Timeline


and

12/18/07:

23:53 Changeset [28856] by bdakin@apple.com

WebCore:

Reviewed by Oliver.

Fix for <rdar://problem/5616982> SVGs with width and height 100%
fail to render when used as <img> or CSS image (16167)

This final part of the work fixes the <img> tag.

This is the real fix.

  • rendering/RenderImage.cpp: (WebCore::RenderImage::calcReplacedWidth): Set the container size on the image. Setting the container size only actually sticks if the values are non-zero, so if the container size really was set, use the imageSize that is calculated using the container size. If it did not stick but the image does have relative width (meaning that the container size is 0), set the width to 0 by hand. We want to avoid setting the width before we have a container size or we will end up incorrectly using the default size of 300x150. (WebCore::RenderImage::calcReplacedHeight): Same as above, but for height.

A few more pieces of information have to be exposed through cached
image to make this happen.

  • loader/CachedImage.cpp: (WebCore::CachedImage::usesImageContainerSize): As mentioned above, when setContainerSize() is called, the container size is only actually set if the values are non-zero. This call tells you if it was set. (WebCore::CachedImage::imageHasRelativeWidth): (WebCore::CachedImage::imageHasRelativeHeight):
  • loader/CachedImage.h:
  • platform/graphics/Image.h: (WebCore::Image::usesContainerSize):
  • svg/graphics/SVGImage.cpp: (WebCore::SVGImage::usesContainerSize):
  • svg/graphics/SVGImage.h:

LayoutTests:

Reviewed by Oliver.

Test for <rdar://problem/5616982> SVGs with width and height 100%
fail to render when used as <img> or CSS image (16167)

  • fast/images/resources/green-relative-size-rect.svg: Added.
  • fast/images/svg-as-relative-image.html: Added.
  • platform/mac/fast/images/svg-as-relative-image-expected.checksum: Added.
  • platform/mac/fast/images/svg-as-relative-image-expected.png: Added.
  • platform/mac/fast/images/svg-as-relative-image-expected.txt: Added.
23:42 Changeset [28855] by oliver@apple.com

Remove dead code due to removal of post-parse declaration discovery.

RS=Geoff.

Due to the removal of the declaration discovery pass after parsing we
no longer need any of the logic used for that discovery.

23:42 Changeset [28854] by oliver@apple.com

Replace post-parse pass to find declarations with logic in the parser itself

Reviewed by Geoff.

Instead of finding declarations in a pass following the initial parsing of
a program, we incorporate the logic directly into the parser. This lays
the groundwork for further optimisations (such as improving performance in
declaration expressions -- var x = y; -- to match that of standard assignment)
in addition to providing a 0.4% performance improvement in SunSpider.

21:04 Changeset [28853] by mrowe@apple.com

Versioning.

21:02 Changeset [28852] by mrowe@apple.com

New tag.

21:02 Changeset [28851] by mrowe@apple.com

New tag.

20:58 Changeset [28850] by mrowe@apple.com

Fix typo in hosted SunSpider page.

20:54 Changeset [28849] by sfalken@apple.com

Correct link to Visual Studio Express 2005.
Old link points to 2008.


Reviewed by Mark.

  • building/tools.html:
19:11 Changeset [28848] by darin@apple.com
  • Scripts/make-js-test-wrappers: Don't generate a wrapper for select-options-remove.js.
18:58 Changeset [28847] by ddkilzer@apple.com

2007-12-18 Xan Lopez <​xan@gnome.org>

Reviewed by Geoff.

Fix ​http://bugs.webkit.org/show_bug.cgi?id=14521
Bug 14521: JavaScriptCore fails to build on Linux/PPC gcc 4.1.2


  • wtf/TCSpinLock.h: (TCMalloc_SpinLock::Unlock):

Use less strict memory operand constraint on inline asm generation.
PLATFORM(DARWIN) left unpatched due to Apple's GCC bug.

Patch by David Kilzer <​ddkilzer@webkit.org>

18:12 Changeset [28846] by mrowe@apple.com

Remove outdated and non-functioning project files for the Apollo port. Rubber-stamped by Maciej Stachowiak and Chris Brichford.

17:41 Changeset [28845] by mitz@apple.com

Reviewed by John Sullivan.

  • remove passing test and reclassify one failing test
  • platform/win/Skipped: Removed fast/text/international/002.html.
17:12 Changeset [28844] by weinig@apple.com

Reviewed by Geoffrey.

  • http/tests/security/cross-frame-access-history-expected.txt:
  • http/tests/security/cross-frame-access-history.html:
16:58 Changeset [28843] by sfalken@apple.com

Add script to run Drosera as part of the nightly.


Reviewed by Kevin M.

  • Scripts/run-drosera.cmd: Added.
16:35 Changeset [28842] by darin@apple.com
  • Scripts/do-webcore-rename: Remove a stray K.
16:34 Changeset [28841] by darin@apple.com
  • fix Windows build
  • pcre/pcre_exec.cpp: (jsRegExpExecute): Change back from false/true to 0/1 -- I probably should not have deleted MATCH_MATCH and MATCH_NOMATCH, but I'm going to leave them out.
15:40 Changeset [28840] by darin@apple.com
  • Scripts/do-webcore-rename: Plan more renaming.
15:38 Changeset [28839] by sfalken@apple.com

<rdar://problem/5649911> REGRESSION (304-306A9): Typing 'p' in popup menu for type-to-select brings up Safari Help


Need to translate the char back into a key code for posting our WM_KEYDOWN.


Reviewed by Ada.

  • platform/win/PopupMenuWin.cpp: (WebCore::PopupWndProc):
14:56 Changeset [28838] by sfalken@apple.com

<rdar://problem/5651534> REGRESSION(r28764-r28765): GDI leak drawing text when no appropriate font is available


Our captured metafile from Uniscribe may contain multiple calls to CreateFontIndirect.
Only create a font with the last one.


Reviewed by Mitz, Darin.

  • platform/graphics/win/FontCacheWin.cpp: (WebCore::metaFileEnumProc): (WebCore::FontCache::getFontDataForCharacters):
14:32 Changeset [28837] by darin@apple.com
  • fix failing regression test
  • fast/regex/resources/TEMPLATE.html: Updated this template -- I accidentally landed the one that was copied without updating the paths.
  • fast/regex/slow.html: Regenerated.
13:50 Changeset [28836] by alice.liu@apple.com

Reviewed by Darin.

Fixed <rdar://problem/5646454> REGRESSION: dom/xhtml/level2/html/HTMLDocument12 fails on Leopard

  • fast/cookies/local-file-can-set-cookies.html: cookies are matched by domain and path, and since layout tests are file URLs, they all match on domain. setting a cookie without specifying the path will make CFNetwork pick up the path from the URL automatically. Since fast/cookies/local-file-can-set-cookies.html set a specific path of /, that cookie will be set for EVERY local file loaded in Safari. removing that component from the cookie string will fix the problem.
13:37 Changeset [28835] by alp@webkit.org

2007-12-17 Brent Fulgham <​bfulgham@gmail.com>

Reviewed by Darin.

​http://bugs.webkit.org/show_bug.cgi?id=16464
Modify WebCore to use win32 thread primitives

Updates to support native windows threading primitives
rather than pthreads emulation library.

  • WebCore.vcproj/WebCore.vcproj:
  • config.h:
  • platform/Threading.h:
  • platform/win/MutexWin.cpp: Added. (WebCore::Mutex::Mutex): (WebCore::Mutex::~Mutex): (WebCore::Mutex::lock): (WebCore::Mutex::tryLock): (WebCore::Mutex::unlock):
  • platform/win/ThreadConditionWin.cpp: Added. (WebCore::ThreadCondition::ThreadCondition): (WebCore::ThreadCondition::~ThreadCondition): (WebCore::ThreadCondition::wait): (WebCore::ThreadCondition::signal): (WebCore::ThreadCondition::broadcast):
  • platform/win/ThreadingWin.cpp: (WebCore::threadMapMutex): (WebCore::threadMap): (WebCore::storeThreadHandleByIdentifier): (WebCore::identifierByThreadHandle): (WebCore::threadHandleForIdentifier): (WebCore::clearThreadHandleForIdentifier): (WebCore::createThread): (WebCore::waitForThreadCompletion): (WebCore::detachThread): (WebCore::currentThread):
13:36 Changeset [28834] by alp@webkit.org

2007-12-18 Rodney Dawes <​dobey@wayofthemonkey.com>

Reviewed by Darin Adler.

Handle EINTR when set by select() and try the select() again
​http://bugs.webkit.org/show_bug.cgi?id=16071

  • platform/network/curl/ResourceHandleManager.cpp: (ResourceHandleManager::downloadTimerCallback):
11:30 Changeset [28833] by darin@apple.com

JavaScriptCore:

Reviewed by Geoff.

Test: fast/regex/slow.html

Slows down SunSpider a bit (about 1.01x); filed a bug to follow up on that:
​http://bugs.webkit.org/show_bug.cgi?id=16503

  • pcre/pcre.h: Changed name of error code to not specifically mention "recursion".
  • pcre/pcre_exec.cpp: (match): Replaced the depth limit, MATCH_RECURSION_LIMIT, with a total match looping limit, matchLimit. Also eliminated the constants for MATCH_MATCH and MATCH_NOMATCH, since they are just true and false (1 and 0). (jsRegExpExecute): More of the MATCH_MATCH change.

LayoutTests:

Reviewed by Geoff.

  • fast/regex/resources: Added.
  • fast/regex/resources/TEMPLATE.html: Copied from fast/js/resources/TEMPLATE.html.
  • fast/regex/resources/slow.js: Added.
  • fast/regex/slow-expected.txt: Added.
  • fast/regex/slow.html: Added.
10:58 Changeset [28832] by mitz@apple.com

Reviewed by Dave Hyatt.

  • avoid the simplified Chinese font linking code for characters that are not in any Windows code page
  • platform/graphics/win/FontCacheWin.cpp: (WebCore::FontCache::getFontDataForCharacters):
10:35 Changeset [28831] by mitz@apple.com

Reviewed by John Sullivan.

  • remove passing test
  • platform/win/Skipped: Removed fast/text/international/wrap-CJK-001.html
10:19 Changeset [28830] by beidson@apple.com

Reviewed by Adele

<rdar://problem/5525770> REGRESSION: HTTP Auth protected favicon request results in a password sheet

Some http-auth protected sites have the main resource(s) unprotected, but many subresources are
protected by authentication. Occasionally one can view the main page of a site but the favicon
is behind the iron curtain - in these cases, we should *not* prompt for a username and password
solely for the favicon.

