Timeline


and

06/11/10:

23:21 Changeset [61053] by eric@webkit.org

2010-06-11 Sheriff Bot <webkit.review.bot@gmail.com>

Unreviewed, rolling out r61052.
http://trac.webkit.org/changeset/61052
https://bugs.webkit.org/show_bug.cgi?id=40517

"Broke fast/forms/caret-rtl.html on Mac bots" (Requested by
eseidel on #webkit).

  • fast/html/hidden-attr-expected.txt: Removed.
  • fast/html/hidden-attr.html: Removed.

2010-06-11 Sheriff Bot <webkit.review.bot@gmail.com>

Unreviewed, rolling out r61052.
http://trac.webkit.org/changeset/61052
https://bugs.webkit.org/show_bug.cgi?id=40517

"Broke fast/forms/caret-rtl.html on Mac bots" (Requested by
eseidel on #webkit).

  • html/HTMLElement.cpp: (WebCore::HTMLElement::mapToEntry): (WebCore::HTMLElement::parseMappedAttribute):
21:54 Applications using WebKit edited by yves@mailhot.com
(diff)
18:28 Changeset [61052] by mjs@apple.com

2010-06-11 Maciej Stachowiak <mjs@apple.com>

Reviewed by Ojan Vafai.

Implement HTML5 hidden attribute
https://bugs.webkit.org/show_bug.cgi?id=40511

Test: fast/html/hidden-attr.html


Note: I used the mapped attribute mechanism for this instead of a rule in the UA stylesheet
to avoid a performance hit from adding a global attribute rule to the UA stylesheet.

  • html/HTMLElement.cpp: (WebCore::HTMLElement::mapToEntry): Pick up hidden as a global mapped attribute. (WebCore::HTMLElement::parseMappedAttribute): Map hidden attribute to display: none.

2010-06-11 Maciej Stachowiak <mjs@apple.com>

Reviewed by Ojan Vafai.

Implement HTML5 hidden attribute
https://bugs.webkit.org/show_bug.cgi?id=40511

  • fast/html/hidden-attr-expected.txt: Added.
  • fast/html/hidden-attr.html: Added. Test static and dynamic cases of hidden attribute.
17:54 Changeset [61051] by simon.fraser@apple.com

2010-06-11 Simon Fraser <simon.fraser@apple.com>

Add expected result missing from the previous commit.

  • svg/text/text-style-invalid-expected.txt: Added.
17:52 Changeset [61050] by simon.fraser@apple.com

2010-06-10 Abhishek Arya <inferno@chromium.org>

Reviewed by Dave Hyatt.

Do not render CSS Styles :first-letter and :first-line in a SVG text element context.
https://bugs.webkit.org/show_bug.cgi?id=40031

Test: svg/text/text-style-invalid.svg

  • rendering/RenderSVGText.cpp: (WebCore::RenderSVGText::firstLineBlock): (WebCore::RenderSVGText::updateFirstLetter):
  • rendering/RenderSVGText.h:
17:24 Changeset [61049] by weinig@apple.com

Use -Os for optimized builds instead of -02. -02 wasn't giving the
right trade off at this time.

Reviewed by Mark Rowe.

  • Configurations/Base.xcconfig:
17:17 Changeset [61048] by weinig@apple.com

Move WKRetain and WKRelease overloaded functions out of WKRetainPtr
and into the files of the type they overload (eg, WKRetain(WKFrameRef
moves to WKFrame.h)).

Reviewed by Anders Carlsson.

  • UIProcess/API/C/WKBase.h:
  • UIProcess/API/C/WKContext.h:
  • UIProcess/API/C/WKFrame.h:
  • UIProcess/API/C/WKFramePolicyListener.h:
  • UIProcess/API/C/WKNavigationData.h:
  • UIProcess/API/C/WKPage.h:
  • UIProcess/API/C/WKPageNamespace.h:
  • UIProcess/API/C/WKPreferences.h:
  • UIProcess/API/C/WKString.h:
  • UIProcess/API/C/WKURL.h:
  • UIProcess/API/cpp/WKRetainPtr.h:
  • UIProcess/API/win/WKView.h:
16:55 Changeset [61047] by ojan@chromium.org

2010-06-10 Ojan Vafai <ojan@chromium.org>

Reviewed by Eric Seidel.

upload all patches that lack in-rietveld to rietveld
https://bugs.webkit.org/show_bug.cgi?id=40444

If it's a patch, then we'll try to upload it as long as it
doesn't have in-rietveld set. No longer set in-rietveld? since
it's not needed.

  • Scripts/webkitpy/common/net/bugzilla.py:
16:44 Changeset [61046] by kbr@google.com

2010-06-11 Kenneth Russell <kbr@google.com>

Reviewed by Dimitri Glazkov.

getParameter with UNPACK_FLIP_Y_WEBGL and UNPACK_PREMULTIPLY_ALPHA_WEBGL is buggy
https://bugs.webkit.org/show_bug.cgi?id=40506

Return the data members of WebGLRenderingContext for these
parameter queries rather than asking OpenGL, which knows nothing
about them.

  • html/canvas/WebGLRenderingContext.cpp: (WebCore::WebGLRenderingContext::getParameter):

2010-06-11 Kenneth Russell <kbr@google.com>

Reviewed by Dimitri Glazkov.

getParameter with UNPACK_FLIP_Y_WEBGL and UNPACK_PREMULTIPLY_ALPHA_WEBGL is buggy
https://bugs.webkit.org/show_bug.cgi?id=40506

Return the data members of WebGLRenderingContext for these
parameter queries rather than asking OpenGL, which knows nothing
about them.

  • fast/canvas/webgl/gl-get-calls-expected.txt:
  • fast/canvas/webgl/gl-get-calls.html:
16:38 Changeset [61045] by simon.fraser@apple.com

2010-06-11 Simon Fraser <simon.fraser@apple.com>

Reviewed by Dr Dan Bernstein.

REGRESSION: crash when unloading an iFrame with Flash from the DOM
https://bugs.webkit.org/show_bug.cgi?id=40161
<rdar://problem/7994710>

Null-check the ownerElement of the RenderView's document when unhooking the compositing
root of an iframe whose layers are parented via the enclosing document. Fixes a crash when
dynamically removing such an iframe.

Test: compositing/iframes/remove-iframe-crash.html

  • rendering/RenderLayerCompositor.cpp: (WebCore::RenderLayerCompositor::detachRootPlatformLayer):
16:33 Changeset [61044] by jhawkins@chromium.org

2010-06-11 Abhishek Arya <inferno@chromium.org>

Reviewed by David Hyatt.

Don't process floats if parent node is not a RenderBlock.
https://bugs.webkit.org/show_bug.cgi?id=40033

Test: svg/text/clear-floats-crash.svg

  • rendering/RenderBlock.cpp: (WebCore::RenderBlock::clearFloats):

2010-06-11 Abhishek Arya <inferno@chromium.org>

Reviewed by David Hyatt.

Tests that we do not crash when clearing floats during SVG load.
https://bugs.webkit.org/show_bug.cgi?id=40033

  • svg/text/clear-floats-crash-expected.txt: Added.
  • svg/text/clear-floats-crash.svg: Added.
15:50 Changeset [61043] by dimich@chromium.org

Unreviewed, fixing Chromium test bots.

  • platform/chromium/test_expectations.txt:

Adding MISSING expectation for the fast/css/nested-rounded-corners.html

15:41 Changeset [61042] by kbr@google.com

2010-06-11 Kenneth Russell <kbr@google.com>

Unreviewed, test fix.

gl-get-calls failing apparently due to UNPACK_ALIGNMENT test
https://bugs.webkit.org/show_bug.cgi?id=40502

Disabled new test of UNPACK_ALIGNMENT which appears to be failing
on some of the bots.

  • fast/canvas/webgl/gl-get-calls-expected.txt:
  • fast/canvas/webgl/gl-get-calls.html:
15:10 Changeset [61041] by eric@webkit.org

2010-06-11 Sheriff Bot <webkit.review.bot@gmail.com>

Unreviewed, rolling out r61036.
http://trac.webkit.org/changeset/61036
https://bugs.webkit.org/show_bug.cgi?id=40499

broke chromium compile (Requested by jamesr on #webkit).

  • bindings/v8/SerializedScriptValue.cpp:
  • bindings/v8/SerializedScriptValue.h: (WebCore::SerializedScriptValue::deserializeAndSetProperty): (WebCore::SerializedScriptValue::create): (WebCore::SerializedScriptValue::createFromWire): (WebCore::SerializedScriptValue::release): (WebCore::SerializedScriptValue::SerializedScriptValue):
15:08 Changeset [61040] by jamesr@google.com

2010-06-11 James Robinson <jamesr@chromium.org>

Unreviewed baselines for new test on Qt/Gtk.

https://bugs.webkit.org/show_bug.cgi?id=40456

Qt and Gtk appear to produce layers of slightly different sizes
for blocks with rounded corners, probably due to roundoff difference
in the graphics layers. Since the diffs are +/- 1 pixel, just
rebaselining.

  • platform/gtk/fast/css/nested-rounded-corners-expected.txt: Added.
  • platform/qt/fast/css/nested-rounded-corners-expected.txt: Added.
15:03 Changeset [61039] by kbr@google.com

2010-06-11 Kenneth Russell <kbr@google.com>

Reviewed by Dimitri Glazkov.

Add texImage2D and texSubImage2D signatures with internal format, etc.
https://bugs.webkit.org/show_bug.cgi?id=40318

Added new texImage2D and texSubImage2D entry points following
WebGL specification changes. Added UNPACK_FLIP_Y_WEBGL and
UNPACK_PREMULTIPLY_ALPHA_WEBGL pixelStorei parameters, honored
them for the new entry points taking HTML elements, and added
support to getParameter. Added warnings to the console for calls
to the obsolete entry points, which will be removed in a few
weeks. Updated all layout tests to use the new entry points and
pixelStorei parameters, and associated expectations. No other new
tests.

  • html/canvas/WebGLRenderingContext.cpp: (WebCore::WebGLRenderingContext::WebGLRenderingContext): (WebCore::WebGLRenderingContext::getParameter): (WebCore::WebGLRenderingContext::pixelStorei): (WebCore::WebGLRenderingContext::texImage2DImpl): (WebCore::WebGLRenderingContext::texImage2D): (WebCore::WebGLRenderingContext::texSubImage2DImpl): (WebCore::WebGLRenderingContext::texSubImage2D): (WebCore::WebGLRenderingContext::printWarningToConsole):
  • html/canvas/WebGLRenderingContext.h:
  • html/canvas/WebGLRenderingContext.idl:
  • platform/graphics/GraphicsContext3D.h: (WebCore::GraphicsContext3D::):

2010-06-11 Kenneth Russell <kbr@google.com>

Reviewed by Dimitri Glazkov.

Add texImage2D and texSubImage2D signatures with internal format, etc.
https://bugs.webkit.org/show_bug.cgi?id=40318

Added new texImage2D and texSubImage2D entry points following
WebGL specification changes. Added UNPACK_FLIP_Y_WEBGL and
UNPACK_PREMULTIPLY_ALPHA_WEBGL pixelStorei parameters, honored
them for the new entry points taking HTML elements, and added
support to getParameter. Added warnings to the console for calls
to the obsolete entry points, which will be removed in a few
weeks. Updated all layout tests to use the new entry points and
pixelStorei parameters, and associated expectations. No other new
tests.

  • fast/canvas/webgl/bug-32364-expected.txt:
  • fast/canvas/webgl/bug-32364.html:
  • fast/canvas/webgl/bug-32888.html:
  • fast/canvas/webgl/gl-get-calls-expected.txt:
  • fast/canvas/webgl/gl-get-calls.html:
  • fast/canvas/webgl/resources/utils3d.js: (doLoadImageTexture):
  • fast/canvas/webgl/script-tests/texImageTest.js:
  • fast/canvas/webgl/tex-image-and-sub-image-2d-with-image-data-expected.txt:
  • fast/canvas/webgl/tex-image-and-sub-image-2d-with-image-data.html:
  • fast/canvas/webgl/tex-image-and-sub-image-2d-with-image-expected.txt:
  • fast/canvas/webgl/tex-image-and-sub-image-2d-with-image.html:
  • fast/canvas/webgl/texImage2DImageDataTest.html:
  • fast/canvas/webgl/texImageTest-expected.txt:
  • fast/canvas/webgl/texture-npot.html:
14:43 Fixing page breaking edited by jamesr@google.com
(diff)
14:42 WikiStart edited by jamesr@google.com
(diff)
14:42 Fixing page breaking created by jamesr@google.com
14:33 Changeset [61038] by zmo@google.com

Roll out the failing part of fast/canvas/webgl/get-active-test.html (failed in Chrome in Mac)

14:24 QtWebKitContrib edited by adawit@kde.org
Added more details on how to track QtWebKit release staging branches (diff)
14:08 Changeset [61037] by jamesr@google.com

2010-06-11 James Robinson <jamesr@chromium.org>

Reviewed by Dimitri Glazkov.

[chromium] Skia mispaints pages with border-radius
https://bugs.webkit.org/show_bug.cgi?id=40456

Skia mispaints pages that have border radius set in some cases. The bug is in the anti aliased
clip path logic used to implement anti aliased curves in Skia. Since Skia internally only supports
1-bit clips, anti aliased clipping is emulated by creating a new alpha layer, storing a set of
clip paths on the side, and then 'erasing' the regions outside the clip. See r49641.
PlatformContextSkia maintains a stack of PlatformContextSkia::State objects that preserve information
like fill color, drawing mode, etc that is manipulated by GraphicsContext::save() /
GraphicsContext::restore() calls as well some internal functions. Whenever a new State object is pushed
a new copy of the current State object is pushed onto the top of this stack using the copy c'tor. The
set of anti alias clip paths is also stored on the State object, but not copied when new entries are
added as the paths only apply to that entry on the stack.

The bug is that the state stack is stored in a WTF::Vector. When this vector exceeds its capacity
(by default at 16 elements) all of the existing State entries are copied into the new buffer using
State's copy constructor. This does not preserve the anti alias clip paths, so when the State entries
are popped the anti aliasing info is lost. This corrupts all further paint operations since it results
in inbalanced save/restore calls to the underlying SkCanvas.

The fix is to make the PlatformContextSkia::State copy constructor copy all fields and to add a new
function PlatformContextSkia::State::cloneInheritedProperties to use when pushing new State entries
that copies everything except for the anti aliased clip paths.

Test: fast/css/nested-rounded-corners.html

  • platform/graphics/skia/PlatformContextSkia.cpp: (PlatformContextSkia::State::State): (PlatformContextSkia::State::cloneInheritedProperties): (PlatformContextSkia::save):

2010-06-11 James Robinson <jamesr@chromium.org>

Reviewed by Dimitri Glazkov.

[chromium] Skia mispaints pages with border-radius
https://bugs.webkit.org/show_bug.cgi?id=40456

Tests deeply nested divs with border-radius.

  • fast/css/nested-rounded-corners-expected.txt: Added.
  • fast/css/nested-rounded-corners.html: Added.
  • platform/chromium/test_expectations.txt
  • platform/chromium-linux/fast/css/nested-rounded-corners-expected.checksum: Added.
  • platform/chromium-linux/fast/css/nested-rounded-corners-expected.png: Added.
13:41 Changeset [61036] by eric@webkit.org

2010-06-11 Jeremy Orlow <jorlow@chromium.org>

Reviewed by Darin Fisher.

[V8] Clean up SerializedScriptValue
https://bugs.webkit.org/show_bug.cgi?id=40482

SerializedScriptValue doesn't follow WebKit's style guidelines very well
and needlessly inlines quite a bit within the .h file. This change cleans
things up. No funcitonal changes.

No change in behavior.

  • bindings/v8/SerializedScriptValue.cpp: (WebCore::SerializedScriptValue::deserializeAndSetProperty): (WebCore::SerializedScriptValue::create): (WebCore::SerializedScriptValue::createFromWire): (WebCore::SerializedScriptValue::release): (WebCore::SerializedScriptValue::SerializedScriptValue):
  • bindings/v8/SerializedScriptValue.h:
13:20 Changeset [61035] by eric@webkit.org

2010-06-11 James Hawkins <jhawkins@chromium.org>

Reviewed by Darin Fisher.

[Chromium] Notify the WebView when the AutoFill popup hides.

https://bugs.webkit.org/show_bug.cgi?id=40459

  • src/AutoFillPopupMenuClient.cpp: (WebKit::AutoFillPopupMenuClient::selectionCleared):
12:57 Changeset [61034] by eric@webkit.org

2010-06-11 Jesus Sanchez-Palencia <jesus@webkit.org>

Reviewed by Kenneth Rohde Christiansen.

[Qt] Typo error in QWebPluginFactory Documentation
https://bugs.webkit.org/show_bug.cgi?id=40490

  • Api/qwebpluginfactory.cpp:
12:37 Changeset [61033] by eric@webkit.org

2010-06-11 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

tables/mozilla/bugs/bug1188.html needlessly depends on HTML Parser text node handling
https://bugs.webkit.org/show_bug.cgi?id=40485

The old HTMLTokenizer handled invalid close tags like "< /A>" as part
of the lexing step. The HTML5 parser specification dictates that
the tokenization/lexing step is to output each character token in sequence
and that the parser is to insert emitted character tokens into the last
parser-created text node. In our current HTML5Lexer-with-old-HTMLParser
setup, we gather up character tokens and output them all in one bunch
(to match the expectations of the old parser).

In the case of "foo< /A>" the HTML5Lexer sees the '<', switched out of the
"data" state and emits all pending characters assuming the start of a close tag.
When it sees the ' ' it switches back to the "data" state. The HTML5Lexer
expects that the parser will coalesce the separate characters it emitted
into one text node. Since we haven't implemented the parser side of the
HTML5 spec, for now we don't. This is a known issue we will address after
enabling the HTML5Lexer, as we start to write an HTML5-compliant parser.

This text-node-coalescing behavior is already tested by numerous html5lib
tests, however this was the only layout test which hit this quirk. I've
added yet another html5lib test of this exact case just for good measure.

  • html5lib/resources/webkit01.dat:
    • Add a test of this exact case to make sure we don't miss it when implementing an HTML5-compliant parser.
  • tables/mozilla/bugs/bug1188.html:
    • Replace "< /A>" with "&lt /A>" to avoid hitting this lexing quirk.
12:23 Changeset [61032] by eric@webkit.org

2010-06-11 Tony Gentilcore <tonyg@chromium.org>

Reviewed by Dimitri Glazkov.

