⚠ Archived content — this site is no longer maintained.   Current WebKit documentation is at docs.webkit.org.

Timeline



Jan 27, 2009:

8:08 PM Changeset in webkit [40306] by andersca@apple.com
  • 2 edits in trunk/WebKit/mac

2009-01-27 Anders Carlsson <​andersca@apple.com>

Reviewed by Oliver Hunt.

Fix two bugs with Core Animation based plug-ins.


  1. The plug-in view was marked as opaque even though it's not. (This would leave garbage in the plug-in view).
  2. The plug-in layer needs to have autoresizing turned on.


  • Plugins/WebNetscapePluginView.mm: (-[WebNetscapePluginView setLayer:]):
5:54 PM Changeset in webkit [40305] by Darin Adler
  • 2 edits in trunk/WebCore

2009-01-27 Darin Adler <Darin Adler>

Reviewed by Sam Weinig.

REGRESSION (r40234): Launching iTunes from a link displays a
"Safari can't find the Internet plug-in" sheet in Safari
rdar://problem/6533678

  • rendering/RenderPartObject.cpp: (WebCore::havePlugin): Put the check for empty string in here. (WebCore::serviceTypeForClassId): Only return activeXType() if we have a plug-in that can handle that type. While it might be logical to return an Active X type for any <object> with a classId in it, even if we don't have a plug-in that implements Active X, existing clients, such as Safari, don't expect that.
5:10 PM Changeset in webkit [40304] by mitz@apple.com
  • 2 edits in trunk/WebCore

Reviewed by Steve Falkenburg.

  • fix <rdar://problem/6524351> Text looks bad when font smoothing is forced off
  • platform/graphics/win/FontCGWin.cpp: (WebCore::Font::drawGlyphs): Changed to use GDI rather than Core Graphics if the client has turned font smoothing off.
3:05 PM Changeset in webkit [40303] by Simon Fraser
  • 17 edits in trunk/WebCore

2009-01-27 Simon Fraser <Simon Fraser>

Reviewed by Dave Hyatt

​https://bugs.webkit.org/show_bug.cgi?id=23360

Use the new container-relative repaint methods to compute
post-layout repaints, and repaints using RenderLayer's
cached repaintRect() relative to that container, rather than
using absolute coords.

Replaced lots of boilerplate old/new rect code in various
layout methods with a stack-based LayoutRepainter class, and
fixed a bug in RenderTable which set didFullRepaint to
'true' every time.

2:40 PM Changeset in webkit [40302] by beidson@apple.com
  • 6 edits in trunk/WebCore

2009-01-27 Brady Eidson <​beidson@apple.com>

Reviewed by Darin Adler

<rdar://problem/3541409> - More groundwork and cleanup for CachedFrames

  • history/CachedFrame.cpp: (WebCore::CachedFrame::CachedFrame): Make the savePlatformDataToCachedFrame() client call here, and add the code for caching subframes. (WebCore::CachedFrame::clear):
  • history/CachedFrame.h: Change to be RefCounted (WebCore::CachedFrame::create): (WebCore::CachedFrame::url):
  • history/CachedPage.cpp: (WebCore::CachedPage::CachedPage): (WebCore::CachedPage::restore):
  • history/CachedPage.h: (WebCore::CachedPage::document): (WebCore::CachedPage::documentLoader): (WebCore::CachedPage::view): (WebCore::CachedPage::url): (WebCore::CachedPage::domWindow): (WebCore::CachedPage::cachedMainFrame):
  • loader/FrameLoader.cpp: (WebCore::FrameLoader::cachePageForHistoryItem): Move the platform data saving into the CachedFrame itself.
2:09 PM Changeset in webkit [40301] by zecke@webkit.org
  • 2 edits in trunk/LayoutTests

2009-01-27 Holger Hans Peter Freyther <​zecke@selfish.org>

Rubber stamped by Darin Adler.

Get the build bot green as discussed on the mailinglist. There is
only one crash after unloading a page with plugins. For the remaining
failing tests DumpRenderTree needs to be looked at.

  • platform/gtk/Skipped:
1:54 PM Changeset in webkit [40300] by Simon Fraser
  • 13 edits in trunk/WebCore

2009-01-27 Simon Fraser <Simon Fraser>

Reviewed by Dave Hyatt

Flip the arguments of computeRectForRepaint() from

computeRectForRepaint(IntRect&, RenderBox* repaintContainer, bool)

to

computeRectForRepaint(RenderBox* repaintContainer, IntRect&, bool)

1:44 PM Changeset in webkit [40299] by ddkilzer@apple.com
  • 2 edits in trunk/WebCore

Add missing header guards to AccessibilityObjectWrapper.h

Reviewed by Anders Carlsson.

  • page/mac/AccessibilityObjectWrapper.h: Added missing header guards.
1:28 PM Changeset in webkit [40298] by Simon Fraser
  • 5 edits in trunk/WebCore

2009-01-27 Simon Fraser <Simon Fraser>

Reviewed by Dave Hyatt

​https://bugs.webkit.org/show_bug.cgi?id=23567

Fix crash of svg/custom/stop-crash.svg testcase because of
infinite recursion, as I missed replacing absoluteClippedOverflowRect()
with clippedOverflowRectForRepaint() in RenderSVGGradientStop.

Also fix a missed conversion in RenderTableCol, and add a couple
of comments.

11:24 AM Changeset in webkit [40297] by adele@apple.com
  • 2 edits in trunk/WebCore

2009-01-27 Adele Peterson <​adele@apple.com>

RS by Dave Hyatt

Use the textfield appearance for search fields in themeWin.css so the border/background
styling will work correctly on these fields.

