15 years agoThe http loader should only ask for http cookies. (not, for example, data: cookies)
staikos [Fri, 25 May 2007 15:37:43 +0000 (15:37 +0000)]
The http loader should only ask for http cookies. (not, for example, data: cookies)

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21754 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoFor the keyDown binding use QStringList instead of QList<QString>. The former is...
hausmann [Fri, 25 May 2007 13:37:22 +0000 (13:37 +0000)]
For the keyDown binding use QStringList instead of QList<QString>. The former is a default
registered metatype. Also print out critical messages in dumprendertree's qt message handler.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21752 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoFix crash in Qt JavaScript bindings when the arguments used on the Qt side are not
hausmann [Fri, 25 May 2007 13:37:09 +0000 (13:37 +0000)]
Fix crash in Qt JavaScript bindings when the arguments used on the Qt side are not
registered with QMetaType.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21751 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by Darin Adler.
ggaren [Fri, 25 May 2007 09:46:53 +0000 (09:46 +0000)]
    Reviewed by Darin Adler.

        Some cleanup to Cache while Fixing <rdar://problem/5055182> The page cache
        has no global cap.

        * loader/Cache.cpp: Removed destructor to help identify this class as a

        * loader/Cache.h: Made constructor and destructor private to match the
        singleton model of PageCache.

        Put structs related to Cache inside its declaration to fix some namespace
        conflicts I had while writing my patch. (They don't exist anymore, but
        I figured this might help future development.)

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21750 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by Darin Adler.
ggaren [Fri, 25 May 2007 09:13:32 +0000 (09:13 +0000)]
    Reviewed by Darin Adler.

        Fixed a bug with Settings where a number of different objects held bare
        Settings* data members, even though the Page owns the Settings object and
        deletes it upon destruction.

        Added NULL checks since the Page can be deleted before the frame/document.

        * page/Settings.cpp: Removed reliance on setNeedsReapplyStylesForSettingsChange.

        * css/cssstyleselector.cpp: CSSStyleSelector was the worst offender. It
        held a number of back pointers that were not guaranteed. I removed the
        Settings*, Frame*, and FrameView* back pointers, but others could probably
        go, too. I replaced them with a single Document* pointer, which is guaranteed
        because the Document owns the CSSStyleSelector.

        * page/Page.cpp: Removed setNeedsReapplyStylesForSettingsChange -- the
        Settings object is now a proper subcontroller of Page, so it doesn't need
        to use a global table to find its page anymore. Also, this function
        confusingly implied that the Frames in a page can have heterogeneous settings,
        which is not true.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21749 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by Geoff.
bdakin [Fri, 25 May 2007 07:42:06 +0000 (07:42 +0000)]
    Reviewed by Geoff.

        Fix for <rdar://problem/5023545> QuickBooks Pro 2007:hang/crash
        after closing QuickBooks Tutorial Center with Leopard9A377

        We can hit a race condition where drawRect will be called after the
        WebView has closed. Quickbooks does not properly close the WebView
        and set the UIDelegate to nil, so the UIDelegate is stale and we
        crash. This is a regression because the code that uses the
        UIDelegate in the drawRect code path was only added recently. The
        method that the UIDelegate calls into is new -- it does not exist
        on Tiger -- so there is no harm in not running this code for
        applications linked against older WebKits. Other applications may
        run into this same bug so I am not doing a bundle
        check...particularly because, as I mentioned, the new UIDelegate
        call would not be implemented by older clients anyway.

        * Misc/WebKitVersionChecks.h:
        * WebView/WebHTMLView.mm:
        (-[WebHTMLView drawSingleRect:]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21748 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Make sure -webkit-border-fit accounts for padding as well as border.
hyatt [Fri, 25 May 2007 07:07:43 +0000 (07:07 +0000)]
    Make sure -webkit-border-fit accounts for padding as well as border.

        Reviewed by aroben

        * rendering/RenderBlock.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21747 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Rework line numbering (again). This time I finally think I have
hyatt [Fri, 25 May 2007 06:58:03 +0000 (06:58 +0000)]
    Rework line numbering (again).  This time I finally think I have
        it right for all the document.write insanity that can occur.

        Reviewed by aroben

        * html/HTMLParser.cpp:
        * html/HTMLParserErrorCodes.cpp:
        * html/HTMLParserErrorCodes.h:
        * html/HTMLTokenizer.cpp:
        * html/HTMLTokenizer.h:
        * platform/SegementedSring.cpp
        * platform/SegmentedString.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21746 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by Adam Roben.
ggaren [Fri, 25 May 2007 06:13:23 +0000 (06:13 +0000)]
    Reviewed by Adam Roben.

        Added a debug-only CachedPage counter. This helped me track down an SVG
        timer bug in my recent page cache work.

        * history/CachedPage.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21745 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Improve <form> error reporting to catch <form>s inside table parts.
hyatt [Fri, 25 May 2007 04:42:11 +0000 (04:42 +0000)]
    Improve <form> error reporting to catch <form>s inside table parts.
        Tweak the wording of some of the error messages to be a bit better.

        Reviewed by olliej

        * html/HTMLParser.cpp:
        * html/HTMLParserErrorCodes.cpp:
        * html/HTMLParserErrorCodes.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21743 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoLayoutTests:
weinig [Fri, 25 May 2007 01:28:55 +0000 (01:28 +0000)]

        Reviewed by Hyatt.

        Test cases for:
        first-letter not updating style with hover
        css-formatted first letter won't hide dynamically

        * fast/css/first-letter-hover-expected.checksum: Added.
        * fast/css/first-letter-hover-expected.png: Added.
        * fast/css/first-letter-hover-expected.txt: Added.
        * fast/css/first-letter-hover.html: Added.
        * fast/css/first-letter-visibility-expected.checksum: Added.
        * fast/css/first-letter-visibility-expected.png: Added.
        * fast/css/first-letter-visibility-expected.txt: Added.
        * fast/css/first-letter-visibility.html: Added.


        Reviewed by Hyatt.

        first-letter not updating style with hover
        css-formatted first letter won't hide dynamically

        Update first letter when RenderBlock::setStyle is called.

        * css/cssstyleselector.cpp:
        * dom/Document.cpp:
        * dom/Document.h:
        * rendering/RenderBlock.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21742 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by Darin.
weinig [Thu, 24 May 2007 23:57:25 +0000 (23:57 +0000)]
    Reviewed by Darin.

        - http://bugs.webkit.org/show_bug.cgi?id=13823
          REGRESSION: fast/forms/select-item-background-clip.html failing pixel test

        Updated the test and the results.

        * fast/forms/select-item-background-clip-expected.checksum:
        * fast/forms/select-item-background-clip-expected.png:
        * fast/forms/select-item-background-clip-expected.txt:
        * fast/forms/select-item-background-clip.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21740 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoLayoutTests:
weinig [Thu, 24 May 2007 23:51:23 +0000 (23:51 +0000)]

        Reviewed by Darin.