  • loader/ResourceLoader.h: Make didReceiveAuthenticationChallenge virtual
  • loader/SubresourceLoader.cpp: (WebCore::SubresourceLoader::didReceiveAuthenticationChallenge): Now that this method is virtual from ResourceLoader, SubresourceLoader can override. First call to the SubresourceLoaderClient. If they cancel the resource load, return early. Otherwise, let ResourceLoader work its magic (resulting in the auth sheet coming down)
  • loader/SubresourceLoader.h:
  • loader/SubresourceLoaderClient.h: (WebCore::SubresourceLoaderClient::didReceiveAuthenticationChallenge):
  • loader/icon/IconLoader.cpp: (WebCore::IconLoader::didReceiveAuthenticationChallenge): Cancel the resource load, since we should never prompt the user for credentials just for a favicon.
  • loader/icon/IconLoader.h:
09:07 Changeset [28829] by sullivan@apple.com

Reviewed by Brady


  • fixed <rdar://problem/5652380> Initial prompt shows "" for databases with no user-visible name
  • storage/DatabaseTracker.cpp: (WebCore::DatabaseTracker::canEstablishDatabase): pass "name" instead of "displayName" if there's no displayName
01:27 Changeset [28828] by mjs@apple.com

Reviewed by Eric.


  • give an extra digit of precision for small differences.
  • resources/sunspider-compare-results.js:

12/17/07:

22:01 Changeset [28827] by mitz@apple.com

Reviewed by Maciej Stachowiak.

  • restore ButtonFace and ThreeDFace to their previous values for non-Mac platforms. The Mac port also uses the same values for now instead of NSColor-based ones.
  • rendering/RenderTheme.cpp: (WebCore::RenderTheme::systemColor):
21:24 Changeset [28826] by mrowe@apple.com

2007-12-17 Rodney Dawes <​dobey@wayofthemonkey.com>

Reviewed by Maciej Stachowiak.

Define WTF_USE_NPOBJECT and WTF_USE_JAVASCRIPTCORE_BINDINGS for GTK+
Add required Frame::createScriptInstanceForWidget to FrameGtk

  • config.h:
  • page/gtk/FrameGtk.cpp (Frame::createScriptInstanceForWidget):
20:39 Changeset [28825] by mjs@apple.com

SunSpider:

Reviewed by Eric.


  • work around IE layout bugs
  • hosted/sunspider.css:

WebKitSite:

Reviewed by Eric.


  • work around IE layout bugs
  • perf/sunspider-0.9/sunspider.css:
18:21 Changeset [28824] by alice.liu@apple.com

Reviewed by Adam.

Fixed <rdar://5566435> window with no scrollbars can be scrolled

  • platform/win/ScrollViewWin.cpp: (WebCore::ScrollView::maximumScroll): corrected maximumScroll() to account for scrolling not allowed. (WebCore::ScrollView::wheelEvent):
18:10 Changeset [28823] by alice.liu@apple.com

Reviewed by Adam.

Test for <rdar://problem/5566435> window with no scrollbars can still be scrolled with wheelevent

  • fast/events/attempt-scroll-with-no-scrollbars-expected.txt: Added.
  • fast/events/attempt-scroll-with-no-scrollbars.html: Added.
  • platform/mac/Skipped: Adding to mac Skipped list because of <rdar://problem/5643675> window.scrollTo scrolls a window with no scrollbars
  • platform/win/Skipped: Adding to win Skipped list because of <rdar://problem/5132009> Windows DRT does not support multiple windows
17:30 Changeset [28822] by darin@apple.com

Reviewed by Sam.

  • <rdar://4516170> Back/Forward Cache on Windows
  • WebFrame.cpp: (WebFrame::forceLayout): Implemented this. Just calling through to Frame::forceLayout matches the Mac pretty well and should be enough for now. Without this we won't repaint the new page when we go back and forward. (WebFrame::dispatchWillSubmitForm): Use COMPtr's AdoptCOM constructor instead of the adoptRef function. (WebFrame::restoreViewState): Removed unneeded notImplemented() call. This is implemented as much as it needs to be. (WebFrame::shouldGoToHistoryItem): Ditto. (WebFrame::saveViewStateToItem): Ditto. (WebFrame::canCachePage): Ditto, also changed it to return true. This is what was preventing the back/forward cache from running, and there's no need for any additional checks in the Windows version.
  • WebView.cpp: (PreferencesChangedOrRemovedObserver::onNotify): Fix backwards null check that was preventing this notification from ever being sent, which in turn meant that page cache had a capacity of 0. (WebView::initWithFrame): Removed unneeded setUsesPageCache(false) code. This actually was ineffective and ended up doing no harm, but we don't need it.
17:07 Known incompatibilities between open-source WebKit and Safari edited by ddkilzer@apple.com
(diff)
17:05 Known incompatibilities between open-source WebKit and Safari edited by ddkilzer@apple.com
(diff)
15:10 Changeset [28821] by sfalken@apple.com

Merged r28816.

14:54 Changeset [28820] by darin@apple.com
  • disabling this test until we can find a way to make it pass on diverse machines
  • fast/css/css2-system-color.html: Removed.
  • fast/css/css2-system-color.html-disabled: Copied from fast/css/css2-system-color.html.
13:31 Changeset [28819] by sfalken@apple.com

New tag.

12:54 Changeset [28818] by alp@webkit.org

2007-12-14 Juan A. Suarez Romero <​jasuarez@igalia.com>

Reviewed by Alp Toker.

​http://bugs.webkit.org/show_bug.cgi?id=16042
[GTK] Eliminate webkit_init()

Move webkit initialization to WebView class init.

12:51 Changeset [28817] by alp@webkit.org

2007-12-17 Luca Bruno <​lethalman88@gmail.com>

Reviewed by Alp Toker.

​http://bugs.webkit.org/show_bug.cgi?id=13542
gdklauncher doesnt change URL in adress GTKEntry.

Let DOMDocument observers do their work before emitting title-changed.
The load-committed signal has been added for both the view and the frame.

12:14 QtWebKitTodo edited by zander@kde.org
(diff)
11:48 Changeset [28816] by jhoneycutt@apple.com

2007-12-17 Jon Honeycutt <​jhoneycutt@apple.com>

Reviewed by Darin.

<rdar://problem/5651291> REGRESSION: Flash content doesn't display
(www.adobe.com)

We were passing NPEvent instead of NPEvent* to NPP_HandleEvent.

  • plugins/win/PluginViewWin.cpp: (WebCore::PluginViewWin::dispatchNPEvent): Changed to take NPEvent&, not NPEvent* (WebCore::PluginViewWin::paint): Pass NPEvent, not NPEvent* (WebCore::PluginViewWin::handleKeyboardEvent): same (WebCore::PluginViewWin::handleMouseEvent): same
  • plugins/win/PluginViewWin.h:
11:44 Changeset [28815] by andersca@apple.com

Fix argument to window.open.

  • http/tests/misc/window-open-then-write.html:
10:42 Changeset [28814] by mrowe@apple.com

2007-12-17 Mark Rowe <​mrowe@apple.com>

Update test results after r28812.

  • platform/mac-tiger/fast/css/css2-system-color-expected.txt:
  • platform/mac/fast/css/css2-system-color-expected.txt:
10:34 Changeset [28813] by ap@webkit.org

Reviewed by Darin.

Live to the promise of never making AppKit special character codes visible via DOM.

Test: fast/events/arrow-keys-on-body.html

  • page/EventHandler.cpp: (WebCore::EventHandler::keyEvent): Check for empty keypress characters after disambiguation, to let quirks-aware code strip special charactrers.
  • platform/mac/KeyEventMac.mm: (WebCore::PlatformKeyboardEvent::disambiguateKeyDownEvent): If not in keyboard event quirks mode, remove the text if it's a special character.
10:21 Changeset [28812] by mrowe@apple.com

2007-12-17 Mark Rowe <​mrowe@apple.com>

Reviewed by Darin Adler.

Fix incorrect array size and incorrect array index in convertNSColorToColor.

  • rendering/RenderThemeMac.mm: (WebCore::convertNSColorToColor):
09:53 Changeset [28811] by darin@apple.com

Reviewed by Mark Rowe.

  • storage/DatabaseThread.cpp: (WebCore::DatabaseThread::dispatchNextTaskIdentifier): Use a RefPtr for the database because there's no guarantee it won't lose its last reference otherwise.
09:09 Changeset [28810] by mitz@apple.com

Reviewed by Maciej Stachowiak.

  • fix <rdar://problem/5333260> Some Chinese characters in Text Encoding menu are bold, others are not and <rdar://problem/5280188> Chinese text looks worse on Safari for Windows cf. Safari for Mac
  • platform/graphics/win/FontCacheWin.cpp: (WebCore::FontCache::getFontDataForCharacters): To ensure that font linking gives consistent results for characters that are exclusive to the simplified Chinese code page and characters that belong to that code page and other code pages, always ask to map to simplified Chinese alone first.
08:18 Changeset [28809] by alp@webkit.org

2007-12-17 Christian Dywan <​christian@twotoasts.de>

Reviewed by Alp Toker.

​http://bugs.webkit.org/show_bug.cgi?id=16378
Implement Icon for Gtk

Icon provides a GdkPixbuf containing a themed icon.
The icon theme is probed for an icon name according to the
Icon Naming Specification or conventional Gnome icon names respectively.

See ​http://standards.freedesktop.org/icon-naming-spec/icon-naming-spec-latest.html

  • platform/graphics/Icon.h:
  • platform/graphics/gtk/IconGtk.cpp: (WebCore::Icon::~Icon): (WebCore::lookupIconName): (WebCore::Icon::newIconForFile): (WebCore::Icon::paint):
03:55 Changeset [28808] by mjs@apple.com

Tag SunSpider 0.9.

03:55 Changeset [28807] by mjs@apple.com

Remove bad tag

03:50 Changeset [28806] by mjs@apple.com

SunSpider:

Reviewed by Mark.


  • fix compare feature
  • hosted/sunspider-results.html:

WebKitSite:

Reviewed by Mark.


  • fix compare feature
  • perf/sunspider-0.9/sunspider-results.html:
03:23 QtWebKitTodo edited by hausmann@webkit.org
(diff)
01:44 Changeset [28805] by mjs@apple.com

Tag SunSpider 0.9.

