WebKit-https.git
13 years agoJavaScriptCore:
darin@apple.com [Mon, 25 Feb 2008 17:06:51 +0000 (17:06 +0000)]
JavaScriptCore:

        Reviewed by Adam.

        * JavaScriptCore.exp: Sort the contents of this file.

WebCore:

        Reviewed by Adam.

        * WebCore.base.exp: Export blankURL, and sort the file.

WebKit/mac:

        Reviewed by Adam.

        * WebView/WebArchiver.mm:
        (+[WebArchiver archiveSelectionInFrame:]): Use blankURL.
        * WebView/WebFrame.mm:
        (-[WebFrame _loadData:MIMEType:textEncodingName:baseURL:unreachableURL:]): Avoid the
        variable name URL to avoid clashing with the renamed KURL in the future. Also use
        blankURL.
        (-[WebFrame loadData:MIMEType:textEncodingName:baseURL:]): Ditto.
        (-[WebFrame _loadHTMLString:baseURL:unreachableURL:]): Ditto.
        (-[WebFrame loadHTMLString:baseURL:]): Ditto.
        (-[WebFrame loadAlternateHTMLString:baseURL:forUnreachableURL:]): Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30568 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago MSVC build fix
aroben@apple.com [Mon, 25 Feb 2008 16:53:15 +0000 (16:53 +0000)]
    MSVC build fix

        * kjs/testkjs.cpp:
        (functionQuit): Don't add a return statement after exit(0) for MSVC.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30567 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago Move many dispatch methods to WebFrameLoaderClient
aroben@apple.com [Mon, 25 Feb 2008 16:52:55 +0000 (16:52 +0000)]
     Move many dispatch methods to WebFrameLoaderClient

         Reviewed by Sam.

         * WebCoreSupport/WebFrameLoaderClient.cpp:
         (WebFrameLoaderClient::dispatchDidHandleOnloadEvents):
         (WebFrameLoaderClient::dispatchDidReceiveServerRedirectForProvisionalLoad):
         (WebFrameLoaderClient::dispatchDidCancelClientRedirect):
         (WebFrameLoaderClient::dispatchWillPerformClientRedirect):
         (WebFrameLoaderClient::dispatchDidChangeLocationWithinPage):
         (WebFrameLoaderClient::dispatchWillClose):
         (WebFrameLoaderClient::dispatchDidReceiveIcon):
         (WebFrameLoaderClient::dispatchDidStartProvisionalLoad):
         (WebFrameLoaderClient::dispatchDidReceiveTitle):
         (WebFrameLoaderClient::dispatchDidCommitLoad):
         (WebFrameLoaderClient::dispatchDidFinishDocumentLoad):
         (WebFrameLoaderClient::dispatchDidFinishLoad):
         (WebFrameLoaderClient::dispatchDidFirstLayout):
         (WebFrameLoaderClient::dispatchShow):
         * WebCoreSupport/WebFrameLoaderClient.h:
         * WebFrame.cpp:
         * WebFrame.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30566 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago Move two more methods to WebFrameLoaderClient
aroben@apple.com [Mon, 25 Feb 2008 16:52:30 +0000 (16:52 +0000)]
     Move two more methods to WebFrameLoaderClient

         Reviewed by Sam.

         * WebCoreSupport/WebFrameLoaderClient.cpp:
         (WebFrameLoaderClient::hasFrameView):
         (WebFrame::forceLayout):
         * WebCoreSupport/WebFrameLoaderClient.h:
         * WebFrame.cpp:
         * WebFrame.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30565 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years agoFix compilation with gcc 4.3, added stdio.h and unistd.h inclusions.
hausmann@webkit.org [Mon, 25 Feb 2008 13:16:40 +0000 (13:16 +0000)]
Fix compilation with gcc 4.3, added stdio.h and unistd.h inclusions.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30564 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago2008-02-25 Johnny Ding <johnnyding.webkit@gmail.com>
mrowe@apple.com [Mon, 25 Feb 2008 13:14:27 +0000 (13:14 +0000)]
2008-02-25  Johnny Ding  <johnnyding.webkit@gmail.com>

        Reviewed by Darin Adler.

        - fix http://bugs.webkit.org/show_bug.cgi?id=17444

        In HTMLTokenizer::write, the code checks 'pendingScripts.isEmpty()' to decide
        whether to save prependingSrc or not. However, in HTMLTokenizer::scriptHandler
        and HTMLTokenizer::scriptExecution, the code checks testBit:LoadingExtScript
        to decide whether to save prependingSrc or not. The later behavior is not right
        because, in scriptHandler and scriptExecution, even the pendingScripts queue is
        empty, the testBit:LoadingExtScript might be TRUE.

        Test: fast/tokenizer/nested-multiple-scripts.html

        * html/HTMLTokenizer.cpp:
        (WebCore::HTMLTokenizer::scriptHandler): check pendingScripts.isEmpty() instead of
        checking state.loadingExtScript().
        (WebCore::HTMLTokenizer::scriptExecution): check pendingScripts.isEmpty() instead of
        checking state.loadingExtScript().

2008-02-25  Johnny Ding  <johnnyding.webkit@gmail.com>

        Reviewed by Darin Adler.

        - bug http://bugs.webkit.org/show_bug.cgi?id=17444
        Test for multiple nested scripts which are in a external script.

        * fast/tokenizer/nested-multiple-scripts-expected.txt: Added.
        * fast/tokenizer/nested-multiple-scripts.html: Added.
        * fast/tokenizer/resources/external-script-1.js: Added.
        * fast/tokenizer/resources/external-script-2.js: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30563 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago * Scripts/do-webcore-rename: Make some updates based on a trial run of
darin@apple.com [Mon, 25 Feb 2008 07:44:58 +0000 (07:44 +0000)]
    * Scripts/do-webcore-rename: Make some updates based on a trial run of
        the renaming script.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30562 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago * Scripts/do-webcore-rename: More renaming plans.
darin@apple.com [Mon, 25 Feb 2008 07:17:12 +0000 (07:17 +0000)]
    * Scripts/do-webcore-rename: More renaming plans.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30561 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago Reviewed by Mark Rowe.
weinig@apple.com [Mon, 25 Feb 2008 07:11:40 +0000 (07:11 +0000)]
    Reviewed by Mark Rowe.

        http://bugs.webkit.org/show_bug.cgi?id=17529
        Add support for reading from stdin from testkjs

        * kjs/testkjs.cpp:
        (GlobalObject::GlobalObject): Add readline function to global object.
        (functionReadline): Added. Reads characters from stdin until a '\n' or
        EOF is encountered. The input is returned as a String to the caller.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30560 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago * dom/Document.h: Removed stray duplicate declaration of
darin@apple.com [Mon, 25 Feb 2008 07:09:45 +0000 (07:09 +0000)]
    * dom/Document.h: Removed stray duplicate declaration of
        RegisteredEventListenerList.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30559 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago * dom/Range.cpp:
darin@apple.com [Mon, 25 Feb 2008 06:52:43 +0000 (06:52 +0000)]
    * dom/Range.cpp:
        (WebCore::Range::surroundContents): Removed incorrect comment.
        I added it a while ago, and I was just mixed up.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30558 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago - quick follow-up to that last check-in
darin@apple.com [Mon, 25 Feb 2008 06:48:07 +0000 (06:48 +0000)]
    - quick follow-up to that last check-in

        * platform/graphics/cg/ImageSourceCG.cpp:
        (WebCore::ImageSource::createFrameAtIndex): Use CFEqual instead of
        WebCore::String's == for speed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30557 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years agoWebCore:
darin@apple.com [Mon, 25 Feb 2008 06:36:19 +0000 (06:36 +0000)]
WebCore:

        Reviewed by Darin.

        Bug 8749: XBM rendered incorrectly as black on white

        xbm images are now created and rendered using black on a
        transparent background (same behavior as in Firefox)

        Test: platform/mac/fast/canvas/canvas-draw-xbm-image.html

        * platform/graphics/cg/ImageSourceCG.cpp:
        (WebCore::ImageSource::createFrameAtIndex):

LayoutTests:

        Reviewed by Darin.

        - test for http://bugs.webkit.org/show_bug.cgi?id=8749

        * platform/mac/fast/canvas/canvas-draw-xbm-image-expected.checksum: Added.
        * platform/mac/fast/canvas/canvas-draw-xbm-image-expected.png: Added.
        * platform/mac/fast/canvas/canvas-draw-xbm-image-expected.txt: Added.
        * platform/mac/fast/canvas/canvas-draw-xbm-image.html: Added.
        * platform/mac/fast/canvas/resources: Added.
        * platform/mac/fast/canvas/resources/smile.xbm: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30556 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago Reviewed by Mark Rowe.
weinig@apple.com [Mon, 25 Feb 2008 05:29:20 +0000 (05:29 +0000)]
    Reviewed by Mark Rowe.

        http://bugs.webkit.org/show_bug.cgi?id=17528
        Give testkjs a bath

        * JavaScriptCore.exp:
        * JavaScriptCore.xcodeproj/project.pbxproj: Make the testkjs.cpp use 4 space indentation.
        * kjs/testkjs.cpp:
        (StopWatch::getElapsedMS):
        (GlobalObject::className):
        (GlobalObject::GlobalObject):
        Rename GlobalImp to GlobalObject and setup the global functions
        in the GlobalObject's constructor. Also, use static functions for
        the implementation so we can use the standard PrototypeFunction
        class and remove TestFunctionImp.
        (functionPrint): Move print() functionality here.
        (functionDebug): Move debug() functionality here.
        (functionGC): Move gc() functionality here.
        (functionVersion): Move version() functionality here.
        (functionRun): Move run() functionality here.
        (functionLoad): Move load() functionality here.
        (functionQuit): Move quit() functionality here.
        (prettyPrintScript): Fix indentation.
        (runWithScripts): Since all the functionality of createGlobalObject is
        now in the GlobalObject constructor, just call new here.
        (parseArguments): Fix indentation.
        (kjsmain): Ditto
        (fillBufferWithContentsOfFile): Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30555 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years agoWebCore:
mitz@apple.com [Mon, 25 Feb 2008 03:08:37 +0000 (03:08 +0000)]
WebCore:

        Reviewed by Darin Adler.

        - fix http://bugs.webkit.org/show_bug.cgi?id=17471
          REGRESSION (r30438): Crash in deleteLineBoxes in cursor-adjusting code browsing Acid3

        Test: http/tests/misc/object-image-error.html

        * html/HTMLObjectElement.cpp:
        (WebCore::HTMLObjectElement::attach): Added an early return in case
        updateFromElement() has changed the object to use fallback content.

LayoutTests:

        Reviewed by Darin Adler.

        - test for http://bugs.webkit.org/show_bug.cgi?id=17471
          REGRESSION (r30438): Crash in deleteLineBoxes in cursor-adjusting code browsing Acid3

        * http/tests/misc/object-image-error-expected.txt: Added.
        * http/tests/misc/object-image-error.html: Added.
        * http/tests/misc/resources/404image.php: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30554 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago Reviewed by Oliver Hunt and Mark Rowe.
weinig@apple.com [Mon, 25 Feb 2008 01:34:09 +0000 (01:34 +0000)]
    Reviewed by Oliver Hunt and Mark Rowe.

        http://bugs.webkit.org/show_bug.cgi?id=17505
        Add support for getting command line arguments in testkjs

        - This slightly changes the behavior of parsing arguments by requiring
          a '-f' before all files.