        - test for http://bugs.webkit.org/show_bug.cgi?id=13800
          REGRESSION: Moving a slider moves another unrelated slider

        * fast/forms/slider-thumb-shared-style-expected.checksum: Added.
        * fast/forms/slider-thumb-shared-style-expected.png: Added.
        * fast/forms/slider-thumb-shared-style-expected.txt: Added.
        * fast/forms/slider-thumb-shared-style.html: Added.


        Reviewed by Darin.

        - fix http://bugs.webkit.org/show_bug.cgi?id=13800
          REGRESSION: Moving a slider moves another unrelated slider

        Test: fast/forms/slider-thumb-shared-style.html

        RenderSlider mutates the thumb's style, therefore it must not be shared
        with another slider's thumb, a situation that could occur if the sliders
        themselves shared style.

        * rendering/RenderSlider.cpp:
        (WebCore::RenderSlider::createThumbStyle): Always use a private copy for
        the thumb style.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21739 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoLayoutTests:
adele [Thu, 24 May 2007 23:48:02 +0000 (23:48 +0000)]

        Reviewed by Darin.

        Test for <rdar://problem/5211921> Form values sometimes get transposed (iGoogle)

        * fast/history/redirect-load-no-form-restoration-expected.txt: Added.
        * fast/history/redirect-load-no-form-restoration.html: Added.
        * fast/history/resources/redirect-load.html: Added.


        Reviewed by Darin.

        Fix for <rdar://problem/5211921> Form values sometimes get transposed (iGoogle)

        Test: fast/history/redirect-load-no-form-restoration.html

        * history/HistoryItem.cpp:
        (WebCore::HistoryItem::setURL): Clear the document state when the url changes for a HistoryItem.
        (WebCore::HistoryItem::isCurrentDocument): Added. Compares the HistoryItem url to the document's url.  This might not be the best way to do this, but it works for now.
        * history/HistoryItem.h:
        * loader/FrameLoader.cpp: (WebCore::FrameLoader::saveDocumentState): Only set the document (and form) state if the document is current for the history item.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21738 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by Darin
weinig [Thu, 24 May 2007 23:28:28 +0000 (23:28 +0000)]
    Reviewed by Darin

        Patch for http://bugs.webkit.org/show_bug.cgi?id=13855
        Port patch 666176 to JavaScriptCore

        - Renamed JSValue::downcast() to JSValue::asCell() which makes the
        function meaning cleaner. It's modeled after Harri Porten change in
        KDE trunk.

        * kjs/collector.cpp:
        * kjs/object.h:
        * kjs/string_object.cpp:
        * kjs/value.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21736 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoUpdate results for better whitespace handling in framesets and tables.
hyatt [Thu, 24 May 2007 23:20:05 +0000 (23:20 +0000)]
Update results for better whitespace handling in framesets and tables.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21735 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoLayoutTests:
weinig [Thu, 24 May 2007 23:05:47 +0000 (23:05 +0000)]

        Reviewed by Adele.

        - test for http://bugs.webkit.org/show_bug.cgi?id=13857
          REGRESSION: onChange function applied to select element executes twice when focus is set

        * fast/forms/select-double-onchange-expected.txt: Added.
        * fast/forms/select-double-onchange.html: Added.


        Reviewed by Adele.

        - fix http://bugs.webkit.org/show_bug.cgi?id=13857
          REGRESSION: onChange function applied to select element executes twice when focus is set

        Test: fast/forms/select-double-onchange.html

        * html/HTMLSelectElement.cpp:
        (WebCore::HTMLSelectElement::menuListOnChange): Changed to update m_lastOnChangeIndex
        before dispatching the change event to avoid reentrancy.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21734 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoUpdate results for better whitespace handling in framesets and tables.
hyatt [Thu, 24 May 2007 22:58:40 +0000 (22:58 +0000)]
Update results for better whitespace handling in framesets and tables.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21733 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by Hyatt.
darin [Thu, 24 May 2007 22:56:33 +0000 (22:56 +0000)]
    Reviewed by Hyatt.

        - fix <rdar://problem/5226451> REGRESSION (21618): Mail block quotes
          are missing the lines on the side

        This patch fixes three problems:

        1) makes the "skip canLoad check" boolean also skip the check
           that prevents any loads while the document is in provisional
           state; this is the proximate cause of the bug

        * loader/SubresourceLoader.cpp: (WebCore::SubresourceLoader::create):
        Don't check the frame's state if skipCanLoadCheck is true.

        2) moves the "skip canLoad check" boolean to the Request object;
           the old implementation would cause that flag to affect the
           new request we served, which might not be the resource with
           that flag set

        3) fixes error-handling code path that would leak requests

        * html/HTMLImageLoader.cpp: (WebCore::HTMLImageLoader::updateFromElement):
        Pass false to the CachedImage constructor to indicate we are not making
        this object for the cache.

        * loader/Cache.cpp:
        (WebCore::createResource): Pass true to the CachedImage constructor to
        indicate we are making this object for the cache.
        (WebCore::Cache::requestResource): Add new code that assumes the object
        will already have the inCache bit set, and that will delete the object
        and return 0 if the cache is disabled and the load failed.

        * loader/CachedImage.h:
        * loader/CachedImage.cpp: (WebCore::CachedImage::CachedImage):
        Added a forCache boolean parameter. Always false for the constructor
        that's only used outside the cache code, and passed in as a boolean
        for the constructor that's used both in cache and outside cache.

        * loader/CachedResource.h:
        * loader/CachedResource.cpp:
        (WebCore::CachedResource::CachedResource): Added a forCache boolean
        parameter that determines the initial state of the m_inCache flag.
        This is needed to prevent a resource from being destroyed if an
        error occurs during the initial load.

        * loader/DocLoader.cpp:
        (WebCore::DocLoader::requestCSSStyleSheet): Added FIXME.
        (WebCore::DocLoader::setLoadInProgress): Added null check.

        * loader/Request.h:
        * loader/Request.cpp: (WebCore::Request::Request):
        Added a shouldSkipCanLoadCheck boolean here, since we need to track
        this for each request.

        * loader/loader.h:
        * loader/loader.cpp:
        (WebCore::Loader::load): Pass the skipCanLoadCheck boolean to the
        Request constructor rather than to the servePendingRequests function.
        (WebCore::Loader::servePendingRequests): Add a loop so we can handle
        cases where the request fails immediately without leaking the request
        and thinking that we're loading forever.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21732 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Add error reporting for residual style.
hyatt [Thu, 24 May 2007 22:29:00 +0000 (22:29 +0000)]
    Add error reporting for residual style.

        Reviewed by aroben

        * html/HTMLParser.cpp:
        * html/HTMLParserErrorCodes.cpp:
        * html/HTMLParserErrorCodes.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21731 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago * English.lproj/StringsNotToBeLocalized.txt: Update for recent changes.
darin [Thu, 24 May 2007 22:13:55 +0000 (22:13 +0000)]
    * English.lproj/StringsNotToBeLocalized.txt: Update for recent changes.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21730 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoLand updated tests.
hyatt [Thu, 24 May 2007 21:49:37 +0000 (21:49 +0000)]
Land updated tests.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21729 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago2007-05-24 Oliver Hunt <oliver@apple.com>
oliver [Thu, 24 May 2007 20:57:00 +0000 (20:57 +0000)]
2007-05-24  Oliver Hunt  <oliver@apple.com>

        Reviewed by Adam, Darin, and Geoff.

