElement): Added. Factored out of hitTestLayer().
        (WebCore::RenderLayer::hitTestLayer): Implemented the fix for <rdar://problem/3552346>
        from r6106 to the self hit test as well, namely: return the enclosing element
        for generated positioned content.
        * rendering/RenderLayer.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21276 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by Darin.
ap [Sun, 6 May 2007 16:05:01 +0000 (16:05 +0000)]
    Reviewed by Darin.

        <script> code wrongly assumes requests can't fail

        Test: fast/loader/unloadable-script.html

        * html/HTMLScriptElement.cpp:
        Check the return value of requestScript().

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21275 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by Darin.
ap [Sun, 6 May 2007 05:17:04 +0000 (05:17 +0000)]
    Reviewed by Darin.

        REGRESSION(14784): WebKit doesn't dynamically reflect changes to the highlight color in System Preferences

        Implemented a notification observer that invalidates cached system colors.

        No automated test possible.

        * rendering/RenderTheme.cpp:
        * rendering/RenderTheme.h:
        * rendering/RenderThemeMac.h:
        * rendering/RenderThemeMac.mm:
        (-[WebCoreRenderThemeNotificationObserver initWithTheme:WebCore::]):
        (-[WebCoreRenderThemeNotificationObserver systemColorsDidChange:]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21274 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by Oliver.
rwlbuis [Sat, 5 May 2007 10:53:58 +0000 (10:53 +0000)]
    Reviewed by Oliver.

        parseMappedAttribute inconsistency

        Do not use local vars in parseMappedAttribute for attr name and value.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21272 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by Hyatt.
oliver [Sat, 5 May 2007 07:10:14 +0000 (07:10 +0000)]
    Reviewed by Hyatt.

        Initialise default fill with Color::black rather than parsing
        the string "black"

        * ksvg2/svg/SVGPaint.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21271 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by Oliver.
sfalken [Sat, 5 May 2007 04:08:29 +0000 (04:08 +0000)]
    Reviewed by Oliver.

        Use TextBreakIterator for sentence breaking instead of TextBoundaries.

        * WebCore.xcodeproj/project.pbxproj: Added TextBreakIeratorInternalICUMac.mm
        * editing/visible_units.cpp:
        (WebCore::startSentenceBoundary): Call TextBreakIterator instead of TextBoundaries.
        (WebCore::endSentenceBoundary): Call TextBreakIterator instead of TextBoundaries.
        (WebCore::previousSentencePositionBoundary): Call TextBreakIterator instead of TextBoundaries.
        (WebCore::nextSentencePositionBoundary): Call TextBreakIterator instead of TextBoundaries.
        * platform/TextBoundaries.h: Removed findSentenceBoundary, findNextSentenceFromIndex.
        * platform/TextBreakIterator.h: Added sentenceBreakIterator.
        * platform/TextBreakIteratorICU.cpp:
        (WebCore::setUpIterator): Add locale parameter (needed to maintain functionality that TextBoundaries code had).
        (WebCore::characterBreakIterator): Pass en_us for locale (same as old code).
        (WebCore::wordBreakIterator): Pass en_us for locale (same as old code).
        (WebCore::lineBreakIterator): Pass en_us for locale (same as old code).
        (WebCore::sentenceBreakIterator): Added.
        * platform/TextBreakIteratorInternalICU.h: Added.
        * platform/gdk/TemporaryLinkStubs.cpp: Removed findNextSentenceFromIndex, findSentenceBoundary.
        (WebCore::currentTextBreakLocaleID): Added stub.
        * platform/mac/TextBoundaries.mm: Removed currentTextBreakLocaleID, findSentenceBoundary, findNextSentenceFromIndex.
        * platform/mac/TextBreakIteratorInternalICUMac.mm: Added.
        (WebCore::currentTextBreakLocaleID): Moved from TextBoundaries.mm
        * platform/qt/TextBoundaries.cpp: Removed findNextSentenceFromIndex, findSentenceBoundary.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21270 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by Hyatt.
adele [Sat, 5 May 2007 00:17:19 +0000 (00:17 +0000)]
    Reviewed by Hyatt.

        Fix for <rdar://problem/5116871> REGRESSION: Can't select options in widget's listbox

        Don't use user-select to determine whether or not options can be selected in listboxes.
        This will match Firefox behavior for -moz-user-select.

        * html/HTMLSelectElement.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21269 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoWebCore:
ggaren [Fri, 4 May 2007 23:39:01 +0000 (23:39 +0000)]

        Reviewed by Tim Hatcher.

        First step in fixing <rdar://problem/5055182> The back cache has no global cap

        Stop giving SnapBack infinite cache-ability. Instead, make SnapBack rely
        on the underlying back cache.

        * WebCore.exp:
        * history/HistoryItem.cpp:
        * history/HistoryItem.h:
        * loader/FrameLoader.cpp:


        Reviewed by Tim Hatcher.

        First step in fixing <rdar://problem/5055182> The back cache has no global cap

        Stop giving SnapBack infinite cache-ability. Instead, make SnapBack rely
        on the underlying back cache.

        I left -setAlwaysAttemptToUsePageCache: as an empty stub because we don't
        want to break Safari 2.0, but I removed its header declaration so nobody
        else starts using it.

        * History/WebHistoryItem.mm:
        (-[WebHistoryItem setAlwaysAttemptToUsePageCache:]):
        * History/WebHistoryItemPrivate.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21261 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoWebCore:
ggaren [Fri, 4 May 2007 21:15:28 +0000 (21:15 +0000)]

        Reviewed by Brady Eidson.

        Removed export of clearPageCache(), since it was unused externally. I plan
        to remove its use from WebCore eventually, too.

        Removed usesPageCache(), since it was unused and it duplicated pageCacheSize().

        * WebCore.exp:
        * history/BackForwardList.cpp:
        * history/BackForwardList.h: Moved statics to the top of the class definition.
        * loader/FrameLoader.cpp:


        Reviewed by Brady Eidson.

        Some cleanup in preparation for fixing <rdar://problem/5055182> The
        back/forward cache has no global cap

        Unified naming of WebKit/WebCore b/f lists -- instead of the potpourri
        of webBackForwardList, backForwardList, list, kitList, coreList, listWrapper,
        and webCoreBackForwardList, we use webBackForwardList for WebKit and
        backForwardList for WebCore, matching their respective class names.

        Removed "private" versions of kit() and core() -- kit() and core() are
        canonically used for converting between WebKit API objects and WebCore
        API objects. I think it's clearer to have only one way to do this.

        Removed COMPUTE_DEFAULT_PAGE_CACHE_SIZE, since it was unused.

        Removed _clearPageCache, since it was unused and it duplicated -setPageCacheSize:0.

        Removed _usesPageCache, since it was unused and it duplicated -pageCacheSize.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21258 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoAdded bugs.webkit.org URL.
darin [Fri, 4 May 2007 17:38:25 +0000 (17:38 +0000)]
Added bugs.webkit.org URL.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21257 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoJavaScriptCore:
darin [Fri, 4 May 2007 17:14:57 +0000 (17:14 +0000)]

        Reviewed by Adele.

        - fix <rdar://problem/5007921> Number.toExponential doesn't work for negative numbers

        * kjs/number_object.cpp: (NumberProtoFunc::callAsFunction):
        Added a call to fabs before calling log10.


        Reviewed by Adele.

        - test for <rdar://problem/5007921> Number.toExponential doesn't work for negative numbers

        * fast/js/kde/resources/Number.js: Added negative number versions of some tests of
        toFixed, toExponential, toPrecision, etc.
        * fast/js/kde/Number-expected.txt: Updated, new tests all passing.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21256 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by Mark Rowe
beidson [Fri, 4 May 2007 17:13:21 +0000 (17:13 +0000)]
    Reviewed by Mark Rowe

        Added main thread assertion to WebHTMLView to help make sure 3rd party clients aren't trying
        to draw on secondary threads

        * WebView/WebHTMLView.mm:
        (-[WebHTMLView drawRect:]): Added ASSERT_MAIN_THREAD()

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21255 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago2007-05-04 Anders Carlsson <andersca@apple.com>
bdash [Fri, 4 May 2007 16:42:11 +0000 (16:42 +0000)]
2007-05-04  Anders Carlsson  <andersca@apple.com>

        Reviewed by Antti.

        <rdar://problem/5179977> Use the correct URLs when dispatching delegate methods for data loads.

        Restore behavior to matching release WebKit by adding a response URL to SubstituteData. If a nil base URL is
        passed, set the _request_ URL to "about:blank", but the response URL to a generated "applewebdata" URL.

        * loader/FrameLoader.cpp:
        First, check for the SubstituteData's response URL.

        No need to special case applewebdata URLs here, they're only used in the response.

        * loader/MainResourceLoader.cpp:
        If the substitute data has a response URL, use it.

        Only load "about:blank" as an empty document if there's no substitute data.

        * loader/SubstituteData.h:
        Add responseURL member to SubstituteData.

2007-05-04  Anders Carlsson  <andersca@apple.com>

        Reviewed by Antti.

        <rdar://problem/5179977> Use the correct URLs when dispatching delegate methods for data loads.

        * WebView/WebFrame.mm:
        (-[WebFrame _loadData:MIMEType:textEncodingName:baseURL:unreachableURL:]):
        If the base URL is nil, set it to "about:blank", and set the response URL to a generated applewebdata URL.

        * WebView/WebView.mm:
        (+[WebView _canHandleRequest:]):
        No need to special-case applewebdata URLs here anymore, they're only used in responses.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21254 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by Zack
lars [Fri, 4 May 2007 12:25:13 +0000 (12:25 +0000)]
    Reviewed by Zack

        Revert on line of the last commit. We still want to keep the RPATH
        support in the pro file.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21253 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by Zack
lars [Fri, 4 May 2007 11:41:21 +0000 (11:41 +0000)]
    Reviewed by Zack

        Remove the old bitmap fonts I tried using to get reliable results from
        DumpRenderTree on X11. Instead we now use the URW Type1 fonts from
        ghostscript. I've added a mirror to simply check them out at

        Fixed DumpRenderTree to make sure these fonts are the only ones we use
        on X11 and added a fonts.conf file to get a well defined fontconfig

        Made sure run-webkit-tests forwards the WEBKIT_TESTFONTS environment variable
        I use to find the fonts to DumpRenderTree.

        Updated results for css1 to work with the new set of fonts.
        Will update the other results once I see this works on all
        machines we have.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21252 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago2007-05-03 Steve Falkenburg <sfalken@apple.com>
sfalken [Fri, 4 May 2007 06:44:48 +0000 (06:44 +0000)]
2007-05-03  Steve Falkenburg  <sfalken@apple.com>

        Reviewed by Oliver.

        Add missing user description parameter to spelling-related editor client method.

        * WebCoreSupport/WebEditorClient.h:

2007-05-03  Steve Falkenburg  <sfalken@apple.com>

        Reviewed by Oliver.

        Add missing user description parameter to spelling-related editor client method.

        * bridge/EditorClient.h:
        * platform/gdk/EditorClientGdk.cpp:
        * platform/gdk/EditorClientGdk.h:
        * platform/graphics/svg/SVGImageEmptyClients.h:

2007-05-03  Steve Falkenburg  <sfalken@apple.com>

        Reviewed by Oliver.

        Add missing user description parameter to spelling-related editor client method.

        * WebCoreSupport/EditorClientQt.cpp:
        * WebCoreSupport/EditorClientQt.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21246 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by Oliver
beidson [Fri, 4 May 2007 04:35:28 +0000 (04:35 +0000)]
    Reviewed by Oliver

        Resolve an outstanding FIXME in Loader::numRequests()

        Before, numRequests() would iterate through the list of requests pending load and the list of currently
        loading requests and tally up a count matching the current DocLoader.

        I noticed while studying and cleaning up the loader code that numRequests() is potentially very hot!
        Indeed load a complex site with many resources and multiple frames, and this method gets called very often,
        tallying up this number every time.

        The FIXME was to keep a collection of Requests mapped to each DocLoader.  In reality, since this map would
        simply be used for retrieving a count, that was overkill.  Keeping a request count in the DocLoader itself
        along with maintaining that count in Loader as requests come and go is a much better way to do this.