        * kjs/testkjs.cpp:
        (createGlobalObject): Add a global property called 'arguments' which
        contains an array with the parsed arguments as strings.
        (runWithScripts): Pass in the arguments vector so that it can be passed
        to the global object.
        (parseArguments): Change parsing rules to require a '-f' before any script
        file. After all '-f' and '-p' arguments have been parsed, the remaining
        are added to the arguments vector and exposed to the script. If there is a
        chance of ambiguity (the user wants to pass the string '-f' to the script),
        the string '--' can be used separate the options from the pass through
        arguments.
        (kjsmain):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30553 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago Rubber-stamped by Oliver Hunt.
mitz@apple.com [Sun, 24 Feb 2008 22:50:02 +0000 (22:50 +0000)]
    Rubber-stamped by Oliver Hunt.

        - replace expected results in platform/mac-tiger and
          platform/mac-leopard which were identical with a single copy in
          platform/mac

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30552 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years agoWebCore:
mitz@apple.com [Sun, 24 Feb 2008 18:48:27 +0000 (18:48 +0000)]
WebCore:

        Reviewed by Darin Adler.

        - fix float positioning when a float that does not fit on the line is
          followed by a float that does

        Test: fast/block/float/narrow-after-wide.html

        * rendering/bidi.cpp:
        (WebCore::RenderBlock::findNextLineBreak): Changed to not position any
        more floats on the line once a float that does not fit is encountered.
        That float should be pushed to the next line, and so should all floats
        that follow, regardless of whether they can fit on the current line.

LayoutTests:

        Reviewed by Darin Adler.

        - test float positioning when a float that does not fit on the line is
          followed by a float that does

        * fast/block/float/narrow-after-wide.html: Added.
        * platform/mac/fast/block/float/narrow-after-wide-expected.checksum: Added.
        * platform/mac/fast/block/float/narrow-after-wide-expected.png: Added.
        * platform/mac/fast/block/float/narrow-after-wide-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30551 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago Reviewed by Darin Adler.
mitz@apple.com [Sun, 24 Feb 2008 18:44:26 +0000 (18:44 +0000)]
    Reviewed by Darin Adler.

        - fix http://bugs.webkit.org/show_bug.cgi?id=17511
          REGRESSION: Reproducible crash in SegmentedSubstring::SegmentedSubstring(SegmentedSubstring const&)

        * wtf/Deque.h:
        (WTF::::expandCapacityIfNeeded): Fixed the case where m_start and m_end
        are both zero but the buffer capacity is non-zero.
        (WTF::::prepend): Added validity checks.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30550 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years agoWebCore:
darin@apple.com [Sun, 24 Feb 2008 08:30:32 +0000 (08:30 +0000)]
WebCore:

        Reviewed by Sam.

        - fix http://bugs.webkit.org/show_bug.cgi?id=16770
          Acid3 expects :visited styled links to restyle on iframe load

        Disentangle global history updating from the back/forward history.
        There are many cases where we don't want to create a new back/forward
        item, but we do still want to add to the global history (used for visited
        link coloring) in those cases.

        Test: fast/history/subframe-is-visited.html

        * loader/FrameLoader.cpp:
        (WebCore::FrameLoader::updateGlobalHistory): Renamed from
        addHistoryForCurrentLocation and removed the back/forward handling.
        (WebCore::FrameLoader::updateHistoryForStandardLoad): Streamlined logic a
        bit. Replaced call to addHistoryForCurrentLocation with a call to
        addBackForwardItemClippedAtTarget. Added an unconditional call to
        updateGlobalHistory.
        (WebCore::FrameLoader::updateHistoryForClientRedirect): Added a FIXME;
        why doesn't this function update global history?
        (WebCore::FrameLoader::updateHistoryForBackForwardNavigation): Ditto.
        (WebCore::FrameLoader::updateHistoryForReload): Replaced the direct call
        the client with a call to the new updateGlobalHistory function.
        (WebCore::FrameLoader::updateHistoryForRedirectWithLockedHistory): Did
        the same changes as for updateHistoryForStandardLoad.
        * loader/FrameLoader.h: More of the same.

        * loader/FrameLoaderClient.h: Removed updateGlobalHistoryForReload and
        renamed updateGlobalHistoryForStandardLoad to updateGlobalHistory.

        * svg/graphics/SVGImageEmptyClients.h:
        (WebCore::SVGEmptyFrameLoaderClient::updateGlobalHistory): Updated to
        match the above.

WebKit/gtk:

        Reviewed by Sam.

        - remove separate client calls for "standard" and "reload' history

        * WebCoreSupport/FrameLoaderClientGtk.cpp:
        (WebKit::FrameLoaderClient::updateGlobalHistory):
        * WebCoreSupport/FrameLoaderClientGtk.h:

WebKit/mac:

        Reviewed by Sam.

        - remove separate client calls for "standard" and "reload' history

        * WebCoreSupport/WebFrameLoaderClient.h:
        * WebCoreSupport/WebFrameLoaderClient.mm:
        (WebFrameLoaderClient::updateGlobalHistory):

WebKit/qt:

        Reviewed by Sam.

        - remove separate client calls for "standard" and "reload' history

        * WebCoreSupport/FrameLoaderClientQt.cpp:
        (WebCore::FrameLoaderClientQt::updateGlobalHistory):
        * WebCoreSupport/FrameLoaderClientQt.h:

WebKit/win:

        Reviewed by Sam.

        - remove separate client calls for "standard" and "reload' history

        * WebFrame.cpp:
        (WebFrame::updateGlobalHistory):
        * WebFrame.h:

WebKit/wx:

        Reviewed by Sam.

        - remove separate client calls for "standard" and "reload' history

        * WebKitSupport/FrameLoaderClientWx.cpp:
        (WebCore::FrameLoaderClientWx::updateGlobalHistory):
        * WebKitSupport/FrameLoaderClientWx.h:

LayoutTests:

        Reviewed by Sam.

        - test for http://bugs.webkit.org/show_bug.cgi?id=16770
          Acid3 expects :visited styled links to restyle on iframe load

        * fast/history/resources/subframe.html: Added.
        * fast/history/subframe-is-visited-expected.txt: Added.
        * fast/history/subframe-is-visited.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30549 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago - another try at fixing the build
darin@apple.com [Sun, 24 Feb 2008 08:09:22 +0000 (08:09 +0000)]
    - another try at fixing the build

        * Api/qwebframe.cpp:
        (QWebFrame::load): Replace new FormData with FormData::create.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30548 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago Reviewed by Sam.
darin@apple.com [Sun, 24 Feb 2008 07:55:40 +0000 (07:55 +0000)]
    Reviewed by Sam.

        - fix http://bugs.webkit.org/show_bug.cgi?id=17508
          REGRESSION (r30535): mailto regression tests failing

        * html/HTMLFormElement.cpp:
        (WebCore::HTMLFormElement::submit): Restore the old logic, with its unpleasant
        trip through String do to the replace operation. It's inefficient, but there's
        no real reason to worry about the efficiency of mailto forms, since that feature
        is nearly obsolete.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30547 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago Reviewed by Sam.
darin@apple.com [Sun, 24 Feb 2008 07:36:57 +0000 (07:36 +0000)]
    Reviewed by Sam.

        - replace SegmentedString's use of DeprecatedValueList with Deque

        Testing indicates this is a slight speed-up for page loading.

        * platform/text/SegmentedString.cpp:
        (WebCore::SegmentedString::length): Update to use Deque.
        (WebCore::SegmentedString::setExcludeLineNumbers): Ditto.
        (WebCore::SegmentedString::append): Ditto.
        (WebCore::SegmentedString::prepend): Ditto.
        (WebCore::SegmentedString::advanceSubstring): Ditto.
        (WebCore::SegmentedString::toString): Ditto.
        * platform/text/SegmentedString.h: Tweak formatting. Don't bother making an entire
        class private with a single friend. Change m_sbstrings from a DeprecatedValueList
        to a Deque.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30546 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago Reviewed by Anders.
darin@apple.com [Sun, 24 Feb 2008 07:26:02 +0000 (07:26 +0000)]
    Reviewed by Anders.

        - fix http://bugs.webkit.org/show_bug.cgi?id=17506
          REGRESSION (r30535): ASSERTION FAILED: i < size()

        * loader/TextResourceDecoder.cpp:
        (WebCore::findTextEncoding): Fix off-by-one in code that null-terminates
        the C string here.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30545 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago Not reviewed, build fix.
pewtermoose@webkit.org [Sun, 24 Feb 2008 07:10:49 +0000 (07:10 +0000)]
    Not reviewed, build fix.

        * platform/network/cf/FormDataStreamCFNet.cpp:
        (WebCore::httpBodyFromRequest):
        * plugins/win/PluginViewWin.cpp:
        (WebCore::PluginView::handlePost):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30544 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago - fix a typo
mitz@apple.com [Sun, 24 Feb 2008 07:07:30 +0000 (07:07 +0000)]
    - fix a typo

        * platform/graphics/win/GlyphPageTreeNodeCGWin.cpp:
        (WebCore::GlyphPage::fill):
        * platform/graphics/win/GlyphPageTreeNodeCairoWin.cpp:
        (WebCore::GlyphPage::fill):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30543 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago2008-02-23 Jan Michael Alonzo <jmalonzo@unpluggable.com>
pewtermoose@webkit.org [Sun, 24 Feb 2008 07:07:03 +0000 (07:07 +0000)]
2008-02-23  Jan Michael Alonzo  <jmalonzo@unpluggable.com>
.:

        Rubber stamped by Darin.

        Add separator '\' after libJavaScriptCore_la_LIBADD and cleanup
        whitespaces introduced in the previous commit.

        * GNUmakefile.am:

JavaScriptCore:

        Rubber stamped by Darin.

        Add separator '\' after libJavaScriptCore_la_LIBADD and cleanup
        whitespaces introduced in the previous commit.

        * GNUmakefile.am:

WebKitTools:

        Rubber stamped by Darin.

        Add separator '\' after libJavaScriptCore_la_LIBADD and cleanup
        whitespaces introduced in the previous commit.

        * GNUmakefile.am:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30542 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago Reviewed by Anders Carlsson.
mitz@apple.com [Sun, 24 Feb 2008 06:59:53 +0000 (06:59 +0000)]
    Reviewed by Anders Carlsson.

        - fix GlyphPage::fill() logic for partial page fills

        * platform/graphics/win/GlyphPageTreeNodeCGWin.cpp:
        (WebCore::GlyphPage::fill): Changed the test for non-BMP characters to
        work correctly when the length parameter is less than a full page size.
        Also updated the comment.
        * platform/graphics/win/GlyphPageTreeNodeCairoWin.cpp:
        (WebCore::GlyphPage::fill): Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30541 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago.:
darin@apple.com [Sun, 24 Feb 2008 05:42:29 +0000 (05:42 +0000)]
.:

        Rubber-stamped by Darin.

        * GNUmakefile.am: Add both GLOBALDEPS and WEBKITDEPS instead of DEPENDENCIES.
        * configure.ac: Break dependent modules into GLOBALDEPS and WEBKITDEPS.

JavaScriptCore:

        * GNUmakefile.am: Add GLOBALDEPS for testkjs and minidom.

WebKitTools:

        Rubber-stamped by Darin.