        Fix for <rdar://problem/5000470> REGRESSION: The IM reconvert
        function returns incorrect symbol due to inconsistent range
        domains in TSM

        Text Services Management uses ranges provided by the
        NSTextInput API to index into the string return by
        -[WebHTMLView string]. As a result some input methods
        incorrectly get their candidate text from the beginning
        of the document instead of from the input element.

        TSM prefers to query -textStorage over -string so as a
        workaround we provide an implementation of -textStorage that
        returns the content of the current text input.  TSM only ever
        queries the result of textStorage as an NSAttributedString so
        we do not need to implement a fake NSTextStorage class

        This should not cause harm to anything else as textStorage is
        actually a method on NSTextView, which we clearly are not. TSM
        only queries the method because it uses respondsToSelector to
        control behaviour.

        * WebView/WebHTMLView.mm:
        (-[WebHTMLView textStorage]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21728 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Report <script/> as a warning instead of an error for now, since we
hyatt [Thu, 24 May 2007 20:45:26 +0000 (20:45 +0000)]
    Report <script/> as a warning instead of an error for now, since we
        actually do close it.  (Sigh.)

        Reviewed by beth

        * html/HTMLParser.cpp:
        * html/HTMLParserErrorCodes.cpp:
        * html/HTMLParserErrorCodes.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21727 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Add error reporting for the </p> -> <p></p> quirk.
hyatt [Thu, 24 May 2007 20:26:31 +0000 (20:26 +0000)]
    Add error reporting for the </p> -> <p></p> quirk.

        Add error reporting when a stray close tag is encountered with no
        corresponding open tag.

        Reviewed by adele

        * html/HTMLParser.cpp:
        * html/HTMLParser.h:
        * html/HTMLParserErrorCodes.cpp:
        * html/HTMLParserErrorCodes.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21726 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by.
kmccullo [Thu, 24 May 2007 20:16:03 +0000 (20:16 +0000)]
    Reviewed by.

        - Do you ever feel like Charlie Brown? *sigh*

        * fast/dom/HTMLHeadElement/resources/head-check.js:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21725 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoLayoutTests:
kmccullo [Thu, 24 May 2007 20:06:56 +0000 (20:06 +0000)]

        Reviewed by.

        - Rolling back change to fix the build until I have a good fix for data: urls.

        * ChangeLog:
        * fast/dom/HTMLHeadElement/resources/head-check.js:
        * fast/dom/frame-loading-via-document-write-expected.txt: Removed.
        * fast/dom/frame-loading-via-document-write.html: Removed.
        * fast/dom/resources/frame-loading-via-document-write.js: Removed.
        * fast/dom/resources/green250x50.png: Removed.
        * fast/dom/resources/success.html: Removed.
        * http/tests/security/frame-loading-via-document-write-expected.txt: Removed.
        * http/tests/security/frame-loading-via-document-write.html: Removed.
        * http/tests/security/resources/frame-loading-via-document-write.js: Removed.
        * http/tests/security/resources/success.html: Removed.


        Reviewed by.

        - Rolling back change to fix the build until I have a good fix for data: urls.

        * ChangeLog:
        * loader/FrameLoader.cpp:
        * platform/String.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21724 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Add two more error messages to the HTMLParser.
hyatt [Thu, 24 May 2007 19:54:47 +0000 (19:54 +0000)]
    Add two more error messages to the HTMLParser.

        Report when the </br> -> <br> quirk is triggered.

        Also report whenever the broken XML self-closing tag syntax is used on
        tags that can actually have an end tag (either optional or required).

        Reviewed by aroben

        * html/HTMLParser.cpp:
        * html/HTMLParserErrorCodes.cpp:
        * html/HTMLParserErrorCodes.h:
        * html/HTMLTokenizer.cpp:
        * html/HTMLTokenizer.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21723 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by Anders.
weinig [Thu, 24 May 2007 18:30:27 +0000 (18:30 +0000)]
    Reviewed by Anders.

        Rename JSHTMLFrameSetCustom.cpp to JSHTMLFrameSetElementCustom.cpp
        for consistency.

        * WebCore.xcodeproj/project.pbxproj:
        * bindings/js/JSHTMLFrameSetCustom.cpp: Removed.
        * bindings/js/JSHTMLFrameSetElementCustom.cpp: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21721 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Fix for <rdar://problem/5208440> (13753)
hyatt [Thu, 24 May 2007 18:26:21 +0000 (18:26 +0000)]
    Fix for <rdar://problem/5208440> (13753)

        REGRESSION: Raw text needs to be pulled outside of tables

        Reviewed by aroben

        * html/HTMLParser.cpp:
        * html/HTMLTableColElement.cpp:
        * html/HTMLTableElement.cpp:
        * html/HTMLTableRowElement.cpp:
        * html/HTMLTableSectionElement.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21720 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoLayoutTests:
adele [Thu, 24 May 2007 17:45:27 +0000 (17:45 +0000)]

        Reviewed by Darin.

        - pixel test for http://bugs.webkit.org/show_bug.cgi?id=13837
          Separator in new <select> widget disappears when block element with border is in document.

        * fast/forms/menulist-separator-painting-expected.checksum: Added.
        * fast/forms/menulist-separator-painting-expected.png: Added.
        * fast/forms/menulist-separator-painting-expected.txt: Added.
        * fast/forms/menulist-separator-painting.html: Added.


        Reviewed by Darin.

        - fix http://bugs.webkit.org/show_bug.cgi?id=13837
          Separator in new <select> widget disappears when block element with border is in document.

        Test: fast/forms/menulist-separator-painting.html

        * platform/graphics/cg/GraphicsContextCG.cpp:
        (WebCore::GraphicsContext::drawConvexPolygon): Changed this function to
        actually do both filling and stroking when needed, even though all existing
        callers set the stroke to none.
        * rendering/RenderThemeMac.mm:
        (WebCore::RenderThemeMac::paintMenuListButton): Added calls to set the stroke
        style and thickness when painting the arrows and the separator.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21719 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by Sam Weinig.
thatcher [Thu, 24 May 2007 17:40:38 +0000 (17:40 +0000)]
    Reviewed by Sam Weinig.

        Remove the OBJC_NEW_PROPERTIES #ifdef and only export the new @property syntax.

        * bindings/objc/DOMObject.h:
        * bindings/scripts/CodeGeneratorObjC.pm:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21718 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoLayoutTests:
weinig [Thu, 24 May 2007 17:20:58 +0000 (17:20 +0000)]

        Reviewed by Anders.

        - update results to account for the new JSHTMLFrameSetElement
          that is being autogenerated and can now produce a constructor.

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


        Reviewed by Anders.