01:44 Changeset [28804] by mjs@apple.com

Remove bad tag

01:44 Changeset [28803] by mjs@apple.com

Rubber stamped by Mark.


  • Add a copy of SunSpider 0.9 to the site
  • perf: Added.
  • perf/sunspider-0.9: Added.
  • perf/sunspider-0.9/3d-cube.html: Added.
  • perf/sunspider-0.9/3d-morph.html: Added.
  • perf/sunspider-0.9/3d-raytrace.html: Added.
  • perf/sunspider-0.9/access-binary-trees.html: Added.
  • perf/sunspider-0.9/access-fannkuch.html: Added.
  • perf/sunspider-0.9/access-nbody.html: Added.
  • perf/sunspider-0.9/access-nsieve.html: Added.
  • perf/sunspider-0.9/bitops-3bit-bits-in-byte.html: Added.
  • perf/sunspider-0.9/bitops-bits-in-byte.html: Added.
  • perf/sunspider-0.9/bitops-bitwise-and.html: Added.
  • perf/sunspider-0.9/bitops-nsieve-bits.html: Added.
  • perf/sunspider-0.9/controlflow-recursive.html: Added.
  • perf/sunspider-0.9/crypto-aes.html: Added.
  • perf/sunspider-0.9/crypto-md5.html: Added.
  • perf/sunspider-0.9/crypto-sha1.html: Added.
  • perf/sunspider-0.9/date-format-tofte.html: Added.
  • perf/sunspider-0.9/date-format-xparb.html: Added.
  • perf/sunspider-0.9/math-cordic.html: Added.
  • perf/sunspider-0.9/math-partial-sums.html: Added.
  • perf/sunspider-0.9/math-spectral-norm.html: Added.
  • perf/sunspider-0.9/regexp-dna.html: Added.
  • perf/sunspider-0.9/string-base64.html: Added.
  • perf/sunspider-0.9/string-fasta.html: Added.
  • perf/sunspider-0.9/string-tagcloud.html: Added.
  • perf/sunspider-0.9/string-unpack-code.html: Added.
  • perf/sunspider-0.9/string-validate-input.html: Added.
  • perf/sunspider-0.9/sunspider-analyze-results.js: Added.
  • perf/sunspider-0.9/sunspider-compare-results.js: Added.
  • perf/sunspider-0.9/sunspider-driver.html: Added.
  • perf/sunspider-0.9/sunspider-record-result.js: Added.
  • perf/sunspider-0.9/sunspider-results.html: Added.
  • perf/sunspider-0.9/sunspider-test-prefix.js: Added.
  • perf/sunspider-0.9/sunspider.css: Added.
  • perf/sunspider-0.9/sunspider.html: Added.
01:43 Changeset [28802] by mjs@apple.com

Tag SunSpider 0.9.

01:43 Changeset [28801] by mjs@apple.com

Reviewed by Mark.


  • Change the format of the output URL to be shorter, to meet IE constraints
  • hosted/sunspider-driver.html:
  • hosted/sunspider-results.html:
01:19 Changeset [28800] by mjs@apple.com

Tag SunSpider 0.9.

00:56 Changeset [28799] by darin@apple.com
  • speculative build fix for non-gcc platforms
  • pcre/pcre_exec.cpp: (match): Remove unused cases from return switch.

12/16/07:

23:19 Changeset [28798] by weinig@apple.com

Reviewed by Mitz.

Fix for ​http://bugs.webkit.org/show_bug.cgi?id=16466
Move the JS Location object to its own file

  • Move Location into its own file and rename it JSLocation.
  • DerivedSources.make:
  • WebCore.pro:
  • WebCore.vcproj/WebCore.vcproj:
  • WebCore.xcodeproj/project.pbxproj:
  • WebCoreSources.bkl:
  • bindings/js/JSDocumentCustom.cpp:
  • bindings/js/JSLocation.cpp: Copied from WebCore/bindings/js/kjs_window.cpp. (WebCore::): (WebCore::JSLocation::JSLocation): (WebCore::JSLocation::getValueProperty): (WebCore::JSLocation::getOwnPropertySlot): (WebCore::JSLocation::put): (WebCore::JSLocationProtoFuncReplace::callAsFunction): (WebCore::JSLocationProtoFuncReload::callAsFunction): (WebCore::JSLocationProtoFuncAssign::callAsFunction): (WebCore::JSLocationProtoFuncToString::callAsFunction):
  • bindings/js/JSLocation.h: Copied from WebCore/bindings/js/kjs_window.h. (WebCore::JSLocation::): (WebCore::JSLocation::frame): (WebCore::JSLocation::classInfo):
  • bindings/js/kjs_window.cpp: (KJS::Window::location):
  • bindings/js/kjs_window.h:
  • history/CachedPage.cpp:
22:26 Changeset [28797] by mitz@apple.com

Reviewed by Sam Weinig.

  • make 'cursor: copy' and 'cursor: none' work.

Already covered by manual-tests/cursor.html

  • rendering/RenderStyle.h: Increase the _cursor_style field to 6 bits, needed for the 33rd and 34th cursor values.
22:01 Changeset [28796] by mrowe@apple.com

Speculative build fix for non-Mac platforms.

22:00 Changeset [28795] by mrowe@apple.com

Refactor Mac plugin stream code to use the shared NetscapePlugInStreamLoader implementation.

Reviewed by Maciej Stachowiak.

20:50 Changeset [28794] by weinig@apple.com

Reviewed by Maciej.

Yet more of ​http://bugs.webkit.org/show_bug.cgi?id=16385
Cleanup kjs_window

  • Move ScheduledAction into its own file and put it in the WebCore namespace.
  • WebCore.pro:
  • WebCore.vcproj/WebCore.vcproj:
  • WebCore.xcodeproj/project.pbxproj:
  • WebCoreSources.bkl:
  • bindings/js/PausedTimeouts.cpp:
  • bindings/js/PausedTimeouts.h:
  • bindings/js/ScheduledAction.cpp: Copied from bindings/js/kjs_window.cpp. (WebCore::ScheduledAction::ScheduledAction): (WebCore::ScheduledAction::execute):
  • bindings/js/ScheduledAction.h: Copied from bindings/js/kjs_window.h. (WebCore::ScheduledAction::ScheduledAction):
  • bindings/js/kjs_window.cpp: (KJS::DOMWindowTimer::DOMWindowTimer): (KJS::DOMWindowTimer::action): (KJS::DOMWindowTimer::takeAction): (KJS::Window::installTimeout): (KJS::Window::timerFired):
  • bindings/js/kjs_window.h:
20:19 Changeset [28793] by darin@apple.com

Reviewed by Maciej.

  • ​http://bugs.webkit.org/show_bug.cgi?id=16438
  • removed some more unused code
  • changed quite a few more names to WebKit-style
  • moved more things out of pcre_internal.h
  • changed some indentation to WebKit-style
  • improved design of the functions for reading and writing 2-byte values from the opcode stream (in pcre_internal.h)
  • pcre/dftables.cpp: (main): Added the kjs prefix a normal way in lieu of using macros.
  • pcre/pcre_compile.cpp: Moved some definitions here from pcre_internal.h. (errorText): Name changes, fewer typedefs. (checkEscape): Ditto. Changed uppercase conversion to use toASCIIUpper. (isCountedRepeat): Name change. (readRepeatCounts): Name change. (firstSignificantOpcode): Got rid of the use of OP_lengths, which is very lightly used here. Hard-coded the length of OP_BRANUMBER. (firstSignificantOpcodeSkippingAssertions): Ditto. Also changed to use the advanceToEndOfBracket function. (getOthercaseRange): Name changes. (encodeUTF8): Ditto. (compileBranch): Name changes. Removed unused after_manual_callout and the code to handle it. Removed code to handle OP_ONCE since we never emit this opcode. Changed to use advanceToEndOfBracket in more places. (compileBracket): Name changes. (branchIsAnchored): Removed code to handle OP_ONCE since we never emit this opcode. (bracketIsAnchored): Name changes. (branchNeedsLineStart): More fo the same. (bracketNeedsLineStart): Ditto. (branchFindFirstAssertedCharacter): Removed OP_ONCE code. (bracketFindFirstAssertedCharacter): More of the same. (calculateCompiledPatternLengthAndFlags): Ditto. (returnError): Name changes. (jsRegExpCompile): Ditto.
  • pcre/pcre_exec.cpp: Moved some definitions here from pcre_internal.h. (matchRef): Updated names. Improved macros to use the do { } while(0) idiom so they expand to single statements rather than to blocks or multiple statements. And refeactored the recursive match macros. (MatchStack::pushNewFrame): Name changes. (getUTF8CharAndIncrementLength): Name changes. (match): Name changes. Removed the ONCE opcode. (jsRegExpExecute): Name changes.
  • pcre/pcre_internal.h: Removed quite a few unneeded includes. Rewrote quite a few comments. Removed the macros that add kjs prefixes to the functions with external linkage; instead renamed the functions. Removed the unneeded typedefs pcre_uint16, pcre_uint32, and uschar. Removed the dead and not-all-working code for LINK_SIZE values other than 2, although we aim to keep the abstraction working. Removed the OP_LENGTHS macro. (put2ByteValue): Replaces put2ByteOpcodeValueAtOffset. (get2ByteValue): Replaces get2ByteOpcodeValueAtOffset. (put2ByteValueAndAdvance): Replaces put2ByteOpcodeValueAtOffsetAndAdvance. (putLinkValueAllowZero): Replaces putOpcodeValueAtOffset; doesn't do the addition, since a comma is really no better than a plus sign. Added an assertion to catch out of range values and changed the parameter type to int rather than unsigned. (getLinkValueAllowZero): Replaces getOpcodeValueAtOffset. (putLinkValue): New function that most former callers of the putOpcodeValueAtOffset function can use; asserts the value that is being stored is non-zero and then calls putLinkValueAllowZero. (getLinkValue): Ditto. (putLinkValueAndAdvance): Replaces putOpcodeValueAtOffsetAndAdvance. No caller was using an offset, which makes sense given the advancing behavior. (putLinkValueAllowZeroAndAdvance): Ditto. (isBracketOpcode): Added. For use in an assertion. (advanceToEndOfBracket): Renamed from moveOpcodePtrPastAnyAlternateBranches, and removed comments about how it's not well designed. This function takes a pointer to the beginning of a bracket and advances to the end of the bracket.
  • pcre/pcre_tables.cpp: Updated names.
  • pcre/pcre_ucp_searchfuncs.cpp: (kjs_pcre_ucp_othercase): Ditto.
  • pcre/pcre_xclass.cpp: (getUTF8CharAndAdvancePointer): Ditto. (kjs_pcre_xclass): Ditto.
  • pcre/ucpinternal.h: Ditto.
  • wtf/ASCIICType.h: (WTF::isASCIIAlpha): Added an int overload, like the one we already have for isASCIIDigit. (WTF::isASCIIAlphanumeric): Ditto. (WTF::isASCIIHexDigit): Ditto. (WTF::isASCIILower): Ditto. (WTF::isASCIISpace): Ditto. (WTF::toASCIILower): Ditto. (WTF::toASCIIUpper): Ditto.
20:08 Changeset [28792] by alp@webkit.org

2007-12-16 Alp Toker <​alp@atoker.com>

Reviewed by Maciej.