  • css/themeWin.css:
10:45 AM Changeset in webkit [40296] by mitz@apple.com
  • 2 edits in trunk/WebCore
  • build fix
  • bindings/js/JSHTMLInputElementCustom.cpp: (WebCore::JSHTMLInputElement::setSelectionStart): (WebCore::JSHTMLInputElement::setSelectionEnd): (WebCore::JSHTMLInputElement::setSelectionRange):
10:26 AM Changeset in webkit [40295] by mitz@apple.com
  • 2 edits in trunk/WebCore
  • build fix
  • platform/graphics/Gradient.cpp: (WebCore::Gradient::setSpreadMethod):
9:30 AM Changeset in webkit [40294] by Dimitri Glazkov
  • 4 edits in trunk/WebCore

2009-01-27 Paul Godavari <​paul@chromium.org>

Reviewed by Eric Seidel.

​https://bugs.webkit.org/show_bug.cgi?id=23561
Explicitly send the MIME type from an image dragged from a web
page to the desktop, to Chromium so that it can properly generate
a file name.

  • platform/chromium/ChromiumDataObject.cpp: (WebCore::ChromiumDataObject::clear): (WebCore::ChromiumDataObject::hasData):
  • platform/chromium/ChromiumDataObject.h:
  • platform/chromium/ClipboardChromium.cpp: (WebCore::writeImageToDataObject):
9:24 AM Changeset in webkit [40293] by Dimitri Glazkov
  • 6 edits in trunk

WebCore:

2009-01-27 Mads Ager <​ager@chromium.org>

Reviewed by Darin Adler.

​https://bugs.webkit.org/show_bug.cgi?id=23300
Follow Firefox in throwing exceptions when using selection
functions on HTMLInputElements that cannot have selection.

This requires custom setSelectionRange, setSelectionStart and
setSelectionEnd implementations, but it gets rid of the custom
getOwnPropertySlot.

  • bindings/js/JSHTMLInputElementCustom.cpp: (WebCore::JSHTMLInputElement::selectionStart): (WebCore::JSHTMLInputElement::selectionEnd): (WebCore::JSHTMLInputElement::setSelectionRange):
  • html/HTMLInputElement.idl:

LayoutTests:

2009-01-27 Mads Ager <​ager@chromium.org>

Reviewed by Darin Adler.

​https://bugs.webkit.org/show_bug.cgi?id=23300
Update test of selection functions on HTMLInputElements to expect
the selection functions to throw exceptions when used on buttons.
This is consistent with the Firefox behavior.

  • fast/forms/selection-functions-expected.txt:
  • fast/forms/selection-functions.html:
9:13 AM Changeset in webkit [40292] by Dimitri Glazkov
  • 16 edits in trunk/WebCore

2009-01-27 Evan Stade <​estade@chromium.org>

Reviewed by Nikolas Zimmermann.

​https://bugs.webkit.org/show_bug.cgi?id=23435
Make spreadMethod a member of Gradient rather than GraphicsContext.
Implement spreadMethod for Skia.

  • platform/graphics/Gradient.cpp: (WebCore::Gradient::Gradient):
  • platform/graphics/Gradient.h: (WebCore::Gradient::setSpreadMethod): (WebCore::Gradient::spreadMethod):
  • platform/graphics/GraphicsContext.cpp:
  • platform/graphics/GraphicsContext.h:
  • platform/graphics/GraphicsContextPrivate.h:
  • platform/graphics/GraphicsTypes.h: (WebCore::):
  • platform/graphics/cairo/GradientCairo.cpp: (WebCore::Gradient::platformGradient):
  • platform/graphics/cairo/GraphicsContextCairo.cpp: (WebCore::GraphicsContext::fillPath): (WebCore::GraphicsContext::strokePath):
  • platform/graphics/qt/GradientQt.cpp: (WebCore::Gradient::platformGradient):
  • platform/graphics/qt/GraphicsContextQt.cpp: (WebCore::GraphicsContext::fillPath): (WebCore::GraphicsContext::strokePath):
  • platform/graphics/skia/GradientSkia.cpp: (WebCore::Gradient::platformGradient):
  • svg/SVGLinearGradientElement.cpp: (WebCore::SVGLinearGradientElement::buildGradient):
  • svg/SVGRadialGradientElement.cpp: (WebCore::SVGRadialGradientElement::buildGradient):
  • svg/graphics/SVGPaintServerGradient.cpp: (WebCore::SVGPaintServerGradient::SVGPaintServerGradient): (WebCore::SVGPaintServerGradient::setup): (WebCore::SVGPaintServerGradient::externalRepresentation):
  • svg/graphics/SVGPaintServerGradient.h:
8:03 AM Changeset in webkit [40291] by ariya@webkit.org
  • 2 edits in trunk/WebCore

2009-01-27 Ariya Hidayat <​ariya.hidayat@trolltech.com>

Rubber-stamped by Simon Hausmann.

​http://www.qtsoftware.com/developer/task-tracker/index_html?id=238662&method=entry

[Qt] Map function keys F1 to F24.

  • platform/qt/PlatformKeyboardEventQt.cpp: (WebCore::windowsKeyCodeForKeyEvent):
7:26 AM Changeset in webkit [40290] by ariya@webkit.org
  • 2 edits in trunk/WebKit/qt

2009-01-27 Ariya Hidayat <​ariya.hidayat@trolltech.com>

Reviewed by Simon Hausmann.

​http://www.qtsoftware.com/developer/task-tracker/index_html?id=238391&method=entry

[Qt] If QPainter fails to start on a QPrinter instance, do not
continue printing.