        Uncomment HTMLFramesSetElement constructor now that it exists.

        * page/DOMWindow.idl:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21716 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Try fixing the buildbot build.
andersca [Thu, 24 May 2007 17:04:37 +0000 (17:04 +0000)]
    Try fixing the buildbot build.

        * loader/ImageDocument.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21715 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by Darin.
weinig [Thu, 24 May 2007 16:39:48 +0000 (16:39 +0000)]
    Reviewed by Darin.

        Patch for http://bugs.webkit.org/show_bug.cgi?id=13853
        Autogenerate the JS bindings for the HTMLFrameSetElement

        * DerivedSources.make:
        * WebCore.pro:
        * WebCore.xcodeproj/project.pbxproj:
        * bindings/js/JSHTMLElementWrapperFactory.cpp:
        * bindings/js/JSHTMLFrameSetCustom.cpp: Added.
        * bindings/js/kjs_html.cpp:
        * bindings/js/kjs_html.h:
        * html/HTMLFrameSetElement.idl:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21714 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoRemove conflict markers.
ddkilzer [Thu, 24 May 2007 16:25:41 +0000 (16:25 +0000)]
Remove conflict markers.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21713 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoFixing check for incorrent font size - it should
zack [Thu, 24 May 2007 15:41:12 +0000 (15:41 +0000)]
Fixing check for incorrent font size - it should
check fixed not normal.


git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21711 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by Tim Hatcher.
harrison [Thu, 24 May 2007 15:11:51 +0000 (15:11 +0000)]
    Reviewed by Tim Hatcher.

        <rdar://problem/5225343> REGRESSION: With View Source window opened, navigating to a different URL in the browser window results in a crash at WebCore::FrameLoader::frameHasLoaded()

        _private->loader->frameLoader() was not being nil checked.
        * WebView/WebDataSource.mm:
        (-[WebDataSource request]):
        Add nil check for _private->loader->frameLoader()

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21710 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoFix multipart HTTP POSTs in the Qt port
hausmann [Thu, 24 May 2007 14:10:38 +0000 (14:10 +0000)]
Fix multipart HTTP POSTs in the Qt port

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21709 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Speculative buildbot build fix.
bdash [Thu, 24 May 2007 12:37:57 +0000 (12:37 +0000)]
    Speculative buildbot build fix.

        * css/cssstyleselector.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21708 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoDidn't mean to commit this
aroben [Thu, 24 May 2007 09:53:26 +0000 (09:53 +0000)]
Didn't mean to commit this

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21707 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoMade Chrome::addMessageToConsole more flexible
aroben [Thu, 24 May 2007 09:42:26 +0000 (09:42 +0000)]
Made Chrome::addMessageToConsole more flexible

addMessageToConsole now takes two additional parameters to specify the
message source (HTML, JS, CSS, XML) and level (error, warning, log). To
maintain our old behavior, all current messages were given
JSMessageSource and ErrorMessageLevel parameters. Messages matching
these characteristics are passed up to the ChromeClient.

Reviewed by Hyatt.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21706 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Rework how HTML handles line numbering so that it is not fooled by
hyatt [Thu, 24 May 2007 08:56:35 +0000 (08:56 +0000)]
    Rework how HTML handles line numbering so that it is not fooled by
        \r without a following \n and so that document.write doesn't lead
        to line numbering confusion.

        Reviewed by aroben

        * html/HTMLParser.cpp:
        * html/HTMLTokenizer.cpp:
        * html/HTMLTokenizer.h:
        * html/HTMLViewSourceDocument.cpp:
        * loader/TextDocument.cpp:
        * platform/SegmentedString.cpp:
        * platform/SegmentedString.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21704 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoUpdate some tests that darin missed.
hyatt [Thu, 24 May 2007 08:47:51 +0000 (08:47 +0000)]
Update some tests that darin missed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21703 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoLand corrected background shorthand test now that I yanked the bogus alert.
hyatt [Thu, 24 May 2007 08:46:26 +0000 (08:46 +0000)]
Land corrected background shorthand test now that I yanked the bogus alert.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21702 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoShare WebCoreHttp and therefore HTTP connections among multiple
hausmann [Thu, 24 May 2007 07:47:03 +0000 (07:47 +0000)]
Share WebCoreHttp and therefore HTTP connections among multiple
QWebNetworkInterface instances by moving the code into

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21701 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoRemove a stray alert in a testcase that I wrote a while back.
hyatt [Thu, 24 May 2007 06:01:05 +0000 (06:01 +0000)]
Remove a stray alert in a testcase that I wrote a while back.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21700 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoAdd a table test for comment nodes.
hyatt [Thu, 24 May 2007 05:58:11 +0000 (05:58 +0000)]
Add a table test for comment nodes.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21699 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Make the delete/InsertRows/Cells methods as well as the rowIndex method
hyatt [Thu, 24 May 2007 05:48:26 +0000 (05:48 +0000)]
    Make the delete/InsertRows/Cells methods as well as the rowIndex method
        able to work properly even when whitespace or comment nodes are present.

        Reviewed by aroben


        * html/HTMLTableRowElement.cpp:
        * html/HTMLTableSectionElement.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21698 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoLand optgroup DTD test.
hyatt [Thu, 24 May 2007 05:39:39 +0000 (05:39 +0000)]
Land optgroup DTD test.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21697 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago <optgroup> should allow <script> and text nodes inside it, just like
hyatt [Thu, 24 May 2007 05:36:51 +0000 (05:36 +0000)]
    <optgroup> should allow <script> and text nodes inside it, just like
        <select> does.

        Reviewed by aroben


        * html/HTMLOptGroupElement.cpp:
        * html/HTMLSelectElement.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21696 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoAdd gigantic z-index test.
hyatt [Thu, 24 May 2007 05:33:02 +0000 (05:33 +0000)]
Add gigantic z-index test.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21695 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Framesets should allow whitespace text nodes as children.
hyatt [Thu, 24 May 2007 05:23:43 +0000 (05:23 +0000)]
    Framesets should allow whitespace text nodes as children.

        Reviewed by aroben

        * html/HTMLFrameSetElement.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21694 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Fix the Qt build after r21685.
bdash [Thu, 24 May 2007 03:28:50 +0000 (03:28 +0000)]
    Fix the Qt build after r21685.

        * platform/qt/CursorQt.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21693 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by Adam.
bdash [Thu, 24 May 2007 03:23:07 +0000 (03:23 +0000)]
    Reviewed by Adam.

        Fix the gdk build following changes in r21685.

        * platform/gdk/CursorGdk.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21692 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Fix for <rdar://problem/5222611> REGRESSION: Paint ordering problems at
hyatt [Thu, 24 May 2007 03:07:43 +0000 (03:07 +0000)]
    Fix for <rdar://problem/5222611> REGRESSION: Paint ordering problems at

        Clamp z-index so that values that exceed maxint (or that are below minint)
        are clamped to those values.