        * GNUmakefile.am: Add both GLOBALDEPS and WEBKITDEPS instead of DEPENDENCIES.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30540 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years agoMove text drawing into wxcode, as we need platform-dependent APIs for drawing non...
kevino@webkit.org [Sun, 24 Feb 2008 05:38:49 +0000 (05:38 +0000)]
Move text drawing into wxcode, as we need platform-dependent APIs for drawing non-kerned text, which wx doesn't yet have.

http://bugs.webkit.org/show_bug.cgi?id=17396

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30539 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago Reviewed by Anders.
darin@apple.com [Sun, 24 Feb 2008 05:19:27 +0000 (05:19 +0000)]
    Reviewed by Anders.

        - http://bugs.webkit.org/show_bug.cgi?id=17496
          make Deque use a circular array; add iterators

        * wtf/Deque.h: Wrote an all-new version of this class that uses a circular
        buffer. Growth policy is identical to vector. Added iterators.

        * wtf/Vector.h: Made two small refinements while using this to implement
        Deque: Made VectorBufferBase derive from Noncopyable, which would have
        saved me some debugging time if it had been there. Renamed Impl and
        m_impl to Buffer and m_buffer.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30538 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago Pointed out by Mitz.
darin@apple.com [Sun, 24 Feb 2008 05:13:57 +0000 (05:13 +0000)]
    Pointed out by Mitz.

        * loader/TextResourceDecoder.cpp:
        (WebCore::find): Fix a "==" that was supposed to be ">".
        (WebCore::findIgnoringCase): Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30537 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago Reviewed by Sam.
darin@apple.com [Sun, 24 Feb 2008 05:12:17 +0000 (05:12 +0000)]
    Reviewed by Sam.

        * html/HTMLFormElement.cpp:
        (WebCore::appendEncodedString): Fix buffer overrun in the code I just landed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30536 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago Reviewed by Sam.
darin@apple.com [Sun, 24 Feb 2008 05:04:38 +0000 (05:04 +0000)]
    Reviewed by Sam.

        - remove all but 1 remaining use of DeprecatedCString
        - change FormData to start refcount at 1

        * html/HTMLFormElement.cpp:
        (WebCore::appendString): Added. Helpers for Vector<char>.
        (WebCore::appendEncodedString): Renamed from encodeCString and changed to
        append to an existing Vector<char> to cut down a bit on memory allocation.
        (WebCore::HTMLFormElement::formData): Replace the DeprecatedCString called
        enc_string with a Vector<char> called encodedData. Change to use the new
        FormData::create function.
        (WebCore::HTMLFormElement::submit): Same thing, for the mailto form code.

        * loader/TextResourceDecoder.cpp:
        (WebCore::find): Added. Helper for searching in char* with length.
        (WebCore::findIgnoringCase): Ditto.
        (WebCore::findTextEncoding): Added. Helper for looking up a text encoding
        when we have a char* with length rather than a null-terminated char*.
        (WebCore::findXMLEncoding): Changed to use char* and length rather than
        DeprecatedCString. Also fixed some things that would be one-character
        buffer overruns in a string that's not null-terminated.
        (WebCore::TextResourceDecoder::checkForCSSCharset): Change to use
        findTextEncoding rather than constructing a CString to pass to the
        TextEncodingconstructor.
        (WebCore::TextResourceDecoder::checkForHeadCharset): Ditto.

        * platform/network/FormData.cpp:
        (WebCore::FormData::FormData): Changed to start refcount at 1.
        (WebCore::FormData::create): Added.
        (WebCore::FormData::copy): Call adoptRef since refcount starts at 1.
        * platform/network/FormData.h: Made constructor private. Added create functions.

        * platform/network/mac/ResourceRequestMac.mm:
        (WebCore::ResourceRequest::doUpdateResourceRequest): Call FormData::create.
        * xml/XMLHttpRequest.cpp:
        (WebCore::XMLHttpRequest::send): Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30535 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years agoJavaScriptCore:
darin@apple.com [Sun, 24 Feb 2008 05:01:27 +0000 (05:01 +0000)]
JavaScriptCore:

        Reviewed by Anders.

        - http://bugs.webkit.org/show_bug.cgi?id=17067
          eliminate attributes parameter from JSObject::put for speed/clarity

        * API/JSCallbackObject.h: Removed attribute arguments.
        * API/JSCallbackObjectFunctions.h:
        (KJS::JSCallbackObject<Base>::put): Ditto.
        * API/JSObjectRef.cpp:
        (JSObjectSetProperty): Use initializeVariable or putDirect when necessary
        to set attribute values.
        * JavaScriptCore.exp: Updated.
        * bindings/objc/objc_runtime.h: Removed attribute arguments.
        * bindings/objc/objc_runtime.mm:
        (ObjcFallbackObjectImp::put): Ditto.
        * bindings/runtime_array.cpp:
        (RuntimeArray::put): Ditto.
        * bindings/runtime_array.h: Ditto.
        * bindings/runtime_object.cpp:
        (RuntimeObjectImp::put): Ditto.
        * bindings/runtime_object.h: Ditto. Also removed canPut which was only
        called from one place in WebCore that can use hasProperty instead.

        * kjs/Activation.h: Removed attribute argument from put and added the new
        initializeVariable function that's used to put variables in variable objects.
        Also made isActivationObject a const member.

        * kjs/JSGlobalObject.cpp:
        (KJS::JSGlobalObject::put): Removed attribute argument.
        (KJS::JSGlobalObject::initializeVariable): Added. Used to give variables
        their initial values, which can include the read-only property.
        (KJS::JSGlobalObject::reset): Removed obsolete comments about flags.
        Removed Internal flag, which is no longer needed.
        * kjs/JSGlobalObject.h: More of the same.

        * kjs/JSVariableObject.h: Added pure virtual initializeVariable function.
        (KJS::JSVariableObject::symbolTablePut): Removed checkReadOnly flag; we always
        check read-only.
        (KJS::JSVariableObject::symbolTableInitializeVariable): Added.

        * kjs/array_instance.cpp:
        (KJS::ArrayInstance::put): Removed attribute argument.
        * kjs/array_instance.h: Ditto.

        * kjs/function.cpp:
        (KJS::FunctionImp::put): Ditto.
        (KJS::Arguments::put): Ditto.
        (KJS::ActivationImp::put): Ditto.
        (KJS::ActivationImp::initializeVariable): Added.
        * kjs/function.h: Removed attribute arguments.

        * kjs/function_object.cpp:
        (KJS::FunctionObjectImp::construct): Removed Internal flag.

        * kjs/lookup.h:
        (KJS::lookupPut): Removed attributes argument. Also changed to use putDirect
        instead of calling JSObject::put.
        (KJS::cacheGlobalObject): Ditto.

        * kjs/nodes.cpp:
        (KJS::ConstDeclNode::handleSlowCase): Call initializeVariable to initialize
        the constant.
        (KJS::ConstDeclNode::evaluateSingle): Ditto.
        (KJS::TryNode::execute): Use putDirect to set up the new object.
        (KJS::FunctionBodyNode::processDeclarations): Removed Internal.
        (KJS::ProgramNode::processDeclarations): Ditto.
        (KJS::EvalNode::processDeclarations): Call initializeVariable to initialize
        the variables and functions.
        (KJS::FuncDeclNode::makeFunction): Removed Internal.
        (KJS::FuncExprNode::evaluate): Ditto.

        * kjs/object.cpp: Removed canPut, which was only being used in one code path,
        not the normal high speed one.
        (KJS::JSObject::put): Removed attribute argument. Moved the logic from
        canPut here, in the one code ath that was still using it.
        * kjs/object.h: Removed Internal attribute, ad canPut function. Removed the
        attributes argument to the put function. Made isActivationObject const.

        * kjs/regexp_object.cpp:
        (KJS::RegExpImp::put): Removed attributes argument.
        (KJS::RegExpImp::putValueProperty): Ditto.
        (KJS::RegExpObjectImp::put): Ditto.
        (KJS::RegExpObjectImp::putValueProperty): Ditto.
        * kjs/regexp_object.h: Ditto.

        * kjs/string_object.cpp:
        (KJS::StringInstance::put): Removed attributes argument.
        * kjs/string_object.h: Ditto.

WebCore:

        Reviewed by Anders.

        - http://bugs.webkit.org/show_bug.cgi?id=17067
          eliminate attributes parameter from JSObject::put for speed/clarity

        * bindings/js/JSCSSStyleDeclarationCustom.cpp:
        (WebCore::JSCSSStyleDeclaration::customPut): Remove attributes argument.

        * bindings/js/JSCanvasPixelArrayCustom.cpp:
        (WebCore::JSCanvasPixelArray::indexGetter): Use early exit idiom.
        (WebCore::JSCanvasPixelArray::indexSetter): Moved length check into the
        CanvasPixelArray object, for consistency with the getter. Removed attributes
        argument.

        * bindings/js/JSDOMWindowCustom.cpp:
        (WebCore::JSDOMWindow::customPut): Removed special case for variable
        initialization, which is not needed since that does use put any more.
        Removed attributes argument.

        * bindings/js/JSEventTargetBase.h:
        (WebCore::JSEventTargetBase::putValueProperty): Removed attributes argument.
        (WebCore::JSEventTargetBase::put): Ditto.
        (WebCore::JSEventTargetPrototype::self): Removed Internal flag.
        * bindings/js/JSEventTargetNode.cpp:
        (WebCore::JSEventTargetNode::put): Removed attributes argument.
        (WebCore::JSEventTargetNode::putValueProperty): Ditto.
        * bindings/js/JSEventTargetNode.h: Ditto.
        * bindings/js/JSHTMLAppletElementCustom.cpp:
        (WebCore::JSHTMLAppletElement::customPut): Ditto.
        * bindings/js/JSHTMLEmbedElementCustom.cpp:
        (WebCore::JSHTMLEmbedElement::customPut): Ditto.
        * bindings/js/JSHTMLInputElementBase.cpp:
        (WebCore::JSHTMLInputElementBase::put): Ditto.
        (WebCore::JSHTMLInputElementBase::putValueProperty): Ditto.
        * bindings/js/JSHTMLInputElementBase.h: Ditto.
        * bindings/js/JSHTMLObjectElementCustom.cpp:
        (WebCore::JSHTMLObjectElement::customPut): Ditto.
        * bindings/js/JSHTMLOptionsCollectionCustom.cpp:
        (WebCore::JSHTMLOptionsCollection::indexSetter): Ditto.
        * bindings/js/JSHTMLSelectElementCustom.cpp:
        (WebCore::JSHTMLSelectElement::indexSetter): Ditto.
        * bindings/js/JSHistoryCustom.cpp:
        (WebCore::JSHistory::customPut): Ditto.
        * bindings/js/JSLocation.cpp:
        (WebCore::JSLocation::put): Ditto.
        * bindings/js/JSLocation.h: Ditto.
        * bindings/js/JSXMLHttpRequest.cpp:
        (WebCore::JSXMLHttpRequest::put): Ditto.
        (WebCore::JSXMLHttpRequest::putValueProperty): Ditto.
        * bindings/js/JSXMLHttpRequest.h: Ditto.

        * bindings/js/kjs_dom.cpp:
        (WebCore::getRuntimeObject): Changed return type to JSObject*.
        * bindings/js/kjs_dom.h: Ditto.

        * bindings/js/kjs_events.cpp:
        (WebCore::JSClipboard::put): Removed attributes argument.
        (WebCore::JSClipboard::putValueProperty): Ditto.
        * bindings/js/kjs_events.h: Ditto.