        * loader/DocLoader.cpp:
        (WebCore::DocLoader::requestCount): Emulate the defunct Loader::numRequests()
        * loader/DocLoader.h:

        * loader/FrameLoader.cpp:
        (WebCore::numRequests): Call DocLoader::requestCount() directly
        (WebCore::FrameLoader::checkCompleted): Use numRequests()

        * loader/loader.cpp:
        (WebCore::Loader::load): Increment the DocLoader's request count
        (WebCore::Loader::servePendingRequests): If the SubresourceLoader failed to create, decrement the count
        (WebCore::Loader::didFinishLoading): If the Request is not Multipart, decrement the count
        (WebCore::Loader::didFail): If the Request is not Multipart, decrement the count
        (WebCore::Loader::didReceiveResponse): If the Request becomes Multipart, decrement the count
        (WebCore::Loader::cancelRequests): Decrement the count for the pending requests being tossed, and ASSERT the
          count is zero after all requests have been cancelled
        * loader/loader.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21245 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by Brady Eidson.
ggaren [Fri, 4 May 2007 03:55:29 +0000 (03:55 +0000)]
    Reviewed by Brady Eidson.

        Some cleanup in preparation for fixing <rdar://problem/5055182> The
        back/forward cache has no global cap

        * loader/FrameLoader.cpp:
        (WebCore::FrameLoader::provisionalLoadStarted): Moved "Can I cache this
        page?" logic to canCachePage(). Moved "I am caching this page" logic to
        (WebCore::FrameLoader::canCachePage): Moved above logic here.
        (WebCore::FrameLoader::cachePageToHistoryItem): Removed bogus NULL check
        and failure return value. Moved above logic here. Don't call setCachedPage()
        until the CachedPage is fully constructed.
        (WebCore::FrameLoader::purgePageCache): Removed comment that duplicated
        the function name.
        * loader/FrameLoader.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21244 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by harrison.
harrison [Fri, 4 May 2007 01:54:27 +0000 (01:54 +0000)]
    Reviewed by harrison.

        <rdar://problem/5120281> CrashTracer: [USER] 1 crash in Mail at WebCore::CSSComputedStyleDeclaration::copyInheritableProperties() const

        Problem is that checkAncestor and lastClosed can be nil when createMarkup is called unrendered nodes, and that was not checked for.

        * editing/markup.cpp:
        Nil-check checkAncestor and lastClosed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21243 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by Kevin.
thatcher [Thu, 3 May 2007 23:08:28 +0000 (23:08 +0000)]
    Reviewed by Kevin.

        <rdar://problem/4975212> REGRESSION: With NetNewsWire 2.1.1, the contextual menu shows extra menu items when focus
        is placed in input or textarea field

        The NetNewsWire UI delegate isn't expecting calls for form controls, so we need to do a linked-on-or-after check.
        If the application was linked against Tiger or earlier and the element is a text form control, just return the
        default menu items and bypass the delegate call completely.

        * WebCoreSupport/WebContextMenuClient.mm:
        (isPreVersion3Client): Cache the result of the WebKitLinkedOnOrAfter call
        (fixMenusToSendToOldClients): Call the new isPreVersion3Client()
        (fixMenusReceivedFromOldClients): Ditto.
        (WebContextMenuClient::getCustomMenuFromDefaultItems): Return the default menu items if the element is a text form control.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21240 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago2007-05-03 Mark Rowe <mrowe@apple.com>
bdash [Thu, 3 May 2007 22:48:45 +0000 (22:48 +0000)]
2007-05-03  Mark Rowe  <mrowe@apple.com>

        Reviewed by Geoff and Kevin.

        <rdar://problem/5141290> WebAssertions.h is still needed by some internal clients

        Second shot at fixing this error.  Stub out the macros rather than forwarding to JavaScriptCore, which would
        leave clients using this header trying to resolve JavaScriptCore symbols against WebKit when linking.  This
        should only happen in production builds when assertions should be disabled anyway as anyone building a
        development configuration should be in a position to move away from using this header.

        * Misc/WebAssertions.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21238 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by Kevin.
thatcher [Thu, 3 May 2007 22:23:18 +0000 (22:23 +0000)]
    Reviewed by Kevin.

        <rdar://problem/5016318> Contextual menu item for opening links is missing in compose windows but present in viewer

        Add the Open Link, Open Link in New Window, Download Linked File and Copy Link menu
        items to the content editable context menu.

        * platform/ContextMenu.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21237 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoGdk build fixes using the Qt build's qmake tool from Holger Freyther, reviewed by...
hausmann [Thu, 3 May 2007 22:04:27 +0000 (22:04 +0000)]
Gdk build fixes using the Qt build's qmake tool from Holger Freyther, reviewed by Zack.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21235 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by Kevin.
thatcher [Thu, 3 May 2007 21:50:56 +0000 (21:50 +0000)]
    Reviewed by Kevin.

        <rdar://problem/5067707> REGRESSION: "Open Link" contextual menu item appears twice in Mail

        Remove the check for Mail in fixMenusToSendToOldClients and fixMenusReceivedFromOldClients
        when linked on or after Leopard. The isAppleMail() function is still used for Tiger Mail fixups.

        * WebCoreSupport/WebContextMenuClient.mm:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21234 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Use BUILDING_ON_TIGER.
adele [Thu, 3 May 2007 21:37:16 +0000 (21:37 +0000)]

        * page/mac/FrameMac.mm: (WebCore::Frame::setUseSecureKeyboardEntry):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21233 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by Darin
beidson [Thu, 3 May 2007 21:35:24 +0000 (21:35 +0000)]
    Reviewed by Darin

        Some more minor cleanup that resolves a FIXME and ditches another pointless method

        No change in functionality == no layout test

        * loader/FrameLoader.cpp:
        (WebCore::FrameLoader::commitProvisionalLoad):  Updated a comment
        (WebCore::FrameLoader::continueLoadAfterWillSubmitForm): Folded startLoading() into this method
        (WebCore::FrameLoader::continueLoadAfterNavigationPolicy): continueAfterWillSubmitForm becomes continueLoadAfterWillSubmitForm
        * loader/FrameLoader.h: Removed startLoading()

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21232 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by Tim Hatcher.
adele [Thu, 3 May 2007 21:29:41 +0000 (21:29 +0000)]
    Reviewed by Tim Hatcher.

        Fix for <rdar://problem/4727607> REGRESSION: KeyScript is deprecated, need a new solution for forcing Roman input for password fields on Leopard

        * page/mac/FrameMac.mm: (WebCore::Frame::setUseSecureKeyboardEntry):
        Use TSMSetDocumentProperty and TSMRemoveDocumentProperty on Leopard disable and enable non-ascii input sources.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21230 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by Anders
beidson [Thu, 3 May 2007 18:50:41 +0000 (18:50 +0000)]
    Reviewed by Anders

        commitProvisionalLoad() was the name of two methods.
        One of them was private and only called from one other private place.
        This was driving me crazy.  I had to change that.

        No change in functionality == no layout test

        * loader/FrameLoader.cpp:
        (WebCore::FrameLoader::transitionToCommitted): roll commitProvisionalLoad() into this method
        * loader/FrameLoader.h: Removed commitProvisionalLoad()

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21229 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Build fix: Add AllInOneFile.o to the ignore list, since it includes files
ggaren [Thu, 3 May 2007 18:37:57 +0000 (18:37 +0000)]
    Build fix: Add AllInOneFile.o to the ignore list, since it includes files
        in the ignore list.

        * Scripts/check-for-global-initializers:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21228 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by Anders.
darin [Thu, 3 May 2007 16:23:53 +0000 (16:23 +0000)]
    Reviewed by Anders.

        - make ICU initialization lazy to speed up application start-up time
          (when it includes an initial webpage load)

        * platform/TextCodecICU.h: Replaced registerEncodingNames and
        registerCodecs with registerBaseEncodingNames, registerBaseCodecs,
        registerExtendedEncodingNames, and registerExtendedCodecs.
        * platform/TextCodecICU.cpp:
        (WebCore::newTextCodecICU): Moved up to the top of the file.
        (WebCore::TextCodecICU::registerBaseEncodingNames): Added. Registers
        just UTF-8.
        (WebCore::TextCodecICU::registerBaseCodecs): Ditto.
        (WebCore::TextCodecICU::registerExtendedEncodingNames): Renamed.
        (WebCore::TextCodecICU::registerExtendedCodecs): Ditto.
        (WebCore::TextCodecICU::createICUConverter): Compare with the name
        "GBK" in a way that does not require extending the maps.

        * platform/TextEncoding.cpp:
        (WebCore::TextEncoding::usesVisualOrdering): Added code so this will
        just return false if we haven't used any extended encoding names yet.
        The result will be correct, and we won't ask for an encoding name that
        will require registering the extended encoding names.
        (WebCore::TextEncoding::isJapanese): Ditto.
        (WebCore::TextEncoding::backslashAsCurrencySymbol): Ditto, but return
        a backslash instead of false.

        * platform/TextEncodingRegistry.h: Added a new function named
        * platform/TextEncodingRegistry.cpp:
        (WebCore::buildBaseTextCodecMaps): Added. Registers the base encodings,
        including Latin-1 because it's the default, and the UTF encodings.
        (WebCore::extendTextCodecMaps): Added. Registers all the other encodings.
        (WebCore::newTextCodec): Removed the all to buildTextCodecMap because
        we always build the map when making the canonical encoding name.
        (WebCore::atomicCanonicalTextEncodingName): Changed code to build only
        the base maps at first, then extend the maps the first time we run into
        a name we don't know.
        (WebCore::noExtendedTextEncodingNameUsed): Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21227 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago2007-05-03 Mark Rowe <mrowe@apple.com>
bdash [Thu, 3 May 2007 04:39:37 +0000 (04:39 +0000)]
2007-05-03  Mark Rowe  <mrowe@apple.com>

        Qt build fix.  In r21185 the Qt version of WebCore::scaleDragImage was
        updated to have the incorrect signature.

        * platform/qt/DragImageQt.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21226 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago * dom/Node.h: Just fixing a variable name because it's driving me
bdakin [Wed, 2 May 2007 22:27:56 +0000 (22:27 +0000)]
    * dom/Node.h: Just fixing a variable name because it's driving me

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21225 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by Hyatt.
bdakin [Wed, 2 May 2007 22:16:37 +0000 (22:16 +0000)]
    Reviewed by Hyatt.

        Fix for <rdar://problem/4513966> Never-ending identical repaint @
        orbitz.com (Safari eating 85% CPU)

        We need to be a little bit more savvy about when we do a FULL style
        recalc since it can be unnecessary and expensive.

        * css/CSSMutableStyleDeclaration.cpp:
        (WebCore::CSSMutableStyleDeclaration::setChanged): Now takes a
        StyleChangeType as a parameter.
        (WebCore::CSSMutableStyleDeclaration::setProperty): Send
        InlineStyleChange as the paramter. This is the optimization!
        * css/CSSMutableStyleDeclaration.h: setChanged takes a
        StyleChangeType as a parameter.
        * dom/Document.cpp: setChanged now takes a StyleChangeType instead
        of a bool.
        (WebCore::Document::recalcStyle): Same.
        * dom/Element.cpp:
        (WebCore::Element::recalcStyle): This is the fix!! Only set change
        to Force is we use descendant rules AND it's a FullStyleChange.
        * dom/Node.cpp: m_hasStyle was an unused bit on Node, so this patch
        gets rid of it and uses that extra bit for m_changed to be a
        StyleChangeType instead of a bool.
        * dom/Node.h: Same.
        (WebCore::): Define the StyleChangeType enum.
        * dom/StyledElement.cpp:
        (WebCore::StyledElement::parseMappedAttribute): Don't bother
        calling setHasStyle() since m_hasStyle is gone now. It was never
        used, and this was the only place it was set.
        * dom/Text.cpp:
        (WebCore::Text::recalcStyle): setChanged now takes a
        StyleChangeType instead of a bool.
        * html/HTMLFrameSetElement.cpp:
        (WebCore::HTMLFrameSetElement::recalcStyle): Same.
        * html/HTMLSelectElement.cpp:
        (WebCore::HTMLSelectElement::restoreState): Same.
        (WebCore::HTMLSelectElement::reset): Same.
        * html/HTMLTextAreaElement.cpp:
        (WebCore::HTMLTextAreaElement::setValue): Same.
        * ksvg2/misc/SVGTimer.cpp:
        (WebCore::SVGTimer::applyAnimations): Same.
        * ksvg2/svg/SVGUseElement.cpp:
        (WebCore::SVGUseElement::recalcStyle): Same.
        * platform/graphics/svg/SVGResource.cpp:
        (WebCore::SVGResource::repaintClients): Same.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21224 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by Brady.
andersca [Wed, 2 May 2007 19:36:22 +0000 (19:36 +0000)]
    Reviewed by Brady.