        Reviewed by darin

        * css/cssstyleselector.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21691 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoLayoutTests:
ddkilzer [Thu, 24 May 2007 01:46:31 +0000 (01:46 +0000)]

        Reviewed by David Hyatt.

        - tests for http://bugs.webkit.org/show_bug.cgi?id=13563
          REGRESSION: Crash loading message in Yahoo! Mail

        * fast/css/stale-style-selector-crash-1-expected.txt: Added.
        * fast/css/stale-style-selector-crash-1.html: Added.
        * fast/css/stale-style-selector-crash-2-expected.txt: Added.
        * fast/css/stale-style-selector-crash-2.html: Added.


        Reviewed by David Hyatt.

        - fix http://bugs.webkit.org/show_bug.cgi?id=13563
          REGRESSION: Crash loading message in Yahoo! Mail

        Tests: fast/css/stale-style-selector-crash-1.html

        * dom/Document.cpp:
        (WebCore::Document::Document): Initialize the m_didCalculateStyleSelector
        flag to 'false'.
        (WebCore::Document::updateStyleSelector): Do not bail out if we have ever
        calculated a non-trivial style selector, even if at the moment we are
        pending style sheets. We could have calculated a non-trivial selector
        already either because at some point there were no pending style sheets
        (so the currently pending ones were added dynamically) or because we were
        forced to do a layout ignoring pending style sheets. Either way, once
        there is a style selector, we need to keep it up to date, otherwise we can
        crash under recalcStyle() during updateRendering() or simply not recalculate
        style correctly for dynamic changes.
        (WebCore::Document::recalcStyleSelector): Set m_didCalculateStyleSelector
        to 'true' when creating a new style selector.
        * dom/Document.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21690 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by Adam.
andersca [Thu, 24 May 2007 01:43:32 +0000 (01:43 +0000)]
    Reviewed by Adam.

        <rdar://problem/5224749> Show a zoom/unzoom cursor for resizable image documents

        * WebCore.xcodeproj/project.pbxproj:
        Install the zoom cursors.

        * loader/ImageDocument.cpp:
        Set the cursor to the zoom in cursor.

        If the image fits, reset the cursor. Otherwise, set it to the zoom in cursor.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21689 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by Geoff.
andersca [Thu, 24 May 2007 01:36:55 +0000 (01:36 +0000)]
    Reviewed by Geoff.

        <rdar://problem/3663808> Resize large images to fit in the browser window

        Add new WebPreferences SPI.

        * WebView/WebPreferenceKeysPrivate.h:
        * WebView/WebPreferences.m:
        (+[WebPreferences initialize]):
        (-[WebPreferences enableAutomaticImageResizing]):
        (-[WebPreferences setEnableAutomaticImageResizing:]):
        * WebView/WebPreferencesPrivate.h:
        * WebView/WebView.mm:
        (-[WebView _updateWebCoreSettingsFromPreferences:]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21688 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoLayoutTests:
darin [Thu, 24 May 2007 01:11:10 +0000 (01:11 +0000)]

        Reviewed by Hyatt.

        - updated the many tests with an extra newline or two at the end
          because of the bug where plain-text serialization will consider
          nodes that are not rendered


        Reviewed by Hyatt.

        - fix problems where unrendered nodes were affecting on plain text serialization

        * editing/TextIterator.cpp:
        (WebCore::TextIterator::advance): Don't call handleXXX functions or exitNode for
        nodes without renderers.
        (WebCore::shouldEmitNewlineAfterNode): Only consider rendered nodes when deciding
        if this is the last node in the document.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21687 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by Geoff.
andersca [Thu, 24 May 2007 01:01:02 +0000 (01:01 +0000)]
    Reviewed by Geoff.

        <rdar://problem/3663808> Resize large images to fit in the browser window

        * WebCore.exp:
        Add new settings method.

        * loader/ImageDocument.cpp:
        New class that handles resize events and click events and forwards them to the image document.

        Let the ImageDocument class keep track of its document structure.

        Returns the scale that should be used when resizing the image.

        Update the image size.

        Toggle between resized and not resized. When restoring the size, scroll the image so that the area under
        the mouse stays the same.

        When the image size is known, resize the image if needed.

        Restore the image size.

        Return whether the image is too big for its window or not.

        Called when the window's size changes. Determine if the window fits or not and resize it if it doesn't fit.

        Forward events to the image document.

        * page/Settings.cpp:
        * page/Settings.h:
        Add new setting.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21686 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by Darin.
andersca [Thu, 24 May 2007 00:50:02 +0000 (00:50 +0000)]
    Reviewed by Darin.

        Add -webkit-zoom-in and -webkit-zoom-out cursors.

        * Resources/zoomInCursor.png: Added.
        * Resources/zoomOutCursor.png: Added.
        * css/CSSComputedStyleDeclaration.cpp:
        * css/CSSValueKeywords.in:
        * page/EventHandler.cpp:
        * platform/Cursor.h:
        * platform/mac/CursorMac.mm:
        * rendering/RenderStyle.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21685 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by Darin.
weinig [Wed, 23 May 2007 23:32:06 +0000 (23:32 +0000)]
    Reviewed by Darin.

        Part one of patch for http://bugs.webkit.org/show_bug.cgi?id=13830
        Auto-generate JS DOM bindings for HTMLDocument and most of the rest
        of HTMLElement

        - Just the HTMLElement part.

        * bindings/js/kjs_html.cpp:
        * bindings/js/kjs_html.h:
        * html/HTMLElement.idl:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21683 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoLayoutTests:
kmccullo [Wed, 23 May 2007 23:19:56 +0000 (23:19 +0000)]

        Reviewed by Darin.

        Update tests for http://bugs.webkit.org/show_bug.cgi?id=13830
        Auto-generate JS DOM bindings for HTMLDocument and most of
        the rest of HTMLElement

        - Corrects prototype chain for JSHTMLDocument and adds test
          for HTMLFormElement.

        * fast/dom/prototype-chain-expected.txt:
        * fast/dom/prototype-chain.html:


        Reviewed by Darin.

        - rdar://problem/5183523
        - We now check the document's URL to see if it can load a resource in the case of a URL click, instead of checking the referrer, which is blank for local files.

        * loader/FrameLoader.cpp:
        * platform/PlatformString.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21682 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoLayoutTests:
weinig [Wed, 23 May 2007 23:00:20 +0000 (23:00 +0000)]

        Reviewed by Darin.

        Update tests for http://bugs.webkit.org/show_bug.cgi?id=13830
        Auto-generate JS DOM bindings for HTMLDocument and most of
        the rest of HTMLElement

        - Corrects prototype chain for JSHTMLDocument and adds test
          for HTMLFormElement.

        * fast/dom/prototype-chain-expected.txt:
        * fast/dom/prototype-chain.html:


        Reviewed by Darin.

        Part two of patch for http://bugs.webkit.org/show_bug.cgi?id=13830
        Auto-generate JS DOM bindings for HTMLDocument and most of the rest
        of HTMLElement

        - Just the HTMLDocument part.