        * bindings/js/kjs_html.cpp:
        (WebCore::runtimeObjectGetter): Updated for change to getRuntimeObject to
        return a JSObject. Used early exit idiom.
        (WebCore::runtimeObjectPropertyGetter): Ditto.
        (WebCore::runtimeObjectCustomGetOwnPropertySlot): Ditto.
        (WebCore::runtimeObjectCustomPut): Use hasProperty to check for properties
        that we should put with the property syntax instead of canPut.
        (WebCore::runtimeObjectImplementsCall): Ditto.
        (WebCore::runtimeObjectCallAsFunction): Ditto.
        * bindings/js/kjs_html.h: Removed attributes argument to runtimeObjectCustomPut.

        * bindings/js/kjs_window.cpp:
        (KJS::Window::put): Removed attributes argument.
        * bindings/js/kjs_window.h: Ditto.

        * bindings/scripts/CodeGeneratorJS.pm: Removed attributes argument from put,
        putValueProperty, customPut, and indexSetter.

        * html/CanvasPixelArray.h:
        (WebCore::CanvasPixelArray::set): Added index checking here, as in the get
        function. Before, the checking was done in the JavaScript bindings for set.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30534 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago2008-02-23 Brent Fulgham <bfulgham@gmail.com>
pewtermoose@webkit.org [Sun, 24 Feb 2008 04:15:23 +0000 (04:15 +0000)]
2008-02-23  Brent Fulgham  <bfulgham@gmail.com>

        Reviewed by Adam and Darin.

        http://bugs.webkit.org/show_bug.cgi?id=17442
        Correct the Windows Cairo implementation of GraphicsContext so
        that Windows 'WorldTransform' is kept in sync with the Cairo
        transormations.

        Also corrects an uninitialized variable in the Cairo Windows
        image drag handler.

        WARNING: NO TEST CASES ADDED OR CHANGED

        * platform/graphics/cairo/GraphicsContextCairo.cpp:  Modify
          methods to call GraphicsContextPlatformPrivate calls, just
          as is done for the CoreGraphics version.  These calls are
          nop's for everything but Windows.
        (WebCore::GraphicsContext::savePlatformState):
        (WebCore::GraphicsContext::restorePlatformState):
        (WebCore::GraphicsContext::clip):
        (WebCore::GraphicsContext::drawLineForMisspellingOrBadGrammar):
        (WebCore::GraphicsContext::translate):
        (WebCore::GraphicsContext::concatCTM): Re-enable for Windows
        (WebCore::GraphicsContext::beginTransparencyLayer):
        (WebCore::GraphicsContext::endTransparencyLayer):
        (WebCore::GraphicsContext::rotate):
        (WebCore::GraphicsContext::scale):
        * platform/graphics/cairo/GraphicsContextPlatformPrivateCairo.h:
        * platform/graphics/win/GraphicsContextCGWin.cpp:  Move common
          code for handling WorldTransform calls to parent class.
        * platform/graphics/win/GraphicsContextCairoWin.cpp:  Add common
          code for handling WorldTransform calls.
          (WebCore::GraphicsContextPlatformPrivate::concatCTM):  Change
          implementation so that it only handles WorldTransform.  The
          Cairo update is done in GraphicsContextCairo.cpp
        * platform/graphics/win/GraphicsContextWin.cpp:  Call platform-private
          methods to keep WorldTransform in sync.
        (WebCore::GraphicsContextPlatformPrivate::save):
        (WebCore::GraphicsContextPlatformPrivate::restore):
        (WebCore::GraphicsContextPlatformPrivate::clip):
        (WebCore::GraphicsContextPlatformPrivate::scale):
        (WebCore::GraphicsContextPlatformPrivate::rotate):
        (WebCore::GraphicsContextPlatformPrivate::translate):
        * platform/win/DragImageCairoWin.cpp:
        (WebCore::createDragImageFromImage):  Correct uninitialized value

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30533 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago2008-02-23 Brent Fulgham <bfulgham@gmail.com>
pewtermoose@webkit.org [Sun, 24 Feb 2008 04:10:14 +0000 (04:10 +0000)]
2008-02-23  Brent Fulgham  <bfulgham@gmail.com>

        Not reviewed, build fix.

        * WebIconDatabase.cpp:
        * WebIconDatabase.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30532 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago2008-02-23 Jan Michael Alonzo <jmalonzo@unpluggable.com>
pewtermoose@webkit.org [Sun, 24 Feb 2008 04:05:05 +0000 (04:05 +0000)]
2008-02-23  Jan Michael Alonzo  <jmalonzo@unpluggable.com>

        Not reviewed, Gtk build fix.

        * kjs/testkjs.pro:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30531 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years agoWebCore:
mitz@apple.com [Sun, 24 Feb 2008 04:00:21 +0000 (04:00 +0000)]
WebCore:

        Reviewed by Dave Hyatt.

        - make non-autowrapping text clear floats

        Test: fast/text/whitespace/nowrap-clear-float.html

        * rendering/RenderBlock.cpp:
        (WebCore::RenderBlock::nextFloatBottomBelow): Renamed nearestFloat() to
        this and changed to avoid comparing bottom to 0 in each iteration.
        (WebCore::RenderBlock::getClearDelta): Updated comment for the rename.
        * rendering/RenderBlock.h:
        * rendering/bidi.cpp:
        (WebCore::RenderBlock::fitBelowFloats): Added. Factored out of
        findNextLineBreak() and simplified.
        (WebCore::RenderBlock::findNextLineBreak): Changed to call
        fitBelowFloats(). Fixed the bug by trying to fit below floats in the
        case of non-wrapping text. Removed some redundancy.

LayoutTests:

        Reviewed by Dave Hyatt.

        - test that non-autowrapping text clear floats

        * fast/text/whitespace/nowrap-clear-float.html: Added.
        * platform/mac-leopard/fast/text/whitespace: Added.
        * platform/mac-leopard/fast/text/whitespace/nowrap-clear-float-expected.checksum: Added.
        * platform/mac-leopard/fast/text/whitespace/nowrap-clear-float-expected.png: Added.
        * platform/mac/fast/text/whitespace/nowrap-clear-float-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30530 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago Reviewed by Oliver Hunt.
weinig@apple.com [Sun, 24 Feb 2008 01:35:07 +0000 (01:35 +0000)]
    Reviewed by Oliver Hunt.

        Fix for http://bugs.webkit.org/show_bug.cgi?id=17504
        Speed up DOM lists array subscription syntax by using the fast
        getOwnPropertySlot and set paths

        - 6x speed improvement on Oliver's ImageData put test.

        * bindings/scripts/CodeGeneratorJS.pm: Add fast path for getOwnPropertySlot
        and put when an indexGetter or indexSetter is used.  We can not use the fast
        path if an overridingNameGetter is used as there would be no way to override.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30529 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years agowx build fix after JSImageData.cpp was added.
kevino@webkit.org [Sun, 24 Feb 2008 01:33:49 +0000 (01:33 +0000)]
wx build fix after JSImageData.cpp was added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30528 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago Windows build fix - move ThreadCondition implementation from WebCore to WTF.
ap@webkit.org [Sat, 23 Feb 2008 21:52:07 +0000 (21:52 +0000)]
    Windows build fix - move ThreadCondition implementation from WebCore to WTF.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30527 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago Build fix.
ap@webkit.org [Sat, 23 Feb 2008 20:49:14 +0000 (20:49 +0000)]
    Build fix.

JavaScriptCore:
        Touch some files, hoping that Windows build bot will create JSC headers.

        * kjs/AllInOneFile.cpp:
        * kjs/array_instance.cpp:
        * wtf/HashTable.cpp:

WebKit/gtk:
        * webkit/webkitprivate.cpp:
        (webkit_init): Renamed initializeThreading to initializeThreadingAndMainThread.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30526 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago Qt/Wx build fix - this file was still in a wrong namespace, too.
ap@webkit.org [Sat, 23 Feb 2008 20:29:58 +0000 (20:29 +0000)]
    Qt/Wx build fix - this file was still in a wrong namespace, too.

        * wtf/ThreadingNone.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30525 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago More build fixing - fix mismatched braces.
ap@webkit.org [Sat, 23 Feb 2008 20:21:15 +0000 (20:21 +0000)]
    More build fixing - fix mismatched braces.

        * JavaScriptCore.pri:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30524 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago Wx and Gtk build fixes.
ap@webkit.org [Sat, 23 Feb 2008 20:10:59 +0000 (20:10 +0000)]
    Wx and Gtk build fixes.

        * JavaScriptCore.pri: Don't try to compile ThreadingPthreads.
        * wtf/ThreadingGtk.cpp: Use a correct namespace.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30523 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago2008-02-23 Alexey Proskuryakov <ap@webkit.org>
ap@webkit.org [Sat, 23 Feb 2008 19:50:04 +0000 (19:50 +0000)]
2008-02-23  Alexey Proskuryakov  <ap@webkit.org>

        Reviewed by Darin.

        Move basic threading support from WebCore to WTF.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30522 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years agoFix typo in comment.
ddkilzer@apple.com [Sat, 23 Feb 2008 18:54:14 +0000 (18:54 +0000)]
Fix typo in comment.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30521 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years agoWebCore:
ddkilzer@apple.com [Sat, 23 Feb 2008 17:08:57 +0000 (17:08 +0000)]
WebCore:

        Please clarify licensing for some files
        <http://bugs.webkit.org/show_bug.cgi?id=14970>

        Reviewed by Darin.

        * bindings/objc/WebScriptObject.h: Added Apple BSD-style license.
        * bindings/objc/WebScriptObjectPrivate.h: Ditto.
        * platform/text/mac/ShapeArabic.c: Added ICU license from WebCore/icu/LICENSE.

WebKit/mac:

        Please clarify licensing for some files
        <http://bugs.webkit.org/show_bug.cgi?id=14970>

        Reviewed by Darin.

        * Plugins/WebNetscapeDeprecatedFunctions.c: Updated copyright statement
        and added Apple BSD-style license.
        * Plugins/WebNetscapeDeprecatedFunctions.h: Ditto.

WebKitTools:

        Please clarify licensing for some files
        <http://bugs.webkit.org/show_bug.cgi?id=14970>

        Reviewed by Darin.

        * DumpRenderTree/TestNetscapePlugIn.subproj/PluginObject.cpp: Added
        copyright statement.  Replaced license with newer Apple BSD-style license.
        * DumpRenderTree/TestNetscapePlugIn.subproj/PluginObject.h: Ditto.
        * DumpRenderTree/TestNetscapePlugIn.subproj/TestObject.cpp: Ditto.
        * DumpRenderTree/TestNetscapePlugIn.subproj/TestObject.h: Ditto.
        * DumpRenderTree/TestNetscapePlugIn.subproj/main.cpp: Ditto.
        * DumpRenderTree/win/TestNetscapePlugin/main.c: Ditto.
        * mangleme/LICENSE: Added (LGPL).

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30520 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago - undo an accidental change to a test's expected results from r30431.
mitz@apple.com [Sat, 23 Feb 2008 08:09:08 +0000 (08:09 +0000)]
    - undo an accidental change to a test's expected results from r30431.

        * fast/js/date-DST-time-cusps-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30519 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago Rubber-stamped by Sam Weinig and Anders Carlsson.
mitz@apple.com [Sat, 23 Feb 2008 08:03:11 +0000 (08:03 +0000)]
    Rubber-stamped by Sam Weinig and Anders Carlsson.