Roll chromium DEPS 49376:49521
https://bugs.webkit.org/show_bug.cgi?id=40488

  • DEPS:
12:11 Changeset [61031] by eric@webkit.org

2010-06-11 Anton Muhin <antonm@chromium.org>

Reviewed by Adam Barth.

[v8] Introduce single element caches for WebCore::String to v8::String conversions
https://bugs.webkit.org/show_bug.cgi?id=40435
Measurements show that for some web apps (GMail, Wave) and some scenarios
(intensive reading and/or keeping a tab open for a long time),
hit rate lies in 30--50% interval.
Inlining fast case gives another minor performance win.

  • bindings/v8/V8Binding.cpp: (WebCore::getStringCache): (WebCore::v8ExternalStringSlow):
  • bindings/v8/V8Binding.h: (WebCore::v8ExternalString):
12:00 Changeset [61030] by eric@webkit.org

2010-06-11 Leandro Pereira <leandro@profusion.mobi>

Reviewed by Gustavo Noronha Silva.

[EFL] Build fix: remove unneeded file (npapi.cpp) and add
other files that got added to the tree.
http://webkit.org/b/40331

  • CMakeLists.txt:
  • CMakeListsEfl.txt:
11:48 Changeset [61029] by dimich@chromium.org

Unreviewed, update Chromium test expectations.

  • platform/chromium/test_expectations.txt:

http://trac.webkit.org/changeset/60999 regressed acid3.html image match on Chromium Mac.
http://trac.webkit.org/changeset/61020 modified get-active-test.html that now fails on Chromium Mac.

11:46 Changeset [61028] by eric@webkit.org

2010-06-11 Tony Gentilcore <tonyg@chromium.org>

Reviewed by Eric Seidel.

Fix fast/forms/preserveFormDuringResidualStyle.html for HTML5 Parser.
https://bugs.webkit.org/show_bug.cgi?id=40454

This required exposing the HTMLParser which we definitely don't want to
do, but the good news is that it can go away when the HTMLParser goes
away.

No new tests because covered by fast/forms/preserve/FormDuringResidualStyle.html.

  • dom/Tokenizer.h: (WebCore::Tokenizer::htmlParser):
  • html/HTML5Tokenizer.cpp: (WebCore::HTML5Tokenizer::htmlParser):
  • html/HTML5Tokenizer.h:
  • html/HTML5TreeBuilder.h: (WebCore::HTML5TreeBuilder::htmlParser):
  • html/HTMLFormControlElement.cpp: (WebCore::HTMLFormControlElement::removedFromTree):
  • html/HTMLTokenizer.h: (WebCore::HTMLTokenizer::htmlParser):
11:22 Changeset [61027] by kbr@google.com

2010-06-11 Kenneth Russell <kbr@google.com>

Reviewed by Dimitri Glazkov.

Delete custom JSC bindings for bufferData, texImage2D and texSubImage2D
https://bugs.webkit.org/show_bug.cgi?id=40443

Deleted custom JSC bindings for bufferData, texImage2D and
texSubImage2D and fixed minor associated issues in code generator
and IDL. Updated fast/canvas/webgl/texImageTest-expected.txt,
which now runs as intended. Ran all layout tests, including WebGL
tests, in Safari.

  • bindings/js/JSWebGLRenderingContextCustom.cpp:
    • Deleted custom bindings.
  • bindings/scripts/CodeGeneratorJS.pm:
    • Made overload helper functions static to fix compiler warnings.
    • Added needed isObject check for overloads between object and non-object arguments.
    • Added JSValue::encode to exception return path, needed on Leopard in particular.
  • html/HTMLCanvasElement.idl:
  • html/HTMLImageElement.idl:
  • html/HTMLVideoElement.idl:
    • Added GenerateNativeConverter extended attribute.
  • html/canvas/WebGLRenderingContext.idl:
    • Deleted #ifdefs for custom JSC bindings.

2010-06-11 Kenneth Russell <kbr@google.com>

Reviewed by Dimitri Glazkov.

Delete custom JSC bindings for bufferData, texImage2D and texSubImage2D
https://bugs.webkit.org/show_bug.cgi?id=40443

Deleted custom JSC bindings for bufferData, texImage2D and
texSubImage2D and fixed minor associated issues in code generator
and IDL. Updated fast/canvas/webgl/texImageTest-expected.txt,
which now runs as intended. Ran all layout tests, including WebGL
tests, in Safari.

  • fast/canvas/webgl/script-tests/texImageTest.js:
  • fast/canvas/webgl/texImageTest-expected.txt:
10:54 Changeset [61026] by ap@apple.com

Written by Dave Moore <davemoore@google.com>.

https://bugs.webkit.org/show_bug.cgi?id=22741
innerHTML, forms and images don't play well together

This was fixed already (in bug 34076), just landing an additional test.

  • fast/forms/form-image-access-by-name-expected.txt: Added.
  • fast/forms/form-image-access-by-name.html: Added.
10:42 Changeset [61025] by kevino@webkit.org

[wx] Build fix, add KillRingNone.cpp until we can implement it.

09:49 Changeset [61024] by ojan@chromium.org

2010-06-10 Ojan Vafai <ojan@chromium.org>

Reviewed by Darin Fisher.

Fix rietveld upload when message length > 100 characters
https://bugs.webkit.org/show_bug.cgi?id=40457

  • Scripts/webkitpy/common/net/rietveld.py:
09:17 Changeset [61023] by eric@webkit.org

2010-06-11 Hans Wennborg <hans@chromium.org>

Reviewed by Jeremy Orlow.

DOM storage should only create databases when needed
https://bugs.webkit.org/show_bug.cgi?id=40301

As soon as a page attempts to use localstorage, StorageAreaSync will
create an empty database if one doesn't already exist. This can lead to
lots of unnecessary database files. In particular, they are created
even when the privacy settings or private browsing mode disallow
localstorage data, which may seem odd to the user.

Database creation should be put off in StorageAreaSync until it is time
to actually write something to the database.

Tests:

manual-tests/localstorage-empty-database.html

  • manual-tests/localstorage-empty-database.html: Added.
  • storage/StorageAreaSync.cpp: (WebCore::StorageAreaSync::StorageAreaSync): (WebCore::StorageAreaSync::openDatabase): (WebCore::StorageAreaSync::performImport): (WebCore::StorageAreaSync::sync):
  • storage/StorageAreaSync.h: (WebCore::StorageAreaSync::):
09:04 Changeset [61022] by eric@webkit.org

2010-06-11 Ilya Tikhonovsky <loislo@chromium.org>

Reviewed by Eric Seidel.

Perl scripts for bindings don't use strict keyword.
As result some problems in the script code were not detected at compile stage.
https://bugs.webkit.org/show_bug.cgi?id=40468

  • bindings/scripts/CodeGenerator.pm:
  • bindings/scripts/IDLParser.pm:
  • bindings/scripts/IDLStructure.pm:
08:28 Changeset [61021] by eric@webkit.org

2010-06-11 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

HTML5Tokenizer should not delete itself while running scripts
https://bugs.webkit.org/show_bug.cgi?id=40458

Tested by fast/dom/document-clear.html

  • html/HTML5Tokenizer.cpp: (WebCore::HTML5Tokenizer::attemptToEnd):
    • Don't ever end() while running scripts. This matches the old HTMLTokenizer.cpp
08:16 Changeset [61020] by eric@webkit.org

2010-06-11 Zhenyao Mo <zmo@google.com>

Reviewed by Dimitri Glazkov.

getActiveUniform must ensure names of arrays end in "[0]"
https://bugs.webkit.org/show_bug.cgi?id=38709

  • fast/canvas/webgl/get-active-test-expected.txt: Copied from LayoutTests/fast/canvas/webgl/getActiveTest-expected.txt, modified to test array names.
  • fast/canvas/webgl/get-active-test.html: Copied from LayoutTests/fast/canvas/webgl/getActiveTest.html.
  • fast/canvas/webgl/getActiveTest-expected.txt: Removed.
  • fast/canvas/webgl/getActiveTest.html: Removed.
  • fast/canvas/webgl/resources/intArrayUniformShader.vert: Added, with array uniform.
  • fast/canvas/webgl/script-tests/get-active-test.js: Copied from LayoutTests/fast/canvas/webgl/script-tests/getActiveTest.js, modified to test array names.
  • fast/canvas/webgl/script-tests/getActiveTest.js: Removed.

2010-06-11 Zhenyao Mo <zmo@google.com>

Reviewed by Dimitri Glazkov.

getActiveUniform must ensure names of arrays end in "[0]"
https://bugs.webkit.org/show_bug.cgi?id=38709

Test: fast/canvas/webgl/get-active-test.html

  • html/canvas/WebGLRenderingContext.cpp: (WebCore::WebGLRenderingContext::getActiveAttrib): Append "[0]" to the name if needed.
08:03 Changeset [61019] by eric@webkit.org

2010-06-11 Zhenyao Mo <zmo@google.com>

Reviewed by Dimitri Glazkov.

readPixels with negative width/height should generate INVALID_VALUE and return
https://bugs.webkit.org/show_bug.cgi?id=39704

  • fast/canvas/webgl/read-pixels-expected.txt: Add test cases for non-positive width/height.
  • fast/canvas/webgl/read-pixels.html: Ditto.

2010-06-11 Zhenyao Mo <zmo@google.com>

Reviewed by Dimitri Glazkov.

readPixels with negative width/height should generate INVALID_VALUE and return
https://bugs.webkit.org/show_bug.cgi?id=39704

  • html/canvas/WebGLRenderingContext.cpp: Deal with negative width/height. (WebCore::WebGLRenderingContext::readPixels):
  • html/canvas/WebGLRenderingContext.h: Change width/height type from unsigned long to long.
  • html/canvas/WebGLRenderingContext.idl: Ditto.
07:50 Changeset [61018] by eric@webkit.org

2010-06-11 Zhenyao Mo <zmo@google.com>

Reviewed by Dimitri Glazkov.

uniform* entry points must ignore the call if a null WebGLUniformLocation is passed
https://bugs.webkit.org/show_bug.cgi?id=38707

  • fast/canvas/webgl/null-uniform-location-expected.txt: No error is generated if a null WebGLUniformLocation is passed to uniform*().
  • fast/canvas/webgl/script-tests/null-uniform-location.js: Ditto.
  • fast/canvas/webgl/script-tests/uniform-location.js: Ditto.
  • fast/canvas/webgl/uniform-location-expected.txt: Ditto.

2010-06-11 Zhenyao Mo <zmo@google.com>

Reviewed by Dimitri Glazkov.

uniform* entry points must ignore the call if a null WebGLUniformLocation is passed
https://bugs.webkit.org/show_bug.cgi?id=38707

  • html/canvas/WebGLRenderingContext.cpp: uniform* entry points ignore the call if a null WebGLUniformLocation is passed. (WebCore::WebGLRenderingContext::uniform1f): (WebCore::WebGLRenderingContext::uniform1fv): (WebCore::WebGLRenderingContext::uniform1i): (WebCore::WebGLRenderingContext::uniform1iv): (WebCore::WebGLRenderingContext::uniform2f): (WebCore::WebGLRenderingContext::uniform2fv): (WebCore::WebGLRenderingContext::uniform2i): (WebCore::WebGLRenderingContext::uniform2iv): (WebCore::WebGLRenderingContext::uniform3f): (WebCore::WebGLRenderingContext::uniform3fv): (WebCore::WebGLRenderingContext::uniform3i): (WebCore::WebGLRenderingContext::uniform3iv): (WebCore::WebGLRenderingContext::uniform4f): (WebCore::WebGLRenderingContext::uniform4fv): (WebCore::WebGLRenderingContext::uniform4i): (WebCore::WebGLRenderingContext::uniform4iv): (WebCore::WebGLRenderingContext::uniformMatrix2fv): (WebCore::WebGLRenderingContext::uniformMatrix3fv): (WebCore::WebGLRenderingContext::uniformMatrix4fv):
07:38 Changeset [61017] by eric@webkit.org

2010-06-11 Zhenyao Mo <zmo@google.com>

Reviewed by Dimitri Glazkov.

drawElements/drawArrays should validate input parameters according to GLES2 spec
https://bugs.webkit.org/show_bug.cgi?id=38700

  • fast/canvas/webgl/draw-arrays-out-of-bounds-expected.txt: Copied from LayoutTests/fast/canvas/webgl/drawArraysOutOfBounds-expected.txt, and fix a few test cases.
  • fast/canvas/webgl/draw-arrays-out-of-bounds.html: Copied from LayoutTests/fast/canvas/webgl/drawArraysOutOfBounds.html, and fix a few test cases.
  • fast/canvas/webgl/draw-elements-out-of-bounds-expected.txt: Copied from LayoutTests/fast/canvas/webgl/drawElementssOutOfBounds-expected.txt, and fix a few test cases.
  • fast/canvas/webgl/draw-elements-out-of-bounds.html: Copied from LayoutTests/fast/canvas/webgl/drawElementssOutOfBounds.html, and fix a few test cases.
  • fast/canvas/webgl/drawArraysOutOfBounds-expected.txt: Removed.
  • fast/canvas/webgl/drawArraysOutOfBounds.html: Removed.
  • fast/canvas/webgl/drawElementssOutOfBounds-expected.txt: Removed.
  • fast/canvas/webgl/drawElementssOutOfBounds.html: Removed.
  • fast/canvas/webgl/resources/webgl-test.js: Add helper functions for easy test of gl function errors. (getGLErrorAsString): (shouldGenerateGLError):

2010-06-11 Zhenyao Mo <zmo@google.com>

Reviewed by Dimitri Glazkov.

drawElements/drawArrays should validate input parameters according to GLES2 spec
https://bugs.webkit.org/show_bug.cgi?id=38700

Tests: fast/canvas/webgl/draw-arrays-out-of-bounds.html

fast/canvas/webgl/draw-elements-out-of-bounds.html

  • html/canvas/WebGLRenderingContext.cpp: (WebCore::WebGLRenderingContext::drawArrays): Validate input parameters. (WebCore::WebGLRenderingContext::drawElements): Ditto. (WebCore::WebGLRenderingContext::validateDrawMode): Validate mode for draw{Arrays/Elements}.
  • html/canvas/WebGLRenderingContext.h: Add validateDrawMode, fix incorrect parameter types.
  • html/canvas/WebGLRenderingContext.idl: Fix incorrect parameter types.
07:25 Changeset [61016] by eric@webkit.org

2010-06-11 Luiz Agostini <luiz.agostini@openbossa.org>, Kenneth Rohde Christiansen <kenneth@webkit.org>

Reviewed by Antti Koivisto.

CSS3 Media Queries are not serialized according to CSSOM
https://bugs.webkit.org/show_bug.cgi?id=39220

Added test for testing the serialization, plus updated current
tests to take serialization into account.

  • fast/css/insertRule-media-expected.txt:
  • fast/css/insertRule-media.html:
  • fast/css/resources/insertRule-media.js:
  • fast/media/media-query-serialization-expected.txt: Added.
  • fast/media/media-query-serialization.html: Added.

2010-06-11 Luiz Agostini <luiz.agostini@openbossa.org>

Reviewed by Antti Koivisto.

CSS3 Media Queries are not serialized according to CSSOM
https://bugs.webkit.org/show_bug.cgi?id=39220

MediaQuery serialization according to specification
http://dev.w3.org/csswg/cssom/#serializing-media-queries.

MediaQuery objects now ignore duplicated expressions. MediaQueryEvaluator::eval()
now does not proccess MediaQuery objects that are known to be invalid.

Test: fast/media/media-query-serialization.html

  • css/MediaQuery.cpp: (WebCore::stringCompare): (WebCore::MediaQuery::serialize): (WebCore::MediaQuery::MediaQuery): (WebCore::MediaQuery::~MediaQuery): (WebCore::MediaQuery::operator==): (WebCore::MediaQuery::cssText): (WebCore::MediaQuery::append):
  • css/MediaQuery.h: (WebCore::MediaQuery::ignored): (WebCore::MediaQuery::begin): (WebCore::MediaQuery::end):
  • css/MediaQueryExp.cpp: (WebCore::MediaQueryExp::serialize):
  • css/MediaQueryExp.h:
  • css/MediaQueryEvaluator.cpp: (WebCore::MediaQueryEvaluator::eval):
07:02 Changeset [61015] by jorlow@chromium.org

2010-06-10 Jeremy Orlow <jorlow@chromium.org>

Reviewed by Steve Block.

Implement more of IndexedDB's Indexes and ObjectStores
https://bugs.webkit.org/show_bug.cgi?id=40424

Clean up both of these classes a bit and add create/open/remove
methods as needed.

Tests in a future patch (promise).

  • storage/IDBDatabase.h:
  • storage/IDBDatabaseImpl.cpp: (WebCore::IDBDatabaseImpl::objectStores): (WebCore::IDBDatabaseImpl::createObjectStore): (WebCore::IDBDatabaseImpl::objectStore): (WebCore::IDBDatabaseImpl::removeObjectStore):
  • storage/IDBDatabaseImpl.h: (WebCore::IDBDatabaseImpl::name): (WebCore::IDBDatabaseImpl::description): (WebCore::IDBDatabaseImpl::version):
  • storage/IDBDatabaseRequest.cpp: (WebCore::IDBDatabaseRequest::IDBDatabaseRequest): (WebCore::IDBDatabaseRequest::createObjectStore): (WebCore::IDBDatabaseRequest::objectStore): (WebCore::IDBDatabaseRequest::removeObjectStore):
  • storage/IDBDatabaseRequest.h: (WebCore::IDBDatabaseRequest::create): (WebCore::IDBDatabaseRequest::name): (WebCore::IDBDatabaseRequest::description): (WebCore::IDBDatabaseRequest::version): (WebCore::IDBDatabaseRequest::objectStores):
  • storage/IDBDatabaseRequest.idl:
  • storage/IDBObjectStore.h:
  • storage/IDBObjectStoreImpl.h:
  • storage/IDBObjectStoreRequest.cpp: (WebCore::IDBObjectStoreRequest::get): (WebCore::IDBObjectStoreRequest::add): (WebCore::IDBObjectStoreRequest::modify): (WebCore::IDBObjectStoreRequest::addOrModify): (WebCore::IDBObjectStoreRequest::remove): (WebCore::IDBObjectStoreRequest::createIndex): (WebCore::IDBObjectStoreRequest::index): (WebCore::IDBObjectStoreRequest::removeIndex):
  • storage/IDBObjectStoreRequest.h:
  • storage/IDBObjectStoreRequest.idl:

2010-06-10 Jeremy Orlow <jorlow@chromium.org>

Reviewed by Steve Block.