        * WebCore.pro:
        * WebCore.xcodeproj/project.pbxproj:
        * bindings/js/JSHTMLDocumentCustom.cpp: Added.
        * bindings/js/kjs_dom.cpp:
        * bindings/js/kjs_dom.h:
        * bindings/js/kjs_html.cpp:
        * bindings/js/kjs_html.h:
        * bindings/objc/DOMPrivate.h:
        * bindings/scripts/CodeGeneratorJS.pm: Add [Deletable] property
        for Document.all and fix HasOverridingNameGetter to allow correct
        prototype chain
        * dom/Document.cpp:
        (WebCore::Document::plugins): adds alias to embeds() for bindings
        * dom/Document.h:
        * html/HTMLDocument.cpp:
        * html/HTMLDocument.h:
        * html/HTMLDocument.idl:


        Reviewed by Darin.

        Patch for http://bugs.webkit.org/show_bug.cgi?id=13830
        Auto-generate JS DOM bindings for HTMLDocument and most of the rest
        of HTMLElement

        * MigrateHeaders.make: add DOMHTMLDocumentPrivate.h
        * WebKit.xcodeproj/project.pbxproj:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21681 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agogit-svn-id: https://svn.webkit.org/repository/webkit/trunk@21680 268f45cc-cd09-0410...
hyatt [Wed, 23 May 2007 22:58:59 +0000 (22:58 +0000)]
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21680 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Fix for <rdar://problem/5221838> REGRESSION (r21581): no content in idisk
hyatt [Wed, 23 May 2007 22:55:01 +0000 (22:55 +0000)]
    Fix for <rdar://problem/5221838> REGRESSION (r21581): no content in idisk
        window at mac.com.

        Back out the change to make self-closed <script/> into a Dashboard quirk

        Reviewed by aroben

        * html/HTMLTokenizer.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21679 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by Geoff.
oliver [Wed, 23 May 2007 20:53:05 +0000 (20:53 +0000)]
    Reviewed by Geoff.

        Fix for <rdar://problem/5223782> REGRESSION: Can't drag and drop a standalone image

        The main resource for a standalone image webarchive has the same mimetype as the
        underlying image.

        * Misc/WebNSPasteboardExtras.mm:
        (-[NSPasteboard _web_writePromisedRTFDFromArchive:containsImage:]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21678 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by.
kmccullo [Wed, 23 May 2007 17:19:24 +0000 (17:19 +0000)]
    Reviewed by.

        - Fixed build breakage.

        * fast/dom/open-and-close-by-DOM-expected.txt: Added.
        * fast/dom/open-and-close-by-DOM.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21671 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago<rdar://problem/5078453> AXNext/PrevLineEndTextMarkerForTextMarker returns kAXErrorNo...
pyeh [Wed, 23 May 2007 17:11:12 +0000 (17:11 +0000)]
<rdar://problem/5078453> AXNext/PrevLineEndTextMarkerForTextMarker returns kAXErrorNoValue for attachment range

        Floating objects, such as aligned images, are not included as part of any line.  So when the given
        text marker is on the floating object, the returned line edge cannot be found and asserts are fired.
        In addition, this paints an inconsistent text layout (in the accessibility sense) of where the floating
        object resides.  Character/word methods report floating object at the beginning of the text line(s).  Yet
        line methods never present floating objects anywhere.
        To fix this, the ax line methods will now do more checking to ensure floating object nodes are included
        when determining the line range/position. That way, assistive apps won't get stuck using the line methods
        to read line-by-line

        * bridge/mac/WebCoreAXObject.mm:
        Added a helper method to adjust line start position to account for floating objects
        (-[WebCoreAXObject doAXLeftLineTextMarkerRangeForTextMarker:]):
        (-[WebCoreAXObject doAXRightLineTextMarkerRangeForTextMarker:]):
        (-[WebCoreAXObject doAXNextLineEndTextMarkerForTextMarker:]):
        (-[WebCoreAXObject doAXPreviousLineStartTextMarkerForTextMarker:]):
        Fixed up line routines to include floating object nodes in line position calculation.
        startOfLine/endOLine may return null for position next to a floating position.  So now there's extra checks to prevent
        returning null line position/ranges back to AX.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21670 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by.
kmccullo [Wed, 23 May 2007 17:02:11 +0000 (17:02 +0000)]
    Reviewed by.

        - Removing failing tests until I've fixed them.

        * fast/dom/open-and-close-by-DOM-expected.txt: Removed.
        * fast/dom/open-and-close-by-DOM.html: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21669 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoAdded support for per-QWebPage network interface handlers, clean up the
hausmann [Wed, 23 May 2007 16:22:18 +0000 (16:22 +0000)]
Added support for per-QWebPage network interface handlers, clean up the
http and file loading and fixed a bug in the HTTP redirect handling.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21668 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by Dave Hyatt.
rwlbuis [Wed, 23 May 2007 14:00:20 +0000 (14:00 +0000)]
    Reviewed by Dave Hyatt.

        line-height in font shorthand does not override a previously stated line-height property

        Make sure line-height is calculated against definite font-size and
        uses the last set line-height, ie. directly or through font shorthand.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21667 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by Zack
lars [Wed, 23 May 2007 12:14:09 +0000 (12:14 +0000)]
    Reviewed by Zack

        Don't do HTTP downloads in a second thread. Simplifies
        the code significantly and fixes crashes on some
        Web pages.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21666 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by Zack and Valgrind
lars [Wed, 23 May 2007 11:07:17 +0000 (11:07 +0000)]
    Reviewed by Zack and Valgrind

        Initialize variable.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21665 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by Darin.
rwlbuis [Wed, 23 May 2007 11:04:14 +0000 (11:04 +0000)]
    Reviewed by Darin.

        REGRESSION (r13744-13750): Crash with empty gradient when drawing to canvas

        Fix crash by testing for zero stops, and returning transparent black
        in this case.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21664 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by Darin.
rwlbuis [Wed, 23 May 2007 10:48:16 +0000 (10:48 +0000)]
    Reviewed by Darin.

        Crash when loading nonexisting symbol

        Check for referenced element and when not found just return.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21663 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoFix Host: line in HTTP headers and resolving of relative URLs
lars [Wed, 23 May 2007 10:01:38 +0000 (10:01 +0000)]
Fix Host: line in HTTP headers and resolving of relative URLs
when redirecting and clean up qHash forward declaration a bit.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21662 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoFix signal/slot connection for load progress.
hausmann [Wed, 23 May 2007 08:48:08 +0000 (08:48 +0000)]
Fix signal/slot connection for load progress.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21661 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Rubber stamped by Adam.
weinig [Wed, 23 May 2007 03:38:10 +0000 (03:38 +0000)]
    Rubber stamped by Adam.

        Fix license.

        * bindings/js/JSHTMLAnchorElementCustom.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21656 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoLayoutTests:
weinig [Wed, 23 May 2007 01:26:00 +0000 (01:26 +0000)]

        Reviewed by Adam.