​http://bugs.webkit.org/show_bug.cgi?id=16356
[GTK] Integrate GStreamer video with the graphics backend

Integrate the GStreamer media backend with the Cairo graphics backend.
There are still some issues: Data is copied more often than necessary,
and repaint() is not called, causing transformed video not to update
sometimes.

  • WebCore.pro:
  • platform/graphics/gtk/MediaPlayerPrivateGStreamer.cpp: (WebCore::MediaPlayerPrivate::MediaPlayerPrivate): (WebCore::MediaPlayerPrivate::~MediaPlayerPrivate): (WebCore::MediaPlayerPrivate::currentTime): (WebCore::MediaPlayerPrivate::setEndTime): (WebCore::MediaPlayerPrivate::seeking): (WebCore::MediaPlayerPrivate::naturalSize): (WebCore::MediaPlayerPrivate::setMuted): (WebCore::MediaPlayerPrivate::setRect): (WebCore::MediaPlayerPrivate::setVisible): (WebCore::MediaPlayerPrivate::repaint): (WebCore::MediaPlayerPrivate::paint): (WebCore::MediaPlayerPrivate::createGSTPlayBin):
  • platform/graphics/gtk/MediaPlayerPrivateGStreamer.h:
  • platform/graphics/gtk/VideoSinkGStreamer.cpp: Added. (webkit_video_sink_base_init): (webkit_video_sink_init): (webkit_video_sink_idle_func): (webkit_video_sink_render): (webkit_video_sink_set_caps): (webkit_video_sink_dispose): (webkit_video_sink_finalize): (webkit_video_sink_set_property): (webkit_video_sink_get_property): (webkit_video_sink_stop): (webkit_video_sink_class_init): (webkit_video_sink_new): (webkit_video_sink_set_surface): (plugin_init):
  • platform/graphics/gtk/VideoSinkGStreamer.h: Added.
18:09 Changeset [28791] by mrowe@apple.com

2007-12-16 Mark Rowe <​mrowe@apple.com>

Mac build fix.

  • WebCore.xcodeproj/project.pbxproj: Remove NetscapePlugInStreamLoader.cpp from the WebCore target. Mac currently has its own implementation in NetscapePlugInStreamLoaderMac.mm which conflicts with the new shared implementation.
17:50 Changeset [28790] by mrowe@apple.com

2007-12-16 Rodney Dawes <​dobey@wayofthemonkey.com>

Reviewed by Maciej Stachowiak.

​http://bugs.webkit.org/show_bug.cgi?id=16389
Bug 16389: Common Implementation of NetscapePlugInStreamLoader

  • WebCore.vcproj/WebCore.vcproj: Remove NetscapePlugInStreamLoaderWin.cpp.
  • loader/NetscapePlugInStreamLoader.cpp: Copy method implementations from NetscapePlugInStreamLoaderWin.cpp.
  • loader/win/NetscapePlugInStreamLoaderWin.cpp: Removed.
17:35 Changeset [28789] by mrowe@apple.com

2007-12-16 Dimitri Glazkov <​dimitri@glazkov.com>

Reviewed by Adam Roben.

Remove the double-quotes around the PATH variable value, in order to make it work in Windows shell.

  • FindSafari/FindSafari.cpp: (_tmain):
17:33 Changeset [28788] by mrowe@apple.com

2007-12-16 Brent Fulgham <​bfulgham@gmail.com>

Reviewed by Adam Roben.

​http://bugs.webkit.org/show_bug.cgi?id=16315
FindSafari needs a path-only option.

  • FindSafari/FindSafari.cpp: (_tmain):
17:31 Changeset [28787] by darin@apple.com
  • fast/js/regexp-overflow-expected.txt: Updated. I accidentally checked in an out-of-date file last time.
17:28 Changeset [28786] by darin@apple.com

Reviewed by Darin Adler.

Fix ​http://bugs.webkit.org/show_bug.cgi?id=16433.
Bug 16433: LOW_BANDWIDTH_DISPLAY build is broken

  • dom/Document.cpp: (WebCore::Document::Document):
  • loader/FrameLoader.cpp: (WebCore::FrameLoader::addLowBandwidthDisplayRequest): (WebCore::FrameLoader::switchOutLowBandwidthDisplayIfReady):
17:26 Changeset [28785] by darin@apple.com

JavaScriptCore:

Reviewed by Maciej.

The problem was that \B was not handled properly in character classes.

Test: fast/js/regexp-overflow.html

  • pcre/pcre_compile.cpp: (check_escape): Added handling of ESC_b and ESC_B in character classes here. Allows us to get rid of the handling of \b in character classes from all the call sites that handle it separately and to handle \B properly as well. (compileBranch): Remove the ESC_b handling, since it's not needed any more. (calculateCompiledPatternLengthAndFlags): Ditto.

LayoutTests:

Reviewed by Maciej.

  • fast/js/regexp-overflow-expected.txt: Updated.
  • fast/js/resources/regexp-overflow.js: Added test cases.
17:26 Changeset [28784] by mrowe@apple.com

2007-12-16 Grace Kloba <​klobag@gmail.com>

Reviewed by Darin Adler.

Fix ​http://bugs.webkit.org/show_bug.cgi?id=16433.
Bug 16433: LOW_BANDWIDTH_DISPLAY build is broken

  • dom/Document.cpp: (WebCore::Document::Document):
  • loader/FrameLoader.cpp: (WebCore::FrameLoader::addLowBandwidthDisplayRequest): (WebCore::FrameLoader::switchOutLowBandwidthDisplayIfReady):
17:20 Changeset [28783] by mrowe@apple.com

2007-12-16 Brent Fulgham <​bfulgham@gmail.com>

Reviewed by Maciej Stachowiak.

​http://bugs.webkit.org/show_bug.cgi?id=16446.

  • building/debug.html: Update to expand the win32 debugging instructions.
17:14 Changeset [28782] by mrowe@apple.com

Fix ​http://bugs.webkit.org/show_bug.cgi?id=16448 ([GTK] Celtic Kane JavaScript performance on Array test is slow relative to Mac).

Reviewed by Maciej Stachowiak.

  • kjs/array_instance.cpp:

(KJS::compareByStringPairForQSort):
(KJS::ArrayInstance::sort): Convert JSValue's to strings once up front and then sort the
results. This avoids calling toString twice per comparison, but requires a temporary buffer
so we only use this approach in cases where the array being sorted is not too large.

16:24 Changeset [28781] by darin@apple.com

WebCore:

Reviewed by Darin.

More of ​http://bugs.webkit.org/show_bug.cgi?id=16385
Cleanup kjs_window

  • Move PausedTimeouts into its own file and put it in the WebCore namespace.
  • WebCore.pro:
  • WebCore.vcproj/WebCore.vcproj:
  • WebCore.xcodeproj/project.pbxproj:
  • WebCoreSources.bkl:
  • bindings/js/PausedTimeouts.cpp: Copied from bindings/js/kjs_window.cpp.
  • bindings/js/PausedTimeouts.h: Copied from bindings/js/kjs_window.h.
  • bindings/js/kjs_window.cpp: (KJS::Window::pauseTimeouts):
  • bindings/js/kjs_window.h:
  • history/CachedPage.cpp:
  • history/CachedPage.h:
  • page/Chrome.cpp:

LayoutTests:

Reviewed by Maciej.

  • tests for the argument handling of the executeSql function
  • storage/execute-sql-args-expected.txt: Added.
  • storage/execute-sql-args.html: Added.
16:17 Changeset [28780] by darin@apple.com
  • try to get the system color test passing on the buildbot

I fear this test may be vulnerable to slight platform differences.

  • platform/mac-tiger/fast: Added.
  • platform/mac-tiger/fast/css: Added.
  • platform/mac-tiger/fast/css/css2-system-color-expected.checksum: Copied from platform/mac/fast/css/css2-system-color-expected.checksum.
  • platform/mac-tiger/fast/css/css2-system-color-expected.png: Copied from platform/mac/fast/css/css2-system-color-expected.png.
  • platform/mac-tiger/fast/css/css2-system-color-expected.txt: Copied from platform/mac/fast/css/css2-system-color-expected.txt. Then updated to match what the buildbot says.
  • platform/mac/fast/css/css2-system-color-expected.txt: Updated to match results I'm getting locally on my Leopard machine.
15:59 Changeset [28779] by weinig@apple.com

Reviewed by Darin.

More of ​http://bugs.webkit.org/show_bug.cgi?id=16385
Cleanup kjs_window

  • Move PausedTimeouts into its own file and put it in the WebCore namespace.
  • WebCore.pro:
  • WebCore.vcproj/WebCore.vcproj:
  • WebCore.xcodeproj/project.pbxproj:
  • WebCoreSources.bkl:
  • bindings/js/PausedTimeouts.cpp: Copied from bindings/js/kjs_window.cpp.
  • bindings/js/PausedTimeouts.h: Copied from bindings/js/kjs_window.h.
  • bindings/js/kjs_window.cpp: (KJS::Window::pauseTimeouts):
  • bindings/js/kjs_window.h:
  • history/CachedPage.cpp:
  • history/CachedPage.h:
  • page/Chrome.cpp:
15:08 Changeset [28778] by bdakin@apple.com

WebCore:

Reviewed by Geoff.

Make relative-size SVGs work in border-image.

  • rendering/RenderBox.cpp: (WebCore::RenderBox::calculateBackgroundSize):
  • rendering/RenderObject.cpp: (WebCore::RenderObject::paintBorderImage):

LayoutTests:

Reviewed by Geoff.

Tests for SVG as border-image.