        <rdar://problem/5151113> Assertion firing in [FrameProgressEntry addChild:forDataSource:] when navigating cnn.com

        The assertion fired because a plug-in was trying to load a subresource when a new load had started but not yet
        committed. The check that would have prevented this was removed in order to fix <rdar://problem/5085897>.

        This puts back the check but changes it to allow loads where the target is the same frame as the plugin's parent frame.

        * Plugins/WebBaseNetscapePluginView.mm:
        (-[WebBaseNetscapePluginView loadRequest:inTarget:withNotifyData:sendNotification:]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21223 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by Antti.
harrison [Wed, 2 May 2007 14:54:20 +0000 (14:54 +0000)]
    Reviewed by Antti.

        <rdar://problem/4859132> Grammar must always be checked in entire-sentence chunks, and shouldn't show markers for current sentence

        Restore the remainder of the original fix now that blocking bug rdar://5174862 is addressed.

        John made the original patch on 2007-04-10, but this fell prey to rdar://5157329.
        Justin backed out the whole patch on 2007-04-30 to avoid the crash.
        Justin then restored most of the original patch, but used word granularity instead, to keep the build working.
        I just addressed the root cause of rdar://5157329 in rdar://5174862.
        This patch switches the temp uses of word granularity back to sentence granularity.

        * editing/mac/EditorMac.mm:
        * page/Frame.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21222 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoJavaScriptCore:
harrison [Wed, 2 May 2007 13:11:05 +0000 (13:11 +0000)]

        Reviewed by Antti.

        <rdar://problem/5174862> Crash resulting from DeprecatedString::insert()

        Added insertion support for more than one value.

        * wtf/Vector.h:
        Added support for inserting multiple values.

        New. Insert at the start of vectors. Convenient for vectors used as strings.


        Reviewed by Antti.

        - test for <rdar://problem/5174862> Crash resulting from DeprecatedString::insert()

        * editing/selection/move-by-sentence-001-expected.checksum: Added.
        * editing/selection/move-by-sentence-001-expected.png: Added.
        * editing/selection/move-by-sentence-001-expected.txt: Added.
        * editing/selection/move-by-sentence-001.html: Added.


        Reviewed by Antti.

        <rdar://problem/5174862> Crash resulting from DeprecatedString::insert()

        Test added: editing/selection/move-by-sentence-001.html

        * editing/visible_units.cpp:
        Use a UChar Vector instead of DeprecatedString.
        Avoid creating an extra string for secure bullet replacement unless it is actually needed.

        * platform/DeprecatedString.cpp:
        Call forceUnicode() before setLength(), so that only the unicode buffer is resized.

        Adjust the unicode buffer even if the ascii buffer is valid.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21221 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago2007-05-02 Holger Freyther <freyther@kde.org>
bdash [Wed, 2 May 2007 09:21:34 +0000 (09:21 +0000)]
2007-05-02  Holger Freyther <freyther@kde.org>

        Reviewed by Mark Rowe.

        * Scripts/build-webkit: Remove the CMake call.
        * Scripts/webkitdirs.pm: Remove isQtWithQMake as QMake is now the only
        buildsystem for the Qt port.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21220 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago2007-05-02 Mitz Pettel <mitz@webkit.org>
bdash [Wed, 2 May 2007 09:20:00 +0000 (09:20 +0000)]
2007-05-02  Mitz Pettel  <mitz@webkit.org>

        Reviewed by Brady.

        - http://bugs.webkit.org/show_bug.cgi?id=13564
          Remove friend class FrameLoader from Frame

        No test added since there is no change in functionality.

        * loader/FrameLoader.cpp:
        * loader/FrameLoader.h: Made clear() private.
        * page/Frame.cpp:
        (WebCore::Frame::setDocument): Changed the argument into a PassRefPtr.
        * page/Frame.h:
        * xml/XSLTProcessor.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21219 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago2007-05-02 Mitz Pettel <mitz@webkit.org>
bdash [Wed, 2 May 2007 09:17:52 +0000 (09:17 +0000)]
2007-05-02  Mitz Pettel  <mitz@webkit.org>

        Reviewed by Dave Hyatt.

        - fix http://bugs.webkit.org/show_bug.cgi?id=9276
          Quirksmode: Image swap resize bug

        Test: fast/replaced/image-resize-width.html

        * rendering/RenderImage.cpp:
        (WebCore::RenderImage::imageChanged): Moved the call to dirtyPrefWidths()
        before the call to calcWidth().

2007-05-02  Mitz Pettel  <mitz@webkit.org>

        Reviewed by Dave Hyatt.

        - test for http://bugs.webkit.org/show_bug.cgi?id=9276
          Quirksmode: Image swap resize bug

        * fast/replaced/image-resize-width-expected.checksum: Added.
        * fast/replaced/image-resize-width-expected.png: Added.
        * fast/replaced/image-resize-width-expected.txt: Added.
        * fast/replaced/image-resize-width.html: Added.
        * fast/replaced/resources/rectangle-red-200x100.png: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21218 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoAdd layout test for dashboard bug.
hyatt [Wed, 2 May 2007 08:44:02 +0000 (08:44 +0000)]
Add layout test for dashboard bug.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21217 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Fix for bug 13566, broken Dashboard widgets. setPrefWidthsDirty did
hyatt [Wed, 2 May 2007 08:39:41 +0000 (08:39 +0000)]
    Fix for bug 13566, broken Dashboard widgets.  setPrefWidthsDirty did
        the wrong thing for text children of a positioned element.

        Reviewed by mitz


        * rendering/RenderObject.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21216 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago * rendering/RenderReplaced.cpp: Fix typo in a comment.
darin [Wed, 2 May 2007 01:40:31 +0000 (01:40 +0000)]
    * rendering/RenderReplaced.cpp: Fix typo in a comment.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21215 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoAdd layout test for 13558.
hyatt [Tue, 1 May 2007 22:31:59 +0000 (22:31 +0000)]
Add layout test for 13558.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21214 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Fix for 13558, crash when using absolutely positioned generated content
hyatt [Tue, 1 May 2007 22:27:03 +0000 (22:27 +0000)]
    Fix for 13558, crash when using absolutely positioned generated content
        inside an inline.

        Reviewed by darin


        * rendering/RenderContainer.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21213 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoLayoutTests:
justing [Tue, 1 May 2007 22:20:46 +0000 (22:20 +0000)]

        Reviewed by darin

        Pressing space key does nothing above quoted content or a signature

        Demonstrates the bug:
        * editing/inserting/5002441-expected.checksum: Added.
        * editing/inserting/5002441-expected.png: Added.
        * editing/inserting/5002441-expected.txt: Added.
        * editing/inserting/5002441.html: Added.

        Fixed.  Spaces passed to execCommand("InsertText", ...)
        are no longer collapsed:
        * editing/inserting/editable-html-element-expected.checksum:
        * editing/inserting/editable-html-element-expected.png:
        * editing/inserting/editable-html-element-expected.txt:
        * editing/pasteboard/4989774-expected.checksum:
        * editing/pasteboard/4989774-expected.png:
        * editing/pasteboard/4989774-expected.txt:
        * editing/selection/4983858-expected.checksum:
        * editing/selection/4983858-expected.png:
        * editing/selection/4983858-expected.txt:


        Reviewed by darin

        Pressing space key does nothing above quoted content or a signature

        Inserting a space under these circumstances inserts a single
        text node containing a regular space and then does a layout.
        That space isn't rendered (which is correct).  Whitespace
        rebalancing is supposed to correct it but failed.  It replaces
        the space with a non-breaking space, but that change doesn't
        dirty line boxes (9441) and so the space isn't rendered.

        This workaround turns all incoming spaces into non-breaking
        spaces before they're inserted (they're rebalanced after
        insertion and turned back into regular spaces if possible).

        * editing/InsertTextCommand.cpp:
        (WebCore::InsertTextCommand::prepareForTextInsertion): Removed
        an old irrelevant FIXME.
        (WebCore::InsertTextCommand::input): Turn incoming spaces into
        non breaking spaces before inserting them.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21212 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoLayoutTests:
darin [Tue, 1 May 2007 22:10:41 +0000 (22:10 +0000)]

        Reviewed by Hyatt.

        - test for http://bugs.webkit.org/show_bug.cgi?id=13559
          REGRESSION: Canvas aspect ratio is incorrect

        * fast/canvas/canvas-size-change-after-layout-expected.checksum: Added.
        * fast/canvas/canvas-size-change-after-layout-expected.png: Added.
        * fast/canvas/canvas-size-change-after-layout-expected.txt: Added.
        * fast/canvas/canvas-size-change-after-layout.html: Added.


        Reviewed by Hyatt.

        - fix http://bugs.webkit.org/show_bug.cgi?id=13559
          REGRESSION: Canvas aspect ratio is incorrect

        - remove intrinsicWidth/Height functions from RenderObject

        Test: fast/canvas/canvas-size-change-after-layout.html

        * html/HTMLCanvasElement.cpp:
        (WebCore::HTMLCanvasElement::createRenderer): Removed code to set
        the intrinsic size, because the renderer now takes care of this
        in its constructor.
        (WebCore::HTMLCanvasElement::reset): Instead of calling the
        setIntrinsicWidth/Height functions, call canvasSizeChanged, which
        takes care of layout, but only if the size actually changed. Also
        call repaint if the contents of the canvas changed.

        * rendering/RenderApplet.h:
        * rendering/RenderApplet.cpp: (WebCore::RenderApplet::intrinsicSize):
        Override the new intrinsicSize function. This override still doesn't
        make sense, so left a FIXME in here.

        * rendering/RenderBox.h: Added an intrinsicSize function here to replace
        the intrinsicWidth/Height functions removed from RenderObject.
        * rendering/RenderBox.cpp:
        (WebCore::RenderBox::calcReplacedWidthUsing): Use intrinsicSize instead
        of intrinsicWidth.
        (WebCore::RenderBox::calcReplacedHeightUsing): Ditto, but for height.

        * rendering/RenderHTMLCanvas.h: Changed the constructor parameter to be
        an HTMLCanvasElement* instead of Node*. Also added a canvasSizeChanged
        * rendering/RenderHTMLCanvas.cpp:
        (WebCore::RenderHTMLCanvas::RenderHTMLCanvas): Pass an initial intrinsic
        size based on the current size of the canvas element. This is now the
        responsibility of the renderer and the DOM element calls the canvasSizeChanged
        function when the size changes.
        (WebCore::RenderHTMLCanvas::paint): Removed checks of the type of the element,
        and used the node() function because we don't have to worry about the anonymous
        case in this class.
        (WebCore::RenderHTMLCanvas::canvasSizeChanged): Added. Changes the size and
        then calls setNeedsLayout and setPrefWidthsDirty as needed. I based the logic
        here on careful reading of RenderImage::imageChanged.

        * rendering/RenderImage.cpp:
        (WebCore::RenderImage::RenderImage): Pass a default intrinsic size of 0,0
        instead of calling setIntrinsicWidth/Height.
        (WebCore::RenderImage::setImageSizeForAltText): Call the new intrinsic size
        functions instead of the width/height ones. Simplifies the logic since we
        handle both dimensions at once.
        (WebCore::RenderImage::imageChanged): Ditto.
        (WebCore::RenderImage::layout): Ditto.
        (WebCore::RenderImage::calcAspectRatioWidth): Ditto.
        (WebCore::RenderImage::calcAspectRatioHeight): Ditto.

        * rendering/RenderObject.h: Removed intrinsicWidth/Height virtual functions.