Implement more of IndexedDB's Indexes and ObjectStores
https://bugs.webkit.org/show_bug.cgi?id=40424

Add plumbing for both of these classes.

  • public/WebIDBDatabase.h: (WebKit::WebIDBDatabase::name): (WebKit::WebIDBDatabase::description): (WebKit::WebIDBDatabase::version): (WebKit::WebIDBDatabase::objectStores): (WebKit::WebIDBDatabase::objectStore): (WebKit::WebIDBDatabase::removeObjectStore):
  • public/WebIDBObjectStore.h: (WebKit::WebIDBObjectStore::indexNames): (WebKit::WebIDBObjectStore::createIndex): (WebKit::WebIDBObjectStore::index): (WebKit::WebIDBObjectStore::removeIndex):
  • src/IDBDatabaseProxy.cpp: (WebCore::IDBDatabaseProxy::name): (WebCore::IDBDatabaseProxy::description): (WebCore::IDBDatabaseProxy::version): (WebCore::IDBDatabaseProxy::objectStores): (WebCore::IDBDatabaseProxy::objectStore): (WebCore::IDBDatabaseProxy::removeObjectStore):
  • src/IDBDatabaseProxy.h:
  • src/IDBObjectStoreProxy.cpp: (WebCore::IDBObjectStoreProxy::indexNames): (WebCore::IDBObjectStoreProxy::createIndex): (WebCore::IDBObjectStoreProxy::index): (WebCore::IDBObjectStoreProxy::removeIndex):
  • src/IDBObjectStoreProxy.h:
  • src/WebIDBDatabaseImpl.cpp: (WebKit::WebIDBDatabaseImpl::WebIDBDatabaseImpl): (WebKit::WebIDBDatabaseImpl::name): (WebKit::WebIDBDatabaseImpl::description): (WebKit::WebIDBDatabaseImpl::version): (WebKit::WebIDBDatabaseImpl::objectStores): (WebKit::WebIDBDatabaseImpl::createObjectStore): (WebKit::WebIDBDatabaseImpl::objectStore): (WebKit::WebIDBDatabaseImpl::removeObjectStore):
  • src/WebIDBDatabaseImpl.h:
  • src/WebIDBObjectStoreImpl.cpp: (WebKit::WebIDBObjectStoreImpl::WebIDBObjectStoreImpl): (WebKit::WebIDBObjectStoreImpl::name): (WebKit::WebIDBObjectStoreImpl::keyPath): (WebKit::WebIDBObjectStoreImpl::indexNames): (WebKit::WebIDBObjectStoreImpl::createIndex): (WebKit::WebIDBObjectStoreImpl::index): (WebKit::WebIDBObjectStoreImpl::removeIndex):
  • src/WebIDBObjectStoreImpl.h:
06:47 Changeset [61014] by apavlov@chromium.org

2010-06-11 Alexander Pavlov <apavlov@chromium.org>

Reviewed by Pavel Feldman.

Web Inspector: Enable serialization/deserialization of the frontend state
https://bugs.webkit.org/show_bug.cgi?id=40228

WebCore:

  • inspector/InspectorBackend.cpp: (WebCore::InspectorBackend::saveApplicationSettings): (WebCore::InspectorBackend::saveSessionSettings):
  • inspector/InspectorBackend.h:
  • inspector/InspectorBackend.idl:
  • inspector/InspectorController.cpp: (WebCore::InspectorController::InspectorController): (WebCore::InspectorController::setSessionSettings): (WebCore::InspectorController::populateScriptObjects): (WebCore::InspectorController::didCommitLoad):
  • inspector/InspectorController.h:
  • inspector/InspectorFrontend.cpp: (WebCore::InspectorFrontend::populateApplicationSettings): (WebCore::InspectorFrontend::populateSessionSettings):
  • inspector/InspectorFrontend.h:
  • inspector/front-end/ConsoleView.js: (WebInspector.ConsoleView.prototype._settingsLoaded): (WebInspector.ConsoleView.prototype._enterKeyPressed.printResult): (WebInspector.ConsoleView.prototype._enterKeyPressed):
  • inspector/front-end/EventListenersSidebarPane.js: (WebInspector.EventListenersSidebarPane.prototype._settingsLoaded): (WebInspector.EventListenersSidebarPane.prototype): ():
  • inspector/front-end/InspectorBackendStub.js: (.WebInspector.InspectorBackendStub.prototype.saveApplicationSettings): (.WebInspector.InspectorBackendStub.prototype.saveSessionSettings):
  • inspector/front-end/ResourceView.js: (WebInspector.ResourceView.prototype._selectTab): (WebInspector.ResourceView.prototype._selectHeadersTab): (WebInspector.ResourceView.prototype.selectContentTab):
  • inspector/front-end/ResourcesPanel.js: (WebInspector.ResourcesPanel.prototype._createStatusbarButtons): (WebInspector.ResourcesPanel.prototype._settingsLoaded): (WebInspector.ResourcesPanel.prototype._toggleLargerResources):
  • inspector/front-end/ScriptsPanel.js: (WebInspector.ScriptsPanel.prototype._showScriptOrResource): (WebInspector.ScriptsPanel.prototype._addScriptToFilesMenu):
  • inspector/front-end/Settings.js: (WebInspector.populateApplicationSettings): (WebInspector.populateSessionSettings): (WebInspector.Settings): (WebInspector.Settings.prototype.reset): (WebInspector.Settings.prototype._load): (WebInspector.Settings.prototype._set):
  • inspector/front-end/StylesSidebarPane.js: (WebInspector.StylesSidebarPane.prototype._settingsLoaded): (WebInspector.StylesSidebarPane.prototype.update): (WebInspector.StylesSidebarPane.prototype._changeSetting): (WebInspector.StylesSidebarPane.prototype._changeColorFormat): (WebInspector.ComputedStyleSidebarPane.settingsLoaded): (WebInspector.ComputedStyleSidebarPane.showInheritedToggleFunction): (WebInspector.ComputedStyleSidebarPane):
  • inspector/front-end/WatchExpressionsSidebarPane.js: (WebInspector.WatchExpressionsSidebarPane): (WebInspector.WatchExpressionsSidebarPane.prototype._settingsLoaded): (WebInspector.WatchExpressionsSection): (WebInspector.WatchExpressionsSection.prototype.saveExpressions):
  • inspector/front-end/inspector.js: (WebInspector.loaded): (WebInspector.reset):

Webkit/chromium:

  • src/js/InspectorControllerImpl.js: (devtools.InspectorBackendImpl):
06:37 Changeset [61013] by pfeldman@chromium.org

2010-06-10 S√∏ren Gjesse <sgjesse@chromium.org>

Reviewed by Pavel Feldman.

V8: Chromium test expectations not matching improved V8 JavaScript stepping
https://bugs.webkit.org/show_bug.cgi?id=40419

Update test expetations to match chaged V8 break locations.

  • src/js/Tests.js:
06:31 Changeset [61012] by pfeldman@chromium.org

2010-06-10 Pavel Feldman <pfeldman@chromium.org>

Reviewed by Yury Semikhatsky.

Web Inspector: Port performSearch from InjectedScript to InspectorDOMAgent.

https://bugs.webkit.org/show_bug.cgi?id=40422

  • inspector/InjectedScriptHost.cpp: (WebCore::InjectedScriptHost::addNodesToSearchResult):
  • inspector/InspectorBackend.cpp: (WebCore::InspectorBackend::performSearch): (WebCore::InspectorBackend::searchCanceled):
  • inspector/InspectorBackend.h:
  • inspector/InspectorBackend.idl:
  • inspector/InspectorDOMAgent.cpp: (WebCore::): (WebCore::InspectorDOMAgent::InspectorDOMAgent): (WebCore::InspectorDOMAgent::~InspectorDOMAgent): (WebCore::InspectorDOMAgent::performSearch): (WebCore::InspectorDOMAgent::searchCanceled): (WebCore::InspectorDOMAgent::onMatchJobsTimer): (WebCore::InspectorDOMAgent::reportNodesAsSearchResults):
  • inspector/InspectorDOMAgent.h: (WebCore::MatchJob::~MatchJob): (WebCore::MatchJob::MatchJob): (WebCore::MatchJob::addNodesToResults):
  • inspector/InspectorFrontend.cpp: (WebCore::InspectorFrontend::addNodesToSearchResult):
  • inspector/InspectorFrontend.h:
  • inspector/InspectorResource.cpp: (WebCore::InspectorResource::updateScriptObject):
  • inspector/front-end/ElementsPanel.js: (WebInspector.ElementsPanel.prototype.searchCanceled): (WebInspector.ElementsPanel.prototype.performSearch): (WebInspector.ElementsPanel.prototype.addNodesToSearchResult):
  • inspector/front-end/InjectedScript.js: (injectedScriptConstructor):
  • inspector/front-end/InjectedScriptAccess.js:

2010-06-10 Pavel Feldman <pfeldman@chromium.org>

Reviewed by Yury Semikhatsky.

Web Inspector: Port performSearch from InjectedScript to InspectorDOMAgent.

https://bugs.webkit.org/show_bug.cgi?id=40422

  • src/js/InspectorControllerImpl.js: (devtools.InspectorBackendImpl):

2010-06-10 Pavel Feldman <pfeldman@chromium.org>

Reviewed by Yury Semikhatsky.

Web Inspector: Port performSearch from InjectedScript to InspectorDOMAgent.

https://bugs.webkit.org/show_bug.cgi?id=40422

  • inspector/elements-panel-search.html:
06:17 Changeset [61011] by mnaganov@chromium.org

2010-06-11 Mikhail Naganov <mnaganov@chromium.org>

Reviewed by Pavel Feldman.

[Chromium] Restore 'console.profiles' access.

https://bugs.webkit.org/show_bug.cgi?id=39840

  • bindings/v8/ScriptProfiler.cpp: (WebCore::ScriptProfiler::stop):
  • bindings/v8/custom/V8ConsoleCustom.cpp: (WebCore::V8Console::profilesAccessorGetter):
  • inspector/InspectorController.cpp: (WebCore::InspectorController::stopUserInitiatedProfiling):
05:53 Changeset [61010] by pfeldman@chromium.org

2010-06-07 Pavel Feldman <pfeldman@chromium.org>

Reviewed by Yury Semikhatsky.

Web Inspector: add a test for dumping global object in the console.

https://bugs.webkit.org/show_bug.cgi?id=40241

  • http/tests/inspector/inspector-test.js: (frontend_dumpTreeOutline): (frontend_dumpTreeItem):
  • inspector/console-dir-global-expected.txt: Added.
  • inspector/console-dir-global.html: Added.
  • inspector/elements-tests.js: (frontend_dumpStyleTreeItem):
05:48 Changeset [61009] by pfeldman@chromium.org

2010-06-10 Nikita Vasilyev <me@elv1s.ru>

Reviewed by Pavel Feldman.

Web Inspector: Make a close brace in Styles pane selectable.
https://bugs.webkit.org/show_bug.cgi?id=40157

  • inspector/front-end/inspector.css: (.styles-selector): (.styles-section .properties li): (.styles-section):
05:39 Changeset [61008] by yurys@chromium.org

2010-06-11 Yury Semikhatsky <yurys@chromium.org>

Reviewed by Pavel Feldman.

Web Inspector: localize missing script source message
https://bugs.webkit.org/show_bug.cgi?id=40467

  • English.lproj/localizedStrings.js:
  • inspector/InspectorController.cpp: (WebCore::InspectorController::didParseSource): fixed typo in comment.
05:28 Changeset [61007] by eric@webkit.org

2010-06-11 Vangelis Kokkevis <vangelis@chromium.org>

Reviewed by David Levin.

[chromium] Adding a build dependency to the gles2 C library to the webkit_unit_tests
project to allow it to link when WebCore is compiled with the ACCELERATED_COMPOSITING
flag.
https://bugs.webkit.org/show_bug.cgi?id=40341

  • WebKit.gyp:
05:16 Changeset [61006] by eric@webkit.org

2010-06-11 Zhenyao Mo <zmo@google.com>

Reviewed by Dimitri Glazkov.

Vertex attributes enabled as arrays but not bound to buffers must generate INVALID_OPERATION
https://bugs.webkit.org/show_bug.cgi?id=40315

  • fast/canvas/webgl/index-validation-expected.txt: Generate INVALID_ERROR when an enabled vertex attrib is not bound to any buffer in drawElements().
  • fast/canvas/webgl/script-tests/index-validation.js: Ditto.

2010-06-11 Zhenyao Mo <zmo@google.com>

Reviewed by Dimitri Glazkov.

Vertex attributes enabled as arrays but not bound to buffers must generate INVALID_OPERATION
https://bugs.webkit.org/show_bug.cgi?id=40315

  • html/canvas/WebGLRenderingContext.cpp: (WebCore::WebGLRenderingContext::validateRenderingState): Check if an enabled vertext attribs is bound to a buffer. (WebCore::WebGLRenderingContext::vertexAttribPointer): Set bound buffer in vertex attrib state.
  • html/canvas/WebGLRenderingContext.h: Add a member in vertex attrib state to track bound buffer.
05:01 QtWebKitJournal edited by hausmann@webkit.org
(diff)
05:00 QtWebKitJournal edited by hausmann@webkit.org
(diff)
04:47 Changeset [61005] by jocelyn.turcotte@nokia.com

[Qt] Second fix attempt for the build break introduced by r61002.

Reviewed by nobody, build fix.

  • Api/qwebpage.cpp:

(QWebPagePrivate::QWebPagePrivate):

03:42 Changeset [61004] by jocelyn.turcotte@nokia.com

[Qt] Fix build break introduced by r61002.

Reviewed by nobody, build fix.

  • Api/qwebpage.cpp:

(QWebPagePrivate::QWebPagePrivate):

03:36 Changeset [61003] by hausmann@webkit.org

[Qt] Explicit conversions from QtScript types to JSC opaque types were removed.
https://bugs.webkit.org/show_bug.cgi?id=40412

Patch by Jedrzej Nowacki <jedrzej.nowacki@nokia.com> on 2010-06-11
Reviewed by Simon Hausmann.

Conversion between a JSC C types and a QtScript private types, takes
main part of the source code. In most cases a mapping between the types
is one to one. New cast operators were added to simplify the code.

The QScriptValuePrivate could be casted to the JSValueRef and the JSObjectRef.
The QScriptEnginePrivate could be casted to the JSGlobalContext.
The QScriptProgramPrivate could be casted to the JSStringRef.

  • qt/api/qscriptengine_p.cpp:

(QScriptEnginePrivate::evaluate):
(QScriptEnginePrivate::newObject):
(QScriptEnginePrivate::globalObject):

  • qt/api/qscriptengine_p.h:

(QScriptEnginePrivate::operator JSGlobalContextRef):

  • qt/api/qscriptprogram_p.h:

(QScriptProgramPrivate::operator JSStringRef):

  • qt/api/qscriptsyntaxcheckresult.cpp:

(QScriptSyntaxCheckResultPrivate::~QScriptSyntaxCheckResultPrivate):
(QScriptSyntaxCheckResultPrivate::errorMessage):
(QScriptSyntaxCheckResultPrivate::errorLineNumber):

  • qt/api/qscriptvalue_p.h:

(QScriptValuePrivate::~QScriptValuePrivate):
(QScriptValuePrivate::QScriptValuePrivate):
(QScriptValuePrivate::isBool):
(QScriptValuePrivate::isNumber):
(QScriptValuePrivate::isNull):
(QScriptValuePrivate::isString):
(QScriptValuePrivate::isUndefined):
(QScriptValuePrivate::isFunction):
(QScriptValuePrivate::toString):
(QScriptValuePrivate::toNumber):
(QScriptValuePrivate::toBool):
(QScriptValuePrivate::toObject):
(QScriptValuePrivate::equals):
(QScriptValuePrivate::strictlyEquals):
(QScriptValuePrivate::instanceOf):
(QScriptValuePrivate::call):
(QScriptValuePrivate::operator JSValueRef):
(QScriptValuePrivate::operator JSObjectRef):
(QScriptValuePrivate::setValue):
(QScriptValuePrivate::inherits):
(QScriptValuePrivate::refinedJSValue):

02:59 Changeset [61002] by jocelyn.turcotte@nokia.com

[Qt] Implement the simple text code path.
https://bugs.webkit.org/show_bug.cgi?id=40077

Reviewed by Simon Hausmann.

Remove the FONT_FAST_PATH macro and use the Qt's
fast text implementation instead of the one of WebKit.

The Qt::TextBypassShaping flag is used to tell Qt to
only use the glyph advances.

Qt 4.7 is needed to get this flag thus the complex path is always
used if QtWebKit is compiled against an earlier version.

Contrary to the WebKit's implementation, the complex code path
is taken if the text is RightToLeft, justified or is formatted
with non-zero letter or word spacing.

JavaScriptCore:

  • wtf/Platform.h:

WebCore:

  • platform/graphics/Font.cpp:

(WebCore::Font::drawText):
(WebCore::Font::floatWidth):
(WebCore::Font::selectionRectForText):
(WebCore::Font::offsetForPosition):

  • platform/graphics/Font.h:

(WebCore::Font::isRoundingHackCharacter):

  • platform/graphics/qt/FontQt.cpp:

(WebCore::fromRawDataWithoutRef):
(WebCore::needComplexCodePath):
(WebCore::setupPlatformContext):
(WebCore::Font::canReturnFallbackFontsForComplexText):
(WebCore::Font::drawSimpleText):
(WebCore::Font::drawComplexText):
(WebCore::Font::floatWidthForSimpleText):
(WebCore::Font::offsetForPositionForSimpleText):
(WebCore::Font::selectionRectForSimpleText):

WebKit\qt:

  • Api/qwebpage.cpp:

(QWebPagePrivate::QWebPagePrivate):

02:58 Changeset [61001] by jocelyn.turcotte@nokia.com

Move the text code path detection code from FontFastPath.cpp to Font.cpp.
https://bugs.webkit.org/show_bug.cgi?id=40077

Reviewed by Simon Hausmann.

This is a patch preliminary to the implementation of the
simple font code path for the Qt port.