        - updated results for <rdar://problem/5714333> Add document.getSelection()

        * fast/dom/Window/window-properties-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30518 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago Reviewed by Darin.
ap@webkit.org [Sat, 23 Feb 2008 07:52:15 +0000 (07:52 +0000)]
    Reviewed by Darin.

        A few tweaks to PCRE-based regex test suite.

        * fast/regex/pcre-test-runner.js: Added support for global regexes. Emulate PCRE test runner
        more precisely by trimming trailing whitespace from some lines.

        * fast/regex/test1-expected.txt:
        * fast/regex/test4-expected.txt:
        * fast/regex/testinput1:
        * fast/regex/testinput4:
        * fast/regex/testoutput1:
        * fast/regex/testoutput4:
        Updated to PCRE 7.6.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30517 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago - update Tiger expected results following r30431
mitz@apple.com [Sat, 23 Feb 2008 07:46:28 +0000 (07:46 +0000)]
    - update Tiger expected results following r30431

        * platform/mac-tiger/svg/W3C-SVG-1.1/text-tselect-02-f-expected.txt:
        * platform/mac-tiger/svg/custom/foreignObject-crash-on-hover-expected.txt:
        * platform/mac-tiger/svg/custom/text-dom-01-f-expected.txt:
        * platform/mac-tiger/svg/custom/text-hit-test-expected.txt:
        * platform/mac-tiger/svg/text/text-align-01-b-expected.txt:
        * platform/mac-tiger/svg/text/text-align-02-b-expected.txt:
        * platform/mac-tiger/svg/text/text-align-03-b-expected.txt:
        * platform/mac-tiger/svg/text/text-align-04-b-expected.txt:
        * platform/mac-tiger/svg/text/text-align-05-b-expected.txt:
        * platform/mac-tiger/svg/text/text-align-06-b-expected.txt:
        * platform/mac-tiger/svg/text/text-altglyph-01-b-expected.txt:
        * platform/mac-tiger/svg/text/text-deco-01-b-expected.txt:
        * platform/mac-tiger/svg/text/text-fonts-01-t-expected.txt:
        * platform/mac-tiger/svg/text/text-fonts-02-t-expected.txt:
        * platform/mac-tiger/svg/text/text-path-01-b-expected.txt:
        * platform/mac-tiger/svg/text/text-spacing-01-b-expected.txt:
        * platform/mac-tiger/svg/text/text-text-01-b-expected.txt:
        * platform/mac-tiger/svg/text/text-text-03-b-expected.txt:
        * platform/mac-tiger/svg/text/text-text-04-t-expected.txt:
        * platform/mac-tiger/svg/text/text-text-05-t-expected.txt:
        * platform/mac-tiger/svg/text/text-text-06-t-expected.txt:
        * platform/mac-tiger/svg/text/text-text-07-t-expected.txt:
        * platform/mac-tiger/svg/text/text-text-08-b-expected.txt:
        * platform/mac-tiger/svg/text/text-tref-01-b-expected.txt:
        * platform/mac-tiger/svg/text/text-tselect-01-b-expected.txt:
        * platform/mac-tiger/svg/text/text-tselect-02-f-expected.txt:
        * platform/mac-tiger/svg/text/text-tspan-01-b-expected.txt:
        * platform/mac-tiger/svg/text/text-ws-01-t-expected.txt:
        * platform/mac-tiger/svg/text/text-ws-02-t-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30516 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years agoWebCore:
justin.garcia@apple.com [Sat, 23 Feb 2008 05:14:55 +0000 (05:14 +0000)]
WebCore:

        Reviewed by Anders Carlsson.

        <rdar://problem/5714333> Add document.getSelection()

        Add a synonym for window.getSelection() that FF supports.  There were requests in
        this bugs dupes to also support the IE only document.selection, but IE's Selection
        object is too different than ours currently for that to be safe (specifically, we
        don't support enough of IE's TextRange methods on our DOM Range object yet). I
        filed <rdar://problem/5761233> to cover that.

        * dom/Document.cpp:
        (WebCore::Document::getSelection): Added.
        * dom/Document.h:
        * dom/Document.idl: Ditto.  JS only since this is about web compatibility.

LayoutTests:

        Reviewed by Anders.

        <rdar://problem/5714333> Add document.getSelection()

        * editing/selection/5714333-expected.txt: Added.
        * editing/selection/5714333.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30515 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years agoWX build fixes
oliver@apple.com [Sat, 23 Feb 2008 05:09:39 +0000 (05:09 +0000)]
WX build fixes

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30514 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years agoRoll out r30500 as it breaks non-Windows use of Cairo.
mrowe@apple.com [Sat, 23 Feb 2008 04:18:40 +0000 (04:18 +0000)]
Roll out r30500 as it breaks non-Windows use of Cairo.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30513 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years agoOnce more with the cairo/qt/wx build fixing
oliver@apple.com [Sat, 23 Feb 2008 03:31:12 +0000 (03:31 +0000)]
Once more with the cairo/qt/wx build fixing

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30512 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years agoWindows build fix. Add missing include of DeprecatedCString.h and sort existing...
mrowe@apple.com [Sat, 23 Feb 2008 03:09:41 +0000 (03:09 +0000)]
Windows build fix.  Add missing include of DeprecatedCString.h and sort existing includes.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30511 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years agoSpeculative Windows build fix. Add missing include of DeprecatedCString.h and sort...
mrowe@apple.com [Sat, 23 Feb 2008 03:03:41 +0000 (03:03 +0000)]
Speculative Windows build fix.  Add missing include of DeprecatedCString.h and sort existing includes.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30510 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years agoBuild fix for windows, wx and qt
oliver@apple.com [Sat, 23 Feb 2008 02:15:36 +0000 (02:15 +0000)]
Build fix for windows, wx and qt

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30509 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago Reviewed by NOBODY (build fix).
oliver@apple.com [Sat, 23 Feb 2008 02:04:49 +0000 (02:04 +0000)]
    Reviewed by NOBODY (build fix).

        Unbreak windows vcproj

        * WebCore.vcproj/WebCore.vcproj:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30508 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago Fix Qt build.
weinig@apple.com [Sat, 23 Feb 2008 01:27:17 +0000 (01:27 +0000)]
    Fix Qt build.

        * Api/qwebhistoryinterface.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30507 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years agoBug 17486: Support HTML5 Canvas.getImageData API
oliver@apple.com [Sat, 23 Feb 2008 01:16:16 +0000 (01:16 +0000)]
Bug 17486: Support HTML5 Canvas.getImageData API
Support Canvas.getImageData and Canvas.createImageData

Reviewed by Sam W.

This patch adds support for all the pixel reading portions
of the HTML5 Canvas spec.  There are two new types ImageData
and CanvasPixelArray which are used to provide the HTML5
ImageData object, and the required semantics for assignment
to the ImageData data array.

We only implement the CG version of ImageBuffer::getImageData,
but the logic is null safe, so this will not introduce any
crashes into other platforms, unfortunately it will result in
JS Object detection "lying" on non-CG platforms.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30506 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years agoWebCore:
weinig@apple.com [Sat, 23 Feb 2008 00:55:20 +0000 (00:55 +0000)]
WebCore:

        Rubber-stamped by Adam Roben.

        Rid the project of the Devil known as DeprecatedString!

        * GNUmakefile.am:
        * WebCore.base.exp:
        * WebCore.pro:
        * WebCore.vcproj/WebCore.vcproj:
        * WebCore.xcodeproj/project.pbxproj:
        * WebCoreSources.bkl:
        * dom/Node.h:
        * html/HTMLFormElement.cpp:
        * platform/DeprecatedString.cpp: Removed.
        * platform/DeprecatedString.h: Removed.
        * platform/graphics/qt/IconQt.cpp:
        * platform/graphics/win/MediaPlayerPrivateQuickTimeWin.cpp:
        * platform/gtk/CookieJarGtk.cpp:
        * platform/gtk/CursorGtk.cpp:
        * platform/gtk/KeyEventGtk.cpp:
        * platform/mac/DeprecatedStringMac.mm: Removed.
        * platform/network/qt/ResourceHandleQt.cpp:
        * platform/network/win/CookieJarWin.cpp:
        (WebCore::cookies):
        * platform/qt/ClipboardQt.cpp:
        * platform/qt/CookieJarQt.cpp:
        * platform/qt/CursorQt.cpp:
        * platform/symbian/DeprecatedStringSymbian.cpp: Removed.
        * platform/text/AtomicString.cpp:
        * platform/text/AtomicString.h:
        * platform/text/PlatformString.h:
        * platform/text/String.cpp:
        * platform/text/StringImpl.cpp:
        * platform/text/qt/StringQt.cpp:
        (WebCore::String::String):
        * platform/text/wx/StringWx.cpp:
        * platform/win/BString.cpp:
        * platform/win/BString.h:
        * platform/win/PasteboardWin.cpp:

WebKit/qt:

        Rubber-stamped by Adam Roben.

        Rid the project of the Devil known as DeprecatedString!

        * Api/qwebhistory.cpp:
        * Api/qwebhistoryinterface.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30501 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago2008-02-22 Brent Fulgham <bfulgham@gmail.com>
pewtermoose@webkit.org [Sat, 23 Feb 2008 00:23:43 +0000 (00:23 +0000)]
2008-02-22  Brent Fulgham  <bfulgham@gmail.com>

        Reviewed by Adam Roben.

        http://bugs.webkit.org/show_bug.cgi?id=17442
        Correct the Windows Cairo implementation of GraphicsContext so
        that Windows 'WorldTransform' is kept in sync with the Cairo
        transformations.

        Also corrects an uninitialized variable in the Cairo Windows
        image drag handler.

        * platform/graphics/cairo/GraphicsContextCairo.cpp:  Modify
          methods to call GraphicsContextPlatformPrivate calls, just
          as is done for the CoreGraphics version.  These calls are
          nop's for everything but Windows.
        (WebCore::GraphicsContext::savePlatformState):
        (WebCore::GraphicsContext::restorePlatformState):
        (WebCore::GraphicsContext::clip):
        (WebCore::GraphicsContext::drawLineForMisspellingOrBadGrammar):
        (WebCore::GraphicsContext::translate):
        (WebCore::GraphicsContext::concatCTM): Re-enable for Windows
        (WebCore::GraphicsContext::beginTransparencyLayer):
        (WebCore::GraphicsContext::endTransparencyLayer):
        (WebCore::GraphicsContext::rotate):
        (WebCore::GraphicsContext::scale):
        * platform/graphics/cairo/GraphicsContextPlatformPrivateCairo.h:
        * platform/graphics/win/GraphicsContextCGWin.cpp:  Move common
          code for handling WorldTransform calls to parent class.
        * platform/graphics/win/GraphicsContextCairoWin.cpp:  Add common
          code for handling WorldTransform calls.
          (WebCore::GraphicsContextPlatformPrivate::concatCTM):  Change
          implementation so that it only handles WorldTransform.  The
          Cairo update is done in GraphicsContextCairo.cpp
        * platform/graphics/win/GraphicsContextWin.cpp:  Call platform-private
          methods to keep WorldTransform in sync.
        (WebCore::GraphicsContextPlatformPrivate::save):
        (WebCore::GraphicsContextPlatformPrivate::restore):
        (WebCore::GraphicsContextPlatformPrivate::clip):
        (WebCore::GraphicsContextPlatformPrivate::scale):
        (WebCore::GraphicsContextPlatformPrivate::rotate):
        (WebCore::GraphicsContextPlatformPrivate::translate):
        * platform/win/DragImageCairoWin.cpp:
        (WebCore::createDragImageFromImage):  Correct uninitialized value

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30500 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago Reviewed by Geoff Garen.
weinig@apple.com [Sat, 23 Feb 2008 00:16:05 +0000 (00:16 +0000)]
    Reviewed by Geoff Garen.