  • fast/borders/resources/green-fixed-size-rect.svg: Added.
  • fast/borders/resources/green-relative-size-rect.svg: Added.
  • fast/borders/svg-as-border-image-2.html: Added.
  • fast/borders/svg-as-border-image.html: Added.
  • platform/mac/fast/borders/svg-as-border-image-2-expected.checksum: Added.
  • platform/mac/fast/borders/svg-as-border-image-2-expected.png: Added.
  • platform/mac/fast/borders/svg-as-border-image-2-expected.txt: Added.
  • platform/mac/fast/borders/svg-as-border-image-expected.checksum: Added.
  • platform/mac/fast/borders/svg-as-border-image-expected.png: Added.
  • platform/mac/fast/borders/svg-as-border-image-expected.txt: Added.
14:53 Changeset [28777] by ggaren@apple.com

Reviewed by Darin Adler and Maciej Stachowiak.


More refactoring to support global variable optimization.


Changed SymbolTable to use RefPtr<UString::Rep> as its key instead of
UString::Rep*. With globals, the symbol table can outlast the
declaration node for any given symbol, so the symbol table needs to ref
its symbol names.


In support, specialized HashMaps with RefPtr keys to allow lookup
via raw pointer, avoiding refcount churn.


SunSpider reports a .6% speedup (prolly just noise).

  • kjs/JSVariableObject.cpp: (KJS::JSVariableObject::getPropertyNames): Symbol table keys are RefPtrs now.
  • kjs/SymbolTable.h: Modified key traits to match RefPtr. Added a static Rep* for null, which helps compute the deletedValue() trait.
  • wtf/HashMap.h: #include the RefPtr specialization so everyone can use it.
  • wtf/RefPtrHashMap.h: Copied from wtf/HashMap.h. Added overloaded versions of find(), contains(), get(), set(), add(), remove(), and take() that take raw pointers as keys.
12:43 QtWebKitTodo edited by hausmann@webkit.org
(diff)
12:14 Changeset [28776] by darin@apple.com
  • fix Tiger build (my fault it was broken)
  • rendering/RenderThemeMac.mm: Define NSUInteger if on Tiger.
11:26 Changeset [28775] by darin@apple.com

WebCore:

Reviewed by Darin.


​http://bugs.webkit.org/show_bug.cgi?id=6129
Incomplete implementation of CSS 2.1 system colors

Test: fast/css/css2-system-color.html


Based on original patch by Rob Buis.


System colors are retrieved from NSColor as appropriate. If the color is a pattern color
(and therefore NSColor won't let us retrieve a color from it) we draw a 1x1 image of the
color and sample that to get a solid color.

  • css/CSSStyleSelector.cpp: (WebCore::): (WebCore::colorForCSSValue):
  • rendering/RenderTheme.cpp: (WebCore::RenderTheme::systemColor):
  • rendering/RenderTheme.h:
  • rendering/RenderThemeMac.h:
  • rendering/RenderThemeMac.mm: (WebCore::getSystemColor): (WebCore::RenderThemeMac::platformColorsDidChange): (WebCore::RenderThemeMac::systemColor):

LayoutTests:

Reviewed by Darin.


​http://bugs.webkit.org/show_bug.cgi?id=6129
Incomplete implementation of CSS 2.1 system colors


Add layout test for this bug and update tests for new system-derived colors

  • fast/css/css2-system-color.html: Added.
  • platform/mac/fast/css/css2-system-color-expected.checksum: Added.
  • platform/mac/fast/css/css2-system-color-expected.png: Added.
  • platform/mac/fast/css/css2-system-color-expected.txt: Added.
  • platform/mac/fast/css/css2-system-fonts-expected.checksum:
  • platform/mac/tables/mozilla/bugs/bug103533-expected.txt:
10:40 Changeset [28774] by mitz@apple.com

WebCore:

Reviewed by Darin Adler.

Test: fast/overflow/scrollbar-position-update.html

  • platform/mac/PlatformScrollBarMac.mm: (WebCore::PlatformScrollbar::updateThumbProportion): Update the NSScroller's value for the new proportions.
  • rendering/RenderLayer.cpp: (WebCore::RenderLayer::updateScrollInfoAfterLayout): Removed unnecessary repaint(). Scrollbars repaint themselves as needed.

LayoutTests:

Reviewed by Darin Adler.

  • fast/overflow/scrollbar-position-update.html: Added.
  • platform/mac-leopard/fast/overflow: Added.
  • platform/mac-leopard/fast/overflow/scrollbar-position-update-expected.checksum: Added.
  • platform/mac-leopard/fast/overflow/scrollbar-position-update-expected.png: Added.
  • platform/mac/fast/overflow/scrollbar-position-update-expected.txt: Added.
10:13 Changeset [28773] by darin@apple.com
  • try to fix the build
  • WebView.cpp: (wxWebView::OnKeyEvents): Add a WebCore prefix.
09:37 Changeset [28772] by ap@webkit.org

Reviewed by Darin.

​http://bugs.webkit.org/show_bug.cgi?id=14140
<rdar://problem/5270958> REGRESSION: Complex system KeyBindings don't work properly

WebCore:

  • dom/KeyboardEvent.h: (WebCore::KeypressCommand::KeypressCommand): (WebCore::KeyboardEvent::keypressCommands): Change stored command class to preserve complete information about commands.
  • editing/EditorCommand.cpp: (WebCore::CommandEntry::): Mark InsertText as a text insertion command, which it is. Previously, we couldn't do it because WebKit didn't really treat insertText: as a command.
  • page/EventHandler.cpp: (WebCore::EventHandler::keyEvent): Copy commands saved while interpreting a keydown event into keypress, to avoid losing state when running interpretKeyEvents: again.

WebKit:

  • WebView/WebHTMLView.mm: (-[WebHTMLView _interceptEditingKeyEvent:shouldSaveCommand:]): Made command replaying work when handling keypress, too. (-[WebHTMLView doCommandBySelector:]): Adapted for the new way to store commands in events. (-[WebHTMLView insertText:]): Append a command, not replace the whole existing vector. Also, restore the state for additional commands to be saved correctly.
09:31 Changeset [28771] by ap@webkit.org

Reviewed by Darin.

​http://bugs.webkit.org/show_bug.cgi?id=16162
Problems with float parsing on Linux (locale-dependent parsing was used).

  • kjs/dtoa.cpp: Removed USE_LOCALE to reduce future confusion.
  • kjs/lexer.cpp: (KJS::Lexer::lex): Parse with kjs_strtod, not the system one.
09:23 QtWebKitTodo edited by zander@kde.org
(diff)
08:16 Changeset [28770] by ap@webkit.org

Reviewed by Darin.

​http://bugs.webkit.org/show_bug.cgi?id=16462
REGRESSION: access keys broken on Windows

05:20 Changeset [28769] by alp@webkit.org

2007-12-16 Xan Lopez <​xan@gnome.org>

Reviewed by Alexey Proskuryakov.

​http://bugs.webkit.org/show_bug.cgi?id=16454
[GTK] Text input doesn't work consistently on PPC

  • platform/gtk/KeyEventGtk.cpp: (WebCore::singleCharacterString):

UChar is 2 bytes (UTF-16), so transform accordingly from
gunichar (UCS-4). Fixes keyboard input on big endian systems.

12/15/07:

22:57 Changeset [28768] by weinig@apple.com

Reviewed by Mark Rowe.

Fix urlSuitableForTestResult to correctly identify wstring::npos as
the case when wstring.find doesn't find anything.

  • DumpRenderTree/win/DumpRenderTree.cpp: (urlSuitableForTestResult):
18:00 Changeset [28767] by darin@apple.com

Reviewed by Mark Rowe.

  • better build fix for the problem affecting GTK and some other platforms "this time for sure"
  • WebCore.xcodeproj/project.pbxproj:
  • bindings/js/JSEventTargetBase.cpp: Move the include of the .lut.h file here.
  • bindings/js/JSEventTargetBase.h: Instead of including the .lut.h file in the header, which won't work, declare the tables that are in the .lut.h file in the header.
17:54 Changeset [28766] by kevino@webkit.org

Fixed the Wx port's FontPlatformData for use in HashTable.

17:33 Changeset [28765] by mitz@apple.com

Reviewed by Darin Adler.

  • fix <rdar://problem/5636090> Text in Georgia, Armenian, Inuktitut, Cree, or Cherokee (KA, HY, IU, CR, CHR) draws as all missing glyphs
  • platform/graphics/win/FontCacheWin.cpp: (WebCore::metaFileEnumProc): Added. Called during metafile record enumeration and extracts the font from the create font record. (WebCore::FontCache::getFontDataForCharacters): If font linking fails, let Uniscribe draw the characters and see what font it chooses.
17:31 Changeset [28764] by darin@apple.com
  • WebCore.pro: Roll my last change out. Was wrong and didn't fix the build.
17:30 Changeset [28763] by darin@apple.com

Disabled the smallFonts.svg test, which has been failing for weeks. I heard that Niko
is going to fix this soon. He should rename it back at that time.

Removed the svn:executable flag from many .svg, .png, and .css files that had it set
for no good reason.

  • svg/batik/text/smallFonts.svg: Removed.
  • svg/batik/text/smallFonts.svg-disabled: Copied from svg/batik/text/smallFonts.svg.
17:11 Changeset [28762] by darin@apple.com

Another try at a GTK build fix.

  • WebCore.pro: Add JSEventTargetBase.cpp to LUT_TABLE_FILES instead of LUT_FILES.
16:07 Changeset [28761] by weinig@apple.com

Force windows to regenerate COM DOM bindings.

  • WebCore.vcproj/build-generated-files.sh: Add license.
  • bindings/scripts/CodeGeneratorCOM.pm: Use shared WK_ucfirst.
15:32 Changeset [28760] by weinig@apple.com

Fix Windows and wx builds.

  • WebCore.vcproj/WebCore.vcproj:
  • WebCoreSources.bkl:
15:31 Changeset [28759] by mrowe@apple.com

Gtk build fix. Add JSEventTargetBase.cpp to SOURCES.

15:24 Changeset [28758] by weinig@apple.com

Fix wx build.

  • WebCoreSources.bkl:
15:23 Changeset [28757] by zimmermann@webkit.org

Not reviewed. Build fix for Qt/Gtk.

14:48 Changeset [28756] by zimmermann@webkit.org

Reviewed by Eric.

Fixes: ​http://bugs.webkit.org/show_bug.cgi?id=16445 (Refactor EventTargetNode & JSEventTargetNode for an upcoming SVG patch)

Split up JSEventTargetNode in JSEventTargetNode & JSEventTargetBase - where most functionality has been moved down
in the base class. Applied the same refactorization to EventTargetNode.