  • platform/graphics/Font.cpp:

(WebCore::Font::setCodePath):
(WebCore::Font::codePath):

  • platform/graphics/Font.h:
  • platform/graphics/FontFastPath.cpp:
02:36 Changeset [61000] by eric@webkit.org

2010-06-11 Sterling Swigart <sswigart@google.com>

Reviewed by Dmitry Titov.

Added AsyncImageResizer and ImageResizerThread classes, which after further
patches will be capable of resizing images.
https://bugs.webkit.org/show_bug.cgi?id=40018

  • Android.mk:
  • CMakeLists.txt:
  • GNUmakefile.am:
  • WebCore.gypi:
  • WebCore.pro:
  • WebCore.vcproj/WebCore.vcproj:
  • WebCore.xcodeproj/project.pbxproj:
  • html/AsyncImageResizer.cpp: Added. (WebCore::AsyncImageResizer::create): (WebCore::AsyncImageResizer::AsyncImageResizer): (WebCore::AsyncImageResizer::~AsyncImageResizer): (WebCore::AsyncImageResizer::notifyFinished):
  • html/AsyncImageResizer.h: Added. (WebCore::AsyncImageResizer::CallbackInfo::CallbackInfo): (WebCore::AsyncImageResizer::): (WebCore::AsyncImageResizer::resizeComplete): (WebCore::AsyncImageResizer::resizeError):
  • html/ImageResizerThread.cpp: Added. (WebCore::returnBlobOrError): (WebCore::ImageResizerThread::start): (WebCore::ImageResizerThread::ImageResizerThread): (WebCore::ImageResizerThread::~ImageResizerThread): (WebCore::ImageResizerThread::imageResizerThreadStart): (WebCore::ImageResizerThread::imageResizerThread):
  • html/ImageResizerThread.h: Added.
02:10 Changeset [60999] by eric@webkit.org

2010-06-11 Tony Gentilcore <tonyg@chromium.org>

Reviewed by Adam Barth.

Update acid3 test with latest from acid3.acidtests.org
https://bugs.webkit.org/show_bug.cgi?id=40428

This is a straight copy from "curl http://acid3.acidtests.org/" with
LayoutTest specific bits reapplied:

  • Rename resources to resources/acid3/...
  • Add window.layoutTestController bits to start and end.

We still fail the same link test. The difference is that its error
changed from "LINKTEST FAILED" to "YOU SHOULD NOT SEE THIS AT ALL".

  • http/tests/misc/acid3-expected.checksum:
  • http/tests/misc/acid3-expected.png:
  • http/tests/misc/acid3-expected.txt:
  • http/tests/misc/acid3.html:
01:50 Changeset [60998] by steveblock@google.com

2010-06-11 Steve Block <steveblock@google.com>

Reviewed by Alexey Proskuryakov.

Client-based Geolocation does not handle multiple simultaneous requests
https://bugs.webkit.org/show_bug.cgi?id=40148

Test: fast/dom/Geolocation/multiple-requests.html

The Geolocation must handle multiple calls to addObserver() from the same Geolocation object.

  • page/GeolocationController.cpp: (WebCore::GeolocationController::addObserver):

2010-06-11 Steve Block <steveblock@google.com>

Reviewed by Alexey Proskuryakov.

Client-based Geolocation does not handle multiple simultaneous requests
https://bugs.webkit.org/show_bug.cgi?id=40148

The change to the GeolocationController is explicitly tested with multiple-requests.html.

This change also fixes reentrant-error.html and reentrant-success.html which were previously
on the mac skipped list. reentrant-error.html required a minor tweak as the mac
GeolocationController can only provide an error code of 2 (POSITION_UNAVAILABLE).

Added to GTK skipped list.

  • fast/dom/Geolocation/multiple-requests-expected.txt: Added.
  • fast/dom/Geolocation/multiple-requests.html: Added.
  • fast/dom/Geolocation/script-tests/multiple-requests.js: Added.
  • fast/dom/Geolocation/script-tests/reentrant-error.js:
  • platform/mac/Skipped:
  • platform/gtk/Skipped:
01:41 Changeset [60997] by eric@webkit.org

2010-06-11 Yael Aharon <yael.aharon@nokia.com>

Reviewed by Laszlo Gombos.

[Qt] There should be a way to enable popups in QtTestBrowser
https://bugs.webkit.org/show_bug.cgi?id=40427

  • QtTestBrowser/main.cpp: (LauncherWindow::toggleJavascriptCanOpenWindows): (LauncherWindow::createChrome):
01:27 Changeset [60996] by eric@webkit.org

2010-06-11 Luiz Agostini <luiz.agostini@openbossa.org>

Reviewed by Kenneth Rohde Christiansen.

add codePointCompare to JavaScriptCore.exp
https://bugs.webkit.org/show_bug.cgi?id=40426

01:12 Changeset [60995] by eric@webkit.org

2010-06-11 Nikita Vasilyev <me@elv1s.ru>

Reviewed by Pavel Feldman.

Web Inspector: When completing using a tab key, select very first value, not a second.
https://bugs.webkit.org/show_bug.cgi?id=40409

  • inspector/front-end/TextPrompt.js: (WebInspector.TextPrompt.prototype._completionsReady):
00:47 Changeset [60994] by hamaji@chromium.org

2010-06-11 Shinichiro Hamaji <hamaji@chromium.org>

Unreviewed.

Skip re-enabled SVG test (svg/W3C-SVG-1.1/fonts-glyph-02-t)

  • platform/chromium/test_expectations.txt:

06/10/10:

23:56 Changeset [60993] by abarth@webkit.org

2010-06-10 Adam Barth <abarth@webkit.org>

Reviewed by Daniel Bates.

Make SheriffBot more chatty
https://bugs.webkit.org/show_bug.cgi?id=40463

People seem to like to talk to SheriffBot, so let's make him chat back.

  • Scripts/webkitpy/thirdparty/init.py:
  • Scripts/webkitpy/tool/bot/irc_command.py:
  • Scripts/webkitpy/tool/bot/irc_command_unittest.py: Added.
  • Scripts/webkitpy/tool/bot/sheriffircbot.py:
  • Scripts/webkitpy/tool/bot/sheriffircbot_unittest.py:
22:51 Changeset [60992] by yuzo@google.com

2010-06-10 Yuzo Fujishima <yuzo@google.com>

Reviewed by Shinichiro Hamaji.

Fix Bug 40452: REGRESSION: printing is broken if stylesheet has @page
https://bugs.webkit.org/show_bug.cgi?id=40452

  • platform/chromium/test_expectations.txt:
  • platform/mac/printing/page-rule-in-media-query-expected.checksum: Added.
  • platform/mac/printing/page-rule-in-media-query-expected.png: Added.
  • platform/mac/printing/page-rule-in-media-query-expected.txt: Added.
  • printing/page-rule-in-media-query.html: Added.

2010-06-10 Yuzo Fujishima <yuzo@google.com>

Reviewed by Shinichiro Hamaji.

Fix Bug 40452: REGRESSION: printing is broken if stylesheet has @page
https://bugs.webkit.org/show_bug.cgi?id=40452

Test: printing/page-rule-in-media-query.html

  • css/CSSStyleSelector.cpp: (WebCore::CSSRuleSet::addRulesFromSheet): (WebCore::CSSRuleSet::addStyleRule):
22:47 Changeset [60991] by eric@webkit.org

2010-06-10 Sheriff Bot <webkit.review.bot@gmail.com>

Unreviewed, rolling out r60989.
http://trac.webkit.org/changeset/60989
https://bugs.webkit.org/show_bug.cgi?id=40462

It makes things crash (Requested by abarth on #webkit).

  • dom/Tokenizer.h: (WebCore::Tokenizer::isHTMLTokenizer):
  • html/HTML5Tokenizer.cpp:
  • html/HTML5Tokenizer.h:
  • html/HTML5TreeBuilder.h:
  • html/HTMLFormControlElement.cpp: (WebCore::HTMLFormControlElement::removedFromTree):
  • html/HTMLTokenizer.h: (WebCore::HTMLTokenizer::isHTMLTokenizer): (WebCore::HTMLTokenizer::htmlParser):
22:36 Changeset [60990] by oliver@apple.com

2010-06-10 Oliver Hunt <oliver@apple.com>

Reviewed by Maciej Stachowiak.

Math Javascript Bug on Safari 5 (webkit 533.16) under "32bit" mode
https://bugs.webkit.org/show_bug.cgi?id=40367

If we're in the slow case of right shift we must write the type tag as
the only reason we hit this code path is because we know we're working
with a double. eg. we are guaranteed that the tag cannot be reused.

  • jit/JITArithmetic32_64.cpp: (JSC::JIT::emitRightShiftSlowCase):

2010-06-10 Oliver Hunt <oliver@apple.com>

Reviewed by Maciej Stachowiak.

Math Javascript Bug on Safari 5 (webkit 533.16) under "32bit" mode
https://bugs.webkit.org/show_bug.cgi?id=40367

Add some tests to ensure correct behaviour of right shift.

  • fast/js/bitops-type-tag.html: Added.
  • fast/js/script-tests/bitops-type-tag.js: Added.
22:28 Changeset [60989] by abarth@webkit.org

2010-06-10 Tony Gentilcore <tonyg@chromium.org>

Reviewed by Eric Seidel.

Fix fast/forms/preserveFormDuringResidualStyle.html for HTML5 Parser.
https://bugs.webkit.org/show_bug.cgi?id=40454

This required exposing the HTMLParser which we definitely don't want to
do, but the good news is that it can go away when the HTMLParser goes
away.

No new tests because covered by fast/forms/preserve/FormDuringResidualStyle.html.

  • dom/Tokenizer.h: (WebCore::Tokenizer::htmlParser):
  • html/HTML5Tokenizer.cpp: (WebCore::HTML5Tokenizer::htmlParser):
  • html/HTML5Tokenizer.h:
  • html/HTML5TreeBuilder.h: (WebCore::HTML5TreeBuilder::htmlParser):
  • html/HTMLFormControlElement.cpp: (WebCore::HTMLFormControlElement::removedFromTree):
  • html/HTMLTokenizer.h: (WebCore::HTMLTokenizer::htmlParser):
21:58 Changeset [60988] by bweinstein@apple.com

Fix the Windows buildbot when WebKitAuxiliaryLibs.zip cannot be found on our server.

Reviewed by Darin Adler.

Robustify update-webkit-auxiliary-libs to be able to handle the case when WebKitAuxiliaryLibs.zip doesn't exist,
and allow the script to fall back to an existing file, only failing if there is no fallback file. Also switch to checking
WEXITSTATUS on $result instead of just comparing $result to 0.

  • Scripts/update-webkit-auxiliary-libs:
21:05 Changeset [60987] by tkent@chromium.org

[DRT/Chromium] Support plugin tests
https://bugs.webkit.org/show_bug.cgi?id=40413

Reviewed by Dimitri Glazkov.

  • DEPS: Bump the Chromium revision to r49376 to have a npapi_layout_test_plugin change and Linux <progress> resources. Add third_party/mesa because app/ depends on them since r49332.
  • WebKit.gyp: Copy a plugin binary for DRT to recognize it.
18:40 Changeset [60986] by jhawkins@chromium.org

2010-06-10 James Hawkins <jhawkins@chromium.org>

Not reviwed.

Build fix for Chromium Mac Release.

  • src/AutoFillPopupMenuClient.cpp: (WebKit::AutoFillPopupMenuClient::getSuggestion): (WebKit::AutoFillPopupMenuClient::valueChanged): (WebKit::AutoFillPopupMenuClient::selectionChanged): (WebKit::AutoFillPopupMenuClient::itemIsSeparator):
17:51 Changeset [60985] by jhawkins@chromium.org

2010-06-09 James Hawkins <jhawkins@chromium.org>

Reviewed by Darin Fisher.

[Chromium] Add the ability to specify a separator in
AutoFillPopupMenuClient.

https://bugs.webkit.org/show_bug.cgi?id=40397

  • public/WebViewClient.h: (WebKit::WebViewClient::didAcceptAutoFillSuggestion):
  • src/AutoFillPopupMenuClient.cpp: (WebKit::AutoFillPopupMenuClient::getSuggestionsCount): (WebKit::AutoFillPopupMenuClient::getSuggestion): (WebKit::AutoFillPopupMenuClient::valueChanged): (WebKit::AutoFillPopupMenuClient::selectionChanged): (WebKit::AutoFillPopupMenuClient::itemIsSeparator): (WebKit::AutoFillPopupMenuClient::initialize): (WebKit::AutoFillPopupMenuClient::setSuggestions):
  • src/AutoFillPopupMenuClient.h:
  • src/WebViewImpl.cpp: (WebKit::WebViewImpl::applyAutoFillSuggestions):
  • src/WebViewImpl.h:
17:49 Changeset [60984] by tony@chromium.org

2010-06-10 Tony Chang <tony@chromium.org>

Reviewed by Kent Tamura.

crash when focus is changed while trying to focus next element
https://bugs.webkit.org/show_bug.cgi?id=40407

  • fast/events/focus-change-crash-expected.txt: Added.
  • fast/events/focus-change-crash.html: Added.

2010-06-10 Tony Chang <tony@chromium.org>

Reviewed by Kent Tamura.

crash when focus is changed while trying to focus next element
https://bugs.webkit.org/show_bug.cgi?id=40407

Test: fast/events/focus-change-crash.html

  • dom/Element.cpp: (WebCore::Element::focus):
17:45 Changeset [60983] by ap@apple.com

Rubber-stamped by Sam Weinig.

https://bugs.webkit.org/show_bug.cgi?id=18830
Assertion failure (using HashMap empty value) in svg/W3C-SVG-1.1/fonts-glyph-02-t.svg

This doesn't happen any more, the code has changed a lot since 2008. Re-enabling the test.

  • platform/mac/svg/W3C-SVG-1.1/fonts-glyph-02-t-expected.checksum:
  • platform/mac/svg/W3C-SVG-1.1/fonts-glyph-02-t-expected.png:
  • platform/mac/svg/W3C-SVG-1.1/fonts-glyph-02-t-expected.txt:
  • svg/W3C-SVG-1.1/fonts-glyph-02-t.svg: Copied from LayoutTests/svg/W3C-SVG-1.1/fonts-glyph-02-t.svg-disabled.
  • svg/W3C-SVG-1.1/fonts-glyph-02-t.svg-disabled: Removed.
17:27 Changeset [60982] by ojan@chromium.org

2010-06-10 Ojan Vafai <ojan@chromium.org>

Reviewed by Alexey Proskuryakov.

don't use Exception.message as it's deprecated in python 2.6+
https://bugs.webkit.org/show_bug.cgi?id=40449

  • Scripts/webkitpy/tool/commands/queues.py:
16:05 Changeset [60981] by yael.aharon@nokia.com

[Qt] Support for loading notification icons
https://bugs.webkit.org/show_bug.cgi?id=40004

Reviewed by Kenneth Rohde Christiansen.

Take into use the icon that was loaded for the notification.

  • WebCoreSupport/NotificationPresenterClientQt.cpp:

(WebCore::NotificationPresenterClientQt::show):

16:05 Changeset [60980] by kbr@google.com

2010-06-10 Sheriff Bot <webkit.review.bot@gmail.com>

Unreviewed, rolling out r60979.
http://trac.webkit.org/changeset/60979
https://bugs.webkit.org/show_bug.cgi?id=40450

Broke build on Leopard (Requested by kbr_google on #webkit).

  • bindings/js/JSWebGLRenderingContextCustom.cpp: (WebCore::JSWebGLRenderingContext::bufferData): (WebCore::JSWebGLRenderingContext::bufferSubData): (WebCore::JSWebGLRenderingContext::texImage2D): (WebCore::JSWebGLRenderingContext::texSubImage2D):
  • bindings/scripts/CodeGeneratorJS.pm:
  • html/HTMLCanvasElement.idl:
  • html/HTMLImageElement.idl:
  • html/HTMLVideoElement.idl:
  • html/canvas/WebGLRenderingContext.idl:

2010-06-10 Sheriff Bot <webkit.review.bot@gmail.com>

Unreviewed, rolling out r60979.
http://trac.webkit.org/changeset/60979
https://bugs.webkit.org/show_bug.cgi?id=40450

Broke build on Leopard (Requested by kbr_google on #webkit).

  • fast/canvas/webgl/script-tests/texImageTest.js:
  • fast/canvas/webgl/texImageTest-expected.txt:
15:44 Changeset [60979] by kbr@google.com

2010-06-10 Kenneth Russell <kbr@google.com>

Reviewed by Dimitri Glazkov.

Delete custom JSC bindings for bufferData, texImage2D and texSubImage2D
https://bugs.webkit.org/show_bug.cgi?id=40443

Deleted custom JSC bindings for bufferData, texImage2D and
texSubImage2D and fixed minor associated issues in code generator
and IDL. Updated fast/canvas/webgl/texImageTest-expected.txt,
which now runs as intended. Ran all layout tests, including WebGL
tests, in Safari.

  • bindings/js/JSWebGLRenderingContextCustom.cpp:
    • Deleted custom bindings.
  • bindings/scripts/CodeGeneratorJS.pm:
    • Made overload helper functions static to fix compiler warnings.
    • Added needed isObject check for overloads between object and non-object arguments.
  • html/HTMLCanvasElement.idl:
  • html/HTMLImageElement.idl:
  • html/HTMLVideoElement.idl:
    • Added GenerateNativeConverter extended attribute.
  • html/canvas/WebGLRenderingContext.idl:
    • Deleted #ifdefs for custom JSC bindings.

2010-06-10 Kenneth Russell <kbr@google.com>

Reviewed by Dimitri Glazkov.

Delete custom JSC bindings for bufferData, texImage2D and texSubImage2D
https://bugs.webkit.org/show_bug.cgi?id=40443

Deleted custom JSC bindings for bufferData, texImage2D and
texSubImage2D and fixed minor associated issues in code generator
and IDL. Updated fast/canvas/webgl/texImageTest-expected.txt,
which now runs as intended. Ran all layout tests, including WebGL
tests, in Safari.

  • fast/canvas/webgl/script-tests/texImageTest.js:
  • fast/canvas/webgl/texImageTest-expected.txt:
15:03 Changeset [60978] by hyatt@apple.com

Update WebKitSystemInterface to fix build.

14:56 Changeset [60977] by dimich@chromium.org

Not reviewed, Chromium build fix from Abhishek Arya.