  • Api/qwebframe.cpp: (QWebFrame::print):
12:05 AM Changeset in webkit [40289] by beidson@apple.com
  • 26 edits in trunk

WebCore:

2009-01-27 Brady Eidson <​beidson@apple.com>

Reviewed by Dan Bernstein

Rework FrameLoaderClient to work on a CachedFrame basis instead of CachedPage

  • WebCore.base.exp:
  • history/CachedPage.cpp:
  • history/CachedPage.h: (WebCore::CachedPage::cachedMainFrame):
  • loader/EmptyClients.h: (WebCore::EmptyFrameLoaderClient::savePlatformDataToCachedFrame): (WebCore::EmptyFrameLoaderClient::transitionToCommittedFromCachedFrame):
  • loader/FrameLoader.cpp: (WebCore::FrameLoader::restoreDocumentState): (WebCore::FrameLoader::commitProvisionalLoad): (WebCore::FrameLoader::transitionToCommitted): (WebCore::FrameLoader::cachePageForHistoryItem):
  • loader/FrameLoaderClient.h:

WebKit/gtk:

2009-01-27 Brady Eidson <​beidson@apple.com>

Reviewed by Dan Bernstein

Rework FrameLoaderClient to work on a CachedFrame basis instead of CachedPage

  • WebCoreSupport/FrameLoaderClientGtk.cpp: (WebKit::FrameLoaderClient::savePlatformDataToCachedFrame): (WebKit::FrameLoaderClient::transitionToCommittedFromCachedFrame):
  • WebCoreSupport/FrameLoaderClientGtk.h:

WebKit/mac:

2009-01-27 Brady Eidson <​beidson@apple.com>

Reviewed by Dan Bernstein

Rework FrameLoaderClient to work on a CachedFrame basis instead of CachedPage

  • History/WebHistoryItem.mm:
  • WebCoreSupport/WebFrameLoaderClient.h:
  • WebCoreSupport/WebFrameLoaderClient.mm: (WebFrameLoaderClient::savePlatformDataToCachedFrame): (WebFrameLoaderClient::transitionToCommittedFromCachedFrame):
  • WebKit.order:

WebKit/qt:

2009-01-27 Brady Eidson <​beidson@apple.com>

Reviewed by Dan Bernstein

Rework FrameLoaderClient to work on a CachedFrame basis instead of CachedPage

  • WebCoreSupport/FrameLoaderClientQt.cpp: (WebCore::FrameLoaderClientQt::savePlatformDataToCachedFrame): (WebCore::FrameLoaderClientQt::transitionToCommittedFromCachedFrame):
  • WebCoreSupport/FrameLoaderClientQt.h:

WebKit/win:

2009-01-27 Brady Eidson <​beidson@apple.com>

Reviewed by Dan Bernstein

Rework FrameLoaderClient to work on a CachedFrame basis instead of CachedPage

  • WebCoreSupport/WebFrameLoaderClient.cpp: (WebFrameLoaderClient::savePlatformDataToCachedFrame): (WebFrameLoaderClient::transitionToCommittedFromCachedFrame):
  • WebCoreSupport/WebFrameLoaderClient.h:
  • WebFrame.cpp:
  • WebFrame.h:

WebKit/wx:

2009-01-27 Brady Eidson <​beidson@apple.com>

Reviewed by Dan Bernstein

Rework FrameLoaderClient to work on a CachedFrame basis instead of CachedPage

  • WebKitSupport/FrameLoaderClientWx.cpp: (WebCore::FrameLoaderClientWx::savePlatformDataToCachedFrame): (WebCore::FrameLoaderClientWx::transitionToCommittedFromCachedFrame):
  • WebKitSupport/FrameLoaderClientWx.h:

Jan 26, 2009:

10:58 PM Changeset in webkit [40288] by Simon Fraser
  • 33 edits in trunk/WebCore

2009-01-26 Simon Fraser <Simon Fraser>

Reviewed by Dave Hyatt

​https://bugs.webkit.org/show_bug.cgi?id=23360

Add variants of all the methods that compute dirty rects for repainting
with versions that compute those rects relative to some container
RenderBox that acts as a painting root.

8:40 PM Changeset in webkit [40287] by Simon Fraser
  • 7 edits in trunk

2009-01-26 Simon Fraser <Simon Fraser>

Reviewed by David Hyatt

Back out r40285, because it was checked in with no bug number, no
testcase, is rendering change that did not get thorough review,
and broke the Mac build.

7:49 PM Changeset in webkit [40286] by beidson@apple.com
  • 4 edits in trunk/WebCore

2009-01-26 Brady Eidson <​beidson@apple.com>

Reviewed by Darin Adler

-Make the recently added back/forward cache logging much better by actually

walking the entire frame tree and indenting the resulting output.

-Fix a null-termination bug in HistoryItem tree logging

  • history/HistoryItem.cpp: (WebCore::HistoryItem::showTreeWithIndent):
  • loader/FrameLoader.cpp: (WebCore::pageCacheLogPrefix): (WebCore::pageCacheLog): (WebCore::FrameLoader::logCanCachePageDecision): (WebCore::FrameLoader::logCanCacheFrameDecision):
  • loader/FrameLoader.h:
7:06 PM Changeset in webkit [40285] by treat@webkit.org
  • 7 edits in trunk

Allow option of hit testing frame content without clipping to the visible
viewport.

6:44 PM Changeset in webkit [40284] by ddkilzer@apple.com
  • 5 edits in trunk/LayoutTests

Fix misspelling of "lowercase" in font-face-locally-installed.html

Reviewed by Dan Bernstein.

  • fast/css/font-face-locally-installed.html: Updated.
  • platform/mac/fast/css/font-face-locally-installed-expected.checksum: Updated.
  • platform/mac/fast/css/font-face-locally-installed-expected.png: Updated.
  • platform/mac/fast/css/font-face-locally-installed-expected.txt: Updated.
5:59 PM Changeset in webkit [40283] by andersca@apple.com
  • 1 edit in trunk/WebKit/mac/ChangeLog

Sam did _not_ review this

5:57 PM Changeset in webkit [40282] by adele@apple.com
  • 2 edits in trunk/JavaScriptCore

2009-01-26 Adele Peterson <​adele@apple.com>

Build fix.