        * rendering/RenderReplaced.h: Replaced intrinsicWidth, intrinsicHeight,
        setIntrinsicWidth, setIntrinsicHeight, m_intrinsicWidth, and m_intrinsicHeight
        with intrinsicSize, setIntrinsicSize, and m_intrinsicSize.
        * rendering/RenderReplaced.cpp:
        (WebCore::RenderReplaced::RenderReplaced): Added a second contructor that takes
        an intrinsic size parameter. Also updated constructor since the intrinsic size
        is now an IntSize instead of two integers.
        (WebCore::RenderReplaced::intrinsicSize): Added. No point in making this inline
        because it's a virtual function.
        (WebCore::RenderReplaced::setIntrinsicSize): Added. Didn't make this inline,
        partly because in the future we might want this to be more than just a simple
        setter function, factoring in common code used in both RenderImage and

        * rendering/RenderWidget.h: Removed a stray using declaration that wasn't needed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21211 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Fix for hang when incorrectly trying to add before/after content to
hyatt [Tue, 1 May 2007 20:11:02 +0000 (20:11 +0000)]
    Fix for hang when incorrectly trying to add before/after content to
        a <select>.  Disallow generated content in all selects and also turn it
        off for <input> buttons.

        Reviewed by aroben

        * rendering/RenderBlock.cpp:
        * rendering/RenderButton.cpp:
        * rendering/RenderButton.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21209 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoLayoutTests:
darin [Tue, 1 May 2007 19:59:53 +0000 (19:59 +0000)]

        Reviewed by Hyatt.

        - added a test for <rdar://problem/5172596> HTML-parser-created tbody elements
          are being ignored when processing CSS child rules

        * fast/css/child-selector-implicit-tbody.html: Added.
        * fast/css/child-selector-implicit-tbody-expected.txt: Added.

        - updated test results for <rdar://problem/5172607> XML parser is creating
          implicit tbody elements; the only change is the tag name that's dumped

        * fast/table/037-expected.txt:
        * tables/mozilla/marvin/x_caption_align_bottom-expected.txt:
        * tables/mozilla/marvin/x_caption_align_top-expected.txt:
        * tables/mozilla/marvin/x_caption_class-expected.txt:
        * tables/mozilla/marvin/x_caption_id-expected.txt:
        * tables/mozilla/marvin/x_caption_style-expected.txt:
        * tables/mozilla/marvin/x_col_align_center-expected.txt:
        * tables/mozilla/marvin/x_col_align_char-expected.txt:
        * tables/mozilla/marvin/x_col_align_justify-expected.txt:
        * tables/mozilla/marvin/x_col_align_left-expected.txt:
        * tables/mozilla/marvin/x_col_align_right-expected.txt:
        * tables/mozilla/marvin/x_col_span-expected.txt:
        * tables/mozilla/marvin/x_col_valign_baseline-expected.txt:
        * tables/mozilla/marvin/x_col_valign_bottom-expected.txt:
        * tables/mozilla/marvin/x_col_valign_middle-expected.txt:
        * tables/mozilla/marvin/x_col_valign_top-expected.txt:
        * tables/mozilla/marvin/x_col_width_pct-expected.txt:
        * tables/mozilla/marvin/x_col_width_px-expected.txt:
        * tables/mozilla/marvin/x_col_width_rel-expected.txt:
        * tables/mozilla/marvin/x_colgroup_align_center-expected.txt:
        * tables/mozilla/marvin/x_colgroup_align_char-expected.txt:
        * tables/mozilla/marvin/x_colgroup_align_justify-expected.txt:
        * tables/mozilla/marvin/x_colgroup_align_left-expected.txt:
        * tables/mozilla/marvin/x_colgroup_align_right-expected.txt:
        * tables/mozilla/marvin/x_colgroup_span-expected.txt:
        * tables/mozilla/marvin/x_colgroup_valign_baseline-expected.txt:
        * tables/mozilla/marvin/x_colgroup_valign_bottom-expected.txt:
        * tables/mozilla/marvin/x_colgroup_valign_middle-expected.txt:
        * tables/mozilla/marvin/x_colgroup_valign_top-expected.txt:
        * tables/mozilla/marvin/x_colgroup_width_rel-expected.txt:
        * tables/mozilla/marvin/x_table_align_center-expected.txt:
        * tables/mozilla/marvin/x_table_align_left-expected.txt:
        * tables/mozilla/marvin/x_table_align_right-expected.txt:
        * tables/mozilla/marvin/x_table_bgcolor_name-expected.txt:
        * tables/mozilla/marvin/x_table_bgcolor_rgb-expected.txt:
        * tables/mozilla/marvin/x_table_cellpadding-expected.txt:
        * tables/mozilla/marvin/x_table_cellpadding_pct-expected.txt:
        * tables/mozilla/marvin/x_table_cellspacing-expected.txt:
        * tables/mozilla/marvin/x_table_cellspacing_pct-expected.txt:
        * tables/mozilla/marvin/x_table_class-expected.txt:
        * tables/mozilla/marvin/x_table_id-expected.txt:
        * tables/mozilla/marvin/x_table_style-expected.txt:
        * tables/mozilla/marvin/x_table_width_pct-expected.txt:
        * tables/mozilla/marvin/x_table_width_px-expected.txt:
        * tables/mozilla/marvin/x_td_align_center-expected.txt:
        * tables/mozilla/marvin/x_td_align_char-expected.txt:
        * tables/mozilla/marvin/x_td_align_justify-expected.txt:
        * tables/mozilla/marvin/x_td_align_left-expected.txt:
        * tables/mozilla/marvin/x_td_align_right-expected.txt:
        * tables/mozilla/marvin/x_td_bgcolor_name-expected.txt:
        * tables/mozilla/marvin/x_td_bgcolor_rgb-expected.txt:
        * tables/mozilla/marvin/x_td_class-expected.txt:
        * tables/mozilla/marvin/x_td_colspan-expected.txt:
        * tables/mozilla/marvin/x_td_height-expected.txt:
        * tables/mozilla/marvin/x_td_id-expected.txt:
        * tables/mozilla/marvin/x_td_nowrap-expected.txt:
        * tables/mozilla/marvin/x_td_rowspan-expected.txt:
        * tables/mozilla/marvin/x_td_style-expected.txt:
        * tables/mozilla/marvin/x_td_valign_baseline-expected.txt:
        * tables/mozilla/marvin/x_td_valign_bottom-expected.txt:
        * tables/mozilla/marvin/x_td_valign_middle-expected.txt:
        * tables/mozilla/marvin/x_td_valign_top-expected.txt:
        * tables/mozilla/marvin/x_td_width-expected.txt:
        * tables/mozilla/marvin/x_th_align_center-expected.txt:
        * tables/mozilla/marvin/x_th_align_char-expected.txt:
        * tables/mozilla/marvin/x_th_align_justify-expected.txt:
        * tables/mozilla/marvin/x_th_align_left-expected.txt:
        * tables/mozilla/marvin/x_th_align_right-expected.txt:
        * tables/mozilla/marvin/x_th_bgcolor_name-expected.txt:
        * tables/mozilla/marvin/x_th_bgcolor_rgb-expected.txt:
        * tables/mozilla/marvin/x_th_class-expected.txt:
        * tables/mozilla/marvin/x_th_colspan-expected.txt:
        * tables/mozilla/marvin/x_th_height-expected.txt:
        * tables/mozilla/marvin/x_th_id-expected.txt:
        * tables/mozilla/marvin/x_th_nowrap-expected.txt:
        * tables/mozilla/marvin/x_th_rowspan-expected.txt:
        * tables/mozilla/marvin/x_th_style-expected.txt:
        * tables/mozilla/marvin/x_th_valign_baseline-expected.txt:
        * tables/mozilla/marvin/x_th_valign_bottom-expected.txt:
        * tables/mozilla/marvin/x_th_valign_middle-expected.txt:
        * tables/mozilla/marvin/x_th_valign_top-expected.txt:
        * tables/mozilla/marvin/x_th_width-expected.txt:
        * tables/mozilla/marvin/x_tr_align_center-expected.txt:
        * tables/mozilla/marvin/x_tr_align_char-expected.txt:
        * tables/mozilla/marvin/x_tr_align_justify-expected.txt:
        * tables/mozilla/marvin/x_tr_align_left-expected.txt:
        * tables/mozilla/marvin/x_tr_align_right-expected.txt:
        * tables/mozilla/marvin/x_tr_bgcolor_name-expected.txt:
        * tables/mozilla/marvin/x_tr_bgcolor_rgb-expected.txt:
        * tables/mozilla/marvin/x_tr_class-expected.txt:
        * tables/mozilla/marvin/x_tr_id-expected.txt:
        * tables/mozilla/marvin/x_tr_style-expected.txt:
        * tables/mozilla/marvin/x_tr_valign_baseline-expected.txt:
        * tables/mozilla/marvin/x_tr_valign_bottom-expected.txt:
        * tables/mozilla/marvin/x_tr_valign_middle-expected.txt:
        * tables/mozilla/marvin/x_tr_valign_top-expected.txt:
        * tables/mozilla_expected_failures/bugs/bug178855-expected.txt:
        * tables/mozilla_expected_failures/marvin/x_caption_align_left-expected.txt:
        * tables/mozilla_expected_failures/marvin/x_caption_align_right-expected.txt:
        * tables/mozilla_expected_failures/marvin/x_colgroup_width_pct-expected.txt:
        * tables/mozilla_expected_failures/marvin/x_colgroup_width_px-expected.txt:


        Reviewed by Hyatt.

        - removed the m_implicit bit from Node, freeing up a bit

        - fixed <rdar://problem/5172607> XML parser is creating implicit tbody elements
        - fixed <rdar://problem/5172596> HTML-parser-created tbody elements are being
          ignored when processing CSS child rules

        Test: fast/css/child-selector-implicit-tbody.html

        * css/cssstyleselector.cpp: (WebCore::CSSStyleSelector::checkSelector):
        Removed code that used to look at implicitNode(), which no longer exists.

        * css/html4.css: Added a style rule to handle cases where we have a tr
        inside a table with no intervening table section.

        * dom/Node.h: Removed the bit.
        * dom/Node.cpp:
        (WebCore::Node::Node): Don't initialize the bit.
        (WebCore::Node::dump): Don't dump the bit.

        * dom/XMLTokenizer.cpp:
        (WebCore::XMLTokenizer::startElementNs): Remove the code that made a
        table body element. This is handled in the render tree, the DOM tree
        should not be mangled.
        (WebCore::XMLTokenizer::endElementNs): Removed code that used to look at
        implicitNode(), which no longer exists.

        * html/HTMLElementFactory.cpp: (WebCore::tableSectionConstructor):
        Removed boolean parameter that was passed to the HTMLTableSectionElement
        * html/HTMLParser.cpp: (WebCore::HTMLParser::handleError): Ditto.
        * html/HTMLTableElement.cpp:
        (WebCore::HTMLTableElement::createTHead): Ditto.
        (WebCore::HTMLTableElement::createTFoot): Ditto.
        (WebCore::HTMLTableElement::insertRow): Changed to return a PassRefPtr,
        because it's possible the row could be removed from the table by JavaScript
        code responding to DOM mutation events before the function returns. Just
        something I noticed by code inspection.

        * html/HTMLTableSectionElement.cpp:
        (WebCore::HTMLTableSectionElement::HTMLTableSectionElement): Removed
        boolean "implicit" parameter to the constructor.
        (WebCore::HTMLTableSectionElement::insertRow): Changed to return a
        PassRefPtr for the same reason cited above.

        * html/HTMLTableElement.h: Changed the return value of insertRow to be
        a PassRefPtr.
        * html/HTMLTableSectionElement.h: Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21208 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago2007-05-01 Jungshik Shin <jungshik.shin@gmail.com>
ap [Tue, 1 May 2007 09:03:04 +0000 (09:03 +0000)]
2007-05-01  Jungshik Shin  <jungshik.shin@gmail.com>

        Reviewed by Alexey.

        - get rid of non-ASCII lteral characters : suppress compiler warnings

        * kjs/testkjs.cpp:
        * pcre/pcre_compile.c:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21200 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoLayoutTests:
hyatt [Tue, 1 May 2007 04:57:01 +0000 (04:57 +0000)]

        Fix for bug 12691, generated content and display: table-cell duplicates
        content on a size change.  Make sure that the code to update before/after
        content is smart enough to drill into anonymous containers to find the
        old child.