  • src/WebImageCG.cpp:

(WebKit::WebImage::WebImage):
(WebKit::WebImage::operator=):

14:55 Changeset [60976] by ojan@chromium.org

2010-06-10 Ojan Vafai <ojan@chromium.org>

Reviewed by Adam Barth.

fix handle_script_error in rietveld upload queue and add testing for handle_script_error
https://bugs.webkit.org/show_bug.cgi?id=40436

  • Scripts/webkitpy/common/system/outputcapture.py:
  • Scripts/webkitpy/tool/commands/earlywarningsystem_unittest.py:
  • Scripts/webkitpy/tool/commands/queues.py:
  • Scripts/webkitpy/tool/commands/queues_unittest.py:
  • Scripts/webkitpy/tool/commands/queuestest.py:
  • Scripts/webkitpy/tool/mocktool.py:
14:40 Changeset [60975] by hyatt@apple.com

Reset layout test results for continuous wheeling.

14:31 Changeset [60974] by hyatt@apple.com

https://bugs.webkit.org/show_bug.cgi?id=40441, back out the original fix for 29601, since it has broken continuous
wheel delta values.

Reviewed by John Sullivan.

WebCore:

  • platform/mac/WebCoreSystemInterface.h:
  • platform/mac/WebCoreSystemInterface.mm:
  • platform/mac/WheelEventMac.mm:

(WebCore::PlatformWheelEvent::PlatformWheelEvent):

WebKit/mac:

  • WebView/WebDynamicScrollBarsView.mm:

(-[WebDynamicScrollBarsView scrollWheel:]):

14:29 Changeset [60973] by hamaji@chromium.org

2010-06-10 Abhishek Arya <inferno@chromium.org>

Reviewed by Dimitri Glazkov.

Add null pointer checks for nativeImageForCurrentFrame
function calls.
https://bugs.webkit.org/show_bug.cgi?id=39797

  • platform/chromium/PasteboardChromium.cpp: (WebCore::Pasteboard::writeImage):
  • platform/graphics/qt/GraphicsContext3DQt.cpp: (WebCore::GraphicsContext3D::getImageData):
  • platform/qt/PasteboardQt.cpp: (WebCore::Pasteboard::writeImage):

2010-06-10 Abhishek Arya <inferno@chromium.org>

Reviewed by Dimitri Glazkov.

Add null pointer checks for nativeImageForCurrentFrame
function calls.
https://bugs.webkit.org/show_bug.cgi?id=39797

  • src/WebImageCG.cpp: (WebKit::WebImage::WebImage): (WebKit::WebImage::operator=):
14:01 Changeset [60972] by sullivan@apple.com
  • mac/WebKit2.exp:

Added another symbol needed by Mac clients.

Reviewed by Dan Bernstein.

13:41 Changeset [60971] by eric@webkit.org

2010-06-10 Jarkko Sakkinen <jarkko.j.sakkinen@gmail.com>

Reviewed by Simon Hausmann.

[Qt] *All* WebGL layout tests fail!
https://bugs.webkit.org/show_bug.cgi?id=40296

Added test for WebKitWebGLEnabled to
LayoutTestController::overridePreference().

  • DumpRenderTree/qt/LayoutTestControllerQt.cpp: (LayoutTestController::overridePreference):
12:34 Changeset [60970] by eric@webkit.org

2010-06-10 Robert Hogan <robert@webkit.org>

Reviewed by Kenneth Rohde Christiansen.

[Qt] Fix Qt-specific event-sender-keydown-frame.html test

It looks like this test was added without elements required for it work.

http://bugs.webkit.org/show_bug.cgi?id=20303

  • platform/qt/Skipped:
  • platform/qt/fast/events/event-sender-keydown-frame.html:
  • platform/qt/fast/events/resources/divs.html: Added.
12:13 Changeset [60969] by andersca@apple.com

Fix build.

  • NetscapeCocoaPlugin/NetscapeCocoaPlugin.xcodeproj/project.pbxproj:
  • NetscapeCocoaPlugin/main.m:

(NPP_New):
(NPP_NewStream):
(NPP_WriteReady):
(NPP_Write):
(NPP_HandleEvent):

11:57 Changeset [60968] by dimich@chromium.org

Unreviewed, update Chromium test expectations.

  • platform/chromium/test_expectations.txt:

Disable fast/text/international/khmer-selection.html on Windows regressed by http://trac.webkit.org/changeset/60953
Skip fast/dom/Window/window-properties-geolocation.html unnecessarily enabled by http://trac.webkit.org/changeset/60956

11:53 Changeset [60967] by kbr@google.com

2010-06-09 Kenneth Russell <kbr@google.com>

Reviewed by Dimitri Glazkov.

Implement TypedArray BYTES_PER_ELEMENT
https://bugs.webkit.org/show_bug.cgi?id=39100

Implemented BYTES_PER_ELEMENT on all ArrayBufferView subclasses.
Required bug fixes to JSC bindings' custom constructors. Updated
fast/canvas/webgl/array-unit-tests.html to verify. Ran all WebGL
layout tests in Safari and Chromium.

  • bindings/js/JSFloat32ArrayConstructor.cpp: (WebCore::JSFloat32ArrayConstructor::JSFloat32ArrayConstructor): (WebCore::JSFloat32ArrayConstructor::getOwnPropertySlot): (WebCore::JSFloat32ArrayConstructor::getOwnPropertyDescriptor):
  • bindings/js/JSFloat32ArrayConstructor.h: (WebCore::JSFloat32ArrayConstructor::createStructure):
  • bindings/js/JSInt16ArrayConstructor.cpp: (WebCore::JSInt16ArrayConstructor::JSInt16ArrayConstructor): (WebCore::JSInt16ArrayConstructor::getOwnPropertySlot): (WebCore::JSInt16ArrayConstructor::getOwnPropertyDescriptor):
  • bindings/js/JSInt16ArrayConstructor.h: (WebCore::JSInt16ArrayConstructor::createStructure):
  • bindings/js/JSInt32ArrayConstructor.cpp: (WebCore::JSInt32ArrayConstructor::JSInt32ArrayConstructor): (WebCore::JSInt32ArrayConstructor::getOwnPropertySlot): (WebCore::JSInt32ArrayConstructor::getOwnPropertyDescriptor):
  • bindings/js/JSInt32ArrayConstructor.h: (WebCore::JSInt32ArrayConstructor::createStructure):
  • bindings/js/JSInt8ArrayConstructor.cpp: (WebCore::JSInt8ArrayConstructor::JSInt8ArrayConstructor): (WebCore::JSInt8ArrayConstructor::getOwnPropertySlot): (WebCore::JSInt8ArrayConstructor::getOwnPropertyDescriptor):
  • bindings/js/JSInt8ArrayConstructor.h: (WebCore::JSInt8ArrayConstructor::createStructure):
  • bindings/js/JSUint16ArrayConstructor.cpp: (WebCore::JSUint16ArrayConstructor::JSUint16ArrayConstructor): (WebCore::JSUint16ArrayConstructor::getOwnPropertySlot): (WebCore::JSUint16ArrayConstructor::getOwnPropertyDescriptor):
  • bindings/js/JSUint16ArrayConstructor.h: (WebCore::JSUint16ArrayConstructor::createStructure):
  • bindings/js/JSUint32ArrayConstructor.cpp: (WebCore::JSUint32ArrayConstructor::JSUint32ArrayConstructor): (WebCore::JSUint32ArrayConstructor::getOwnPropertySlot): (WebCore::JSUint32ArrayConstructor::getOwnPropertyDescriptor):
  • bindings/js/JSUint32ArrayConstructor.h: (WebCore::JSUint32ArrayConstructor::createStructure):
  • bindings/js/JSUint8ArrayConstructor.cpp: (WebCore::JSUint8ArrayConstructor::JSUint8ArrayConstructor): (WebCore::JSUint8ArrayConstructor::getOwnPropertySlot): (WebCore::JSUint8ArrayConstructor::getOwnPropertyDescriptor):
  • bindings/js/JSUint8ArrayConstructor.h: (WebCore::JSUint8ArrayConstructor::createStructure):
  • html/canvas/Float32Array.idl:
  • html/canvas/Int16Array.idl:
  • html/canvas/Int32Array.idl:
  • html/canvas/Int8Array.idl:
  • html/canvas/Uint16Array.idl:
  • html/canvas/Uint32Array.idl:
  • html/canvas/Uint8Array.idl:

2010-06-09 Kenneth Russell <kbr@google.com>

Reviewed by Dimitri Glazkov.

Implement TypedArray BYTES_PER_ELEMENT
https://bugs.webkit.org/show_bug.cgi?id=39100

Implemented BYTES_PER_ELEMENT on all ArrayBufferView subclasses.
Required bug fixes to JSC bindings' custom constructors. Updated
fast/canvas/webgl/array-unit-tests.html to verify. Ran all WebGL
layout tests in Safari and Chromium.

  • fast/canvas/webgl/array-unit-tests.html:
11:17 Changeset [60966] by abarth@webkit.org

2010-06-10 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

HTML5 Parser should continue parsing after script execution
https://bugs.webkit.org/show_bug.cgi?id=40416

The previous code was just wrong. We were always blocking
the parser after any script execution. Now we correctly continue
parsing after a successful script execution and only block the
parser when we weren't able to execute the script immediately.

Added a new haveParsingBlockingScript() function to HTML5ScriptRunner
to make some of the code more self-documenting.

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

fast/js/implicit-call-with-global-reentry.html

  • html/HTML5ScriptRunner.cpp: (WebCore::HTML5ScriptRunner::execute):
    • Remove some old (now bogus) FIXMEs and notImplemented() calls.
    • Fix the logic to match the HTML5 spec by continuing parsing after script execution and only blocking the parser when the script wasn't able to immediately run.

(WebCore::HTML5ScriptRunner::haveParsingBlockingScript):

  • New function to make some of the code read better.

(WebCore::HTML5ScriptRunner::executeParsingBlockingScripts):

  • Use haveParsingBlockingScript().

(WebCore::HTML5ScriptRunner::executeScriptsWaitingForLoad):

  • Use haveParsingBlockingScript().

(WebCore::HTML5ScriptRunner::runScript):

  • Use haveParsingBlockingScript().
  • html/HTML5ScriptRunner.h:

2010-06-10 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

HTML5 Parser should continue parsing after script execution
https://bugs.webkit.org/show_bug.cgi?id=40416

Test the ordering of script execution with multiple calls to
document.write().

  • fast/tokenizer/write-multiple-scripts-expected.txt: Added.
  • fast/tokenizer/write-multiple-scripts.html: Added.
10:59 Changeset [60965] by yurys@chromium.org

2010-06-10 Yury Semikhatsky <yurys@chromium.org>

Reviewed by Pavel Feldman.

Web Inspector: to reduce the front end start up time don't push script
sources to frontend if debugger is always enabled, instead request
script content lazily at the moment it should be displyed. It is critical for
always enabled debugger because in that case lots of script may have
already been parsed when the front end is opening and pushing all
of them at once may slow things down(even though the script sources will never
be used if scripts panel isn't open).
https://bugs.webkit.org/show_bug.cgi?id=40364

  • inspector/InspectorBackend.cpp: (WebCore::InspectorBackend::getScriptSource):
  • inspector/InspectorBackend.h:
  • inspector/InspectorBackend.idl:
  • inspector/InspectorController.cpp: (WebCore::InspectorController::didCommitLoad): (WebCore::InspectorController::getScriptSource): (WebCore::InspectorController::didParseSource):
  • inspector/InspectorController.h:
  • inspector/InspectorFrontend.cpp: (WebCore::InspectorFrontend::didGetScriptSource):
  • inspector/InspectorFrontend.h:
  • inspector/front-end/InspectorBackendStub.js: (.WebInspector.InspectorBackendStub.prototype.editScriptSource): (.WebInspector.InspectorBackendStub.prototype.getScriptSource):
  • inspector/front-end/ScriptView.js: (WebInspector.ScriptView.prototype.setupSourceFrameIfNeeded): (WebInspector.ScriptView.prototype._didGetScriptSource): (WebInspector.ScriptView.prototype._sourceFrameSetupFinished):

2010-06-10 Yury Semikhatsky <yurys@chromium.org>

Reviewed by Pavel Feldman.

Web Inspector: to reduce the front end start up time don't push script
sources to frontend if debugger is always enabled, instead request
script content lazily at the moment it should be displyed. It is critical for
always enabled debugger because in that case lots of script may have
already been parsed when the front end is opening and pushing all
of them at once may slow things down(even though the script sources will never
be used if scripts panel isn't open).
https://bugs.webkit.org/show_bug.cgi?id=40364

  • src/js/DevTools.js:
  • src/js/InspectorControllerImpl.js: (devtools.InspectorBackendImpl): (.devtools.InspectorBackendImpl.prototype.getScriptSource):
10:39 Changeset [60964] by abarth@webkit.org

2010-06-10 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

Use allowRequestIfNoIllegalURICharacters instead of context for XSSAuditor::canLoadExternalScriptFromSrc
https://bugs.webkit.org/show_bug.cgi?id=40404

We originally added the context parameter to
canLoadExternalScriptFromSrc to work around some false positives caused
by folks checking external script URLs on the server. Our thought was
that we could tell these were not real XSS attacks because the
surrounding context wouldn't match in the URL and the document.

Implementing this feature in the HTML5 parser is hard because it
pierces a layer of abstraction (the token abstraction of the input
stream). We could hack this into the new parser, but instead I think
it's better to switch to using the allowRequestIfNoIllegalURICharacters
heuristic.

We designed the allowRequestIfNoIllegalURICharacters after the context
heuristic to deal with other cases where the server was validating
input before echoing it. However, we never tried applying it to
canLoadExternalScriptFromSrc.

It's possible that this will cause false positives and will need to be
reverted, which is why I've left in some of the infrustructure for
computing context. We don't have a good way to know if that will
happen except to try. We do know, however, that this heuristic will
work for the original false positives we saw.

  • html/HTML5Tokenizer.cpp: (WebCore::HTML5Tokenizer::shouldLoadExternalScriptFromSrc):
  • html/HTMLTokenizer.cpp: (WebCore::HTMLTokenizer::parseTag):
  • page/XSSAuditor.cpp: (WebCore::XSSAuditor::canLoadExternalScriptFromSrc):
  • page/XSSAuditor.h:
10:24 Changeset [60963] by eric@webkit.org

2010-06-10 Mahesh Kulkarni <mahesh.kulkarni@nokia.com>

Reviewed by Simon Hausmann.

[Qt] Memory leak in qwebpage unit test cases
https://bugs.webkit.org/show_bug.cgi?id=40405

  • tests/qwebpage/tst_qwebpage.cpp: (tst_QWebPage::infiniteLoopJS):
10:10 Changeset [60962] by eric@webkit.org

2010-06-10 Kwang Yul Seo <skyul@company100.net>

Reviewed by Kent Tamura.

Add ENABLE(DATABASE) guard for DatabaseAuthorizer.cpp
https://bugs.webkit.org/show_bug.cgi?id=40399

Build fix for ENABLE(DATABASE)=0.

  • storage/DatabaseAuthorizer.cpp:
09:49 Changeset [60961] by eric@webkit.org

2010-06-10 Chang Shu <chang.shu@nokia.com>

Reviewed by Kenneth Rohde Christiansen.

[Qt] Unskip canvas/philip/tests/2d.shadow.enable.blur.html.

  • platform/qt/Skipped:

2010-06-10 Shu Chang <chang.shu@nokia.com>

Reviewed by Kenneth Rohde Christiansen.

Change the type of ShadowBlur from int to float in GraphicsContext.
Using int for ShadowBlur loses precision and fails the test.
Note: This code change fixes Qt port but Mac is still failing due to
platform issue. Function CGContextSetShadowWithColor() does not take
a blur value less than 0.5.

https://bugs.webkit.org/show_bug.cgi?id=40370

  • platform/graphics/GraphicsContext.cpp: (WebCore::GraphicsContext::setShadow): (WebCore::GraphicsContext::getShadow):
  • platform/graphics/GraphicsContext.h:
  • platform/graphics/GraphicsContextPrivate.h:
  • platform/graphics/cairo/FontCairo.cpp: (WebCore::Font::drawGlyphs):
  • platform/graphics/cairo/GraphicsContextCairo.cpp: (WebCore::GraphicsContext::calculateShadowBufferDimensions): (WebCore::drawPathShadow): (WebCore::drawBorderlessRectShadow): (WebCore::GraphicsContext::setPlatformShadow):
  • platform/graphics/cairo/ImageCairo.cpp: (WebCore::BitmapImage::draw):
  • platform/graphics/cg/GraphicsContextCG.cpp: (WebCore::GraphicsContext::setPlatformShadow):
  • platform/graphics/chromium/FontChromiumWin.cpp: (WebCore::TransparencyAwareFontPainter::TransparencyAwareGlyphPainter::drawGlyphs): (WebCore::Font::drawComplexText):
  • platform/graphics/gtk/FontGtk.cpp: (WebCore::Font::drawComplexText):
  • platform/graphics/haiku/GraphicsContextHaiku.cpp: (WebCore::GraphicsContext::setPlatformShadow):
  • platform/graphics/mac/FontMac.mm: (WebCore::Font::drawGlyphs):
  • platform/graphics/openvg/GraphicsContextOpenVG.cpp: (WebCore::GraphicsContext::setPlatformShadow):
  • platform/graphics/qt/FontQt.cpp: (WebCore::Font::drawComplexText):
  • platform/graphics/qt/GraphicsContextQt.cpp: (WebCore::GraphicsContext::drawRect): (WebCore::GraphicsContext::drawLine): (WebCore::GraphicsContext::strokeArc): (WebCore::GraphicsContext::drawConvexPolygon): (WebCore::drawFilledShadowPath): (WebCore::GraphicsContext::strokePath): (WebCore::drawBorderlessRectShadow): (WebCore::GraphicsContext::setPlatformShadow):
  • platform/graphics/qt/ImageQt.cpp: (WebCore::BitmapImage::draw):
  • platform/graphics/skia/GraphicsContextSkia.cpp: (WebCore::GraphicsContext::setPlatformShadow):
  • platform/graphics/skia/SkiaFontWin.cpp: (WebCore::windowsCanHandleDrawTextShadow):
  • platform/graphics/win/FontCGWin.cpp: (WebCore::drawGDIGlyphs): (WebCore::Font::drawGlyphs):
  • platform/graphics/wince/GraphicsContextWince.cpp: (WebCore::GraphicsContext::fillRoundedRect): (WebCore::GraphicsContext::setPlatformShadow): (WebCore::GraphicsContext::drawText):
  • platform/graphics/wx/GraphicsContextWx.cpp: (WebCore::GraphicsContext::setPlatformShadow):
09:32 Changeset [60960] by eric@webkit.org

2010-06-10 Yael Aharon <yael.aharon@nokia.com>

Reviewed by Kenneth Rohde Christiansen.