This makes it possible for the upcoming patch implementing the EventTarget interface for SVGElementInstance
to share as much code as possible with the EventTargetNode classes.

12:03 Changeset [28755] by eric@webkit.org

2007-12-15 Eric Seidel <​eric@webkit.org>

Reviewed by Oliver.

Width of SVG elements not applied to container elements
​http://bugs.webkit.org/show_bug.cgi?id=16305
Added calcPrefWidths for RenderSVGRoot (copied from RenderReplaced)
This is caused by poor factoring in the render tree. RenderContainer probably should be a template.

svg/css/css-box-min-width.html

  • rendering/RenderSVGRoot.cpp: (WebCore::RenderSVGRoot::calcPrefWidths): Added.
  • rendering/RenderSVGRoot.h:
  • css/SVGCSSStyleSelector::applySVGProperty: fixed ASSERT in debug builds
06:46 Changeset [28754] by ddkilzer@apple.com

Fix comment after isSafeScript() was renamed to allowsAccessFrom().

  • bindings/js/kjs_window.cpp: (KJS::createWindow):
06:13 Changeset [28753] by alp@webkit.org

2007-12-15 Alp Toker <​alp@atoker.com>

GTK+ DRT build fix for GLib < 2.14.

  • DumpRenderTree/gtk/LayoutTestControllerGtk.cpp: (LayoutTestController::setWaitToDump):
05:40 Changeset [28752] by alp@webkit.org

2007-12-15 Alp Toker <​alp@atoker.com>

​http://bugs.webkit.org/show_bug.cgi?id=16449
cairo_arc() functions hang or crash when passed inf as radius or start/end angle

Fixed in r28751. Remove these tests from the skip list:

fast/canvas/arc-crash.html
fast/canvas/canvas-with-incorrect-args.html

  • platform/gtk/Skipped:
05:12 Changeset [28751] by alp@webkit.org

2007-12-15 Alp Toker <​alp@atoker.com>

Reviewed by Mark Rowe.

​http://bugs.webkit.org/show_bug.cgi?id=16449
cairo_arc() functions hang or crash when passed inf as radius or start/end angle

Add checks. This matches a similar workaround for a CG bug in the CG
graphics backend: <rdar://problem/5189233>

Fixes:

fast/canvas/arc-crash.html
fast/canvas/canvas-with-incorrect-args.html

  • platform/graphics/cairo/PathCairo.cpp: (WebCore::Path::addArc):
04:39 Changeset [28750] by mrowe@apple.com

Add a watchdog timer to Gtk DumpRenderTree, and implement alert/prompt/confirm.
This prevents many layout tests from hanging while waiting on user responses to dialogs.

Reviewed by Alp Toker.

04:39 Changeset [28749] by mrowe@apple.com

Add a skeleton Gtk skipped list containing only tests that hang run-webkit-tests.

03:00 Changeset [28748] by mrowe@apple.com

Flesh out DumpRenderTree for Gtk. After these changes, the majority of the tests in fast/js pass.

Reviewed by Alp Toker.

01:30 Changeset [28747] by ap@webkit.org

Correct a mistake in ChangeLog.

01:07 Changeset [28746] by ap@webkit.org

Reviewed by Maciej.

​http://bugs.webkit.org/show_bug.cgi?id=16436
Alt+Space works incorrectly

Test: platform/win/fast/events/alt-space-scroll.html

  • WebView.cpp: (WebView::keyDown): (WebView::keyPress): We do not handle WM_SYSCHAR events.
01:03 Changeset [28745] by ap@webkit.org

Reviewed by Darin.

​http://bugs.webkit.org/show_bug.cgi?id=16078
Google Maps zooming via the scroll wheel (almost) always zooms IN only.

  • dom/WheelEvent.cpp: (WebCore::WheelEvent::WheelEvent): Ensure that delta is never rounded down to zero - we are getting values less than 1 from many mice on OS X, and Google Maps code assumes scrolling down if event.wheelDelta is zero.

12/14/07:

22:24 Changeset [28744] by alp@webkit.org

2007-12-14 Alp Toker <​alp@atoker.com>

Reviewed by Mark Rowe.

Enable the AllInOneFile.cpp optimization for the GTK+ port.

20:35 Changeset [28743] by mrowe@apple.com

Remove commented out fprintf's that were for debugging purposes only.

20:31 Changeset [28742] by mrowe@apple.com

Don't use the MADV_DONTNEED code path for now as it has no effect on Mac OS X and is currently untested on other platforms.

Reviewed by Maciej Stachowiak.

  • wtf/TCSystemAlloc.cpp:

(TCMalloc_SystemRelease): Return after releasing memory rather than potentially falling
through into another mechanism if multiple are supported.

20:21 Changeset [28741] by alp@webkit.org

2007-12-14 Alp Toker <​alp@atoker.com>

Build fix for GTK+/Qt and ports that don't use AllInOneFile.cpp.

Include UnusedParam.h.

19:53 Changeset [28740] by oliver@apple.com

Fix build on windows

Reviewd by Stephanie

  • wtf/FastMalloc.cpp:

(WTF::TCMalloc_PageHeap::IncrementalScavenge):

19:26 Changeset [28739] by mitz@apple.com
  • Windows build fix
  • WebCore.vcproj/WebCore.vcproj:
19:02 Changeset [28738] by mitz@apple.com
  • yet another build fix
  • DOMHTMLClasses.cpp: (DOMHTMLDocument::URL):
19:00 Changeset [28737] by alp@webkit.org

2007-12-14 Alp Toker <​alp@atoker.com>

Fix inconsistent indentation in the license header. Maciej was right!

  • WebCoreSupport/FrameLoaderClientGtk.cpp:
18:51 Changeset [28736] by alp@webkit.org

2007-12-14 Alp Toker <​alp@atoker.com>

GTK+/Qt build fix. Track added files in r28722.

  • WebCore.pro:
18:16 Changeset [28735] by ddkilzer@apple.com

<rdar://problem/5647272> Remove user agent string hack for flickr.com

Reviewed by Darin.

  • WebView/WebView.mm: (-[WebView _userAgentForURL:]): Removed hack.
18:15 Changeset [28734] by ddkilzer@apple.com

<rdar://problem/5647261> Remove user agent string hack for yahoo.com

Reviewed by Darin.

  • WebView/WebView.mm: (-[WebView _userAgentForURL:]): Removed hack.
18:09 Changeset [28733] by mitz@apple.com
  • try again to fix the Windows build
  • wtf/TCSystemAlloc.cpp: (TCMalloc_SystemRelease):
18:00 Changeset [28732] by mitz@apple.com
  • try to fix the Windows build
  • wtf/TCSystemAlloc.cpp: (TCMalloc_SystemRelease):
17:35 Changeset [28731] by mitz@apple.com
  • build fix
  • WebIconDatabase.h:
17:19 Changeset [28730] by weinig@apple.com

Fix ChangeLog

17:17 Changeset [28729] by weinig@apple.com

Reviewed by Geoff.

Speed up getElementByClassName.

  • platform/text/StringImpl.cpp: (WebCore::StringImpl::foldCase): Optimize the case when all the characters are ASCII.
17:03 Changeset [28728] by alp@webkit.org

2007-12-14 Alp Toker <​alp@atoker.com>

Reviewed by Maciej.

​http://bugs.webkit.org/show_bug.cgi?id=16432
[GTK] Update license headers

Consent has been given by the authors of these files to change license
to the LGPL as outlined in the bug report.

16:25 Changeset [28727] by oliver@apple.com

Add logic for TCMalloc to release memory to the system

Reviewed by Maciej and Oliver.

Add final changes to make TCMalloc release memory to the system.
This results in a 0.4% regression against ToT, but this is offset
against the gains made by the original TCMalloc r38 merge - in fact
we retain around 0.3-0.4% progression overall.

15:42 Changeset [28726] by darin@apple.com

Reviewed by Adele.

  • loader/MainResourceLoader.cpp: (WebCore::MainResourceLoader::willSendRequest): Removed call to checkNavigationPolicy. That's handled by FrameLoader.
  • loader/MainResourceLoader.h: Removed callContinueAfterNavigationPolicy and continueAfterNavigationPolicy.
15:41 Changeset [28725] by darin@apple.com

Reviewed by Adele.

  • DOMHTMLClasses.cpp: (DOMHTMLDocument::URL): Implemented.
  • WebActionPropertyBag.cpp: (WebActionPropertyBag::Read): Fixed the WebActionElementKey case.
15:06 Changeset [28724] by mitz@apple.com
  • share the results of this test between Tiger and Leopard
  • platform/mac-leopard/fast/text/international/bidi-control-chars-treated-as-ZWS-expected.txt: Removed.
  • platform/mac/fast/text/international/bidi-control-chars-treated-as-ZWS-expected.txt: Copied from LayoutTests/platform/mac-leopard/fast/text/international/bidi-control-chars-treated-as-ZWS-expected.txt.
14:06 Changeset [28723] by andersca@apple.com

WebCore:

Reviewed by Brady.

Make document.open count as committing a document load, so that -[WebFrame dataSource:] won't
return nil in that case.


  • loader/FrameLoader.cpp: (WebCore::FrameLoader::didExplicitOpen):

WebKitTools:

Reviewed by Brady.

Assert that a web frame that's loading a resource always has either a data source or
a provisional data source.


  • DumpRenderTree/mac/ResourceLoadDelegate.mm: (-[ResourceLoadDelegate webView:identifierForInitialRequest:fromDataSource:]):

LayoutTests:

Reviewed by Brady.

Add test that calls window.open and then immediately writes into the document
of the newly opened window.


  • http/tests/misc/resources/compass.jpg: Added.
  • http/tests/misc/window-open-then-write-expected.txt: Added.
  • http/tests/misc/window-open-then-write.html: Added.
  • platform/win/Skipped:
13:48 Changeset [28722] by weinig@apple.com

WebCore:

Reviewed by Mitz.

This patch also renames AtomicStringList to ClassNames to better reflect its actual use,
and takes advantage of admitting that it's class-specific to encapsulate class attribute
parsing so it can be shared. It also changes the class to use a Vector, rather than a linked
list to store the class names.