        Reviewed by olliej

        * fast/css-generated-content/table-cell-before-content-expected.checksum: Added.
        * fast/css-generated-content/table-cell-before-content-expected.png: Added.
        * fast/css-generated-content/table-cell-before-content-expected.txt: Added.
        * fast/css-generated-content/table-cell-before-content.html: Added.


        Fix for bug 12691, generated content and display: table-cell duplicates
        content on a size change.  Make sure that the code to update before/after
        content is smart enough to drill into anonymous containers to find the
        old child.

        Reviewed by olliej

        Added fast/css-generated-content/table-cell-before-content.html

        * rendering/RenderBlock.cpp:
        * rendering/RenderButton.cpp:
        * rendering/RenderButton.h:
        * rendering/RenderContainer.cpp:
        * rendering/RenderContainer.h:
        * rendering/RenderInline.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21199 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by justin
justing [Tue, 1 May 2007 04:32:03 +0000 (04:32 +0000)]
    Reviewed by justin

        comment to Range::comparePoint in dom/Range.cpp has
        '0xD0' in place of '-': leading to a compiler warning

        * dom/Range.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21198 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoLayoutTests:
justing [Mon, 30 Apr 2007 23:45:40 +0000 (23:45 +0000)]

        Reviewed by darin

        <rdar://problem/4875189> List gets deleted when replacing contents

        Replacing the contents of a list:
        * editing/inserting/4875189-1-expected.checksum: Added.
        * editing/inserting/4875189-1-expected.png: Added.
        * editing/inserting/4875189-1-expected.txt: Added.
        * editing/inserting/4875189-1.html: Added.
        Of a floating div:
        * editing/inserting/4875189-2-expected.checksum: Added.
        * editing/inserting/4875189-2-expected.png: Added.
        * editing/inserting/4875189-2-expected.txt: Added.
        * editing/inserting/4875189-2.html: Added.
        Pasting a list on top of itself now creates a nested list:
        * editing/pasteboard/paste-4039777-fix-expected.checksum:
        * editing/pasteboard/paste-4039777-fix-expected.png:
        * editing/pasteboard/paste-4039777-fix-expected.txt:


        Reviewed by darin

        <rdar://problem/4875189> List gets deleted when replacing contents

        * editing/ReplaceSelectionCommand.cpp:
        (WebCore::ReplaceSelectionCommand::doApply): Don't expand
        to include elements, like lists, when deleting a selection
        to replace it.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21197 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by ggaren
justing [Mon, 30 Apr 2007 22:06:27 +0000 (22:06 +0000)]
    Reviewed by ggaren

        Buildfix.  Leave in grammar checking machinery to
        keep WebKit building on other platforms but don't
        operate on sentence sized chunks so that we don't
        re-introduce 5157329.

        * editing/Editor.h:
        * editing/mac/EditorMac.mm:
        * page/Frame.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21196 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoLayoutTests:
justing [Mon, 30 Apr 2007 21:14:09 +0000 (21:14 +0000)]

        Reviewed by darin

        <rdar://problem/5168598> Crash when deleting a link inside an input field

        * editing/deleting/5168598-expected.checksum: Added.
        * editing/deleting/5168598-expected.png: Added.
        * editing/deleting/5168598-expected.txt: Added.
        * editing/deleting/5168598.html: Added.


        Reviewed by darin

        Crash when deleting a link inside an input field

        Deletion code tries to update one of its positions to keep
        it valid and nullifies it.  Later in mergeParagraphs we try
        to find out if it's valid by checking to see if it's still in
        the document and crash.

        The fix is just to not bother keeping the position,
        m_upstreamStart, valid, since after the point in the code
        under examination, it doesn't need to be valid anymore.
        Besides, code to update DeleteSelectionCommand's positions
        should go in the functions that handle the content removal
        that might invalid them (there are already FIXMEs about this).

        Also, users shouldn't be able to insert links (or any rich content)
        into plaintext-only regions like text fields, even if those
        text fields are inside richly editable regions.  This is a
        separate bug and is filed as 5171552.

        * editing/DeleteSelectionCommand.cpp:
        (WebCore::DeleteSelectionCommand::handleGeneralDelete): Don't
        bother updating m_upstreamStart.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21195 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago<rdar://problem/5108546> REGRESSION: Zoom follows keyboard focus does not work
pyeh [Mon, 30 Apr 2007 21:03:01 +0000 (21:03 +0000)]
<rdar://problem/5108546> REGRESSION: Zoom follows keyboard focus does not work

        * WebCore.xcodeproj/project.pbxproj:
        Add new file SelectionControllerMac.mm
        * dom/Document.cpp:
        Move the code to fire off AXSelectedTextChanged notification to a new method notifyAccessibilityForSelectionChange in SelectionControllerMac.mm
        * editing/SelectionController.cpp:
        call notifyAccessibilityForSelectionChange to handle firing of accessibility notification and zoomFocusChange
        * editing/SelectionController.h:
        Add notifyAccessibilityForSelectionChange method for Mac platform
        * platform/Widget.h:
        * platform/mac/WidgetMac.mm:
        Add a helper routine that takes IntPoint in local coordinate and convert that to screen coordinate

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21194 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by harrison
justing [Mon, 30 Apr 2007 19:43:27 +0000 (19:43 +0000)]
    Reviewed by harrison

        REGRESSION: Crash at WTF::fastFree() when arrowing or deleting backward through editable text (13493)

        Backing out the changes for 4859132.

        * editing/Editor.h:
        * editing/mac/EditorMac.mm:
        * page/Frame.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21193 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago - Removed debugging statements and took out the removal of the deletion
kmccullo [Mon, 30 Apr 2007 17:43:13 +0000 (17:43 +0000)]
    - Removed debugging statements and took out the removal of the deletion
          of the symlink.  This is because multiple instances of run-webkit-tests
          can be running at the same time.
        * Scripts/run-webkit-tests:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21192 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by Oliver.
antti [Mon, 30 Apr 2007 10:44:25 +0000 (10:44 +0000)]
    Reviewed by Oliver.

        Fix <rdar://problem/5169261>
        REGRESSION: Google Maps zooming too sensitive when using two fingers on trackpad

        Use line based delta values when generating DOM wheel events, use pixel deltas
        for scrolling only.

        No test case, requires user interaction and specific hardware.

        * page/EventHandler.cpp:
        * platform/PlatformWheelEvent.h:
        * platform/mac/WheelEventMac.mm:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21188 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by Brady.
oliver [Mon, 30 Apr 2007 06:27:02 +0000 (06:27 +0000)]
    Reviewed by Brady.

        Build fix for experimental features build

        * rendering/RenderForeignObject.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21187 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago2007-04-29 Mitz Pettel <mitz@webkit.org>
bdash [Mon, 30 Apr 2007 03:29:00 +0000 (03:29 +0000)]
2007-04-29  Mitz Pettel  <mitz@webkit.org>

        Reviewed by Dave Hyatt.

        - fix http://bugs.webkit.org/show_bug.cgi?id=13544
          REGRESSION (r21120): Member profile frame not erased at Yahoo! Answers

        Test: fast/repaint/layer-hide-when-needs-layout.html

        * rendering/RenderObject.cpp:
        (WebCore::RenderObject::setStyle): If we get a layout hint, we count on
        layout() or updateLayerPositions() to repaint. In the case of a layer
        going invisible, however, we need to repaint here since neither of the
        above will do it.

2007-04-29  Mitz Pettel  <mitz@webkit.org>

        Reviewed by Dave Hyatt.

        - repaint test for http://bugs.webkit.org/show_bug.cgi?id=13544
          REGRESSION (r21120): Member profile frame not erased at Yahoo! Answers

        * fast/repaint/layer-hide-when-needs-layout-expected.checksum: Added.
        * fast/repaint/layer-hide-when-needs-layout-expected.png: Added.
        * fast/repaint/layer-hide-when-needs-layout-expected.txt: Added.
        * fast/repaint/layer-hide-when-needs-layout.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21186 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by Steve.
oliver [Sun, 29 Apr 2007 22:31:40 +0000 (22:31 +0000)]
    Reviewed by Steve.

        Fix for <rdar://problem/5126393> REGRESSION (r19672-19697): Safari drag image drags at original size (13261)

        Added logic to ensure that when we create the drag image
        we're looking at the rendered image size.

        * page/DragController.cpp:
          Pass the rendered image size to fitDragImageToMaxSize

        * platform/DragImage.cpp:
           Need to do scaling relative to rendered image size

        * platform/DragImage.h:
           scaleDragImage needs x and y scales to allow for them to be
           scaled seperately on a page

        * platform/gdk/DragImageGdk.cpp:
           Prevent build bustage

        * platform/mac/DragImageMac.mm:
           scale width and height independently

        * platform/qt/DragImageQt.cpp:
           Prevent build bustage

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21185 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by Adele.
darin [Sun, 29 Apr 2007 20:32:51 +0000 (20:32 +0000)]
    Reviewed by Adele.

        - fix the Gmail part of http://bugs.webkit.org/show_bug.cgi?id=12798
          and <rdar://problem/4118559>

        This patch changes things so that resizing can make things larger than their
        initial size, but not smaller. I believe this resolves the problem for Gmail,
        but you can still get strange behavior if you have a small right-aligned <textarea>.

        * rendering/RenderLayer.cpp:
        (WebCore::RenderLayer::resize): Rewrote to use more IntPoint and IntSize and add
        in a new minimum-size algorithm. This function now has the side effect of storing
        a new minimumSizeForResizing in the element being resized, and respects any existing
        minimum as well.
        (WebCore::RenderLayer::offsetFromResizeCorner): Reversed this so it converts the
        point into coordinates relative to the resize corner, hence negative numbers mean
        the point is above and to the left. This is used only in the algorithm above.

        * dom/Element.h: Removed unneeded include of Attr.h, and added forward declarations
        as needed. Removed all friend declarations; there was no need for any of them.
        Tweaked formatting of various declarations and removed unneeded parameter names.
        Made as much as possible private, and a couple things public to avoid the need for
        friend classes. Added minimumSizeForResizing, setMinimumSizeForResizing, rareData,
        and createRareData functions.

        * dom/Element.cpp:
        (WebCore::rareDataMap): Added. Function to return the single global map.
        (WebCore::rareDataFromMap): Added. For use by inlined functions to get the
        data from the map in the case where it's present.
        (WebCore::ElementRareData::ElementRareData): Added.
        (WebCore::Element::~Element): Added code to delete the element from the rare
        data map when appropriate.
        (WebCore::Element::rareData): Added.
        (WebCore::Element::createRareData): Added.
        (WebCore::Element::setAttributeNodeNS): Got rid of the old inline version of this,
        since it was the sole reason we had to include Attr.h in Element.h.
        (WebCore::Element::minimumSizeForResizing): Added. Accesses rare data.
        (WebCore::Element::setMinimumSizeForResizing): Ditto.

        * dom/Node.h: Renamed the m_specified bit since it's now used for two purposes.
        Its new long name is m_attrWasSpecifiedOrElementHasRareData.

        * dom/Attr.cpp: (WebCore::Attr::Attr): Updated for bit name change.
        * dom/Attr.h:
        (WebCore::Attr::specified): Ditto.
        (WebCore::Attr::setSpecified): Added, since setting the bit directly is pretty ugly now.
        * dom/Document.cpp: (WebCore::Document::adoptNode): Updated for bit name change.
        * dom/Node.cpp:
        (WebCore::Node::Node): Ditto.
        (WebCore::Node::dump): Ditto.

        * dom/NamedAttrMap.cpp: (WebCore::NamedAttrMap::isReadOnlyNode): Moved out of line,
        since this was the sole reason we had to include Element.h in NamedAttrMap.h.

        * html/HTMLElement.cpp:
        (WebCore::HTMLElement::nodeName): Changed to use tagQName() since that's a public
        way to do something that's now private.
        (WebCore::HTMLElement::cloneNode): Ditto.