        - Remove use of DeprecatedString in CSSStyleSelector.

        * css/CSSStyleSelector.cpp:
        (WebCore::findHash): Removed.  Use find instead.
        (WebCore::findSlashDotDotSlash): Changed to take a UChar* and a length.
        (WebCore::findSlashSlash): Ditto.
        (WebCore::findSlashDotSlash): Ditto.
        (WebCore::containsColonSlashSlash): Ditto.
        (WebCore::cleanPath): Change to operate on a String.
        (WebCore::checkPseudoState): Changed to use a Vector as a buffer.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30499 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago Reviewed by Sam Weinig.
ggaren@apple.com [Fri, 22 Feb 2008 23:16:57 +0000 (23:16 +0000)]
    Reviewed by Sam Weinig.

        Fixed <rdar://problem/5057509> Repro leak of JSXMLHttpRequest and
        associated objects @ www.viamichelin.it, which was probably an underlying
        cause of <rdar://problem/5744037> Gmail out of memory (17455)

        If SubresourceLoader::create returned NULL, we would ref() / gcProtect()
        the XMLHttpRequest but think we hadn't, therefore never
        calling deref() / gcUnprotect().

        This could happen at gmail.com, since gmail.com attempts to send
        XMLHttpRequests from unload handlers in order to gather usage statistics.
        (According to comments in the code, SubresourceLoader::create returns
        NULL when called from an unload handler.)

        The solution is to ref() / gcProtect() only if SubresourceLoader::create
        doesn't return NULL. This make sense, since we only need to protect the
        request as long as it has an outstanding network transaction.

        * xml/XMLHttpRequest.cpp:
        (WebCore::XMLHttpRequest::XMLHttpRequest):
        (WebCore::XMLHttpRequest::send):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30493 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years agoJavaScriptCore:
ggaren@apple.com [Fri, 22 Feb 2008 23:16:40 +0000 (23:16 +0000)]
JavaScriptCore:

        Reviewed by Sam Weinig.

        Partial fix for <rdar://problem/5744037> Gmail out of memory (17455)

        I'm removing KJS_MEM_LIMIT for the following reasons:

        - We have a few reports of KJS_MEM_LIMIT breaking important web
        applications, like GMail and Google Reader. (For example, if you
        simply open 12 GMail tabs, tab #12 will hit the limit.)

        - Firefox has no discernable JS object count limit, so any limit, even
        a large one, is a potential compatibility problem.

        - KJS_MEM_LIMIT does not protect against malicious memory allocation,
        since there are many ways to maliciously allocate memory without
        increasing the JS object count.

        - KJS_MEM_LIMIT is already mostly broken, since it only aborts the
        script that breaches the limit, not any subsequent scripts.

        - We've never gotten bug reports about websites that would have
        benefited from an unbroken KJS_MEM_LIMIT. The initial check-in of
        KJS_MEM_LIMIT (KJS revision 80061) doesn't mention a website that
        needed it.

        - Any website that brings you anywhere close to crashing due to the
        number of live JS objects will almost certainly put up the "slow
        script" dialog at least 20 times beforehand.

        * kjs/collector.cpp:
        (KJS::Collector::collect):
        * kjs/collector.h:
        * kjs/nodes.cpp:
        (KJS::TryNode::execute):

LayoutTests:

        Reviewed by Sam Weinig.

        Removing the test for KJS_MEM_LIMIT, since I removed KJS_MEM_LIMIT.

        * fast/js/out-of-memory-expected.txt: Removed.
        * fast/js/out-of-memory.html: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30492 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago Reviewed, tweaked and landed by Sam.
weinig@apple.com [Fri, 22 Feb 2008 22:30:57 +0000 (22:30 +0000)]
    Reviewed, tweaked and landed by Sam.

        - Make RegularExpression operate on Strings instead of DeprecatedStrings.

        * dom/DOMImplementation.cpp:
        (WebCore::DOMImplementation::isXMLMIMEType): Use string instead of DeprecatedString
        to build up the RegularExpression.
        * page/Frame.cpp:
        (WebCore::createRegExpForLabels): Use String instead of DeprecatedString.
        (WebCore::Frame::searchForLabelsAboveCell): Ditto.
        (WebCore::Frame::searchForLabelsBeforeElement): Ditto.
        (WebCore::Frame::matchLabelsAgainstElement): Ditto.
        * page/Frame.h:
        * page/mac/FrameMac.mm:
        (WebCore::regExpForLabels): Ditto.
        (WebCore::Frame::searchForNSLabelsAboveCell): Ditto.
        (WebCore::Frame::searchForLabelsBeforeElement): Ditto.
        (WebCore::Frame::matchLabelsAgainstElement): Ditto.

        * platform/text/RegularExpression.cpp:
        (WebCore::RegularExpression::Private::Private):
        (WebCore::RegularExpression::Private::compile):
        (WebCore::RegularExpression::RegularExpression):
        (WebCore::RegularExpression::operator=):
        (WebCore::RegularExpression::pattern):
        (WebCore::RegularExpression::match):
        (WebCore::RegularExpression::search):
        (WebCore::RegularExpression::searchRev):
        (WebCore::replace): Added.
        * platform/text/RegularExpression.h:
        Change functions to take Strings as input instead of DeprecatedStrings and
        reduce the complexity of the class by removing unneeded globbing support.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30490 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago2008-02-22 Jon Honeycutt <jhoneycutt@apple.com>
jhoneycutt@apple.com [Fri, 22 Feb 2008 22:15:31 +0000 (22:15 +0000)]
2008-02-22  Jon Honeycutt  <jhoneycutt@apple.com>

        Reviewed by Anders.

        <rdar://problem/5760360> REGRESSION(r30376): Crash loading plugin page
        during stress test (after only 5 min) - null dereference

        Full-frame plug-ins create PluginStream objects without loaders, as the
        PluginView receives the loading callbacks. We were trying to call
        setDefersLoading on these null pointers.

        * plugins/PluginStream.cpp:
        (WebCore::PluginStream::startStream): Add null check.
        (WebCore::PluginStream::destroyStream): Same.
        (WebCore::PluginStream::deliverData): Same.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30489 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago Reviewed and tweaked by Jon Honeycutt. Reviewed and landed by Sam.
weinig@apple.com [Fri, 22 Feb 2008 21:39:27 +0000 (21:39 +0000)]
    Reviewed and tweaked by Jon Honeycutt. Reviewed and landed by Sam.

        - Remove uses of DeprecatedString in Windows plugin code.

        * plugins/win/PluginPackageWin.cpp:
        (WebCore::getVersionInfo): Cleanup formatting.
        (WebCore::PluginPackage::freeLibraryTimerFired): Remove un-needed variable name.
        (WebCore::PluginPackage::storeFileVersion): Move casts.
        (WebCore::PluginPackage::fetchInfo): Use OwnArrayPtr and switch to more efficient
        use of Vectors.
        (WebCore::PluginPackage::load): Fix whitespace.
        (WebCore::PluginPackage::hash): Make the hashCodes const
        * plugins/win/PluginViewWin.cpp:
        (WebCore::makeURL): Use String instead of DeprecatedString.
        (WebCore::parseRFC822HeaderFields): Ditto.
        (WebCore::PluginView::handlePost): Ditto.
        (WebCore::PluginView::status): Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30488 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago Reviewed, tweaked and landed by Sam.
weinig@apple.com [Fri, 22 Feb 2008 21:16:18 +0000 (21:16 +0000)]
    Reviewed, tweaked and landed by Sam.

        - Don't use DeprecatedString in HTMLTokenizer.

        * html/HTMLTokenizer.cpp:
        (WebCore::HTMLTokenizer::finish):
        * platform/text/PlatformString.h:
        (WebCore::find):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30486 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago Move hasWebView to WebFrameLoaderClient
aroben@apple.com [Fri, 22 Feb 2008 20:58:36 +0000 (20:58 +0000)]
     Move hasWebView to WebFrameLoaderClient

         I added an m_webFrame member to WebFrameLoaderClient. This is slightly
         strange because WebFrame inherits from WebFrameLoaderClient, but this
         member will be needed once we remove the inheritance, so we might as
         well prepare for that now.

         Reviewed by Anders.

         * WebCoreSupport/WebFrameLoaderClient.cpp:
         (WebFrameLoaderClient::WebFrameLoaderClient): Changed to take a
         WebFrame* parameter.
         (WebFrameLoaderClient::hasWebView): Moved here from WebFrame.cpp.
         * WebCoreSupport/WebFrameLoaderClient.h: Added an m_webFrame
         parameter.
         * WebFrame.cpp: Removed hasWebView.
         (WebFrame::WebFrame): Updated for WebFrameLoaderClient constructor
         change.
         * WebFrame.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30485 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago Reviewed by Adam.
andersca@apple.com [Fri, 22 Feb 2008 20:05:46 +0000 (20:05 +0000)]
    Reviewed by Adam.

        <rdar://problem/5748781>
        https://bugs.webkit.org/show_bug.cgi?id=17413
        REGRESSION: Latest Nightly doesn't load Java plugin w/Safari 3.1b

        Copy the Java plug-in over to the new location.

        * FindSafari/FindSafari.cpp:
        (_tmain):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30482 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago Start to move FrameLoaderClient methods off WebFrame onto a separate class
aroben@apple.com [Fri, 22 Feb 2008 19:53:56 +0000 (19:53 +0000)]
    Start to move FrameLoaderClient methods off WebFrame onto a separate class

        WebFrame now inherits from a new WebFrameLoaderClient class, which
        will gradually assume all FrameLoaderClient responsibilities. Once
        that process is complete, WebFrame will no longer inherit from
        WebFrameLoaderClient.

        In this first patch, I've only moved createDocumentLoader up to the
        WebFrameLoaderClient class.

        Reviewed by Anders.

        * WebCoreSupport/WebFrameLoaderClient.cpp: Added.
        (WebFrameLoaderClient::WebFrameLoaderClient):
        (WebFrameLoaderClient::~WebFrameLoaderClient):
        (WebFrameLoaderClient::createDocumentLoader): Moved here from
        WebFrame.cpp.
        * WebCoreSupport/WebFrameLoaderClient.h: Added.
        * WebFrame.cpp: Removed createDocumentLoader implementation.
        * WebFrame.h: Changed to inherit from WebFrameLoaderClient.
        * WebKit.vcproj/WebKit.vcproj: Added new files to the project.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30481 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago Reviewed by Adam.
andersca@apple.com [Fri, 22 Feb 2008 19:51:49 +0000 (19:51 +0000)]
    Reviewed by Adam.

        <rdar://problem/5747325> REGRESSION: HTTP layout tests hang

        * DumpRenderTree/win/DumpRenderTree.cpp:
        (runTest):
        Init the URL request with the correct timeout.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30480 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago Move FormValuesPropertyBag into its own files
aroben@apple.com [Fri, 22 Feb 2008 19:07:53 +0000 (19:07 +0000)]
     Move FormValuesPropertyBag into its own files

         Reviewed by Sam.