Tests: fast/dom/getElementsByClassName/001.html

fast/dom/getElementsByClassName/002.html
fast/dom/getElementsByClassName/003.html
fast/dom/getElementsByClassName/004.html
fast/dom/getElementsByClassName/005.html
fast/dom/getElementsByClassName/006.html
fast/dom/getElementsByClassName/007.html
fast/dom/getElementsByClassName/008.html
fast/dom/getElementsByClassName/009.html
fast/dom/getElementsByClassName/010.xml
fast/dom/getElementsByClassName/011.xml
fast/dom/getElementsByClassName/012.html
fast/dom/getElementsByClassName/013.html
fast/dom/getElementsByClassName/014.html
fast/dom/getElementsByClassName/array/001.html
fast/dom/getElementsByClassName/array/002.html
fast/dom/getElementsByClassName/array/003.html
fast/dom/getElementsByClassName/array/004.html
fast/dom/getElementsByClassName/dumpNodeList.html

  • WebCore.xcodeproj/project.pbxproj:
  • css/CSSStyleSelector.cpp: (WebCore::CSSStyleSelector::matchRules): (WebCore::CSSStyleSelector::checkOneSelector):
  • dom/AtomicStringList.h: Removed.
  • dom/ClassNames.cpp: Added. (WebCore::ClassNames::contains): (WebCore::ClassNames::parseClassAttribute):
  • dom/ClassNames.h: Copied from WebCore/dom/AtomicStringList.h. (WebCore::ClassNames::ClassNames): (WebCore::ClassNames::size): (WebCore::ClassNames::clear): (WebCore::ClassNames::operator[]): (WebCore::isClassWhitespace):
  • dom/ClassNodeList.cpp: Added. (WebCore::ClassNodeList::ClassNodeList): (WebCore::ClassNodeList::length): (WebCore::ClassNodeList::item): (WebCore::ClassNodeList::nodeMatches):
  • dom/ClassNodeList.h: Added.
  • dom/Document.cpp: (WebCore::Document::getElementsByName): (WebCore::Document::getElementsByClassName):
  • dom/Document.h:
  • dom/Document.idl:
  • dom/Element.cpp: (WebCore::Element::getClassNames): (WebCore::Element::getElementsByClassName):
  • dom/Element.h:
  • dom/Element.idl:
  • dom/NameNodeList.cpp: (WebCore::NameNodeList::NameNodeList): (WebCore::NameNodeList::item): (WebCore::NameNodeList::nodeMatches):
  • dom/NameNodeList.h:
  • dom/NamedMappedAttrMap.cpp: (WebCore::NamedMappedAttrMap::clearAttributes): (WebCore::NamedMappedAttrMap::parseClassAttribute):
  • dom/NamedMappedAttrMap.h: (WebCore::NamedMappedAttrMap::getClassNames):
  • dom/StyledElement.cpp: (WebCore::StyledElement::getClassNames):
  • dom/StyledElement.h:

LayoutTests:

Reviewed by Mitz.

Tests added from ​http://tc.labs.opera.com/apis/getElementsByClassName/

  • fast/dom/Window/window-properties-expected.txt:
  • fast/dom/getElementsByClassName: Added.
  • fast/dom/getElementsByClassName/001-expected.txt: Added.
  • fast/dom/getElementsByClassName/001.html: Added.
  • fast/dom/getElementsByClassName/002-expected.txt: Added.
  • fast/dom/getElementsByClassName/002.html: Added.
  • fast/dom/getElementsByClassName/003-expected.txt: Added.
  • fast/dom/getElementsByClassName/003.html: Added.
  • fast/dom/getElementsByClassName/004-expected.txt: Added.
  • fast/dom/getElementsByClassName/004.html: Added.
  • fast/dom/getElementsByClassName/005-expected.txt: Added.
  • fast/dom/getElementsByClassName/005.html: Added.
  • fast/dom/getElementsByClassName/006-expected.txt: Added.
  • fast/dom/getElementsByClassName/006.html: Added.
  • fast/dom/getElementsByClassName/007-expected.txt: Added.
  • fast/dom/getElementsByClassName/007.html: Added.
  • fast/dom/getElementsByClassName/008-expected.txt: Added.
  • fast/dom/getElementsByClassName/008.html: Added.
  • fast/dom/getElementsByClassName/009-expected.txt: Added.
  • fast/dom/getElementsByClassName/009.html: Added.
  • fast/dom/getElementsByClassName/010-expected.txt: Added.
  • fast/dom/getElementsByClassName/010.xml: Added.
  • fast/dom/getElementsByClassName/011-expected.txt: Added.
  • fast/dom/getElementsByClassName/011.xml: Added.
  • fast/dom/getElementsByClassName/012-expected.txt: Added.
  • fast/dom/getElementsByClassName/012.html: Added.
  • fast/dom/getElementsByClassName/013-expected.txt: Added.
  • fast/dom/getElementsByClassName/013.html: Added.
  • fast/dom/getElementsByClassName/014-expected.txt: Added.
  • fast/dom/getElementsByClassName/014.html: Added.
  • fast/dom/getElementsByClassName/array: Added.
  • fast/dom/getElementsByClassName/array/001-expected.txt: Added.
  • fast/dom/getElementsByClassName/array/001.html: Added.
  • fast/dom/getElementsByClassName/array/002-expected.txt: Added.
  • fast/dom/getElementsByClassName/array/002.html: Added.
  • fast/dom/getElementsByClassName/array/003-expected.txt: Added.
  • fast/dom/getElementsByClassName/array/003.html: Added.
  • fast/dom/getElementsByClassName/array/004-expected.txt: Added.
  • fast/dom/getElementsByClassName/array/004.html: Added.
  • fast/dom/getElementsByClassName/dumpNodeList-expected.txt: Added.
  • fast/dom/getElementsByClassName/dumpNodeList.html: Added.
  • fast/dom/getElementsByClassName/resources: Added.
  • fast/dom/getElementsByClassName/resources/common.js: Added.
12:50 Changeset [28721] by darin@apple.com

JavaScriptCore:

Reviewed by Sam.

  • removed unnecessary includes of "Vector.h"
  • wtf/HashMap.h: (WTF::copyKeysToVector): Make the type of the vector be a template parameter. This allows copying keys into a vector of a base class or one with an inline capacity. (WTF::copyValuesToVector): Ditto.
  • wtf/HashSet.h: (WTF::copyToVector): Ditto.

WebCore:

Reviewed by Sam.


  • platform/graphics/FontFallbackList.h: Added include of <wtf/Forward.h>
  • platform/graphics/GlyphPageTreeNode.h: Removed an unneeded include.
12:42 Changeset [28720] by darin@apple.com
  • fix mistake causing nearly all tests to fail on Windows
  • DumpRenderTree/win/UIDelegate.cpp: (UIDelegate::webViewAddMessageToConsole): Need to compare the result of find with npos, not 0. Also pass URL in to URL function rather than passing the entire message.
  • DumpRenderTree/mac/UIDelegate.mm: (-[UIDelegate webView:addMessageToConsole:]): Pass path only to path function rather than passing the entire message.
12:34 Changeset [28719] by darin@apple.com
  • fix Tiger build
  • DumpRenderTree/mac/UIDelegate.mm: (-[UIDelegate webView:addMessageToConsole:]): Re-implement this without using any new-to-Leopard methods.
12:25 Changeset [28718] by darin@apple.com

WebKit/mac:

Reviewed by Brady.

  • WebView/WebView.mm: (-[WebView setFrameLoadDelegate:]): Call [WebIconDatabase sharedIconDatabase] if the a didReceiveIcon method is present.

WebKitTools:

  • DumpRenderTree/mac/FrameLoadDelegate.mm: Don't implement didReceiveIcon delegate method since it now triggers unwanted icon loading. We only had it because we implemented "all" delegate methods here.
12:03 Changeset [28717] by darin@apple.com

WebCore:

Reviewed by Alexey.

Add a few more operations to document.execCommand.

Finished up the transition to the new Editor::Command, including removing
the Editor::execCommand function.

  • WebCore.base.exp: Added Editor::Command::isSupported.
  • editing/Editor.h: Removed execCommand.
  • editing/EditorCommand.cpp: (WebCore::expandSelectionToGranularity): Added. (WebCore::verticalScrollDistance): Added; replaces canScroll. (WebCore::executeDeleteBackward): Added. Moved code here from WebHTMLView. (WebCore::executeDeleteBackwardByDecomposingPreviousCharacter): Ditto. (WebCore::executeDeleteForward): Ditto. (WebCore::executeDeleteToBeginningOfLine): Ditto. (WebCore::executeDeleteToBeginningOfParagraph): Ditto. (WebCore::executeDeleteToEndOfLine): Ditto. (WebCore::executeDeleteToEndOfParagraph): Ditto. (WebCore::executeMovePageDown): Renamed this command to be consistent with all the other Move commands. They all modify the caret. Reimplemented to match the Mac OS X version by removing the explicit scrolling, and letting it be done automatically by code that makes the caret visible. In some cases the old code would scroll twice which was harmless but unnecessary. (WebCore::executeMovePageDownAndModifySelection): Added. Moved code here from WebHTMLView. (WebCore::executeMovePageUp): See MovePageDown above. (WebCore::executeMovePageUpAndModifySelection): Added. Moved code here from WebHTMLView. (WebCore::executeSelectLine): Ditto. (WebCore::executeSelectParagraph): Ditto. (WebCore::executeSelectSentence): Ditto. (WebCore::executeSelectWord): Ditto. (WebCore::executeSwapWithMark): Some small tweaks.
  • page/ContextMenuController.cpp: (WebCore::ContextMenuController::contextMenuItemSelected): Changed to use Editor::command() instead of Editor::execCommand(). This code could be changed to use Editor::Command quite a bit more, but I didn't do that this time.
  • Removed some obsolete unused code.
  • page/mac/EventHandlerMac.mm: (WebCore::EventHandler::passMouseDownEventToWidget): Removed the special case code for NSTextView. This was left over from when we used NSTextField and NSTextView for form elements and is no longer used at all.
  • page/mac/WebCoreFrameBridge.h: Removed 20 unused methods that were still on one side of the bridge or another. We really need to find a time to tear down the rest of the bridge, but that's not this patch.
  • page/mac/WebCoreFrameBridge.mm: (-[WebCoreFrameBridge addData:]): Changed to get at the _shouldCreateRenderers field directly instead of using a method. (-[WebCoreFrameBridge rangeByAlteringCurrentSelection:SelectionController::direction:SelectionController::granularity:]): Took out obsolete comment about matching enums (we use a single enum now and have for some time). (-[WebCoreFrameBridge selectionGranularity]): Ditto.

WebKit:

  • StringsNotToBeLocalized.txt: Updated.