        * dom/Attribute.cpp:
        * dom/NamedAttrMap.h:
        * dom/NamedMappedAttrMap.cpp:
        * dom/StyledElement.h:
        * html/HTMLTokenizer.h:
        * xml/XPathFunctions.cpp:
        * xml/XPathStep.cpp:
        Updated includes, and other administrivia.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21184 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoLayoutTests:
weinig [Sun, 29 Apr 2007 20:09:08 +0000 (20:09 +0000)]

        Reviewed by Darin and Hyatt.

        - test and updated results for http://bugs.webkit.org/show_bug.cgi?id=13487
          Implement O(1) absoluteClippedOverflowRect and absoluteOutlineBox during layout for a possible speed gain

        * fast/repaint/bugzilla-3509-expected.checksum:
        * fast/repaint/bugzilla-3509-expected.png:
        * fast/repaint/intermediate-layout-position-clip-expected.checksum: Added.
        * fast/repaint/intermediate-layout-position-clip-expected.png: Added.
        * fast/repaint/intermediate-layout-position-clip-expected.txt: Added.
        * fast/repaint/intermediate-layout-position-clip.html: Added.
        * fast/repaint/intermediate-layout-position-expected.checksum:
        * fast/repaint/intermediate-layout-position-expected.png:


        Reviewed by Darin and Hyatt.

        - http://bugs.webkit.org/show_bug.cgi?id=13487
          Implement O(1) absoluteClippedOverflowRect and absoluteOutlineBox during layout for a possible speed gain

        Test for a bug fixed by this patch: fast/repaint/intermediate-layout-position-clip.html

        By keeping track of the total translation and clip during layout, absolutePosition and
        computeAbsoluteRepaintRect become O(1).

        * WebCore.pro: Added LayoutState.cpp
        * WebCore.vcproj/WebCore/WebCore.vcproj: Added LayoutState.{cpp,h}
        * WebCore.xcodeproj/project.pbxproj: Added LayoutState.{cpp,h}
        * WebCoreSources.bkl: Added LayoutState.cpp
        * page/FrameView.cpp:
        (WebCore::FrameView::layout): When doing subtree layout, push an initial layout state for the
        layout root.
        * rendering/LayoutState.cpp: Added. A LayoutState corresponding to a box consists of the origin of
        its coordinate system in view coordinates and possibly the clip rect, in view coordinates, which
        applies to its children.
        (WebCore::LayoutState::operator delete):
        * rendering/LayoutState.h: Added.
        * rendering/RenderBlock.cpp:
        (WebCore::RenderBlock::layoutBlock): Factored out the "only positioned objects" case into
        layoutOnlyPositionedObjects(). Added code to push/pop LayoutState, unless we have columns, in
        which case we disable/enable LayoutState. Removed the translation by the layout delta
        of the old absolute rects, since that is now integral to absoluteClippedOverflowRect() and
        (WebCore::RenderBlock::layoutOnlyPositionedObjects): Added this helper function that checks for
        and handles the case where only positioned objects need layout. Returns true in that case.
        (WebCore::RenderBlock::repaintOverhangingFloats): Disabled LayoutState when repainting floats, since
        they may belong to other containers.
        (WebCore::RenderBlock::updateFirstLetter): Disabled LayoutState when adding and removing from the
        first letter container, since it may be different than ourselves.
        * rendering/RenderBlock.h:
        * rendering/RenderBox.cpp:
        (WebCore::RenderBox::setStyle): Minor cleanup: changed element()->document() to the equivalent document().
        (WebCore::RenderBox::offsetForPositionedInContainer): Added. Factored out common code from
        absolutePosition, computeAbsoluteRepaintRect and RenderLayer::updateLayerPosition into this function.
        (WebCore::RenderBox::absolutePosition): Added an O(1) code path when this function is called during
        layout (and LayoutState has not been disabled). Factored out some code into
        (WebCore::RenderBox::absoluteClippedOverflowRect): Added the layout delta here, so callers don't need
        (WebCore::RenderBox::computeAbsoluteRepaintRect): Added an O(1) code path when this function is called
        during layout (and LayoutState has not been disabled). Factored out some code into
        * rendering/RenderContainer.cpp:
        (WebCore::RenderContainer::layout): Added LayoutState push/pop.
        * rendering/RenderFlexibleBox.cpp:
        (WebCore::RenderFlexibleBox::layoutBlock): Factored out the "only positioned objects" case into
        layoutOnlyPositionedObjects(). Added LayoutState push/pop. Removed the translation by the layout delta
        of the old absolute rects, since that is now integral to absoluteClippedOverflowRect() and
        * rendering/RenderFlow.cpp:
        (WebCore::RenderFlow::absoluteClippedOverflowRect): Added an ASSERT.
        * rendering/RenderForeignObject.cpp:
        (WebCore::RenderForeignObject::layout): Disabled LayoutState because it is incompatible with arbitrary
        affine transforms.
        * rendering/RenderHTMLCanvas.cpp:
        (WebCore::RenderHTMLCanvas::layout): Removed the translation by the layout delta of the old absolute
        rects, since that is now integral to absoluteClippedOverflowRect() and absoluteOutlineBox().
        * rendering/RenderImage.cpp:
        (WebCore::RenderImage::layout): Ditto.
        * rendering/RenderLayer.cpp:
        (WebCore::RenderLayer::updateLayerPositions): Added ASSERTs. Factored out some code into
        (WebCore::RenderLayer::updateScrollInfoAfterLayout): Disabled LayoutState around the call to
        * rendering/RenderListItem.cpp:
        (WebCore::RenderListItem::updateMarkerLocation): Disabled LayoutState when moving the list marker, since
        it can trigger repainting in other containers.
        * rendering/RenderObject.cpp:
        (WebCore::RenderObject::isRoot): Minor cleanup: changed element()->document() to the equivalent
        (WebCore::RenderObject::computeAbsoluteRepaintRect): Removed 'return' at the end of the function.
        (WebCore::RenderObject::container): Reordered to avoid calling isText() twice.
        (WebCore::RenderObject::absoluteOutlineBox): Added the layout delta here, so callers don't need to.
        * rendering/RenderSVGContainer.cpp:
        (WebCore::RenderSVGContainer::layout): Disabled LayoutState because it is incompatible with arbitrary
        affine transforms.
        * rendering/RenderTable.cpp:
        (WebCore::RenderTable::layout):  Factored out the "only positioned objects" case into
        layoutOnlyPositionedObjects(). Added LayoutState push/pop. Removed the translation by the layout delta
        of the old absolute rects, since that is now integral to absoluteClippedOverflowRect() and
        * rendering/RenderTableCell.cpp:
        (WebCore::RenderTableCell::absoluteClippedOverflowRect): Added the layout delta here, so callers
        don't need to.
        (WebCore::RenderTableCell::computeAbsoluteRepaintRect): For the O(1) code path, skipped the correction
        for the parent row, because RenderTableRow doesn't push a translation into LayoutState.
        (WebCore::RenderTableCell::absolutePosition): Ditto.
        * rendering/RenderTableRow.cpp:
        (WebCore::RenderTableRow::layout): Added LayoutState push/pop.
        * rendering/RenderTableSection.cpp:
        (WebCore::RenderTableSection::setCellWidths): Added LayoutState push/pop if cells are repainted or
        receive layout.
        (WebCore::RenderTableSection::calcRowHeight): Ditto.
        (WebCore::RenderTableSection::layoutRows): Added LayoutState push/pop.
        * rendering/RenderView.cpp:
        (WebCore::RenderView::layout): Added initial LayoutState setup for the layout.
        (WebCore::RenderView::paintBoxDecorations): Minor cleanup: changed element()->document() to the
        equivalent document().
        (WebCore::RenderView::repaintViewRectangle): Ditto.
        (WebCore::RenderView::pushLayoutState): Added. Pushes initial layout state for subtree layout.
        * rendering/RenderView.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21183 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoAdd layout test for previous checkin.
hyatt [Sun, 29 Apr 2007 10:33:17 +0000 (10:33 +0000)]
Add layout test for previous checkin.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21182 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Fix for bugzilla 13535, regression from positioning cleanup a while
hyatt [Sun, 29 Apr 2007 10:30:48 +0000 (10:30 +0000)]
    Fix for bugzilla 13535, regression from positioning cleanup a while
        back.  Some code got added to compute available width for a relpositioned
        inline, only the code meant to be used for inline flows and not inline blocks.

        Reviewed by mjs


        * rendering/RenderBox.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21181 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago2007-04-29 Oliver Hunt <oliver@apple.com>
oliver [Sun, 29 Apr 2007 10:17:19 +0000 (10:17 +0000)]
2007-04-29  Oliver Hunt  <oliver@apple.com>
        Drag and drop support for WebKit/Qt


        Reviewed by Zack.

        Implement basic drag and drop support for qt
        this allows content and link dragging but does not
        provide the DOM data access methods.

        * platform/DragData.h:
        * platform/qt/ClipboardQt.cpp:
        * platform/qt/ClipboardQt.h:
        * platform/qt/DragDataQt.cpp:


        Reviewed by Zack.

        Tie QT drag events to the DragController logic
        to allow drag and drop events to be handled by

        * Api/qwebframe.cpp:
        * Api/qwebframe.h:
        * Api/qwebpage.cpp:
        * Api/qwebpage.h:
        * WebCoreSupport/DragClientQt.cpp:
        * WebCoreSupport/DragClientQt.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21180 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by Darin and Oliver.
mjs [Sun, 29 Apr 2007 07:33:46 +0000 (07:33 +0000)]
    Reviewed by Darin and Oliver.

        - change Document to hold onto Frame* instead of FrameView*

        This is an architectural cleanup and a prerequisite to fixing citibank login
        in an architecturally sound way.

        Some fixups for the page cache:

        * history/CachedPage.cpp:
        (WebCore::CachedPage::restore): Add an assert that the document's view
        matches the cached page's view - this needs to be true by the time this
        function is called.
        (WebCore::CachedPage::clear): Change assert to require frame matching, not
        view matching.
        * history/CachedPage.h:
        (WebCore::CachedPage::view): Add a way to get the view out.
        * loader/FrameLoader.cpp:
        (WebCore::FrameLoader::begin): pass Frame, not FrameView, when creating document
        (WebCore::FrameLoader::open): restore view based on CachedPage, not Document.

        Remaining changes are just straightforward updates based on which
        pointer is held or passed:

        * dom/DOMImplementation.cpp:
        * dom/DOMImplementation.h:
        * dom/Document.cpp:
        * dom/Document.h:
        * html/HTMLDocument.cpp:
        * html/HTMLDocument.h:
        * html/HTMLViewSourceDocument.cpp:
        * html/HTMLViewSourceDocument.h:
        * ksvg2/svg/SVGDocument.cpp:
        * ksvg2/svg/SVGDocument.h:
        * loader/ImageDocument.cpp:
        * loader/ImageDocument.h:
        * loader/PluginDocument.cpp:
        * loader/PluginDocument.h:
        * loader/TextDocument.cpp:
        * loader/TextDocument.h:
        * xml/XSLTProcessor.cpp:
        * xml/XSLTProcessor.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21179 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago2007-04-28 Jungshik Shin <jungshik.shin@gmail.com>
bdash [Sun, 29 Apr 2007 06:42:32 +0000 (06:42 +0000)]
2007-04-28  Jungshik Shin  <jungshik.shin@gmail.com>

        Reviewed by Sam Weinig.

        - Replace copyright sign in Latin-1 (0xA9) with '(C)'

        * bindings/npruntime.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21178 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoLayoutTests:
adele [Sun, 29 Apr 2007 05:32:01 +0000 (05:32 +0000)]

        Added test to cover updated fix for <rdar://problem/5152299>

        * fast/events/no-blur-on-enter-button-expected.txt: Added.
        * fast/events/no-blur-on-enter-button.html: Added.


        RS by Darin.

        Update to fix for <rdar://problem/5152299>

        Test: fast/events/no-blur-on-enter-button.html

        * html/HTMLButtonElement.cpp: (WebCore::HTMLButtonElement::defaultEventHandler):
        Don't blur the button when handling the enter key.  This matches our new behavior for input elements.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21177 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoImproved the ChangeLog comment. Maybe the code needs a comment like this too.
darin [Sun, 29 Apr 2007 04:35:06 +0000 (04:35 +0000)]
Improved the ChangeLog comment. Maybe the code needs a comment like this too.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21176 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoJavaScriptCore:
darin [Sun, 29 Apr 2007 04:25:33 +0000 (04:25 +0000)]

        Reviewed by Maciej.