        Update tests for http://bugs.webkit.org/show_bug.cgi?id=13833
        Add ObjC DOM binding for HTMLMarqeeElement

        * fast/dom/wrapper-classes-expected.txt:
        * fast/dom/wrapper-classes.html: HTMLMarqueeElement now is a


        Reviewed by Adam.

        Patch for http://bugs.webkit.org/show_bug.cgi?id=13833
        Add ObjC DOM binding for HTMLMarqeeElement

        - Also adds missing DOMHTMLFramePrivate.

        * DerivedSources.make:
        * WebCore.xcodeproj/project.pbxproj:
        * bindings/objc/DOM.mm:
        * bindings/objc/DOMHTML.h:
        * bindings/objc/DOMInternal.h:
        * bindings/objc/DOMPrivate.h:
        * bindings/objc/PublicDOMInterfaces.h:


        Reviewed by Adam.

        Patch for http://bugs.webkit.org/show_bug.cgi?id=13833
        Add ObjC DOM binding for HTMLMarqeeElement

        - Also adds missing DOMHTMLFramePrivate.

        * MigrateHeaders.make:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21655 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoLayoutTests:
justing [Wed, 23 May 2007 01:19:43 +0000 (01:19 +0000)]

        Reviewed by darin

        REGRESSION(r21611): Several new layout test pixel failures (misspelling markers)

        * editing/selection/13804-expected.checksum: Added.
        * editing/selection/13804-expected.png: Added.
        * editing/selection/13804-expected.txt: Added.
        * editing/selection/13804.html: Added.
        These were missing:
        * editing/selection/5076323-1-expected.png: Added.
        * editing/selection/5076323-2-expected.png: Added.
        * editing/selection/5076323-3-expected.png: Added.


        Reviewed by darin, who found the fix independently.

        REGRESSION(r21611): Several new layout test pixel failures (misspelling markers)

        r21611 exposed a bug in the way removeMarkers computes
        the end of the range to remove markers from.  It also
        introduced a bug: removeMarkers expects a startOffset
        and a length, not a startOffset and an endOffset (added
        a layout test).

        * dom/Document.cpp:
        (WebCore::Document::removeMarkers): The other removeMarkers
        expects a startOffset and a length, not a startOffset and
        an endOffset.
        (WebCore::Document::removeMarkers): Fixed the endOffset
        from length and startOffset calculation.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21654 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoAdd layout test for better background dumping.
hyatt [Wed, 23 May 2007 00:55:52 +0000 (00:55 +0000)]
Add layout test for better background dumping.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21653 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Fix for bug 13802, background colors can't be reset to default
hyatt [Wed, 23 May 2007 00:43:11 +0000 (00:43 +0000)]
    Fix for bug 13802, background colors can't be reset to default
        color.  Fix the value cssText method to do the right thing for background
        and background-position values.

        Reviewed by beth


        * css/CSSInitialValue.h:
        * css/CSSMutableStyleDeclaration.cpp:
        * css/CSSMutableStyleDeclaration.h:
        * css/CSSValue.h:
        * css/cssparser.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21652 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoLayoutTests:
darin [Tue, 22 May 2007 23:47:07 +0000 (23:47 +0000)]

        Reviewed by Geoff.

        - test for <rdar://problem/5215830> REGRESSION: ebay.com layout is wrong

        * fast/dom/iframe-document.html: Updated test to expect no document property.
        * fast/dom/iframe-document-expected.txt: Updated.

        * fast/forms/focus2.html: Changed test to use ownerDocument instead of document.


        Reviewed by Geoff.

        - fix <rdar://problem/5215830> REGRESSION: ebay.com layout is wrong

        The code on ebay was looking for the "document" property on a iframe.
        We changed that to be the ownerDocument rather than the contentDocument,
        which was not what ebay was expecting. The best fix seems to be removing
        the document property altogether.

        Test: fast/dom/iframe-document.html

        * bindings/js/kjs_html.h: Remove ElementDocument.
        * bindings/js/kjs_html.cpp: (KJS::JSHTMLElement::getValueProperty): Ditto.


        Reviewed by Geoff.

        * WebInspector/webInspector/treeoutline.js: Use ownerDocument instead of non-standard
        document property.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21651 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by Darin.
adele [Tue, 22 May 2007 23:43:07 +0000 (23:43 +0000)]
    Reviewed by Darin.

        A little cleanup from my last checkin.

        * page/Chrome.h: Formatting change.
        * page/mac/ChromeMac.mm: Fixed copyright.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21650 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoLayoutTests:
adele [Tue, 22 May 2007 23:21:21 +0000 (23:21 +0000)]

        Reviewed by Darin.

        Test for http://bugs.webkit.org/show_bug.cgi?id=13796
        Crash in [NSInputContext currentInputContext] when you click on this map

        * fast/frames/frame-display-none-focus-expected.txt: Added.
        * fast/frames/frame-display-none-focus.html: Added.
        * fast/frames/resources/iframe-focus.html: Added.


        Reviewed by Darin.

        Fix for http://bugs.webkit.org/show_bug.cgi?id=13796
        Crash in [NSInputContext currentInputContext] when you click on this map

        This is another case of <rdar://problem/5171145> Safari crashed closing tab in NSInputContext updateInputContexts

        Test: fast/frames/frame-display-none-focus.html

        * WebCore.xcodeproj/project.pbxproj: Added ChromeMac.mm
        * page/Chrome.h:
        * page/mac/ChromeMac.mm: Added.
        (WebCore::Chrome::focusNSView): Moved code from Widget::setFocus so it can be used whenever we need to make a view the first responder.
        * page/mac/EventHandlerMac.mm: (WebCore::EventHandler::focusDocumentView): Calls focusNSView.
        * platform/mac/WidgetMac.mm: (WebCore::Widget::setFocus): Calls focusNSView.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21649 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoLayoutTests:
justing [Tue, 22 May 2007 23:18:38 +0000 (23:18 +0000)]

        Reviewed by darin

        <rdar://problem/4875189> List is deleted when replacing contents with Chinese/Japanese characters

        Moved from unsupported-content because we've supported editing lists
        for a while now:
        * editing/deleting/4875189-expected.checksum: Added.
        * editing/deleting/4875189-expected.png: Added.
        * editing/deleting/4875189-expected.txt: Added.
        * editing/deleting/4875189.html: Added.
        * editing/unsupported-content/list-delete-002-expected.checksum: Removed.
        * editing/unsupported-content/list-delete-002-expected.png: Removed.
        * editing/unsupported-content/list-delete-002-expected.txt: Removed.
        * editing/unsupported-content/list-delete-002.html: Removed.

        Reflects new behavior:
        * editing/deleting/delete-select-all-003-expected.checksum:
        * editing/deleting/delete-select-all-003-expected.png:
        * editing/deleting/delete-select-all-003-expected.txt:

        Ditto and also removed a FIXME:
        * editing/deleting/list-item-1-expected.checksum:
        * editing/deleting/list-item-1-expected.png:
        * editing/deleting/list-item-1-expected.txt:
        * editing/deleting/list-item-1.html:


        Reviewed by darin

        List is deleted when replacing contents with Chinese/Japanese characters

        * editing/htmlediting.cpp:
        (WebCore::isSpecialElement): Lists are no longer special.
        They can be removed by removing their content and then
        then hitting delete inside of the empty list item that's
        left over. Even when that list item is the first thing
        in the editable region.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21648 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by Darin.
adele [Tue, 22 May 2007 23:02:34 +0000 (23:02 +0000)]
    Reviewed by Darin.