  • debugger/Debugger.cpp:
5:56 PM Changeset in webkit [40281] by andersca@apple.com
  • 4 edits in trunk/WebKit/mac

2009-01-26 Anders Carlsson <​andersca@apple.com>

Reviewed by Sam Weinig.

Add the ability for plug-ins to make WebKit operate in "modal mode"


  • Plugins/Hosted/NetscapePluginHostProxy.h:
  • Plugins/Hosted/NetscapePluginHostProxy.mm: (WebKit::NetscapePluginHostProxy::NetscapePluginHostProxy): (WebKit::NetscapePluginHostProxy::pluginHostDied): If the plug-in crashes while we're modal, make sure to leave the modal mode.


(WebKit::NetscapePluginHostProxy::setModal):
(WKPCSetModal):

  • Plugins/Hosted/WebKitPluginClient.defs:
5:33 PM Changeset in webkit [40280] by sullivan@apple.com
  • 3 edits in trunk/WebKit/mac

2009-01-26 John Sullivan <​sullivan@apple.com>

fixed <rdar://problem/6530053> REGRESSION (Leopard): Shift-tab in http authentication window gets
stuck in the Name field rather than cycling around

Reviewed by Dan Bernstein

  • Panels/English.lproj/WebAuthenticationPanel.nib/designable.nib:
  • Panels/English.lproj/WebAuthenticationPanel.nib/keyedobjects.nib: The two static text fields and the last button all had their "next key view" outlets set to the name field, which caused shift-tab from the name field to do the wrong thing. Fixed by making each selectable view have exactly one "next key view" set to it.
5:08 PM Changeset in webkit [40279] by barraclough@apple.com
  • 2 edits in trunk/JavaScriptCore

2009-01-26 Gavin Barraclough <​barraclough@apple.com>

Reviewed by Darin Adler.

Fixes for eq null & neq null, on 64-bit JIT.
​https://bugs.webkit.org/show_bug.cgi?id=23559

This patch degrades 64-bit JIT performance on some benchmarks,
due to the whole not-being-incorrect thing.

  • jit/JIT.cpp: (JSC::JIT::privateCompileMainPass):
4:45 PM Changeset in webkit [40278] by weinig@apple.com
  • 6 edits
    3 adds in trunk

WebCore:

2009-01-26 Sam Weinig <​sam@webkit.org>

Reviewed by Dan Bernstein.

Fix for ​https://bugs.webkit.org/show_bug.cgi?id=18141
Acid3 tests 77 and 78 fail on reload due to SVG font race
<rdar://problem/6504899>

Match Opera's behavior of loading SVGFonts as soon as <font-face-uri>
tag is inserted into the document and not at the first use of the font
as is done for CSS web fonts.

This fixes a race condition noticeable by some in the Acid3 test that
happened because the onload event handler for an iframe containing a
<font-face-uri> with an external reference was not delayed by the pending
load of the font.

Test: http/tests/misc/SVGFont-delayed-load.html

  • svg/SVGFontFaceUriElement.cpp: (WebCore::SVGFontFaceUriElement::~SVGFontFaceUriElement): (WebCore::SVGFontFaceUriElement::parseMappedAttribute): Call loadFont if the href attribute changes. (WebCore::SVGFontFaceUriElement::insertedIntoDocument): Call loadFont. (WebCore::SVGFontFaceUriElement::loadFont): Force a load of the font specified in href attribute.
  • svg/SVGFontFaceUriElement.h:

LayoutTests:

2009-01-26 Sam Weinig <​sam@webkit.org>

Reviewed by Dan Bernstein.

Test for ​https://bugs.webkit.org/show_bug.cgi?id=18141
Acid3 tests 77 and 78 fail on reload due to SVG font race
<rdar://problem/6504899>

  • http/tests/misc/SVGFont-delayed-load-expected.txt: Added.
  • http/tests/misc/SVGFont-delayed-load.html: Added.
  • http/tests/misc/resources/SVGFont-delayed-loader.svg: Added.
  • svg/custom/acid3-test-77.html:
  • svg/custom/resources/Acid3Font-loader.svg:
4:38 PM Changeset in webkit [40277] by andersca@apple.com
  • 4 edits in trunk/WebKit/mac

2009-01-26 Anders Carlsson <​andersca@apple.com>

Reviewed by Sam Weinig.

Add the ability for a plug-in to show or hide the menu bar.


  • Plugins/Hosted/NetscapePluginHostProxy.h:
  • Plugins/Hosted/NetscapePluginHostProxy.mm: (WebKit::NetscapePluginHostProxy::NetscapePluginHostProxy): (WebKit::NetscapePluginHostProxy::pluginHostDied): (WebKit::NetscapePluginHostProxy::setMenuBarVisible): (WKPCSetMenuBarVisible):
  • Plugins/Hosted/WebKitPluginClient.defs:
4:09 PM Changeset in webkit [40276] by pol@apple.com
  • 13 edits in trunk/LayoutTests

2009-01-26 Pierre-Olivier Latour <​pol@apple.com>

Updated the reference images for the video tests that rely on the H264 test movie,
as QuickTime 7.6 apparently slightly changed the results from the 7.5.5 version.

  • platform/mac/media/video-aspect-ratio-expected.checksum:
  • platform/mac/media/video-aspect-ratio-expected.png:
  • platform/mac/media/video-controls-rendering-expected.checksum:
  • platform/mac/media/video-controls-rendering-expected.png:
  • platform/mac/media/video-display-toggle-expected.checksum:
  • platform/mac/media/video-display-toggle-expected.png:
  • platform/mac/media/video-layer-crash-expected.checksum:
  • platform/mac/media/video-layer-crash-expected.png:
  • platform/mac/media/video-transformed-expected.checksum:
  • platform/mac/media/video-transformed-expected.png:
  • platform/mac/media/video-zoom-expected.checksum:
  • platform/mac/media/video-zoom-expected.png:
4:00 PM Changeset in webkit [40275] by timothy@apple.com
  • 2 edits in trunk/WebCore

Change Web Archive serialization back to the binary property list format.