WebKit/gtk:

Reviewed by Alexey.

  • WebCoreSupport/EditorClientGtk.cpp: (WebKit::EditorClient::handleKeyboardEvent): Changed to use Editor::command instead of Editor::execCommand.
  • WebView/webkitwebview.cpp: Ditto.

WebKit/mac:

Reviewed by Alexey.

  • Changed a few more editing operations to use WebCore instead of WebKit.
  • Removed some obsolete unused code.
  • WebCoreSupport/WebFrameBridge.h: Moved declarations of methods that are both defined and used on the WebKit side to here. These no longer belong on the bridge and should be moved to the WebFrame class (or elsewhere).
  • WebCoreSupport/WebFrameBridge.mm: Removed some unused methods.
  • WebView/WebFrameView.mm: (+[WebFrameView _viewTypesAllowImageTypeOmission:]): Fix typo in comment.
  • WebView/WebHTMLView.mm: (-[WebHTMLViewPrivate dealloc]): Removed unused firstResponderTextViewAtMouseDownTime. (-[WebHTMLViewPrivate clear]): Ditto. (-[WebHTMLView _setMouseDownEvent:]): Ditto. (commandNameForSelector): Added special cases for pageDown:, pageDownAndModifySelection:, pageUp:, and pageUpAndModifySelection:, since those names probably aren't specific enough to be used in WebCore (what AppKit calls scrollPageDown: vs. pageDown: needs to be disambiguated with the word "Move"). Added deleteBackward:, deleteBackwardByDecomposingPreviousCharacter:, deleteForward:, deleteToBeginningOfLine:, deleteToBeginningOfParagraph:, deleteToEndOfLine:, deleteToEndOfParagraph:, pageDown:, pageDownAndModifySelection:, pageUp:, pageUpAndModifySelection:, selectLine:, selectParagraph:, selectSentence:, and selectWord: to the list of commands that are forwarded to WebCore. (-[WebHTMLView validateUserInterfaceItemWithoutDelegate:]): Eliminated the long list of operations that we forward to WebCore. Instead, look up any command that WebCore can handle, after any that we handle specially in WebHTMLView. Also fixed a bug where an item that's not a menu item with changeBaseWritingDirection:NSWritingDirectionNatural would end up enabled instead of disabled and streamlined the logic for toggleGrammarChecking:. (-[WebHTMLView mouseDown:]): Removed unused firstResponderTextViewAtMouseDownTime. (-[WebHTMLView becomeFirstResponder]): Removed unused willBecomeFirstResponderForNodeFocus. (-[WebHTMLView resignFirstResponder]): Ditto. (-[WebHTMLView checkSpelling:]): Took unneeded extra initialization of NSSpellChecker.
  • WebView/WebHTMLViewInternal.h: Removed unused willBecomeFirstResponderForNodeFocus, firstResponderTextViewAtMouseDownTime, _textViewWasFirstResponderAtMouseDownTime: and _willMakeFirstResponderForNodeFocus.

WebKit/qt:

Reviewed by Alexey.

  • Api/qwebpage.cpp: (QWebPage::triggerAction): Removed some use of Editor member functions we plan to eventually eliminate. Switch from Editor::execCommand to Editor::command.
  • WebCoreSupport/EditorClientQt.cpp: (WebCore::EditorClientQt::handleKeyboardEvent): Ditto. Also updated name from MoveUpByPageAndModifyCaret to MovePageUp and from MoveDownByPageAndModifyCaret to MovePageDown.

WebKit/win:

Reviewed by Alexey.

  • WebView.cpp: (WebView::execCommand): Switched from Editor::execCommand to Editor:command. Updated name from MoveUpByPageAndModifyCaret to MovePageUp and from MoveDownByPageAndModifyCaret to MovePageDown. (WebView::copy): Switched from Editor::execCommand to Editor:command. (WebView::cut): Ditto. (WebView::paste): Ditto. (WebView::delete_): Ditto.

WebKit/wx:

Reviewed by Alexey.

  • WebKitSupport/EditorClientWx.cpp: (WebCore::EditorClientWx::handleKeyboardEvent): Switched from Editor::execCommand to Edtor::command.

LayoutTests:

Reviewed by Alexey.

To avoid changing results at this time, I didn't correct the many tests that now have
misleading "this test cannot be run manually" text or references to "NSResponder" or
uneeded instructions for running the test manually. We can do that on a later pass.

  • editing/deleting/5300379.html: Use execCommand instead of doCommand.
  • editing/deleting/delete-to-end-of-paragraph.html: Ditto.
  • editing/deleting/smart-delete-003.html: Ditto.
  • editing/deleting/smart-delete-004.html: Ditto.
  • editing/execCommand/nsresponder-indent.html: Ditto.
  • editing/execCommand/nsresponder-outdent.html: Ditto.
  • editing/selection/4947387.html: Ditto.
  • editing/selection/5195166-1.html: Ditto.
  • editing/selection/5195166-2.html: Ditto.
  • editing/selection/select-line.html: Ditto.
  • editing/selection/selection-actions.html: Ditto.
11:55 Changeset [28716] by justin.garcia@apple.com

WebCore:

Reviewed by Darin Adler.

<rdar://problem/5575101> GoogleDocs: Hang in SplitElementCommand::doApply when outdenting a list item in a particular list

  • editing/CompositeEditCommand.cpp: (WebCore::CompositeEditCommand::splitTreeToNode): Moved here.
  • editing/CompositeEditCommand.h:
  • editing/IndentOutdentCommand.cpp: Moved splitTreeToNode.
  • editing/IndentOutdentCommand.h: Ditto, and removed unimplemented splitTreeTo.
  • editing/InsertListCommand.cpp: (WebCore::InsertListCommand::doApply): Split ancestors of listChildNode between it and listNode, if they exists, so that moving listChildNode doesn't put it out of order. Added a test case to cover each change.
  • editing/SplitElementCommand.cpp: (WebCore::SplitElementCommand::doApply): Added an ASSERT to catch code that tries to split a container at a bogus child, and an early return to avoid a hang in that case.

LayoutTests:

Reviewed by Darin Adler.


<rdar://problem/5575101> Hang in SplitElementCommand::doApply when outdenting a list item in a particular list

  • editing/execCommand/5575101-1-expected.txt: Added.
  • editing/execCommand/5575101-1.html: Added.
  • editing/execCommand/5575101-2-expected.txt: Added.
  • editing/execCommand/5575101-2.html: Added.
  • editing/execCommand/5575101-3-expected.txt: Added.
  • editing/execCommand/5575101-3.html: Added.
11:47 Changeset [28715] by andersca@apple.com

JavaScriptCore:

Reviewed by Darin and Geoff.

<rdar://problem/5619295>
REGRESSION: 303-304: Embedded YouTube video fails to render- JS errors (16150) (Flash 9)


Get rid of unnecessary and incorrect security checks for plug-ins accessing JavaScript objects.


The way this used to work was that each NPObject that wrapped a JSObject would have a root object
corresponding to the frame object (used for managing the lifecycle) and an origin root object (used for
doing security checks).


This would prevent a plug-in from accessing a frame's window object if it's security origin was different
(some parts of the window, such as the location object, can be accessed from frames with different security
origins, and those checks are being done in WebCore).


Also, if a plug-in were to access a window object of a frame that later went away, it could lead to that
Window JSObject being garbage collected and the NPObject pointing to freed memory.


How this works now is that there is no origin root object anymore, and all NPObject wrappers that are created
for a plug-in will have the root object of the containing frame of that plug-in.


  • bindings/NP_jsobject.cpp: (jsDeallocate): Don't free the origin root object.


(_NPN_CreateScriptObject):
Remove the origin root object parameter.


(_NPN_InvokeDefault):
(_NPN_Invoke):
(_NPN_Evaluate):
(_NPN_GetProperty):
(_NPN_SetProperty):
(_NPN_RemoveProperty):
(_NPN_HasProperty):
(_NPN_HasMethod):
(_NPN_Enumerate):
Get rid of all security checks.


  • bindings/NP_jsobject.h: Remove originRootObject from the JavaScriptObject struct.


  • bindings/c/c_utility.cpp: (KJS::Bindings::convertValueToNPVariant): Always use the root object from the ExecState.

WebCore:

Reviewed by Darin and Geoff.

<rdar://problem/5619295>
REGRESSION: 303-304: Embedded YouTube video fails to render- JS errors (16150) (Flash 9)

_NPN_CreateScriptObject doesn't take an origin root object anymore.


  • html/HTMLPlugInElement.cpp: (WebCore::HTMLPlugInElement::createNPObject):
  • page/Frame.cpp: (WebCore::Frame::windowScriptNPObject):

WebKitTools:

Reviewed by Darin and Geoff.

<rdar://problem/5619295>
REGRESSION: 303-304: Embedded YouTube video fails to render- JS errors (16150) (Flash 9)


Add property getting methods to the plug-in.


  • DumpRenderTree/TestNetscapePlugIn.subproj/PluginObject.cpp: (pluginInvoke):

LayoutTests:

Reviewed by Darin and Geoff.

<rdar://problem/5619295>
REGRESSION: 303-304: Embedded YouTube video fails to render- JS errors (16150) (Flash 9)


Add cross frame plug/in test where a plug/in inside an iframe tries to access properties of the
top-level frame.


  • http/tests/plugins/cross-frame-object-access-expected.txt: Added.
  • http/tests/plugins/cross-frame-object-access.html: Added.
  • http/tests/plugins/resources/cross-frame-object-access.html: Added.
11:37 Changeset [28714] by mitz@apple.com

Reviewed by Darin Adler.

  • platform/graphics/cg/GraphicsContextCG.cpp: (WebCore::GraphicsContext::setShadow): Slightly increase the magnitude of the offsets passed to CGContextSetShadow* to ensure that the end result after truncation is the desired integer offsets.
10:34 Changeset [28713] by oliver@apple.com

Move layout test to mac skipped list until mac frt bug fixed.

Reviewed by Adam.

Fix test output, and move test to mac skipped list until
<rdar://problem/5647952> fast/events/mouseout-on-window.html
is fixed.

10:19 Changeset [28712] by kmccullough@apple.com

Reviewed by Darin.

  • Layout test fix for mac. When dumped to the console local file paths now only show the name of the resource not the whole path. This is to make the results machine and OS independent.
  • DumpRenderTree/mac/UIDelegate.mm: (-[UIDelegate webView:addMessageToConsole:]):
Note: See TracTimeline for information about the timeline view.