        - fix <rdar://problem/5154144> Hamachi test fails: assertion failure in ListHashSet

        Test: fast/forms/add-remove-form-elements-stress-test.html

        * wtf/ListHashSet.h:
        (WTF::ListHashSetNodeAllocator::ListHashSetNodeAllocator): Initialize
        m_isDoneWithInitialFreeList to false.
        (WTF::ListHashSetNodeAllocator::allocate): Added assertions based on a debug-only
        m_isAllocated flag that make sure we don't allocate a block that's already allocated.
        These assertions helped pinpoint the bug. Set m_isDoneWithInitialFreeList when we
        allocate the last block of the initial free list. Once we're done with the initial
        free list, turn off the rule that says that the next node in the pool after the last
        node in the free list is also free. This rule works because any free nodes are added
        to the head of the free list, so a node that hasn't been allocated even once is always
        at the tail of the free list. But it doesn't work any longer once the entire pool
        has been used at least once.
        (WTF::ListHashSetNodeAllocator::deallocate): Set the node's m_isAllocated to false.
        (WTF::ListHashSetNodeAllocator::pastPool): Added. Used above.
        (WTF::ListHashSetNodeAllocator::inPool): Changed to use the pastPool function.
        (WTF::ListHashSetNode::ListHashSetNode): Initialize m_isAllocated to true.
        (WTF::ListHashSetNode::operator new): Removed variable name for unused size
        (WTF::ListHashSetNode::destroy): Changed to call the destructor rather than
        delete -- this gets rid of the need to define an operator delete.


        Reviewed by Maciej.

        - test for <rdar://problem/5154144> Hamachi test fails: assertion failure in ListHashSet

        * fast/forms/add-remove-form-elements-stress-test-expected.txt: Added.
        * fast/forms/add-remove-form-elements-stress-test.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21175 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoLayoutTests:
adele [Sun, 29 Apr 2007 04:09:26 +0000 (04:09 +0000)]

        Tests for <rdar://problem/5152299> REGRESSION: <button> elements do not respond to space and return key as similar <input> elements do

        * fast/forms/button-enter-click-expected.txt: Added.
        * fast/forms/button-enter-click.html: Added.
        * fast/forms/button-spacebar-click-expected.txt: Added.
        * fast/forms/button-spacebar-click.html: Added.


        Fix by Darin, reviewed by me.

        Fix for <rdar://problem/5152299> REGRESSION: <button> elements do not respond to space and return key as similar <input> elements do


        * html/HTMLButtonElement.h: (WebCore::HTMLButtonElement::): Made type enum private.
        Removed unnecessary m_value and m_currValue, and unused m_dirty.
        * html/HTMLButtonElement.cpp:
        (WebCore::HTMLButtonElement::HTMLButtonElement): Removed initialization of m_dirty.
        (WebCore::HTMLButtonElement::parseMappedAttribute): Removed case to parse value and accesskey attributes, which don't do anything.
        (WebCore::HTMLButtonElement::defaultEventHandler): Removed unnecessary check for button type.
         Added cases to handle spacebar and enter keys in the same way that they're handled for <input type=button/reset/submit> buttons.
        (WebCore::HTMLButtonElement::isSuccessfulSubmitButton): Shortened comment.
        (WebCore::HTMLButtonElement::appendFormData): Use value() instead of m_currValue.  Renamed encoding argument to formData.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21174 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoLayoutTests:
darin [Sun, 29 Apr 2007 01:36:11 +0000 (01:36 +0000)]

        Reviewed by Darin.

        - test for http://bugs.webkit.org/show_bug.cgi?id=13501
          REGRESSION (r21093): Pasting into text field fails and makes the caret disappear

        * fast/dynamic/anonymous-block-layer-lost-expected.checksum: Added.
        * fast/dynamic/anonymous-block-layer-lost-expected.png: Added.
        * fast/dynamic/anonymous-block-layer-lost-expected.txt: Added.
        * fast/dynamic/anonymous-block-layer-lost.html: Added.


        Reviewed by Darin.

        - fix http://bugs.webkit.org/show_bug.cgi?id=13501
          REGRESSION (r21093): Pasting into text field fails and makes the caret disappear

        Test: fast/dynamic/anonymous-block-layer-lost.html

        * rendering/RenderBlock.cpp:
        (WebCore::RenderBlock::removeChild): Avoid cutting off layers when removing
        the anonymous block.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21173 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by Darin.
darin [Sun, 29 Apr 2007 01:34:03 +0000 (01:34 +0000)]
    Reviewed by Darin.

        - fix http://bugs.webkit.org/show_bug.cgi?id=13471
          REGRESSION (r21045): Secure keyboard entry mode remains in effect after leaving a password field by submitting

        * loader/FrameLoader.cpp:
        (WebCore::FrameLoader::clear): Replaced direct access to Frame's d->m_doc
        with a call to Frame::setDocument.
        (WebCore::FrameLoader::begin): Ditto.
        (WebCore::FrameLoader::open): Ditto.
        (WebCore::FrameLoader::switchOutLowBandwidthDisplayIfReady): Ditto.
        (WebCore::FrameLoader::canCachePage): Added a comment.
        * manual-tests/secure-keyboard-enabled-after-submit.html: Added.
        * page/Frame.cpp:
        (WebCore::Frame::setDocument): Reset secure keyboard entry, in case the
        old document's focused node has set it.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21172 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by Adele.
darin [Sun, 29 Apr 2007 01:30:19 +0000 (01:30 +0000)]
    Reviewed by Adele.

        - fix http://bugs.webkit.org/show_bug.cgi?id=13379
          REGRESSION: Cannot add an option to a selected popup

        * manual-tests/update-empty-popup.html: Added.
        * platform/mac/PopupMenuMac.mm:
        (WebCore::PopupMenu::show): Added a call to the client's hidePopup()
        callback before returning in the case that the menu is empty. Otherwise
        the client thinks that the menu stayed open.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21171 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoLayoutTests:
adele [Sat, 28 Apr 2007 19:00:38 +0000 (19:00 +0000)]

        Reviewed by Darin.

        Test for <rdar://problem/4806988> REGRESSION: history.go(0) doesn't reload page (affects buttons at facebook.com)

        * fast/history/history_reload-expected.txt: Added.
        * fast/history/history_reload.html: Added.
        * fast/history/resources/history_reload_window.html: Added.


        Reviewed by Darin.

        Fix for <rdar://problem/4806988> REGRESSION: history.go(0) doesn't reload page (affects buttons at facebook.com)

        Test: fast/history/history_reload.html

        * loader/FrameLoader.cpp: (WebCore::FrameLoader::scheduleHistoryNavigation):
        We avoid doing a redirection when going to the same url in a history navigation, but we have to still
        schedule a redirection when the number of steps is zero so we can force a reload.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21170 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by Darin.
ap [Sat, 28 Apr 2007 18:29:19 +0000 (18:29 +0000)]
    Reviewed by Darin.

        REGRESSION (r20997-21003): Crash in WebCore::Node::document

        Test: fast/dom/NodeList/childNodes-reset-cache.html

        * dom/Node.cpp:
        (WebCore::Node::notifyLocalNodeListsChildrenChanged): Reset the common caches for child node
        lists. We cannot rely on notifications, since there may be no list objects left around.

        * dom/ChildNodeList.cpp:
        * dom/ChildNodeList.h:
        Override rootNodeChildrenChanged() with an empty implementation, now that caches are reset
        in Node::notifyLocalNodeListsChildrenChanged().

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21169 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago2007-04-28 Mark Rowe <mrowe@apple.com>
bdash [Sat, 28 Apr 2007 11:36:32 +0000 (11:36 +0000)]
2007-04-28  Mark Rowe  <mrowe@apple.com>

        Build fix.  Change use of nil to 0 in C++ source file.

        * editing/Editor.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21168 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoLayoutTests:
darin [Sat, 28 Apr 2007 08:11:43 +0000 (08:11 +0000)]

        Reviewed by Adele.

        - test for http://bugs.webkit.org/show_bug.cgi?id=6216
          REGRESSION: getElementsByTagNameNS with empty namespace doesn't work

        * fast/dom/namespaces-1-expected.txt: Added.
        * fast/dom/namespaces-1.html: Added.


        Reviewed by Adele.

        - fix http://bugs.webkit.org/show_bug.cgi?id=6216
          REGRESSION: getElementsByTagNameNS with empty namespace doesn't work

        * dom/Node.cpp:
        (WebCore::TagNodeList::TagNodeList): Changed parameter to PassRefPtr, tweaked formatting.
        (WebCore::TagNodeList::item): Tweaked formatting.
        (WebCore::TagNodeList::nodeMatches): Tweaked formatting.
        (WebCore::Node::setPrefix): Tweaked formatting.
        (WebCore::Node::getElementsByTagNameNS): Added special case to turn empty namespaces into
        null namespaces.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21167 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoLayoutTests:
darin [Sat, 28 Apr 2007 08:07:08 +0000 (08:07 +0000)]

        Reviewed by Kevin Decker.

        - update test result for fix for <rdar://problem/5148271> Chinese numbers
          are incorrect in numbered lists (including pasting from TextEdit to Mail)

        * fast/lists/w3-list-styles-expected.txt: Updated. The new test results match
        the behavior of Firefox.


        Reviewed by Kevin Decker.

        - fix <rdar://problem/5148271> Chinese numbers are incorrect in numbered lists
          (including pasting from TextEdit to Mail)

        Covered by an existing regression test, which had wrong results checked in.

        Hyatt told me not to use the algorithm in the CSS 3 specification, but I didn't listen!
        This corrects mistakes there and makes our results match current Firefox.

        * rendering/RenderListMarker.cpp:
        (WebCore::toCJKIdeographic): Added code to strip off trailing zero digits.
        Added special case for zero, since that's the one case where we don't want
        to strip off trailing zero digits. Fixed mixed-up logic that would put the
        digits in each group backwards.
        (WebCore::listMarkerText): Corrected character values for the CJK_IDEOGRAPHIC case.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21166 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by Oliver.
darin [Sat, 28 Apr 2007 08:02:40 +0000 (08:02 +0000)]
    Reviewed by Oliver.

        - fix <rdar://problem/5138250> REGRESSION (r18245): Changing any one aspect
          of a font applies all the aspects to selected text

        Since this bug is only visible with the font panel, I don't
        think there's any way to make a DumpRenderTree-driven test.

        * editing/Editor.cpp: (WebCore::Editor::fontForSelection):
        Set hasMultipleFonts properly. The old code had an early exit
        that was left over from when hasMultipleFonts was a pointer.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21165 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by Maciej.
aroben [Sat, 28 Apr 2007 06:57:49 +0000 (06:57 +0000)]
    Reviewed by Maciej.

        Cleaned up prepare-ChangeLog

        * Scripts/prepare-ChangeLog: No code changes, but style now matches
        the prevalent style of our perl scripts.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21164 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago2007-04-27 Mitz Pettel <mitz@webkit.org>
bdash [Sat, 28 Apr 2007 06:48:29 +0000 (06:48 +0000)]
2007-04-27  Mitz Pettel  <mitz@webkit.org>

        Reviewed by Antti.

        - http://bugs.webkit.org/show_bug.cgi?id=13521
          Update fast/css/percentage-non-integer for the new empty table cell behavior

        * fast/css/percentage-non-integer-expected.checksum:
        * fast/css/percentage-non-integer-expected.png:
        * fast/css/percentage-non-integer-expected.txt:
        * fast/css/percentage-non-integer.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21163 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago2007-04-27 Mitz Pettel <mitz@webkit.org>
bdash [Sat, 28 Apr 2007 06:40:41 +0000 (06:40 +0000)]
2007-04-27  Mitz Pettel  <mitz@webkit.org>

        Reviewed by Dave Hyatt.

        - fix http://bugs.webkit.org/show_bug.cgi?id=13525
          Repainting a listbox schedules a relayout

        No test possible as there is no change to functionality.