<rdar://problem/6127865> Web archives saved with Safari 4 cannot be read by Safari 3

Reviewed by Brady Eidson.

  • loader/archive/cf/LegacyWebArchive.cpp: (WebCore::LegacyWebArchive::rawDataRepresentation):
3:57 PM Changeset in webkit [40274] by cwzwarich@webkit.org
  • 8 edits
    1 add in trunk

2009-01-26 Cameron Zwarich <​cwzwarich@uwaterloo.ca>

Reviewed by Gavin Barraclough.

Bug 23552: Dashcode evaluator no longer works after making ExecStates actual call frames
<​https://bugs.webkit.org/show_bug.cgi?id=23552>
<rdar://problem/6398839>

Dashcode will crash when using the evaluator because it saves a global call
frame, even after global code has finished executing, and then uses this as
a launching pad to execute new JS in the evaluator. The fix is to detect
when Dashcode is attempting to do this and execute code from a global call
frame instead.

JavaScriptCore:

  • JavaScriptCore.exp:
  • debugger/Debugger.cpp: (JSC::evaluateInGlobalCallFrame): Added so that WebScriptCallFrame can evaluate JS starting from a global call frame.
  • debugger/Debugger.h:

WebKit/mac:

  • ForwardingHeaders/runtime/Protect.h: Added.
  • WebView/WebScriptDebugDelegate.mm: (-[WebScriptCallFrame _initWithGlobalObject:debugger:caller:debuggerCallFrame:]): Added debugger, a WebScriptDebugger* argument. (-[WebScriptCallFrame evaluateWebScript:]): Detect when Dashcode is using a stale WebScriptCallFrame to execute new JS and evaluate it starting from the global object's global call frame instead.
  • WebView/WebScriptDebugger.h: (WebScriptDebugger::globalObject): Added. (WebScriptDebugger::globalCallFrame): Added.
  • WebView/WebScriptDebugger.mm: (WebScriptDebugger::WebScriptDebugger): Initialize m_globalObject. (WebScriptDebugger::initGlobalCallFrame): Created as a clone of callEvent so that the global call frame can be saved immediately after being created. (WebScriptDebugger::callEvent): Pass 'this' as the debugger argument of WebScriptCallFrame's _initWithGlobalObject method.
3:27 PM Changeset in webkit [40273] by andersca@apple.com
  • 3 edits in trunk/WebKit/mac

2009-01-26 Anders Carlsson <​andersca@apple.com>

Reviewed by Oliver Hunt.

Make WKPCInvoke a simpleroutine.


  • Plugins/Hosted/NetscapePluginHostProxy.mm: (WKPCInvoke):
  • Plugins/Hosted/WebKitPluginClient.defs:
3:16 PM Changeset in webkit [40272] by andersca@apple.com
  • 2 edits in trunk/WebCore

2009-01-26 Anders Carlsson <​andersca@apple.com>

Fix 64-bit build.


  • WebCore.LP64.exp:
3:14 PM Changeset in webkit [40271] by Dimitri Glazkov
  • 7 edits
    2 moves in trunk/WebCore

2009-01-26 Dmitry Titov <​dimich@chromium.org>

Reviewed by Alexey Proskuryakov.

​https://bugs.webkit.org/show_bug.cgi?id=23511
Move DOMTimer from bindings/js to page/.
This was prepared by moving all JSC dependencies from DOMTimer
to ScheduledAction earlier, and helps unfork the Chromuim version
of the DOMTimer.

  • GNUmakefile.am:
  • WebCore.pro:
  • WebCore.scons:
  • WebCore.vcproj/WebCore.vcproj:
  • WebCore.xcodeproj/project.pbxproj:
  • WebCoreSources.bkl:
  • bindings/js/DOMTimer.cpp: Removed.
  • bindings/js/DOMTimer.h: Removed.
  • page/DOMTimer.cpp: Copied from WebCore/bindings/js/DOMTimer.cpp.
  • page/DOMTimer.h: Copied from WebCore/bindings/js/DOMTimer.h.
3:07 PM Changeset in webkit [40270] by hyatt@apple.com
  • 4 edits in trunk/WebCore

2009-01-26 David Hyatt <​hyatt@apple.com>

Eliminate rightmostPosition, lowestPosition and leftmostPosition from RenderFlow. Move the code into
RenderBlock's methods.

Reviewed by Eric Seidel

  • rendering/RenderBlock.cpp: (WebCore::RenderBlock::lowestPosition): (WebCore::RenderBlock::rightmostPosition): (WebCore::RenderBlock::leftmostPosition):
  • rendering/RenderFlow.cpp:
  • rendering/RenderFlow.h:
3:06 PM Changeset in webkit [40269] by darin@chromium.org
  • 3 edits in trunk/WebCore

2009-01-26 Darin Fisher <​darin@chromium.org>

Reviewed by Eric Seidel.

​https://bugs.webkit.org/show_bug.cgi?id=23553
Upstream changes from jcampan to fix chromium bugs 6692 and 6694.