        Adding some asserts to help detect other cases of <rdar://problem/5171145>

        * WebCoreSupport/WebFrameBridge.mm: (-[WebFrameBridge makeFirstResponder:]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21647 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by Tim Hatcher.
darin [Tue, 22 May 2007 22:59:02 +0000 (22:59 +0000)]
    Reviewed by Tim Hatcher.

        - fix <rdar://problem/5206089> REGRESSION: hovering over amazon.com "product categories"
          tab makes page content disappear

        The contentWindow property was doing a cross-frame security check, but it shouldn't.
        Window objects are responsible for protecting themselves from cross-frame activity.
        Because contentWindow returned undefined, we then ended up setting visibility on the
        body of the main frame to "hidden".

        No layout test yet. When we make cross-domain tests, we should include this one.

        * html/HTMLFrameElement.idl: Removed CheckFrameSecurity attribute from contentWindow.
        * html/HTMLIFrameElement.idl: Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21646 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by.
kmccullo [Tue, 22 May 2007 22:52:10 +0000 (22:52 +0000)]
    Reviewed by.

        - Forgot to checkin for fix for rdar://problem/4981886

        * fast/dom/open-and-close-by-DOM-expected.txt: Added.
        * fast/dom/open-and-close-by-DOM.html: Added.
        * fast/dom/resources/close.html: Added.
        * fast/dom/resources/middle.html: Added.
        * http/tests/security/local-image-from-remote.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21645 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoLayoutTests:
justing [Tue, 22 May 2007 19:02:58 +0000 (19:02 +0000)]

        Reviewed by oliver

        REGRESSION(r21467): Active selection remains in text field after clicking on a button

        Demonstrates bug:
        * editing/selection/5213963-expected.checksum: Added.
        * editing/selection/5213963-expected.png: Added.
        * editing/selection/5213963-expected.txt: Added.
        * editing/selection/5213963.html: Added.
        We now clear the selection on mouse down (but not before
        the mousedown event is fired):
        * fast/forms/focus-selection-input-expected.txt:
        * fast/forms/focus-selection-textarea-expected.txt:


        Reviewed by oliver

        REGRESSION(r21467): Active selection remains in text field after clicking on a button

        * page/EventHandler.cpp:
        (WebCore::EventHandler::dispatchMouseEvent): Call
        FocusController::setFocusedNode, because Document::setFocusedNode
        no longer clears the selection.
        Begin passing the new focused frame to
        FocusController::setFocusedNode, because when the
        new focus node is null, the selection should only be
        cleared if the focused frame isn't changing.
        (WebCore::Element::focus): Begin passing the new focused
        (WebCore::Element::blur): Ditto.
        * html/HTMLInputElement.cpp:
        (WebCore::HTMLInputElement::focus): Ditto.
        * html/HTMLTextAreaElement.cpp:
        (WebCore::HTMLTextAreaElement::focus): Ditto.
        * page/Frame.cpp:
        (WebCore::Frame::setFocusedNodeIfNeeded): Ditto.
        * page/FocusController.cpp:
        (WebCore::clearSelectionIfNeeded): Take in the new
        focused frame.  If it's not changing, don't clear the
        (WebCore::FocusController::setFocusedNode): Take
        in the new focused frame.
        * page/FocusController.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21642 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by Kevin (Sarge)
beidson [Tue, 22 May 2007 19:02:37 +0000 (19:02 +0000)]
    Reviewed by Kevin (Sarge)

        <rdar://problem/5219885> - Crash after closing a inline popup ad at http://news.yahoo.com/

        This regressed in http://trac.webkit.org/projects/webkit/changeset/21618

        * WebView/WebFrame.mm:
        (-[WebFrame dataSource]): Null check the frameloader

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21641 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoback this out for zack
staikos [Tue, 22 May 2007 15:11:53 +0000 (15:11 +0000)]
back this out for zack

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21640 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoPath implementing more Cliboard from Marius Bugge Monsen.
zack [Tue, 22 May 2007 08:53:55 +0000 (08:53 +0000)]
Path implementing more Cliboard from Marius Bugge Monsen.

Adding implementations of some of the missing methods.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21639 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoFix http headers for POST.
hausmann [Tue, 22 May 2007 07:31:11 +0000 (07:31 +0000)]
Fix http headers for POST.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21638 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years agoUpdating expected results for r21074.
justing [Tue, 22 May 2007 06:53:49 +0000 (06:53 +0000)]
Updating expected results for r21074.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21637 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago2007-05-21 Oliver Hunt <oliver@apple.com>
oliver [Tue, 22 May 2007 03:10:06 +0000 (03:10 +0000)]
2007-05-21  Oliver Hunt  <oliver@apple.com>


        Reviewed by Sam.

        Layout test for http://bugs.webkit.org/show_bug.cgi?id=13789

        * svg/custom/getscreenctm-in-mixed-content-expected.checksum: Added.
        * svg/custom/getscreenctm-in-mixed-content-expected.png: Added.
        * svg/custom/getscreenctm-in-mixed-content-expected.txt: Added.
        * svg/custom/getscreenctm-in-mixed-content.xhtml: Added.


        Reviewed by Sam.

        Fix for http://bugs.webkit.org/show_bug.cgi?id=13789
        -- SVGLocatable::getScreenCTM() faulty

        Needed to update SVGLength to handle the case where the SVGElement is
        not the root document element -- we do this by falling back on the
        renderer for the context.

        For <svg> elements embedded as mixed content in xhtml we consider the
        absolutePosition of the parent to be the origin for the <svg> element.

        * ksvg2/svg/SVGLength.cpp:
        * ksvg2/svg/SVGSVGElement.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21636 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by Adam.
thatcher [Tue, 22 May 2007 02:14:03 +0000 (02:14 +0000)]
    Reviewed by Adam.

        <rdar://problem/5084400> -[DOMRange lineBoxRects] returns an empty array when the DOM nodes have just been created

        Do a updateLayoutIgnorePendingStylesheets before asking the Node or Range for it's box rects.

        * bindings/objc/DOM.mm:
        (-[DOMNode boundingBox]):
        (-[DOMNode lineBoxRects]):
        (-[DOMRange boundingBox]):
        (-[DOMRange lineBoxRects]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21635 268f45cc-cd09-0410-ab3c-d52691b4dbfc

15 years ago Reviewed by George Staikos.
weinig [Tue, 22 May 2007 01:42:21 +0000 (01:42 +0000)]
    Reviewed by George Staikos.

        Fix Qt build.

        * WebCore.pro: add missing IDL files.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@21634 268f45cc-cd09-0410-ab3c-d52691b4dbfc