Support for loading notification icons
https://bugs.webkit.org/show_bug.cgi?id=40396

  • http/tests/notifications: Added.
  • http/tests/notifications/icon-does-not-exist-expected.txt: Added.
  • http/tests/notifications/icon-does-not-exist.html: Added.
  • http/tests/notifications/icon-exists-cancel-expected.txt: Added.
  • http/tests/notifications/icon-exists-cancel.html: Added.
  • http/tests/notifications/icon-exists-expected.txt: Added.
  • http/tests/notifications/icon-exists-show-alert-during-load-expected.txt: Added.
  • http/tests/notifications/icon-exists-show-alert-during-load.html: Added.
  • http/tests/notifications/icon-exists.html: Added.
  • http/tests/notifications/icon-requires-auth-expected.txt: Added.
  • http/tests/notifications/icon-requires-auth.html: Added.
  • http/tests/notifications/resources: Added.
  • http/tests/notifications/resources/icon-exists.png: Added.
  • http/tests/notifications/resources/icon-requires-auth.php: Added.
  • platform/chromium/test_expectations.txt:
  • platform/gtk/Skipped:
  • platform/mac/Skipped:
  • platform/win/Skipped:

2010-06-10 Yael Aharon <yael.aharon@nokia.com>

Reviewed by Kenneth Rohde Christiansen.

Support for loading notification icons
https://bugs.webkit.org/show_bug.cgi?id=40396

Make notification objects download the icon needed for displaying the
notification before calling the NotificationPresenter to display the
notification.

An error during the download would cause the notification to be displayed
without an icon.

If a notification is in the process of download, and a new notification
is created with the same ReplaceId, the download is not cancelled
immediately and the notification is removed only after the download is
complete.

Tests: http/tests/notifications/icon-does-not-exist.html

http/tests/notifications/icon-exists-cancel.html
http/tests/notifications/icon-exists-show-alert-during-load.html
http/tests/notifications/icon-exists.html
http/tests/notifications/icon-requires-auth.html

  • notifications/Notification.cpp: (WebCore::Notification::Notification): (WebCore::Notification::~Notification): (WebCore::Notification::show): (WebCore::Notification::cancel): (WebCore::Notification::startLoading): (WebCore::Notification::stopLoading): (WebCore::Notification::didReceiveResponse): (WebCore::Notification::didReceiveData): (WebCore::Notification::didFinishLoading): (WebCore::Notification::didFail): (WebCore::Notification::didFailRedirectCheck): (WebCore::Notification::didReceiveAuthenticationCancellation): (WebCore::Notification::finishLoading):
  • notifications/Notification.h: (WebCore::Notification::iconData): (WebCore::Notification::releaseIconData): (WebCore::Notification::):
08:45 Changeset [60959] by eric@webkit.org

2010-06-10 Diego Gonzalez <diegohcg@webkit.org>

Reviewed by Kenneth Rohde Christiansen.

[Qt] DRT EventSender support to graphics mouse events activation
https://bugs.webkit.org/show_bug.cgi?id=40017

  • DumpRenderTree/qt/EventSenderQt.cpp: (EventSender::mouseDown): (EventSender::mouseUp): (EventSender::mouseMoveTo): (EventSender::keyDown): (EventSender::contextClick): (EventSender::sendTouchEvent): (EventSender::sendOrQueueEvent): (EventSender::eventFilter): (EventSender::createGraphicsSceneMouseEvent): (EventSender::sendEvent):
  • DumpRenderTree/qt/EventSenderQt.h: (EventSender::isGraphicsBased):
08:34 Changeset [60958] by eric@webkit.org

2010-06-10 Raine Makelainen <raine.makelainen@nokia.com>

Reviewed by Kenneth Rohde Christiansen.

Impossible to set input method hints based HTML5 input types
https://bugs.webkit.org/show_bug.cgi?id=40107

Helper methods for checking "tel", "number", "email",
and "url" input element types.

  • html/HTMLInputElement.h: (WebCore::HTMLInputElement::isTelephoneField): (WebCore::HTMLInputElement::isNumberField): (WebCore::HTMLInputElement::isEmailField): (WebCore::HTMLInputElement::isUrlField):

2010-06-10 Raine Makelainen <raine.makelainen@nokia.com>

Reviewed by Kenneth Rohde Christiansen.

Impossible to set input method hints based HTML5 input types
https://bugs.webkit.org/show_bug.cgi?id=40107

EditorClientQt to set input method hints for "number", "tel",
"email", and "url" HTML input elements.

Tests for HTML input elements and input method hints added for
QGraphicsWebView and QWebView.

  • WebCoreSupport/EditorClientQt.cpp: (WebCore::EditorClientQt::setInputMethodState):
  • tests/qgraphicswebview/resources/input_types.html: Added.
  • tests/qgraphicswebview/tst_qgraphicswebview.cpp: (GraphicsWebView::GraphicsWebView): (GraphicsWebView::fireMouseClick): (tst_QGraphicsWebView::focusInputTypes):
  • tests/qgraphicswebview/tst_qgraphicswebview.qrc: Added.
  • tests/qwebview/resources/input_types.html: Added.
  • tests/qwebview/tst_qwebview.cpp: (WebView::fireMouseClick): (tst_QWebView::focusInputTypes):
  • tests/qwebview/tst_qwebview.qrc:
07:31 Changeset [60957] by eric@webkit.org

2010-06-10 Daniel Cheng <dcheng@chromium.org>

Reviewed by Jian Li.

Don't convert filenames to URLs in edit drags.
https://bugs.webkit.org/show_bug.cgi?id=38826

For security reasons, we don't want to expose file system paths to web
content, so we filter them out of edit drags.

  • editing/pasteboard/file-drag-to-editable-expected.txt: Added.
  • editing/pasteboard/file-drag-to-editable.html: Added.
  • editing/pasteboard/script-tests/file-drag-to-editable.js: Added.
  • platform/gtk/Skipped:
  • platform/mac/Skipped:
  • platform/qt/Skipped:
  • platform/win/Skipped:

2010-06-10 Daniel Cheng <dcheng@chromium.org>

Reviewed by Jian Li.

Don't convert filenames to URLs in edit drags.
https://bugs.webkit.org/show_bug.cgi?id=38826

For security reasons, we don't want to expose file system paths to web
content, so we filter them out of edit drags.

Test: editing/pasteboard/file-drag-to-editable.html

  • page/DragController.cpp: (WebCore::documentFragmentFromDragData):
  • platform/DragData.h: (WebCore::DragData::):
  • platform/android/DragDataAndroid.cpp: (WebCore::DragData::containsURL): (WebCore::DragData::asURL):
  • platform/chromium/DragDataChromium.cpp: (WebCore::DragData::containsURL): (WebCore::DragData::asURL):
  • platform/efl/DragDataEfl.cpp: (WebCore::DragData::containsURL): (WebCore::DragData::asURL):
  • platform/gtk/DragDataGtk.cpp: (WebCore::DragData::containsURL): (WebCore::DragData::asURL):
  • platform/haiku/DragDataHaiku.cpp: (WebCore::DragData::containsURL): (WebCore::DragData::asURL):
  • platform/mac/DragDataMac.mm: (WebCore::DragData::containsURL): (WebCore::DragData::asURL):
  • platform/qt/DragDataQt.cpp: (WebCore::DragData::asPlainText): (WebCore::DragData::containsURL): (WebCore::DragData::asURL):
  • platform/win/ClipboardUtilitiesWin.cpp: (WebCore::getURL): (WebCore::getPlainText):
  • platform/win/ClipboardUtilitiesWin.h:
  • platform/win/ClipboardWin.cpp: (WebCore::ClipboardWin::getData):
  • platform/win/DragDataWin.cpp: (WebCore::DragData::containsURL): (WebCore::DragData::asURL):
  • platform/wince/DragDataWince.cpp: (WebCore::DragData::containsURL): (WebCore::DragData::asURL):
  • platform/wx/DragDataWx.cpp: (WebCore::DragData::containsURL): (WebCore::DragData::asURL):
07:16 Changeset [60956] by eric@webkit.org

2010-06-10 Marcus Bulach <bulach@google.com>

Reviewed by Jeremy Orlow.

[chromium] Upstream layout tests expectations for Geolocation.
Rebaseline fast/dom/Geolocation/callback-exception-expected.html due to JSC x V8 messages.
https://bugs.webkit.org/show_bug.cgi?id=39721

  • platform/chromium/fast/dom/Geolocation/callback-exception-expected.txt: Added.
  • platform/chromium/test_expectations.txt:
06:56 Changeset [60955] by eric@webkit.org

2010-06-10 Mike Belshe <mbelshe@chromium.org>

Reviewed by David Levin

Track whether a resource is loaded via a proxy.

https://bugs.webkit.org/show_bug.cgi?id=40312

  • platform/network/chromium/ResourceResponse.h: (WebCore::ResourceResponse::wasFetchedViaProxy): (WebCore::ResourceResponse::setWasFetchedViaProxy):

2010-06-10 Mike Belshe <mbelshe@google.com>

Reviewed by David Levin

Track whether a resource is loaded via a proxy.

https://bugs.webkit.org/show_bug.cgi?id=40312

  • public/WebURLResponse.h:
  • src/WebURLResponse.cpp: (WebKit::WebURLResponse::wasFetchedViaProxy): (WebKit::WebURLResponse::setWasFetchedViaProxy):
06:44 Changeset [60954] by eric@webkit.org

2010-06-10 Vangelis Kokkevis <vangelis@chromium.org>

Reviewed by Dimitri Glazkov.

[chromium] Adding the ANGLE project to the DEPS file as it's required
by the gpu-accelerated compositor.
https://bugs.webkit.org/show_bug.cgi?id=40379

  • DEPS:
06:31 Changeset [60953] by eric@webkit.org

2010-06-10 Jungshik Shin <jshin@chromium.org>

Reviewed by Kent Tamura

Bug 38224 - [chromium] Enable rendering of Ethiopic, Lao, Tibetan

and a few other scripts on Win XP

https://bugs.webkit.org/show_bug.cgi?id=38224

Make it possible to specify a list of fonts for per-script
fallback instead of a single font per script. This is
necessary for Malayalam (for which Windows font doesn't
support Unicode 5.x fully on Vista or earlier), Ethiopic
(we want to support on XP with a 3rd party font because XP
doesn't have any Ethiopic font out of the box) and some other
scripts. This is a short-term 'fix' until we have per-script
font preferences.

No layout test is added because the test results would be
dependent on which fonts are present.

  • platform/graphics/chromium/FontUtilsChromiumWin.cpp: (WebCore::): (WebCore::FontMap::ScriptToFontFamilies::):
04:23 Changeset [60952] by andreip@google.com

2010-06-09 Andrei Popescu <andreip@google.com>

Reviewed by Adam Barth.

CodeGeneratorJS.pm incorrectly increments $paramIndex when a method is declared with [CallWith]
https://bugs.webkit.org/show_bug.cgi?id=40372

Use two variables to keep track of the number of arguments passed from JavaScript vs the number
of arguments passed to the corresponding C++ method. These numbers can be different since
the parameter declared via [CallWith] is generated in the bindings instead of being passed
from JavaScript.

The layout tests added here test both that the parameters are read correctly as well as
tests some basic indexed database functionality. There are also bindings specific
tests added in TestObj.idl.

  • storage/indexeddb/idb-objectstore-request-expected.txt: Added.
  • storage/indexeddb/idb-objectstore-request.html: Added.
  • storage/indexeddb/script-tests/idb-objectstore-request.js: Added. (openSuccess): (createSuccess): (createObjectStore): (doCreateOrOpen): (test):

2010-06-09 Andrei Popescu <andreip@google.com>

Reviewed by Adam Barth.

CodeGeneratorJS.pm incorrectly increments $paramIndex when a method is declared with [CallWith]
https://bugs.webkit.org/show_bug.cgi?id=40372

Use two variables to keep track of the number of arguments passed from JavaScript vs the number
of arguments passed to the corresponding C++ method. These numbers can be different since
the parameter declared via [CallWith] is generated in the bindings instead of being passed
from JavaScript.

Test: storage/indexeddb/idb-objectstore-request.html

  • bindings/scripts/CodeGeneratorJS.pm:
  • bindings/scripts/test/CPP/WebDOMTestObj.cpp: (WebDOMTestObj::withScriptExecutionContext):
  • bindings/scripts/test/CPP/WebDOMTestObj.h:
  • bindings/scripts/test/JS/JSTestObj.cpp: (WebCore::jsTestObjPrototypeFunctionWithDynamicFrameAndArg): (WebCore::jsTestObjPrototypeFunctionWithDynamicFrameAndOptionalArg): (WebCore::jsTestObjPrototypeFunctionWithDynamicFrameAndUserGesture): (WebCore::jsTestObjPrototypeFunctionWithDynamicFrameAndUserGestureASAD):
03:34 Changeset [60951] by morrita@google.com

2010-06-09 MORITA Hajime <morrita@google.com>

Reviewed by Kent Tamura.

Refactoring: Simple shadow elements should be factored out .
https://bugs.webkit.org/show_bug.cgi?id=40400

Pulled ShadowElement class up from SliderThumbElement,
ProgressValueElement, ShadowInputElement.
And renamed ProgressValueElement to ShadowBlockElement.

No new tests. Because there is no functional change.

  • CMakeLists.txt:
  • GNUmakefile.am:
  • WebCore.gypi:
  • WebCore.pro:
  • WebCore.vcproj/WebCore.vcproj:
  • WebCore.xcodeproj/project.pbxproj:
  • rendering/RenderFileUploadControl.cpp:
  • rendering/RenderProgress.cpp: (WebCore::RenderProgress::updateValuePartState):
  • rendering/RenderProgress.h:
  • rendering/RenderSlider.cpp: (WebCore::SliderThumbElement::SliderThumbElement): (WebCore::SliderThumbElement::defaultEventHandler): (WebCore::SliderThumbElement::detach):
  • rendering/ShadowElement.cpp: Added. (WebCore::ShadowBlockElement::create): (WebCore::ShadowBlockElement::ShadowBlockElement): (WebCore::ShadowInputElement::create): (WebCore::ShadowInputElement::ShadowInputElement):
  • rendering/ShadowElement.h: Added. (WebCore::ShadowElement::ShadowElement): (WebCore::ShadowElement::isShadowNode): (WebCore::ShadowElement::shadowParentNode):
02:31 Changeset [60950] by eric@webkit.org

2010-06-10 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

Reduce FrameView.h includes to speed up build times
https://bugs.webkit.org/show_bug.cgi?id=40408

Another fix for Windows.

  • WebView.cpp:
02:21 Changeset [60949] by eric@webkit.org

2010-06-10 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

Reduce FrameView.h includes to speed up build times
https://bugs.webkit.org/show_bug.cgi?id=40408

  • Api/qwebframe.cpp:
    • Include RenderLayer.h since it's used in this file.
02:14 Changeset [60948] by eric@webkit.org

2010-06-10 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

Reduce FrameView.h includes to speed up build times
https://bugs.webkit.org/show_bug.cgi?id=40408

Another fix for Qt.

  • page/Frame.h:
    • Include CSSMutableStyleDeclaration.h since it's used by m_typingStyle = 0;
02:06 Changeset [60947] by eric@webkit.org

2010-06-10 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

Reduce FrameView.h includes to speed up build times
https://bugs.webkit.org/show_bug.cgi?id=40408

More attempted build fixes for Tiger and Qt.

  • page/Frame.cpp:
  • page/Frame.h:

2010-06-10 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

Reduce FrameView.h includes to speed up build times
https://bugs.webkit.org/show_bug.cgi?id=40408

  • webkit/webkitwebinspector.cpp:
    • Include RenderLayer.h
01:58 Changeset [60946] by eric@webkit.org

2010-06-10 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

Reduce FrameView.h includes to speed up build times
https://bugs.webkit.org/show_bug.cgi?id=40408

Add a few more RenderLayer.h includes to fix non-mac builds.

  • page/FrameView.cpp:
  • rendering/RenderObject.cpp:
  • rendering/RenderTreeAsText.cpp:
01:49 Changeset [60945] by eric@webkit.org

2010-06-10 Kwang Yul Seo <skyul@company100.net>

Reviewed by Eric Seidel.

Remove weakRandomNumber
https://bugs.webkit.org/show_bug.cgi?id=40291

weakRandomNumber is used nowhere. Currently, WeakRandom is used instead.

  • wtf/RandomNumber.cpp:
  • wtf/RandomNumber.h:
01:35 Changeset [60944] by abarth@webkit.org

2010-06-10 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

fast/repaint/opacity-change-on-overflow-float.html depends on invalid HTML parsing
https://bugs.webkit.org/show_bug.cgi?id=40262

See comment in opacity-change-on-overflow-float.html to understand why
we made this goofy change.

  • fast/repaint/opacity-change-on-overflow-float.html:
  • html5lib/resources/webkit01.dat:
  • html5lib/runner-expected.txt:
  • html5lib/webkit-resumer-expected.txt:
  • platform/chromium-win/fast/repaint/opacity-change-on-overflow-float-expected.txt:
  • platform/mac/fast/repaint/opacity-change-on-overflow-float-expected.txt:
01:33 Changeset [60943] by eric@webkit.org

2010-06-10 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

Reduce FrameView.h includes to speed up build times
https://bugs.webkit.org/show_bug.cgi?id=40408

Cleaned up FrameView.h includes and then propagated the
cascade of includes to the necessary .cpp files.
This should reduce the set of files rebuilt every
time FrameView.h changes, and should also reduce the size
of some .o files.

No functional changes, thus no tests.