  • platform/chromium/PopupMenuChromium.cpp: (WebCore::PopupListBox::setLoopSelectionNavigation): (WebCore::PopupListBox::PopupListBox): (WebCore::PopupContainer::setLoopSelectionNavigation): (WebCore::PopupListBox::handleKeyEvent): (WebCore::PopupListBox::paintRow): (WebCore::PopupListBox::clearSelection): (WebCore::PopupListBox::selectNextRow): (WebCore::PopupListBox::selectPreviousRow):
  • platform/chromium/PopupMenuChromium.h:
2:54 PM Changeset in webkit [40268] by Dimitri Glazkov
  • 1 edit in trunk/WebCore/WebCore.xcodeproj/project.pbxproj

Remove extraneous WebCore XCode project setting, introduced in previous commit.

2:52 PM Changeset in webkit [40267] by beidson@apple.com
  • 4 edits in trunk/WebCore

2009-01-26 Brady Eidson <​beidson@apple.com>

Reviewed by Sam Weinig

Remove some historical CachedPage time stamp fluff.

  • history/CachedPage.cpp: (WebCore::CachedPage::CachedPage):
  • history/CachedPage.h: (WebCore::CachedPage::timeStamp):
  • loader/FrameLoader.cpp: (WebCore::FrameLoader::cachePageForHistoryItem):
2:45 PM Changeset in webkit [40266] by hyatt@apple.com
  • 8 edits in trunk/WebCore

2009-01-26 David Hyatt <​hyatt@apple.com>

Move absoluteClippedOverflowRect, linesBoundingBox and borderBoundingBox from RenderFlow to RenderInline.

Reviewed by Eric Seidel

  • rendering/RenderBox.cpp: (WebCore::RenderBox::containingBlockHeightForPositioned):
  • rendering/RenderFlow.cpp:
  • rendering/RenderFlow.h:
  • rendering/RenderInline.cpp: (WebCore::RenderInline::linesBoundingBox): (WebCore::RenderInline::absoluteClippedOverflowRect):
  • rendering/RenderInline.h: (WebCore::RenderInline::borderBoundingBox):
  • rendering/RenderLayer.cpp: (WebCore::RenderLayer::updateLayerPosition):
  • rendering/RenderTreeAsText.cpp: (WebCore::operator<<):
2:40 PM Changeset in webkit [40265] by Dimitri Glazkov
  • 5 edits in trunk/WebCore

2009-01-26 Dimitri Glazkov <Dimitri Glazkov>

Reviewed by Eric Seidel.

​https://bugs.webkit.org/show_bug.cgi?id=22650
Implement support for "Immutable" attribute to CodeGeneratorJS.pm, thus
allowing to explicitly mark methods as immutable (i.e. returning a new
instance rather than a reference to an existing instance).

  • bindings/js/JSSVGMatrixCustom.cpp: Removed custom methods.
  • bindings/scripts/CodeGeneratorJS.pm: Added check for Immutable

attribute when generating function call.

  • svg/SVGMatrix.idl: Replaced Custom attribute with Immutable on

methods.

2:29 PM Changeset in webkit [40264] by hyatt@apple.com
  • 22 edits in trunk/WebCore

2009-01-26 David Hyatt <​hyatt@apple.com>

Eliminate isInlineFlow() now that a RenderBlock can never be an inline flow. isInlineFlow() is now
equivalent to isRenderInline(), so all calls to isInlineFlow() have been replaced with isRenderInline().

Reviewed by Eric Seidel

  • dom/Text.cpp: (WebCore::Text::rendererIsNeeded):
  • rendering/InlineFlowBox.cpp: (WebCore::InlineFlowBox::placeBoxesHorizontally): (WebCore::shouldDrawTextDecoration):
  • rendering/RenderBlock.cpp: (WebCore::RenderBlock::hitTestContents): (WebCore::InlineMinMaxIterator::next): (WebCore::RenderBlock::calcInlinePrefWidths):
  • rendering/RenderBlock.h: (WebCore::RenderBlock::isBlockFlow):
  • rendering/RenderBox.cpp: (WebCore::RenderBox::clientWidth): (WebCore::RenderBox::clientHeight): (WebCore::RenderBox::nodeAtPoint): (WebCore::RenderBox::imageChanged): (WebCore::RenderBox::offsetForPositionedInContainer): (WebCore::RenderBox::containingBlockWidthForPositioned): (WebCore::RenderBox::containingBlockHeightForPositioned):
  • rendering/RenderBox.h: (WebCore::RenderBox::width): (WebCore::RenderBox::height): (WebCore::RenderBox::size): (WebCore::RenderBox::frameRect):
  • rendering/RenderContainer.cpp: (WebCore::RenderContainer::updateBeforeAfterContentForContainer):
  • rendering/RenderFlow.cpp: (WebCore::RenderFlow::dirtyLinesFromChangedChild): (WebCore::RenderFlow::createInlineBox): (WebCore::RenderFlow::paintLines): (WebCore::RenderFlow::hitTestLines): (WebCore::RenderFlow::absoluteClippedOverflowRect): (WebCore::RenderFlow::lowestPosition): (WebCore::RenderFlow::rightmostPosition): (WebCore::RenderFlow::leftmostPosition):
  • rendering/RenderFlow.h: (WebCore::RenderFlow::borderBoundingBox):
  • rendering/RenderInline.h: (WebCore::RenderInline::isRenderInline):
  • rendering/RenderLayer.cpp: (WebCore::RenderLayer::updateLayerPosition): (WebCore::RenderLayer::intersectsDamageRect): (WebCore::RenderLayer::boundingBox):
  • rendering/RenderListItem.cpp: (WebCore::getParentOfFirstLineBox):
  • rendering/RenderObject.cpp: (WebCore::RenderObject::getAbsoluteRepaintRectWithOutline): (WebCore::RenderObject::styleWillChange): (WebCore::RenderObject::firstLineStyle):
  • rendering/RenderObject.h:
  • rendering/RenderSVGInline.cpp: (WebCore::RenderSVGInline::createInlineBox):
  • rendering/RenderSVGText.cpp: (WebCore::RenderSVGText::createInlineBox):
  • rendering/RenderTableRow.cpp: (WebCore::RenderTableRow::nodeAtPoint):
  • rendering/RenderTableSection.cpp: (WebCore::RenderTableSection::nodeAtPoint):
  • rendering/RenderText.cpp: (WebCore::isInlineFlowOrEmptyText):
  • rendering/RenderTreeAsText.cpp: (WebCore::operator<<):
  • rendering/bidi.cpp: (WebCore::bidiNext): (WebCore::bidiFirst): (WebCore::appendRunsForObject): (WebCore::RenderBlock::createLineBoxes): (WebCore::RenderBlock::computeHorizontalPositionsForLine): (WebCore::RenderBlock::layoutInlineChildren): (WebCore::inlineFlowRequiresLineBox): (WebCore::requiresLineBox): (WebCore::RenderBlock::skipTrailingWhitespace): (WebCore::RenderBlock::skipLeadingWhitespace): (WebCore::RenderBlock::findNextLineBreak):
2:27 PM Changeset in webkit [40263] by andersca@apple.com
  • 9 edits in trunk/WebKit/mac