         * WebCoreSupport/FormValuesPropertyBag.cpp: Added.
         (FormValuesPropertyBag::QueryInterface):
         (FormValuesPropertyBag::AddRef):
         (FormValuesPropertyBag::Release):
         (FormValuesPropertyBag::Read):
         (FormValuesPropertyBag::Write):
         (FormValuesPropertyBag::CountProperties):
         (FormValuesPropertyBag::GetPropertyInfo):
         (FormValuesPropertyBag::LoadObject):
         * WebCoreSupport/FormValuesPropertyBag.h: Added.
         (FormValuesPropertyBag::FormValuesPropertyBag):
         * WebFrame.cpp: Deleted FormValuesPropertyBag code.
         * WebKit.vcproj/WebKit.vcproj: Added new files to the project.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30479 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago Remove some unused WebFrame methods
aroben@apple.com [Fri, 22 Feb 2008 19:07:40 +0000 (19:07 +0000)]
     Remove some unused WebFrame methods

         FrameWinClient was deleted back in r22965, but these methods were
         never deleted.

         Reviewed by Sam.

         * WebFrame.cpp:
         * WebFrame.h: Made one createFrame overload protected, since it's only
         called internally by WebFrame.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30478 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago Move Client implementations into a WebCoreSupport directory
aroben@apple.com [Fri, 22 Feb 2008 19:06:18 +0000 (19:06 +0000)]
     Move Client implementations into a WebCoreSupport directory

         Reviewed by Anders.

         * WebKit.vcproj/WebKit.vcproj: Updated file paths, and added
         WebCoreSupport directory to the include path.
         * WebCoreSupport/WebChromeClient.cpp: Renamed from WebKit/win/WebChromeClient.cpp.
         * WebCoreSupport/WebChromeClient.h: Renamed from WebKit/win/WebChromeClient.h.
         * WebCoreSupport/WebContextMenuClient.cpp: Renamed from WebKit/win/WebContextMenuClient.cpp.
         * WebCoreSupport/WebContextMenuClient.h: Renamed from WebKit/win/WebContextMenuClient.h.
         * WebCoreSupport/WebDragClient.cpp: Renamed from WebKit/win/WebDragClient.cpp.
         * WebCoreSupport/WebDragClient.h: Renamed from WebKit/win/WebDragClient.h.
         * WebCoreSupport/WebEditorClient.cpp: Renamed from WebKit/win/WebEditorClient.cpp.
         * WebCoreSupport/WebEditorClient.h: Renamed from WebKit/win/WebEditorClient.h.
         * WebCoreSupport/WebInspectorClient.cpp: Renamed from WebKit/win/WebInspectorClient.cpp.
         * WebCoreSupport/WebInspectorClient.h: Renamed from WebKit/win/WebInspectorClient.h.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30477 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago Reviewed by Adam Roben
sullivan@apple.com [Fri, 22 Feb 2008 18:32:12 +0000 (18:32 +0000)]
    Reviewed by Adam Roben

        Reverted the changed from yesterday to add pasteAndMatchStyle:, as the existing
        pasteAsPlainText: has the same behavior.

        * WebView/WebHTMLView.mm:
        (-[WebHTMLView _pasteWithPasteboard:allowPlainText:]):
        (-[WebHTMLView readSelectionFromPasteboard:]):
        (-[WebHTMLView validateUserInterfaceItemWithoutDelegate:]):
        (-[WebHTMLView pasteAsRichText:]):
        (-[WebHTMLView paste:]):
        * WebView/WebView.mm:
        * WebView/WebViewPrivate.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30476 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago<rdar://problem/5759327> REGRESSION: while(NaN) acts like while(true)
oliver@apple.com [Fri, 22 Feb 2008 10:27:12 +0000 (10:27 +0000)]
<rdar://problem/5759327> REGRESSION: while(NaN) acts like while(true)

Reviewed by Alexey P.

Fix yet another case where we incorrectly relied on implicit double
to bool coercion.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30475 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years agoAdd null check to fix crash in media query code I checked in.
hyatt@apple.com [Fri, 22 Feb 2008 07:04:16 +0000 (07:04 +0000)]
Add null check to fix crash in media query code I checked in.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30474 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago2008-02-21 Alp Toker <alp@atoker.com>
alp@webkit.org [Fri, 22 Feb 2008 06:10:42 +0000 (06:10 +0000)]
2008-02-21  Alp Toker  <alp@atoker.com>

        Fix a configure script typo.

        * configure.ac:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30473 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years agoWebCore:
weinig@apple.com [Fri, 22 Feb 2008 03:35:40 +0000 (03:35 +0000)]
WebCore:

        Reviewed by Anders Carlsson.

        Fix for <rdar://problem/5757946>

        - Parse URLs before checking whether they are javascript: urls
          (which require security checks).

        * bindings/js/JSAttrCustom.cpp:
        (WebCore::JSAttr::setValue):
        * bindings/js/JSElementCustom.cpp:
        (WebCore::allowSettingSrcToJavascriptURL):
        * bindings/js/JSHTMLFrameElementCustom.cpp:
        (WebCore::allowSettingJavascriptURL):
        * bindings/js/JSHTMLIFrameElementCustom.cpp:
        (WebCore::JSHTMLIFrameElement::setSrc):

LayoutTests:

        Reviewed by Anders Carlsson.

        Update tests and results for <rdar://problem/5757946>

        * http/tests/security/javascriptURL/javascriptURL-execution-context-frame-location-htmldom-expected.txt:
        * http/tests/security/javascriptURL/javascriptURL-execution-context-frame-location-htmldom.html:
        * http/tests/security/javascriptURL/javascriptURL-execution-context-frame-src-getAttribute-value-expected.txt:
        * http/tests/security/javascriptURL/javascriptURL-execution-context-frame-src-getAttribute-value.html:
        * http/tests/security/javascriptURL/javascriptURL-execution-context-frame-src-htmldom-expected.txt:
        * http/tests/security/javascriptURL/javascriptURL-execution-context-frame-src-htmldom.html:
        * http/tests/security/javascriptURL/javascriptURL-execution-context-frame-src-setAttribute-expected.txt:
        * http/tests/security/javascriptURL/javascriptURL-execution-context-frame-src-setAttribute.html:
        * http/tests/security/javascriptURL/javascriptURL-execution-context-frame-src-setAttributeNS-expected.txt:
        * http/tests/security/javascriptURL/javascriptURL-execution-context-frame-src-setAttributeNS.html:
        * http/tests/security/javascriptURL/javascriptURL-execution-context-frame-src-setAttributeNode-expected.txt:
        * http/tests/security/javascriptURL/javascriptURL-execution-context-frame-src-setAttributeNode.html:
        * http/tests/security/javascriptURL/javascriptURL-execution-context-frame-src-setAttributeNodeNS-expected.txt:
        * http/tests/security/javascriptURL/javascriptURL-execution-context-frame-src-setAttributeNodeNS.html:
        * http/tests/security/javascriptURL/javascriptURL-execution-context-iframe-src-getAttribute-value-expected.txt:
        * http/tests/security/javascriptURL/javascriptURL-execution-context-iframe-src-getAttribute-value.html:
        * http/tests/security/javascriptURL/javascriptURL-execution-context-iframe-src-htmldom-expected.txt:
        * http/tests/security/javascriptURL/javascriptURL-execution-context-iframe-src-htmldom.html:
        * http/tests/security/javascriptURL/javascriptURL-execution-context-iframe-src-setAttribute-expected.txt:
        * http/tests/security/javascriptURL/javascriptURL-execution-context-iframe-src-setAttribute.html:
        * http/tests/security/javascriptURL/javascriptURL-execution-context-iframe-src-setAttributeNS-expected.txt:
        * http/tests/security/javascriptURL/javascriptURL-execution-context-iframe-src-setAttributeNS.html:
        * http/tests/security/javascriptURL/javascriptURL-execution-context-iframe-src-setAttributeNode-expected.txt:
        * http/tests/security/javascriptURL/javascriptURL-execution-context-iframe-src-setAttributeNode.html:
        * http/tests/security/javascriptURL/javascriptURL-execution-context-iframe-src-setAttributeNodeNS-expected.txt:
        * http/tests/security/javascriptURL/javascriptURL-execution-context-iframe-src-setAttributeNodeNS.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30472 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago <rdar://problem/5757873> Buffer overrun in DeprecatedCString::find() in WebCore
adachan@apple.com [Fri, 22 Feb 2008 01:07:35 +0000 (01:07 +0000)]
    <rdar://problem/5757873> Buffer overrun in DeprecatedCString::find() in WebCore
        We could get a buffer overrun in DeprecatedCString::find() if the end of the
        string matches a beginning portion of the substring, for example, if string is
        "a" but the substring is "ab".
        The code as is also will not match things correctly under certain situations
        since the inner while loop increments the index. For example, we wouldn't find
        a match if the string is "aab..." and the substring is "ab".  Changed the
        inner while loop to increment a temporary index into str.

        Test: fast/loader/charset-parse.html

        Reviewed by Dan Berstein.

        * platform/DeprecatedCString.cpp:
        (WebCore::DeprecatedCString::find):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30468 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years agoAdd test case for viewport media queries.
hyatt@apple.com [Fri, 22 Feb 2008 00:15:12 +0000 (00:15 +0000)]
Add test case for viewport media queries.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30467 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago Fix for bug 17301. CSS media queries need to use the correct viewport
hyatt@apple.com [Fri, 22 Feb 2008 00:14:47 +0000 (00:14 +0000)]
    Fix for bug 17301.  CSS media queries need to use the correct viewport
        when contained in documents inside iframes (rather than always using the
        top-level document's viewport).  CSS media queries based on the viewport
        also needed to be dynamic and update as you resize the window (this is
        a HOT feature). :)

        This patch gets Acid3 up to 86/100 with 3 colored boxes filled in.

        Reviewed by olliej

        Added fast/media/viewport-media-query.html

        * css/CSSStyleSelector.cpp:
        (WebCore::CSSStyleSelector::CSSStyleSelector):
        (WebCore::CSSStyleSelector::addViewportDependentMediaQueryResult):
        (WebCore::CSSStyleSelector::affectedByViewportChange):
        * css/CSSStyleSelector.h:
        (WebCore::MediaQueryResult::MediaQueryResult):
        * css/MediaQueryEvaluator.cpp:
        (WebCore::MediaQueryEvaluator::MediaQueryEvaluator):
        (WebCore::MediaQueryEvaluator):
        (WebCore::MediaQueryEvaluator::eval):
        (WebCore::colorMediaFeatureEval):
        (WebCore::monochromeMediaFeatureEval):
        (WebCore::device_aspect_ratioMediaFeatureEval):
        (WebCore::device_pixel_ratioMediaFeatureEval):
        (WebCore::gridMediaFeatureEval):
        (WebCore::device_heightMediaFeatureEval):
        (WebCore::device_widthMediaFeatureEval):
        (WebCore::heightMediaFeatureEval):
        (WebCore::widthMediaFeatureEval):
        (WebCore::min_colorMediaFeatureEval):
        (WebCore::max_colorMediaFeatureEval):
        (WebCore::min_monochromeMediaFeatureEval):
        (WebCore::max_monochromeMediaFeatureEval):
        (WebCore::min_device_aspect_ratioMediaFeatureEval):
        (WebCore::max_device_aspect_ratioMediaFeatureEval):
        (WebCore::min_device_pixel_ratioMediaFeatureEval):
        (WebCore::max_device_pixel_ratioMediaFeatureEval):
        (WebCore::min_heightMediaFeatureEval):
        (WebCore::max_heightMediaFeatureEval):
        (WebCore::min_widthMediaFeatureEval):
        (WebCore::max_widthMediaFeatureEval):
        (WebCore::min_device_heightMediaFeatureEval):
        (WebCore::max_device_heightMediaFeatureEval):
        (WebCore::min_device_widthMediaFeatureEval):
        (WebCore::max_device_widthMediaFeatureEval):
        * css/MediaQueryEvaluator.h:
        * css/MediaQueryExp.cpp:
        (WebCore::MediaQueryExp::~MediaQueryExp):
        * css/MediaQueryExp.h:
        (WebCore::MediaQueryExp::value):
        (WebCore::MediaQueryExp::isViewportDependent):
        * html/HTMLMediaElement.cpp:
        (WebCore::HTMLMediaElement::pickMedia):
        * page/FrameView.cpp:
        (WebCore::FrameView::layout):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30466 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years agoWebCore:
andersca@apple.com [Thu, 21 Feb 2008 21:23:08 +0000 (21:23 +0000)]
WebCore:

        Reviewed by Sam.