  • accessibility/AccessibilityRenderObject.cpp:
  • dom/ContainerNode.cpp:
  • dom/Document.cpp:
  • dom/Element.cpp:
  • dom/EventTarget.h:
  • dom/Node.cpp:
  • editing/EditorCommand.cpp:
  • editing/SelectionController.cpp:
  • html/HTML5Tokenizer.cpp: (WebCore::HTML5Tokenizer::script):
    • No need for this to be inline. Being inline required Frame.h in the header, which causes huge include cascade.
  • html/HTML5Tokenizer.h:
  • page/DragController.cpp:
  • page/EventHandler.cpp:
  • page/Frame.h:
  • page/FrameView.h:
  • page/PrintContext.cpp:
  • page/SpatialNavigation.cpp:
  • rendering/RenderBlock.cpp:
  • rendering/RenderBlockLineLayout.cpp:
  • rendering/RenderImage.cpp:
  • rendering/RenderInline.cpp:
  • rendering/RenderObject.h:
  • rendering/RenderTextControlSingleLine.cpp:
  • rendering/RenderWidget.cpp:
  • rendering/SVGRenderSupport.cpp:
01:15 QtWebKitBackportingFixes edited by hausmann@webkit.org
(diff)
00:59 Changeset [60942] by hausmann@webkit.org

REGRESSION: [Qt] When dragging onto a page that handles the drag in Javascript it will be a move and not a copy by default
https://bugs.webkit.org/show_bug.cgi?id=40401

Patch by Andy Shaw <andy.shaw@nokia.com> on 2010-06-10
Reviewed by Simon Hausmann.

The correct pattern in Qt for Dnd events is to use acceptProposedAction() instead
of accept().

  • Api/qwebpage.cpp:

(QWebPagePrivate::dragEnterEvent):
(QWebPagePrivate::dragMoveEvent):
(QWebPagePrivate::dropEvent):

00:48 Changeset [60941] by abarth@webkit.org

2010-06-10 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

Move HTML5 entity parser to its own file
https://bugs.webkit.org/show_bug.cgi?id=40406

Moving this algorithm to its own file better encapsulates its
dependencies and makes it callable from other parts of WebCore.

  • Android.mk:
  • CMakeLists.txt:
  • GNUmakefile.am:
  • WebCore.gypi:
  • WebCore.pro:
  • WebCore.vcproj/WebCore.vcproj:
  • WebCore.xcodeproj/project.pbxproj:
  • html/HTML5EntityParser.cpp: Copied from WebCore/html/HTML5Lexer.cpp. (WebCore::): (WebCore::consumeHTML5Entity):
  • html/HTML5EntityParser.h: Copied from WebCore/html/HTML5Lexer.h.
  • html/HTML5Lexer.cpp: (WebCore::HTML5Lexer::processEntity): (WebCore::HTML5Lexer::nextToken):
  • html/HTML5Lexer.h:

06/09/10:

23:34 Changeset [60940] by abarth@webkit.org

2010-06-09 Tony Gentilcore <tonyg@chromium.org>

Reviewed by Adam Barth.

HTML5 Parser: Fix fast/profiler tests that depend on event handler line numbers
https://bugs.webkit.org/show_bug.cgi?id=40393

This emulated the old behavior in HTMLTokenizer:processToken()

No new tests because covered by:

  • fast/profiler/dead-time.html
  • fast/profiler/inline-event-handler.html
  • fast/profiler/stop-profiling-after-setTimeout.html
  • fast/profiler/throw-exception-from-eval.html
  • html/HTML5Tokenizer.cpp: (WebCore::HTML5Tokenizer::pumpLexer):
23:19 Changeset [60939] by ap@apple.com

Undo an accidentally committed change.

  • fast/events/pageshow-pagehide-on-back-cached.html:
22:38 Changeset [60938] by eric@webkit.org

2010-06-09 Adam Barth <abarth@webkit.org>

Reviewed by Ojan Vafai.

webkit-patch upload fails for security bugs because there's no rietveld flag
https://bugs.webkit.org/show_bug.cgi?id=40309

There's probably a more elegant way of seeing whether the form control
exists, but this seems to work.

  • Scripts/webkitpy/common/net/bugzilla.py:
22:33 Changeset [60937] by ap@apple.com

Reviewed by Dan Bernstein.

https://bugs.webkit.org/show_bug.cgi?id=9504
img tag rewritten by innerHTML doesn't use image map

Test: fast/dom/replaced-image-map.html

  • html/HTMLMapElement.cpp: (WebCore::HTMLMapElement::~HTMLMapElement): Destructor is not a good place to make changes that affect page behavior, because destructors are called by garbage collector. Don't undregister the image map from here. (WebCore::HTMLMapElement::parseMappedAttribute): Only tell document about the map if it's actually in document. A map in a detached subtree isn't used by Firefox at least. (WebCore::HTMLMapElement::insertedIntoDocument): Register the map. (WebCore::HTMLMapElement::removedFromDocument): Unregister the map.
  • html/HTMLMapElement.h: Added insertedIntoDocument/removedFromDocument overrides.
22:22 Changeset [60936] by rolandsteiner@chromium.org

Bug 35632 - htmlediting.cpp : isEmptyTableCell() is incomplete
https://bugs.webkit.org/show_bug.cgi?id=35632

Reviewed by Ojan Vafai.

WebCore:

Correct isEmptyTableCell to check for the presence of other renderer
children.

Test: editing/deleting/delete-br-in-last-table-cell.html

  • editing/htmlediting.cpp:

(WebCore::isEmptyTableCell):

LayoutTests:

Correct isEmptyTableCell to check for the presence of other renderer
children.

Layout test checks that deleting a <br> within the last table cell works.

  • editing/deleting/delete-br-in-last-table-cell-expected.txt: Added.
  • editing/deleting/delete-br-in-last-table-cell.html: Added.
  • platform/gtk/Skipped:
  • platform/qt/Skipped:
21:39 Changeset [60935] by eric@webkit.org

2010-06-09 Pierre Rossi <pierre.rossi@nokia.com>

Reviewed by Kenneth Rohde Christiansen.

QWebPage::findText() does not clear selection when passed empty string
https://bugs.webkit.org/show_bug.cgi?id=31779

  • Api/qwebpage.cpp: (QWebPage::findText):
  • tests/qwebpage/tst_qwebpage.cpp: (tst_QWebPage::findText):
21:26 Changeset [60934] by eric@webkit.org

2010-06-09 Bo Liu <boliu@chromium.org>

Reviewed by David Levin.

[Chromium] Pass <video> full screen calls to Chromium port client
https://bugs.webkit.org/show_bug.cgi?id=40317

  • public/WebViewClient.h: (WebKit::WebViewClient::supportsFullscreen): (WebKit::WebViewClient::enterFullscreenForNode): (WebKit::WebViewClient::exitFullscreenForNode):
  • src/ChromeClientImpl.cpp: (WebKit::ChromeClientImpl::supportsFullscreenForNode): (WebKit::ChromeClientImpl::enterFullscreenForNode): (WebKit::ChromeClientImpl::exitFullscreenForNode):
  • src/ChromeClientImpl.h:
20:47 Changeset [60933] by eric@webkit.org

2010-06-09 Qi Zhang <qi.2.zhang@nokia.com>

Reviewed by Laszlo Gombos.

[Qt] Failed at http://philip.html5.org/tests/canvas/suite/tests/2d.path.arcTo.transformation.html
https://bugs.webkit.org/show_bug.cgi?id=38598

Remove canvas/philip/tests/2d.path.arcTo.scale.html, canvas/philip/tests/2d.path.arcTo.transformation.html
and canvas/philip/tests/2d.path.transformation.changing.html from Skipped.

  • platform/qt/Skipped:

2010-06-09 Qi Zhang <qi.2.zhang@nokia.com>

Reviewed by Laszlo Gombos.

[Qt] Failed at http://philip.html5.org/tests/canvas/suite/tests/2d.path.arcTo.transformation.html
https://bugs.webkit.org/show_bug.cgi?id=38598

In path transform function handle path only have moveElement case.

  • platform/graphics/qt/PathQt.cpp: (WebCore::Path::transform):
20:01 Changeset [60932] by eric@webkit.org

2010-06-09 Anton Muhin <antonm@chromium.org>

Reviewed by Nate Chapin.

[v8] First phase of switching to new named property query API
https://bugs.webkit.org/show_bug.cgi?id=40303

To allow better management of attributes of intercepted properties,
we're starting to switch to new named property query API which
now could return attributes instead of simple property present/absent
flag. The next step would remove USE_NEW_QUERY_CALLBACK, then
v8 would have them enabled by default.

  • bindings/scripts/CodeGeneratorV8.pm:
  • bindings/v8/NPV8Object.cpp:
  • bindings/v8/ScriptArray.cpp:
  • bindings/v8/ScriptCallStack.cpp:
  • bindings/v8/SerializedScriptValue.cpp:
  • bindings/v8/V8Binding.cpp:
  • bindings/v8/V8DOMWrapper.cpp:
  • bindings/v8/V8GCController.cpp:
  • bindings/v8/V8IsolatedContext.cpp:
  • bindings/v8/V8NPObject.cpp: (WebCore::npObjectQueryProperty):
  • bindings/v8/V8Proxy.cpp:
  • bindings/v8/custom/V8PopStateEventCustom.cpp:
  • bindings/v8/custom/V8StorageCustom.cpp: (WebCore::V8Storage::namedPropertyQuery):
  • config.h:
19:40 Changeset [60931] by eric@webkit.org

2010-06-09 Steve Block <steveblock@google.com>

Reviewed by Jeremy Orlow.

Need to include V8Proxy.h in V8 generated bindings for toV8Context()
https://bugs.webkit.org/show_bug.cgi?id=40290

No new tests, build fix only.

  • bindings/scripts/CodeGeneratorV8.pm:
19:29 Changeset [60930] by eric@webkit.org

2010-06-09 Jarkko Sakkinen <jarkko.j.sakkinen@gmail.com>

Reviewed by Kenneth Rohde Christiansen.

WebGL rendering context does not activate unless accelerated
compositing is enabled from settings.
https://bugs.webkit.org/show_bug.cgi?id=37772

  • html/HTMLCanvasElement.cpp: (WebCore::HTMLCanvasElement::getContext):
19:13 Changeset [60929] by eric@webkit.org

2010-06-09 Anders Bakken <agbakken@gmail.com>

Reviewed by David Levin.

[Qt] PageClientQt.h has coding-style errors
https://bugs.webkit.org/show_bug.cgi?id=40257

  • WebCoreSupport/PageClientQt.h:
19:00 Changeset [60928] by eric@webkit.org

2010-06-09 Anders Bakken <agbakken@gmail.com>

Reviewed by David Levin.

[Qt] InspectorClientQt.cpp has coding-style errors
https://bugs.webkit.org/show_bug.cgi?id=40255

  • WebCoreSupport/InspectorClientQt.cpp:
18:48 Changeset [60927] by eric@webkit.org

2010-06-09 Yong Li <yoli@rim.com>

Reviewed by George Staikos.

https://bugs.webkit.org/show_bug.cgi?id=40252
Make image down-sampling threshold runtime adjustable.

No new test needed.

  • platform/graphics/ImageSource.cpp: (WebCore::ImageSource::setData):
  • platform/graphics/ImageSource.h: (WebCore::ImageSource::maxPixelsPerDecodedImage): (WebCore::ImageSource::setMaxPixelsPerDecodedImage):
17:49 Changeset [60926] by abarth@webkit.org

2010-06-09 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

Fix handling of bytes received from the network while in document.write
https://bugs.webkit.org/show_bug.cgi?id=40356

The old tokenizer has specially logic for handling the case of
receiving bytes from the network while in a nested call to
document.write. This patch implements similar logic for the HTML5
tokenizer. Also, this patch abstracts the tricky shuffling of
SegmentedStrings behind a simple API.

I'm not sure how to trigger this case. My guess is we can trigger it
using a nested event loop, e.g., via alert(), but I'm not sure how to
test that in a LayoutTest. There don't appear to be any LayoutTests
that currently test this behavior despite it being present in the old
tokenizer.

  • html/HTML5Tokenizer.cpp: (WebCore::HTML5Tokenizer::pumpLexer): (WebCore::HTML5Tokenizer::write):
    • Added a branch for the |append| argument.

(WebCore::HTML5Tokenizer::end):
(WebCore::HTML5Tokenizer::finish):
(WebCore::HTML5Tokenizer::executeScript):

  • Switch over to using a RAII pattern for recording and restoring insertion points.
  • html/HTML5Tokenizer.h: (WebCore::HTML5Tokenizer::InputStream::InputStream): (WebCore::HTML5Tokenizer::InputStream::appendToEnd): (WebCore::HTML5Tokenizer::InputStream::insertAtCurrentInsertionPoint): (WebCore::HTML5Tokenizer::InputStream::close):
    • Putting the close() method on InputStream makes it much easier to handle EOF. We now just close the last buffer in the stream when the network says it's done.

(WebCore::HTML5Tokenizer::InputStream::current):

  • This class could be moved to its own file, but it shouldn't be used outside of the tokenizer.

(WebCore::HTML5Tokenizer::InsertionPointRecord::InsertionPointRecord):
(WebCore::HTML5Tokenizer::InsertionPointRecord::~InsertionPointRecord):

  • A simple RAII class for managing saved insertion points.
  • platform/text/SegmentedString.cpp: (WebCore::SegmentedString::operator=):
    • Fix a related bug where m_closed was not being copied properly in the assignment operator.
17:30 Changeset [60925] by eric@webkit.org

2010-06-09 Anders Bakken <agbakken@gmail.com>

Reviewed by David Levin.

[Qt] FrameLoaderClientQt.h has coding-style errors
https://bugs.webkit.org/show_bug.cgi?id=40251

  • WebCoreSupport/FrameLoaderClientQt.h: (WebCore::FrameLoaderClientQt::dispatchWillSendSubmitEvent):
17:28 Changeset [60924] by ojan@chromium.org

2010-06-09 Ojan Vafai <ojan@chromium.org>

Reviewed by Adam Barth.

make rietveld upload faster and avoid posting to bug on errors
https://bugs.webkit.org/show_bug.cgi?id=40389

Only grab the first item of the upload queue instead of trying
to compute the whole list upfront (which is O(n) bugzilla lookups!).

Also, don't post comments to the bug when uploading fails.

  • Scripts/webkitpy/common/net/bugzilla.py:
  • Scripts/webkitpy/tool/commands/queues.py:
  • Scripts/webkitpy/tool/commands/queues_unittest.py:
  • Scripts/webkitpy/tool/commands/stepsequence.py:
  • Scripts/webkitpy/tool/mocktool.py:
16:25 Changeset [60923] by eric@webkit.org

2010-06-09 Anders Bakken <agbakken@gmail.com>

Reviewed by Kenneth Rohde Christiansen.

[Qt] EditorClientQt.h has coding-style errors
https://bugs.webkit.org/show_bug.cgi?id=40249

  • WebCoreSupport/EditorClientQt.h:
16:18 Changeset [60922] by zmo@google.com

Unreviewed. Add myself to the committer list.

15:55 Changeset [60921] by eric@webkit.org

2010-06-09 Anders Bakken <agbakken@gmail.com>

Reviewed by David Levin.

[Qt] EditCommandQt.cpp/h have coding-style errors
https://bugs.webkit.org/show_bug.cgi?id=40248

  • WebCoreSupport/EditCommandQt.cpp: (EditCommandQt::EditCommandQt): (EditCommandQt::~EditCommandQt): (EditCommandQt::redo): (EditCommandQt::undo):
  • WebCoreSupport/EditCommandQt.h:
15:45 Changeset [60920] by abarth@webkit.org

2010-06-09 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

Fix typo in fast/lists/ordered-list-with-no-ol-tag.html
https://bugs.webkit.org/show_bug.cgi?id=40391

This test contains an untermianted end tag, which parses differently in
the HTML5 parser than in the legacy parser. Having read the bug that
caused this test to be added, making the text valid HTML seems to
respect the intent of the test (which is about numbering).

As discussed in other patches, we have the parsing behavior covered in
parsing tests.

  • fast/lists/ordered-list-with-no-ol-tag.html:
15:43 Changeset [60919] by abarth@webkit.org

2010-06-09 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

Make invalid comment in fast/css/word-space-extra.html really invalid
https://bugs.webkit.org/show_bug.cgi?id=40388

This test tries to have a comment inside a comment. To make that work
in the old parser, the author added a space between the -- and the >.
However, that's a valid end-of-comment in the new parser. I've change
the space to a * to really make it not a valid end of comment.

  • fast/css/word-space-extra.html:
15:30 Changeset [60918] by ap@apple.com

Reviewed by Brady Eidson.

Export StringImpl::ascii(). It might be not very useful, but it's a public function.

15:07 Changeset [60917] by eric@webkit.org

2010-06-09 Anders Bakken <agbakken@gmail.com>

Reviewed by David Levin.

[Qt] DumpRenderTreeSupportQt.cpp has coding-style errors
https://bugs.webkit.org/show_bug.cgi?id=40247

  • WebCoreSupport/DumpRenderTreeSupportQt.cpp: (DumpRenderTreeSupportQt::firstRectForCharacterRange):
15:05 Changeset [60916] by abarth@webkit.org

2010-06-09 Tony Gentilcore <tonyg@chromium.org>

Reviewed by Adam Barth.

Fix fast/parser/hex-entities-length.html for HTML5 parser
https://bugs.webkit.org/show_bug.cgi?id=40385

Stripping surrogate pair values appears to be a problem with the spec
in "Preprocessing the input stream." Minefield doesn't appear to
implement the part in question. So this patch removes that guard and
adds a FIXME to track the issue.

No new tests because covered by fast/parser/hex-entities-length.html

  • html/HTML5Lexer.h: (WebCore::HTML5Lexer::InputStreamPreprocessor::peek):
14:51 Changeset [60915] by eric@webkit.org

2010-06-09 Anders Bakken <agbakken@gmail.com>

Reviewed by David Levin.

[Qt] DragClientQt.cpp has coding-style errors
https://bugs.webkit.org/show_bug.cgi?id=40246

  • WebCoreSupport/DragClientQt.cpp: (WebCore::DragClientQt::startDrag):
14:39 Changeset [60914] by kbr@google.com

2010-06-09 Kenneth Russell <kbr@google.com>

Reviewed by Dimitri Glazkov.

Update readPixels to take ArrayBufferView rather than returning it
https://bugs.webkit.org/show_bug.cgi?id=40322

No new tests; covered by existing tests, which have been modified.

  • html/canvas/WebGLRenderingContext.cpp: (WebCore::WebGLRenderingContext::readPixels):
  • html/canvas/WebGLRenderingContext.h:
  • html/canvas/WebGLRenderingContext.idl:

2010-06-09 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

Test how XML directives get attached to the DOM
https://bugs.webkit.org/show_bug.cgi?id=40384

In the old parser, the XML directive doesn't get added to the DOM. In
new parser (and in Minefield), it gets added as a comment, which shows
up in this test.

This behavior was tested more obliquely by
fast/css-generated-content/hover-style-change.html in some
DumpRenderTree output. I've removed that coverage from the CSS test
and moved it to a new parser test that can actually be run in other
browsers.