2009-01-26 Anders Carlsson <​andersca@apple.com>

Reviewed by Sam Weinig.

Implement using plug-in objects as constructors, and setting and getting properties from a plug-in object.


  • Plugins/Hosted/NetscapePluginHostProxy.mm: (WKPCBooleanAndDataReply):
  • Plugins/Hosted/NetscapePluginInstanceProxy.h: (WebKit::NetscapePluginInstanceProxy::Reply::): (WebKit::NetscapePluginInstanceProxy::BooleanAndDataReply::BooleanAndDataReply): Rename NPObjectInvokeReply to BooleanAndDataReply.


  • Plugins/Hosted/NetscapePluginInstanceProxy.mm: (WebKit::NetscapePluginInstanceProxy::addValueToArray): Fix a cut and paste error.


(WebKit::NetscapePluginInstanceProxy::demarshalValueFromArray):
Handle NPObjects.


  • Plugins/Hosted/ProxyInstance.h:
  • Plugins/Hosted/ProxyInstance.mm: (WebKit::ProxyField::valueFromInstance): (WebKit::ProxyField::setValueToInstance): Call the ProxyInstance method.


(WebKit::ProxyInstance::~ProxyInstance):
Release the NPObject.


(WebKit::ProxyInstance::supportsConstruct):
Ask the plug-in host if an instance supports construct.


(WebKit::ProxyInstance::fieldValue):
(WebKit::ProxyInstance::setFieldValue):
Call the plug-in host methods.


  • Plugins/Hosted/WebKitPluginHostTypes.h: Rename ObjectValueType to JSObjectValueType, and add NPObjectValueType.
1:53 PM Changeset in webkit [40262] by hyatt@apple.com
  • 1 edit in trunk/LayoutTests/platform/mac/fast/table/form-with-table-style-expected.txt

Land new layout tests now that display:compact support has been removed.

1:52 PM Changeset in webkit [40261] by hyatt@apple.com
  • 1 edit in trunk/LayoutTests/fast/leaks/001-expected.txt

Land new layout tests now that display:compact support has been removed.

1:51 PM Changeset in webkit [40260] by hyatt@apple.com
  • 1 edit in trunk/LayoutTests/platform/mac/fast/forms/form-hides-table-expected.txt

Land new layout tests now that display:compact support has been removed.

1:51 PM Changeset in webkit [40259] by hyatt@apple.com
  • 7 edits in trunk/LayoutTests/platform/mac/fast/compact

Land new layout tests now that display:compact support has been removed.

1:50 PM Changeset in webkit [40258] by hyatt@apple.com
  • 10 edits in trunk/WebCore

2009-01-26 David Hyatt <​hyatt@apple.com>

Drop support for display:compact.

(1) display:compact is no longer part of CSS2.1.
(2) Shipping Safari has completely broken compact support with severely broken rendering
(all of the compact text just piles on top of the text in the inserted block and on top of itself as well).
(3) This has been broken in WebKit for years (ever since blocks converted to painting lines directly) without
anyone even noticing.

Reviewed by Eric Seidel

  • rendering/InlineFlowBox.cpp: (WebCore::InlineFlowBox::placeBoxesHorizontally):
  • rendering/RenderBlock.cpp: (WebCore::RenderBlock::handleSpecialChild): (WebCore::RenderBlock::handleRunInChild): (WebCore::RenderBlock::layoutBlockChildren): (WebCore::RenderBlock::paint): (WebCore::RenderBlock::paintObject): (WebCore::RenderBlock::nodeAtPoint): (WebCore::RenderBlock::positionForCoordinates): (WebCore::shouldCheckLines): (WebCore::getHeightForLineCount): (WebCore::RenderBlock::renderName):
  • rendering/RenderBlock.h:
  • rendering/RenderBox.cpp: (WebCore::RenderBox::sizesToIntrinsicWidth):
  • rendering/RenderBox.h: (WebCore::RenderBox::width): (WebCore::RenderBox::height): (WebCore::RenderBox::size): (WebCore::RenderBox::frameRect):
  • rendering/RenderFlow.cpp: (WebCore::RenderFlow::paintLines): (WebCore::RenderFlow::hitTestLines): (WebCore::RenderFlow::absoluteClippedOverflowRect): (WebCore::RenderFlow::linesBoundingBox):
  • rendering/RenderObject.h: (WebCore::RenderObject::isInline):
  • rendering/bidi.cpp: (WebCore::RenderBlock::computeHorizontalPositionsForLine): (WebCore::RenderBlock::layoutInlineChildren):
  • rendering/bidi.h: (WebCore::BidiRun::BidiRun):
1:44 PM Changeset in webkit [40257] by mrowe@apple.com
  • 2 edits in trunk/WebKit/mac

Fix the build.