        - fix the root cause of http://bugs.webkit.org/show_bug.cgi?id=10862
          REGRESSION: Collapsing form input on www.myspace.com

        No test possible since text fields are no longer implemented as flexboxes.

        * rendering/RenderObject.cpp:
        (WebCore::RenderObject::scheduleRelayout): Added a check that the subtree
        root is in the tree and not free-floating.
        * rendering/bidi.cpp:
        (WebCore::RenderBlock::bidiReorderCharacters): Changed addChild and removeChild
        to appendChildNode and removeChildNode to save some unnecessary overhead.
        These calls still mark the tree for relayout, but the change in scheduleRelayout
        ensures that a layout won't be scheduled.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21162 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoAdd back in WebInspectorPanel.
thatcher [Sat, 28 Apr 2007 05:04:23 +0000 (05:04 +0000)]
Add back in WebInspectorPanel.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21161 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoRevert some unintended changes and fix the build.
thatcher [Sat, 28 Apr 2007 04:54:16 +0000 (04:54 +0000)]
Revert some unintended changes and fix the build.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21160 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago2007-04-27 Timothy Hatcher <timothy@apple.com>
thatcher [Sat, 28 Apr 2007 04:48:58 +0000 (04:48 +0000)]
2007-04-27 Timothy Hatcher <timothy@apple.com>

        Reviewed by Darin.

        There were only a couple calls to ResourceLoadDelegate forwarder left,
        this removed the calls and added a new cached method for didFailLoad.

        * WebKit/DefaultDelegates/WebDefaultResourceLoadDelegate.h: Removed.
        * WebKit/DefaultDelegates/WebDefaultResourceLoadDelegate.m: Removed.
        * WebKit/Plugins/WebNullPluginView.mm: Call the resource load delegate directly.
        * WebKit/WebCoreSupport/WebFrameLoaderClient.mm: Call the cached didFailLoad delegate function.
        * WebKit/WebKit.xcodeproj/project.pbxproj: Remove WebDefaultResourceLoadDelegate.
        * WebKit/WebView/WebDataSource.mm: Remove the #import for WebDefaultResourceLoadDelegate.h
        * WebKit/WebView/WebView.mm: Remove the ResourceLoadDelegate forwarder, and remove a method that isn't used.
        * WebKit/WebView/WebViewInternal.h: Ditto.
        * WebKit/WebView/WebViewPrivate.h: Remove a method that is no longer used.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21159 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago<rdar://problem/5153017> REPRODUCIBLE ASSERT: range != nil in WebViewFactory.mm:415...
pyeh [Sat, 28 Apr 2007 01:41:56 +0000 (01:41 +0000)]
<rdar://problem/5153017> REPRODUCIBLE ASSERT: range != nil in WebViewFactory.mm:415 -[WebViewFactory startOfTextMarkerRange:] (13354)

        * bridge/mac/WebCoreAXObject.mm:
        (-[WebCoreAXObject textMarkerRangeForRange:]):
        Fix the range length check.  It was incorrectly assuming that the max range requested cannot be equal
        to the length of the string.  Since NSRange location is index based, it's perfectly valid to have the max
        range equal to the the text length.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21158 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoWebCore:
harrison [Sat, 28 Apr 2007 01:04:51 +0000 (01:04 +0000)]

        Reviewed by Darin.

        <rdar://problem/5160627> Export JS list creation support as ObjC SPI for Mail

        * WebCore.exp:
        * editing/Editor.h:
        * editing/Editor.cpp:
        Add insertOrderedList() and insertUnorderedList().
        Move existing list management functions from WebCoreFrameBridge.
        The level increasing functions now return a PassRefPtr<Node> instead of Node.
        Add check for canEditRichly.

        - in many other functions, condense the check for nil client().

        * editing/InsertListCommand.h:
        * editing/InsertListCommand.cpp:
        Add insertList(), which returns the new list element.
        Don't set the list element's ID if the specified ID was an empty string.

        * editing/ModifySelectionListLevel.h:
        * editing/ModifySelectionListLevel.cpp:
        Return PassRefPtr<Node> instead of Node.

        * page/mac/WebCoreFrameBridge.h:
        * page/mac/WebCoreFrameBridge.mm:
        Move list management functions to Editor.h and Editor.cpp.


        Reviewed by Darin.

        <rdar://problem/5160627> Export JS list creation support as ObjC SPI for Mail

        * WebKit.xcodeproj/project.pbxproj:
        * WebView/WebHTMLView.mm:
        (-[WebHTMLView _selectedRange]):
        (-[WebHTMLView _shouldDeleteRange:]):
        (-[WebHTMLView _canEdit]):
        (-[WebHTMLView _canEditRichly]):
        (-[WebHTMLView _hasSelection]):
        (-[WebHTMLView _hasSelectionOrInsertionPoint]):
        (-[WebHTMLView _hasInsertionPoint]):
        (-[WebHTMLView _isEditable]):
        Condense the check for nil [self frame].
        Remove canEditRichly checks and rely on the editor to do the check instead.

        (-[WebHTMLView _insertOrderedList]):
        (-[WebHTMLView _insertUnorderedList]):

        (-[WebHTMLView _canIncreaseSelectionListLevel]):
        (-[WebHTMLView _canDecreaseSelectionListLevel]):
        (-[WebHTMLView _increaseSelectionListLevel]):
        (-[WebHTMLView _increaseSelectionListLevelOrdered]):
        (-[WebHTMLView _increaseSelectionListLevelUnordered]):
        (-[WebHTMLView _decreaseSelectionListLevel]):
        Moved from bridge to frame editor.

        * WebView/WebHTMLViewPrivate.h:
        Add _insertOrderedList and _insertUnorderedList to WebHTMLView(WebPrivate)

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21157 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Fix for <rdar://problem/5134254>, toolbar buttons don't work in gmail after...
hyatt [Sat, 28 Apr 2007 00:48:43 +0000 (00:48 +0000)]
    Fix for <rdar://problem/5134254>, toolbar buttons don't work in gmail after double-click.  Make sure
        to clear out the capturing frame, since you don't get a second up when a double-click happens.

        Reviewed by mjs

        * page/EventHandler.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21156 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoLayoutTests:
justing [Fri, 27 Apr 2007 21:59:53 +0000 (21:59 +0000)]

        Reviewed by oliver

        Enable execCommand("CreateLink", ...) for caret selections

        * editing/execCommand/5164796-expected.checksum: Added.
        * editing/execCommand/5164796-expected.png: Added.
        * editing/execCommand/5164796-expected.txt: Added.
        * editing/execCommand/5164796.html: Added.


        Reviewed by oliver

        Enable execCommand("CreateLink", ...) for caret selections

        * editing/CreateLinkCommand.cpp:
        (WebCore::CreateLinkCommand::doApply): For a caret
        selection, insert a link at the caret position
        with the url as the link text.
        * editing/JSEditor.cpp: Enable CreateLink for caret

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21155 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Rubberstamped by Mark
beidson [Fri, 27 Apr 2007 17:02:13 +0000 (17:02 +0000)]
    Rubberstamped by Mark

        Remove default implementation of UIDelegate method that was removed

        I have also been instructed to give Tim a hard time about this one - apparently
        it was his job to clean it out and he failed...  failed miserably.


        * DefaultDelegates/WebDefaultUIDelegate.m: Removed webViewPrint:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21149 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago2007-04-27 Nazar Kulyk <schamane@myeburg.net>
bdash [Fri, 27 Apr 2007 14:55:19 +0000 (14:55 +0000)]
2007-04-27  Nazar Kulyk  <schamane@myeburg.net>

        Reviewed by Mark Rowe.

        Basic auto-correction of user-entered URLs.

        * GdkLauncher/main.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21148 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Forgot this png from previous commit.
antti [Fri, 27 Apr 2007 13:08:36 +0000 (13:08 +0000)]
    Forgot this png from previous commit.

        * fast/table/empty-cells-expected.png: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21147 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago2007-04-28 Mark Rowe <mrowe@apple.com>
bdash [Fri, 27 Apr 2007 12:36:52 +0000 (12:36 +0000)]
2007-04-28  Mark Rowe  <mrowe@apple.com>

        Rubber-stamped by Anders.

        Rename some methods and variables to match the style guidelines.

        * GdkLauncher/main.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21146 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoLayoutTests:
antti [Fri, 27 Apr 2007 11:32:29 +0000 (11:32 +0000)]

        Reviewed by Darin, Hyatt.

        Test for http://bugs.webkit.org/show_bug.cgi?id=13081
        Empty table cells not handled correctly

        Added test case and checked in updated results for others. Changes represent
        progressions or are not visible in rendering output.

        * fast/block/positioning/negative-right-pos-expected.txt:
        * fast/css/acid2-expected.txt:
        * fast/css/acid2-pixel-expected.txt:
        * fast/css/percentage-non-integer-expected.checksum:
        * fast/css/percentage-non-integer-expected.png:
        * fast/css/percentage-non-integer-expected.txt:
        * fast/repaint/table-cell-move-expected.checksum:
        * fast/repaint/table-cell-move-expected.png:
        * fast/repaint/table-cell-move-expected.txt:
        * fast/table/012-expected.txt:
        * fast/table/032-expected.txt:
        * fast/table/empty-cells-expected.checksum: Added.
        * fast/table/empty-cells-expected.txt: Added.
        * fast/table/empty-cells.html: Added.
        * fast/table/empty-section-crash-expected.txt:
        * http/tests/misc/acid2-expected.txt:
        * http/tests/misc/acid2-pixel-expected.txt:
        * tables/mozilla/bugs/bug100334-expected.txt:
        * tables/mozilla/bugs/bug1188-expected.txt:
        * tables/mozilla/bugs/bug16012-expected.checksum:
        * tables/mozilla/bugs/bug16012-expected.png:
        * tables/mozilla/bugs/bug16012-expected.txt:
        * tables/mozilla/bugs/bug1818-6-expected.txt:
        * tables/mozilla/bugs/bug222336-expected.txt:
        * tables/mozilla/bugs/bug222467-expected.txt:
        * tables/mozilla/bugs/bug78162-expected.txt:
        * tables/mozilla/core/misc-expected.checksum:
        * tables/mozilla/core/misc-expected.png:
        * tables/mozilla/core/misc-expected.txt:
        * tables/mozilla_expected_failures/bugs/bug14007-1-expected.checksum:
        * tables/mozilla_expected_failures/bugs/bug14007-1-expected.png:
        * tables/mozilla_expected_failures/bugs/bug14007-1-expected.txt:
        * tables/mozilla_expected_failures/bugs/bug72393-expected.checksum:
        * tables/mozilla_expected_failures/bugs/bug72393-expected.png:
        * tables/mozilla_expected_failures/bugs/bug72393-expected.txt:
        * tables/mozilla_expected_failures/other/empty_cells-expected.txt:


        Reviewed by Darin, Hyatt.

        Fix http://bugs.webkit.org/show_bug.cgi?id=13081
        Empty table cells not handled correctly

        To match Firefox and IE, empty table cells with variable width need to be
        treated as special case. They generally get zero width.

        * rendering/AutoTableLayout.cpp:
        * rendering/AutoTableLayout.h:
        * rendering/RenderStyle.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21145 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by Oliver.
rwlbuis [Fri, 27 Apr 2007 11:20:09 +0000 (11:20 +0000)]
    Reviewed by Oliver.

        Text not rendered in a second inline SVG element

        Correct the rendering of svg text when embedded in non-svg container.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21144 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by Mark.
mjs [Fri, 27 Apr 2007 08:43:19 +0000 (08:43 +0000)]
    Reviewed by Mark.

        <rdar://problem/5154113> Repro ASSERT (would be crash) in KJS::GCLock::GCLock (13462)

        * WebInspector/WebInspector.m:
        (-[WebInspectorPrivate dealloc]): Delay release of WebView to avoid GC re-entrancy.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21143 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago2007-04-27 Mark Rowe <mrowe@apple.com>
bdash [Fri, 27 Apr 2007 08:40:27 +0000 (08:40 +0000)]
2007-04-27  Mark Rowe  <mrowe@apple.com>

        Reviewed by Maciej.

        * WebCore.pro: Remove non-existent paths from qmake files.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21142 268f45cc-cd09-0410-ab3c-d52691b4dbfc