        Make more classes start out with a refcount of 1.

        * dom/QualifiedName.cpp:
        (WebCore::QNameComponentsTranslator::translate):
        (WebCore::QualifiedName::QualifiedName):
        * dom/QualifiedName.h:
        (WebCore::QualifiedName::QualifiedNameImpl::create):
        (WebCore::QualifiedName::QualifiedNameImpl::QualifiedNameImpl):
        * history/BackForwardList.cpp:
        (WebCore::BackForwardList::BackForwardList):
        * history/BackForwardList.h:
        (WebCore::BackForwardList::create):
        * page/Page.cpp:
        (WebCore::Page::Page):
        * platform/text/CString.cpp:
        (WebCore::CString::init):
        (WebCore::CString::newUninitialized):
        (WebCore::CString::copyBufferIfNeeded):
        * platform/text/CString.h:
        (WebCore::CStringBuffer::create):
        (WebCore::CStringBuffer::CStringBuffer):

WebKit/mac:

        Reviewed by Sam.

        Use BackForwardList::create instead.

        * History/WebBackForwardList.mm:
        (-[WebBackForwardList init]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30465 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years agowx build fix for the domString() -> string() rename in r30443.
kevino@webkit.org [Thu, 21 Feb 2008 21:21:23 +0000 (21:21 +0000)]
wx build fix for the domString() -> string() rename in r30443.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30464 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago Reviewed by Jessica Kahn
sullivan@apple.com [Thu, 21 Feb 2008 20:35:16 +0000 (20:35 +0000)]
    Reviewed by Jessica Kahn

        support for pasteAndMatchStyle: command (see <rdar://problem/5723952>)

        * WebView/WebHTMLView.mm:
        (-[WebHTMLView _pasteWithPasteboard:allowPlainText:matchStyle:]):
        added matchStyle parameter, passed along to bridge (formerly always passed NO to bridge)
        (-[WebHTMLView readSelectionFromPasteboard:]):
        pass NO for new matchStyle parameter to match old behavior
        (-[WebHTMLView validateUserInterfaceItemWithoutDelegate:]):
        validate pasteAndMatchStyle the same way as pasteAsRichText
        (-[WebHTMLView pasteAndMatchStyle:]):
        just like pasteAsRichText but passes YES for matchStyle
        (-[WebHTMLView pasteAsRichText:]):
        pass NO for new matchStyle parameter to match old behavior
        (-[WebHTMLView paste:]):
        ditto

        * WebView/WebView.mm:
        added macro(pasteAndMatchStyle)

        * WebView/WebViewPrivate.h:
        added pasteAndMatchStyle: to WebViewEditingActionsPendingPublic category

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30463 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago2008-02-21 Alp Toker <alp@atoker.com>
alp@webkit.org [Thu, 21 Feb 2008 20:16:43 +0000 (20:16 +0000)]
2008-02-21  Alp Toker  <alp@atoker.com>

        GTK+/autotools SVG experimental build fix

        Don't enable SVG filters in --enable-svg-experimental.

        This feature isn't supported at all yet. Developers can enable it
        explicitly by passing --enable-svg-filters if needed.

        * configure.ac:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30462 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago Reviewed by Sam Weinig.
antti@apple.com [Thu, 21 Feb 2008 20:10:00 +0000 (20:10 +0000)]
    Reviewed by Sam Weinig.

        <rdar://problem/5753789>
        REGRESSION: 1.5% -2% Sunspider regression from r30009 (ebay photo upload hang)

        Ensure all versions of allowsAccessFrom are inlined to single functions.
        This is a 2% win in browser hosted Sunspider.

        * bindings/js/kjs_window.cpp:
        (KJS::Window::allowsAccessFrom):
        (KJS::Window::allowsAccessFromPrivate):
        * bindings/js/kjs_window.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30461 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago Reviewed by Sam Weinig.
mitz@apple.com [Thu, 21 Feb 2008 19:59:40 +0000 (19:59 +0000)]
    Reviewed by Sam Weinig.

        - minor cleanup

        * rendering/bidi.cpp:
        (WebCore::bidiNext): Removed redundant isBR() check -- isText() returns
        true for RenderLineBreak.
        (WebCore::bidiFirst): Ditto.
        (WebCore::shouldSkipWhitespaceAfterStartObject):
        (WebCore::RenderBlock::findNextLineBreak):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30460 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago2008-02-21 Mike Auty <mike.auty@gmail.com>
alp@webkit.org [Thu, 21 Feb 2008 19:43:44 +0000 (19:43 +0000)]
2008-02-21  Mike Auty  <mike.auty@gmail.com>

        Reviewed by Alp Toker.

        http://bugs.webkit.org/show_bug.cgi?id=17445
        [GTK] WebKit doesn't compile with LDFLAGS="-Wl,--as-needed"

        The GNUmakefile.am files make use of the LDFLAGS variable to include library
        additions such as -ljpeg etc.  Unfortunately, if these inclusions aren't made
        in LIBADD/LDADD variables, then they are mis-ordered during the linking.

        The as-needed flag discards libraries whose functions have not been needed by
        earlier libraries, which therefore makes the ordering important.

        This moves all -l library inclusion statements from LDFLAGS variables to
        LIBADD/LDADD variables.

        * GNUmakefile.am:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30459 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years agoWebCore:
ggaren@apple.com [Thu, 21 Feb 2008 18:09:48 +0000 (18:09 +0000)]
WebCore:

        Reviewed by David Harrison.

        Fixed <rdar://problem/5756125> REGRESSION: A crash occurs at
        WebCore::Frame::scriptProxy() when completing a search with Package Tracker widget

        Test: fast/dom/script-element-without-frame-crash.html

        * html/HTMLTokenizer.cpp:
        (WebCore::HTMLTokenizer::parseTag): Added back a NULL check that was
        accidentally removed in r30325.

LayoutTests:

        Reviewed by David Harrison.

        Test for <rdar://problem/5756125> REGRESSION: A crash occurs at
        WebCore::Frame::scriptProxy() when completing a search with Package Tracker widget

        * fast/dom/script-element-without-frame-crash-expected.txt: Added.
        * fast/dom/script-element-without-frame-crash.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30458 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago2008-02-21 Rodney Dawes <dobey@wayofthemonkey.com>
pewtermoose@webkit.org [Thu, 21 Feb 2008 17:30:01 +0000 (17:30 +0000)]
2008-02-21  Rodney Dawes  <dobey@wayofthemonkey.com>

        GTK+ build fix. s/domString()/string()/

        * platform/graphics/gtk/FontPlatformDataGtk.cpp:
        (FontPlatformData::FontPlatformData):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30457 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years ago Rubber-stamped by Dan Bernstein.
weinig@apple.com [Thu, 21 Feb 2008 07:54:57 +0000 (07:54 +0000)]
    Rubber-stamped by Dan Bernstein.

        Gratuitous change to this test to use the instanceof operator
        instead of the isPrototypeOf method.

        * fast/tokenizer/doctype-search-reset.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30456 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years agoWebCore:
mitz@apple.com [Thu, 21 Feb 2008 07:38:13 +0000 (07:38 +0000)]
WebCore:

        Reviewed by Dave Hyatt.

        - fix http://bugs.webkit.org/show_bug.cgi?id=17465
          REGRESSION: <DIV> tokenized into Div if still searching for DOCTYPE

        Test: fast/tokenizer/doctype-search-reset.html

        * html/HTMLTokenizer.cpp:
        (WebCore::HTMLTokenizer::parseTag):

LayoutTests:

        Reviewed by Dave Hyatt.

        - test for http://bugs.webkit.org/show_bug.cgi?id=17465
          REGRESSION: <DIV> tokenized into Div if still searching for DOCTYPE

        * fast/tokenizer/doctype-search-reset-expected.txt: Added.
        * fast/tokenizer/doctype-search-reset.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30455 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years agoWebCore:
mitz@apple.com [Thu, 21 Feb 2008 05:00:10 +0000 (05:00 +0000)]
WebCore:

        Reviewed by Dave Hyatt.

        - fix http://bugs.webkit.org/show_bug.cgi?id=17464
          REGRESSION: Crash in RenderBlock::findNextLineBreak reading r30444 commit email in GMail

        Test: fast/text/wbr-in-pre-crash.html

        * rendering/bidi.cpp:
        (WebCore::RenderBlock::findNextLineBreak):

LayoutTests:

        Reviewed by Dave Hyatt.

        - test for http://bugs.webkit.org/show_bug.cgi?id=17464
          REGRESSION: Crash in RenderBlock::findNextLineBreak reading r30444 commit email in GMail

        * fast/text/wbr-in-pre-crash.html: Added.
        * platform/mac-leopard/fast/text/wbr-in-pre-crash-expected.checksum: Added.
        * platform/mac-leopard/fast/text/wbr-in-pre-crash-expected.png: Added.
        * platform/mac/fast/text/wbr-in-pre-crash-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30454 268f45cc-cd09-0410-ab3c-d52691b4dbfc

13 years agoJavaScriptCore:
weinig@apple.com [Thu, 21 Feb 2008 04:50:47 +0000 (04:50 +0000)]
JavaScriptCore:

        Reviewed by Darin.

        Fix for Bug 16753: date set methods with no args should result in NaN (Acid3 bug)
        The set values result in NaN now when called with no args, NaN or +/- inf values.
        The setYear, setFullYear and setUTCFullYear methods used on NaN dates work as
        descripted in the standard.

        * kjs/date_object.cpp:
        (KJS::fillStructuresUsingTimeArgs):
        (KJS::fillStructuresUsingDateArgs):
        (KJS::setNewValueFromTimeArgs):
        (KJS::setNewValueFromDateArgs):
        (KJS::dateProtoFuncSetYear):

LayoutTests:

        Reviewed by Darin.

        - test for Bug 16753: date set methods with no args should result in NaN (Acid3 bug)

        * fast/js/date-set-to-nan-expected.txt: Added.
        * fast/js/date-set-to-nan.html: Added.
        * fast/js/resources/date-set-to-nan.js: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@30453 268f45cc-cd09-0410-ab3c-d52691b4dbfc