Remove -Wformat=2 from the warning flags as newer versions of GCC emit
warnings about non-literal format strings for uses of our UI_STRING macro.

  • Configurations/Base.xcconfig:
1:42 PM Changeset in webkit [40256] by pol@apple.com
  • 2 edits in trunk/WebKitTools

2009-01-26 Pierre-Olivier Latour <​pol@apple.com>

Tweaked again earlier fix, this time just to print a warning and not abort if
attempting to generate pixel results and Perian is installed.

​https://bugs.webkit.org/show_bug.cgi?id=22615

  • Scripts/run-webkit-tests:
12:55 PM Changeset in webkit [40255] by darin@chromium.org
  • 2 edits in trunk/WebCore

2009-01-26 Adam Langley <​agl@google.com>

Reviewed by Nikolas Zimmermann.

gtk2drawing: fix compiler warnings

g_object_set_data expects a gpointer as the last argument. Cast the

TRUE/FALSE values to avoid a warning.

gtk_tree_view_column_new doesn't return a GtkWidget.

  • platform/gtk/gtk2drawing.c: (setup_widget_prototype): (moz_gtk_get_combo_box_inner_button): (moz_gtk_get_combo_box_button_inner_widgets): (ensure_combo_box_widgets): (moz_gtk_get_combo_box_entry_inner_widgets): (moz_gtk_get_combo_box_entry_arrow): (ensure_combo_box_entry_widgets): (ensure_toolbar_widget): (ensure_menu_bar_item_widget): (ensure_menu_popup_widget): (ensure_menu_item_widget): (ensure_image_menu_item_widget): (ensure_menu_separator_widget): (ensure_check_menu_item_widget): (ensure_tree_header_cell_widget): (moz_gtk_entry_paint):
12:14 PM Changeset in webkit [40254] by hyatt@apple.com
  • 4 edits in trunk

2009-01-26 David Hyatt <​hyatt@apple.com>

Change display:run-in handling so that we no longer reuse RenderBlock as though it is an inline flow.
Instead we create a new RenderInline, move the RenderBlock's children underneath it, and then set the
inline as the new element's renderer.

Reviewed by Dan Bernstein

  • rendering/RenderBlock.cpp: (WebCore::RenderBlock::handleRunInChild):
  • rendering/RenderInline.cpp: (WebCore::RenderInline::renderName):
11:43 AM Changeset in webkit [40253] by mrowe@apple.com
  • 5 edits in trunk/WebKit

Clean up after r40240.

Rubber-stamped by Sam Weinig.

  • WebKit.xcodeproj/project.pbxproj: Reinstate WebKitPluginHost.defs.
  • Configurations/Base.xcconfig: Don't dead code strip in debug builds for now as it leads to link errors.
  • Plugins/Hosted/HostedNetscapePluginStream.mm: Revert change that is no longer needed now that WebKitPluginHost.defs

is back in the build.

10:48 AM Changeset in webkit [40252] by Adam Roben
  • 2 edits in trunk/WebKit/win

Fix an assertion failure when Safari loads an error page

Reviewed by Alexey Proskuryakov.

  • WebFrame.cpp: (WebFrame::loadData): Use the two-argument version of KURL and MarshallingHelpers::BSTRToKURL, for reasons mentioned in r40248.
10:46 AM Changeset in webkit [40251] by christian@webkit.org
  • 2 edits in trunk/WebKitTools

2009-01-26 Christian Dywan <​christian@twotoasts.de>

Rubber stamped by Holger Freyther.

  • GtkLauncher/main.c: (main): Initialize threads, which is required for libSoup.
8:46 AM Changeset in webkit [40250] by Simon Hausmann
  • 3 edits in trunk/WebKit/qt

2009-01-26 David Boddie <​dboddie@trolltech.com>

Reviewed by Simon Hausmann.

Doc: qdoc fix for a link to a property and added a basic description of the offline caching and storage features from HTML 5.

8:44 AM Changeset in webkit [40249] by Dimitri Glazkov
  • 2 edits in trunk/WebCore

2009-01-26 Dimitri Glazkov <Dimitri Glazkov>

Reviewed by Darin Adler.

​https://bugs.webkit.org/show_bug.cgi?id=23480
Add missing method stubs to BackForwardListChromium.

  • history/BackForwardListChromium.cpp: (WebCore::BackForwardList::backItem): (WebCore::BackForwardList::forwardItem):
7:19 AM Changeset in webkit [40248] by Adam Roben
  • 5 edits in trunk/WebKit/win

Fix an assertion on launch in KURL::KURL

All URL strings passed in to the WebKit API need to be passed to
MarshallingHelpers::BSTRToKURL so that they may be properly parsed.

Reviewed by Alexey Proskuryakov.

  • MarshallingHelpers.cpp: (MarshallingHelpers::BSTRToKURL): Changed to use the two-argument constructor for KURL so that the strings will be parsed rather than assumed to be in the correct encoding/form.
  • WebResource.cpp: (WebResource::initWithData):
  • WebURLResponse.cpp: (WebURLResponse::initWithURL):
  • WebView.cpp: (WebView::userAgentForURL): (WebView::copyURL): Changed to use MarshallingHelpers::BSTRToKURL instead of trying to do the work manually.
6:03 AM QtWebKitJournal edited by Simon Hausmann
(diff)
2:16 AM Changeset in webkit [40247] by Darin Adler
  • 3 edits in trunk/LayoutTests

2009-01-26 Darin Adler <Darin Adler>

Requested by David Hyatt.

  • platform/mac/fast/table/form-with-table-style-expected.checksum: Updated to reflect the progression; we now render things we used to fail to render.
  • platform/mac/fast/table/form-with-table-style-expected.png: Ditto.
Note: See TracTimeline for information about the timeline view.