  • fast/css-generated-content/hover-style-change.html:
    • Remove XML directive that was causing this test to depend on out-of-spec parser behavior w.r.t. whether to attach the XML directive to the DOM.
  • fast/parser/xml-directive-in-dom-expected.txt: Added.
  • fast/parser/xml-directive-in-dom.html: Added.
14:19 Changeset [60913] by abarth@webkit.org

2010-06-09 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

Test how XML directives get attached to the DOM
https://bugs.webkit.org/show_bug.cgi?id=40384

In the old parser, the XML directive doesn't get added to the DOM. In
new parser (and in Minefield), it gets added as a comment, which shows
up in this test.

This behavior was tested more obliquely by
fast/css-generated-content/hover-style-change.html in some
DumpRenderTree output. I've removed that coverage from the CSS test
and moved it to a new parser test that can actually be run in other
browsers.

  • fast/css-generated-content/hover-style-change.html:
    • Remove XML directive that was causing this test to depend on out-of-spec parser behavior w.r.t. whether to attach the XML directive to the DOM.
  • fast/parser/xml-directive-in-dom-expected.txt: Added.
  • fast/parser/xml-directive-in-dom.html: Added.
14:07 Changeset [60912] by eric@webkit.org

2010-06-09 Anders Bakken <agbakken@gmail.com>

Reviewed by Laszlo Gombos.

[Qt] ContextMenuClientQt.h has coding-style errors
https://bugs.webkit.org/show_bug.cgi?id=40242

  • WebCoreSupport/ContextMenuClientQt.h:
14:02 Changeset [60911] by jchaffraix@webkit.org

WebCore: Fix drawing zero-sized gradients on Canvas.
https://bugs.webkit.org/show_bug.cgi?id=40340

Patch by Enrico Ros <eros@codeaurora.org> on 2010-06-09
Reviewed by Ariya Hidayat.

HTML5 draft says that nothing must be painted if the gradient start
point equals the stop point.

This commit fixes the following canvas test:
http://philip.html5.org/tests/canvas/suite/tests/2d.gradient.interpolate.zerosize.html

  • html/canvas/CanvasRenderingContext2D.cpp:

(WebCore::CanvasRenderingContext2D::fillRect): skip zero-sized linear gradients

  • platform/graphics/Gradient.h:

(WebCore::Gradient::isRadial): made public
(WebCore::Gradient::isZeroSize): true if start == stop

LayoutTests: Fixed zero-sized gradients on canvas.
https://bugs.webkit.org/show_bug.cgi?id=40340

Patch by Enrico Ros <eros@codeaurora.org> on 2010-06-09
Reviewed by Ariya Hidayat.

  • platform/mac/Skipped: Unskip the test that is passing now.
  • platform/qt/Skipped: Unskip the test that is passing now.
13:53 Changeset [60910] by eric@webkit.org

2010-06-09 Anders Bakken <agbakken@gmail.com>

Reviewed by David Levin.

[Qt] ChromeClientQt.cpp has coding-style errors
https://bugs.webkit.org/show_bug.cgi?id=40240

  • WebCoreSupport/ChromeClientQt.cpp: (WebCore::ChromeClientQt::pageRect): (WebCore::ChromeClientQt::createWindow): (WebCore::ChromeClientQt::runJavaScriptAlert): (WebCore::ChromeClientQt::runJavaScriptConfirm): (WebCore::ChromeClientQt::runJavaScriptPrompt): (WebCore::ChromeClientQt::print): (WebCore::ChromeClientQt::attachRootGraphicsLayer): (WebCore::ChromeClientQt::visibleRectForTiledBackingStore):
13:31 Changeset [60909] by eric@webkit.org

2010-06-09 Leandro Pereira <leandro@profusion.mobi>

Reviewed by Adam Treat.

[EFL] Allow building core libraries as shared objects to speed up
linking time on machines with small amounts of memory.
http://webkit.org/b/39899

  • CMakeLists.txt: Add SHARED_CORE option.
  • cmake/FindGlib.cmake: Add GObject to the list of GLib libraries to search.
  • cmake/OptionsEfl.cmake: Remove default library type definitions.
  • cmake/WebKitHelpers.cmake: -fvisibility=hidden did not behave well when SHARED_CORE is used.

2010-06-09 Leandro Pereira <leandro@profusion.mobi>

Reviewed by Adam Treat.

[EFL] Allow building core libraries as shared objects to speed up
linking time on machines with small amounts of memory.
http://webkit.org/b/39899

  • CMakeLists.txt: If building with shared core, install the lib.
  • jsc/CMakeListsEfl.txt: Needs Glib and Ecore to link dynamically.
  • wtf/CMakeLists.txt: If building with shared core, install the lib.

2010-06-09 Leandro Pereira <leandro@profusion.mobi>

Reviewed by Adam Treat.

[EFL] Allow building core libraries as shared objects to speed up
linking time on machines with small amounts of memory.
http://webkit.org/b/39899

  • CMakeLists.txt: If building with shared core, install the lib. Remove npapi.cpp (causes conflicts when linking dynamically).
  • CMakeListsEfl.txt: EFL libraries are needed to link dynamically.
13:03 Changeset [60908] by mrowe@apple.com

<rdar://problem/8070662> REGRESSION (r51629): WebBackForwardList created via -init crashes when -addItem: is called.

Reviewed by Adele Peterson.

A WebBackForwardList created via -init results in a BackForwardList being created with a null m_page.
BackForwardList needs to be careful not to dereference m_page without first ensuring it's not null.

  • history/BackForwardList.cpp:

(WebCore::BackForwardList::addItem): Null-check m_page.
(WebCore::BackForwardList::goBack): Ditto.
(WebCore::BackForwardList::goForward): Ditto.
(WebCore::BackForwardList::goToItem): Ditto.
(WebCore::BackForwardList::setCapacity): Ditto.

12:37 Changeset [60907] by eric@webkit.org

2010-06-09 Kwang Yul Seo <skyul@company100.net>

Reviewed by Kent Tamura.

[BREWMP] Add dummy Context Menu implementation
https://bugs.webkit.org/show_bug.cgi?id=40223

Brew MP does not use Context Menu. Add dummy implementation.

  • platform/brew/ContextMenuBrew.cpp: Added. (WebCore::ContextMenu::ContextMenu): (WebCore::ContextMenu::~ContextMenu): (WebCore::ContextMenu::itemCount): (WebCore::ContextMenu::insertItem): (WebCore::ContextMenu::appendItem): (WebCore::ContextMenu::itemWithAction): (WebCore::ContextMenu::itemAtIndex): (WebCore::ContextMenu::setPlatformDescription): (WebCore::ContextMenu::platformDescription): (WebCore::ContextMenu::releasePlatformDescription):
  • platform/brew/ContextMenuItemBrew.cpp: Added. (WebCore::ContextMenuItem::ContextMenuItem): (WebCore::ContextMenuItem::~ContextMenuItem): (WebCore::ContextMenuItem::releasePlatformDescription): (WebCore::ContextMenuItem::type): (WebCore::ContextMenuItem::action): (WebCore::ContextMenuItem::title): (WebCore::ContextMenuItem::platformSubMenu): (WebCore::ContextMenuItem::setType): (WebCore::ContextMenuItem::setAction): (WebCore::ContextMenuItem::setTitle): (WebCore::ContextMenuItem::setSubMenu): (WebCore::ContextMenuItem::setChecked): (WebCore::ContextMenuItem::setEnabled): (WebCore::ContextMenuItem::enabled):
12:25 Changeset [60906] by eric@webkit.org

2010-06-09 Kwang Yul Seo <skyul@company100.net>

Reviewed by Kent Tamura.

[BREWMP] Port Clipboard
https://bugs.webkit.org/show_bug.cgi?id=35734

Add dummy Clipboard.

  • platform/brew/ClipboardBrew.cpp: Added. (WebCore::ClipboardBrew::ClipboardBrew): (WebCore::ClipboardBrew::~ClipboardBrew): (WebCore::ClipboardBrew::clearData): (WebCore::ClipboardBrew::clearAllData): (WebCore::ClipboardBrew::getData): (WebCore::ClipboardBrew::setData): (WebCore::ClipboardBrew::types): (WebCore::ClipboardBrew::files): (WebCore::ClipboardBrew::setDragImage): (WebCore::ClipboardBrew::setDragImageElement): (WebCore::ClipboardBrew::createDragImage): (WebCore::ClipboardBrew::declareAndWriteDragImage): (WebCore::ClipboardBrew::writeURL): (WebCore::ClipboardBrew::writeRange): (WebCore::ClipboardBrew::writePlainText): (WebCore::ClipboardBrew::hasData):
  • platform/brew/ClipboardBrew.h: Added.
12:12 Changeset [60905] by eric@webkit.org

2010-06-09 Anders Bakken <agbakken@gmail.com>

Reviewed by David Levin.

[Qt] ClipboardQt.cpp has coding-style errors
https://bugs.webkit.org/show_bug.cgi?id=39781

  • platform/qt/ClipboardQt.cpp: (WebCore::ClipboardQt::declareAndWriteDragImage):
12:06 Changeset [60904] by jparent@chromium.org

Bugzilla: Don't send mail for cancel/deny/granted for in-rietveld.

Reviewed by David Kilzer.

https://bugs.webkit.org/show_bug.cgi?id=40345

  • Bugzilla/Flag.pm:
11:32 Changeset [60903] by ojan@chromium.org

2010-06-09 Ojan Vafai <ojan@chromium.org>

Reviewed by Adam Barth.

Rietveld upload queue fails when setting in-rietveld flag
https://bugs.webkit.org/show_bug.cgi?id=40371

Make the comment arguments to set_flag_on_attachment optional
and add it to the MockBugzilla.

  • Scripts/webkitpy/common/net/bugzilla.py:
  • Scripts/webkitpy/tool/bot/queueengine.py:
  • Scripts/webkitpy/tool/commands/download_unittest.py:
  • Scripts/webkitpy/tool/commands/queues.py:
  • Scripts/webkitpy/tool/commands/queues_unittest.py:
  • Scripts/webkitpy/tool/mocktool.py:
11:24 Changeset [60902] by kbr@google.com

2010-06-09 Kenneth Russell <kbr@google.com>

Reviewed by Dimitri Glazkov.

Rename FloatArray to Float32Array
https://bugs.webkit.org/show_bug.cgi?id=40323

Used do-webcore-rename to perform renaming. Manually undid
incorrect changes to WebCore/bridge/jni/jsc/JNIBridgeJSC.cpp,
WebCore/manual-tests/resources/ArrayParameterTestApplet.java and
ArrayParameterTestApplet.class. Updated LayoutTests. Built and ran
all layout tests on Safari; built Chromium and ran selected WebGL
tests.

  • src/GraphicsContext3D.cpp:

2010-06-09 Kenneth Russell <kbr@google.com>

Reviewed by Dimitri Glazkov.

Rename FloatArray to Float32Array
https://bugs.webkit.org/show_bug.cgi?id=40323

Used do-webcore-rename to perform renaming. Manually undid
incorrect changes to WebCore/bridge/jni/jsc/JNIBridgeJSC.cpp,
WebCore/manual-tests/resources/ArrayParameterTestApplet.java and
ArrayParameterTestApplet.class. Updated LayoutTests. Built and ran
all layout tests on Safari; built Chromium and ran selected WebGL
tests.

  • CMakeLists.txt:
  • DerivedSources.make:
  • GNUmakefile.am:
  • WebCore.gypi:
  • WebCore.pri:
  • WebCore.pro:
  • WebCore.xcodeproj/project.pbxproj:
  • bindings/generic/RuntimeEnabledFeatures.h: (WebCore::RuntimeEnabledFeatures::float32ArrayEnabled):
  • bindings/js/JSArrayBufferViewCustom.cpp: (WebCore::toJS):
  • bindings/js/JSDOMWindowCustom.cpp: (WebCore::JSDOMWindow::float32Array): (WebCore::JSDOMWindow::webGLFloatArray):
  • bindings/js/JSFloat32ArrayConstructor.cpp: Copied from WebCore/bindings/js/JSFloatArrayConstructor.cpp. (WebCore::): (WebCore::JSFloat32ArrayConstructor::JSFloat32ArrayConstructor): (WebCore::constructCanvasFloatArray): (WebCore::JSFloat32ArrayConstructor::getConstructData):
  • bindings/js/JSFloat32ArrayConstructor.h: Copied from WebCore/bindings/js/JSFloatArrayConstructor.h.
  • bindings/js/JSFloat32ArrayCustom.cpp: Copied from WebCore/bindings/js/JSFloatArrayCustom.cpp. (WebCore::JSFloat32Array::indexSetter): (WebCore::toJS): (WebCore::JSFloat32Array::set):
  • bindings/js/JSFloatArrayConstructor.cpp: Removed.
  • bindings/js/JSFloatArrayConstructor.h: Removed.
  • bindings/js/JSFloatArrayCustom.cpp: Removed.
  • bindings/js/JSWebGLRenderingContextCustom.cpp: (WebCore::dataFunctionf): (WebCore::dataFunctionMatrix):
  • bindings/v8/custom/V8ArrayBufferViewCustom.cpp: (WebCore::toV8):
  • bindings/v8/custom/V8DOMWindowCustom.cpp: (WebCore::V8DOMWindow::WebGLFloatArrayAccessorGetter):
  • bindings/v8/custom/V8Float32ArrayCustom.cpp: Copied from WebCore/bindings/v8/custom/V8FloatArrayCustom.cpp. (WebCore::V8Float32Array::constructorCallback): (WebCore::V8Float32Array::setCallback): (WebCore::toV8):
  • bindings/v8/custom/V8FloatArrayCustom.cpp: Removed.
  • bindings/v8/custom/V8WebGLRenderingContextCustom.cpp: (WebCore::vertexAttribAndUniformHelperf): (WebCore::uniformMatrixHelper):
  • html/canvas/Float32Array.cpp: Copied from WebCore/html/canvas/FloatArray.cpp. (WebCore::Float32Array::create): (WebCore::Float32Array::Float32Array): (WebCore::Float32Array::slice):
  • html/canvas/Float32Array.h: Copied from WebCore/html/canvas/FloatArray.h.
  • html/canvas/Float32Array.idl: Copied from WebCore/html/canvas/FloatArray.idl.
  • html/canvas/FloatArray.cpp: Removed.
  • html/canvas/FloatArray.h: Removed.
  • html/canvas/FloatArray.idl: Removed.
  • html/canvas/WebGLGetInfo.cpp: (WebCore::WebGLGetInfo::WebGLGetInfo): (WebCore::WebGLGetInfo::getWebGLFloatArray):
  • html/canvas/WebGLGetInfo.h:
  • html/canvas/WebGLRenderingContext.cpp: (WebCore::WebGLRenderingContext::getUniform): (WebCore::WebGLRenderingContext::getVertexAttrib): (WebCore::WebGLRenderingContext::uniform1fv): (WebCore::WebGLRenderingContext::uniform2fv): (WebCore::WebGLRenderingContext::uniform3fv): (WebCore::WebGLRenderingContext::uniform4fv): (WebCore::WebGLRenderingContext::uniformMatrix2fv): (WebCore::WebGLRenderingContext::uniformMatrix3fv): (WebCore::WebGLRenderingContext::uniformMatrix4fv): (WebCore::WebGLRenderingContext::vertexAttrib1fv): (WebCore::WebGLRenderingContext::vertexAttrib2fv): (WebCore::WebGLRenderingContext::vertexAttrib3fv): (WebCore::WebGLRenderingContext::vertexAttrib4fv): (WebCore::WebGLRenderingContext::getWebGLFloatArrayParameter):
  • html/canvas/WebGLRenderingContext.h:
  • html/canvas/WebGLRenderingContext.idl:
  • page/DOMWindow.idl:
  • platform/graphics/GraphicsContext3D.h:
  • platform/graphics/mac/GraphicsContext3DMac.cpp:
  • platform/graphics/qt/GraphicsContext3DQt.cpp:

2010-06-09 Kenneth Russell <kbr@google.com>

Reviewed by Dimitri Glazkov.

Rename FloatArray to Float32Array
https://bugs.webkit.org/show_bug.cgi?id=40323

Used do-webcore-rename to perform renaming. Manually undid
incorrect changes to WebCore/bridge/jni/jsc/JNIBridgeJSC.cpp,
WebCore/manual-tests/resources/ArrayParameterTestApplet.java and
ArrayParameterTestApplet.class. Updated LayoutTests. Built and ran
all layout tests on Safari; built Chromium and ran selected WebGL
tests.

  • fast/canvas/webgl/array-unit-tests-expected.txt:
  • fast/canvas/webgl/array-unit-tests.html:
  • fast/dom/Window/script-tests/window-property-descriptors.js:
  • fast/dom/Window/window-properties.html:
  • fast/dom/script-tests/constructed-objects-prototypes.js: (constructorPropertiesOnWindow):
  • fast/dom/script-tests/prototype-inheritance-2.js: (constructorNamesForWindow):
  • fast/dom/script-tests/prototype-inheritance.js:
  • fast/js/script-tests/global-constructors.js:

2010-06-09 Kenneth Russell <kbr@google.com>

Reviewed by Dimitri Glazkov.

Rename FloatArray to Float32Array
https://bugs.webkit.org/show_bug.cgi?id=40323

Used do-webcore-rename to perform renaming. Manually undid
incorrect changes to WebCore/bridge/jni/jsc/JNIBridgeJSC.cpp,
WebCore/manual-tests/resources/ArrayParameterTestApplet.java and
ArrayParameterTestApplet.class. Updated LayoutTests. Built and ran
all layout tests on Safari; built Chromium and ran selected WebGL
tests.

  • Scripts/do-webcore-rename:
11:12 Changeset [60901] by loki@webkit.org

Remove some unused variable warnings from JITOpcodes
https://bugs.webkit.org/show_bug.cgi?id=40298

Reviewed by David Levin.

  • jit/JITOpcodes.cpp:

(JSC::JIT::privateCompileCTINativeCall):

  • jit/JITOpcodes32_64.cpp:

(JSC::JIT::privateCompileCTINativeCall):

10:49 Changeset [60900] by abarth@webkit.org

2010-06-09 Tony Gentilcore <tonyg@chromium.org>

Reviewed by Adam Barth.

video-timeupdate-duringplayback.html missing closing angle
https://bugs.webkit.org/show_bug.cgi?id=40311

This patch removes invalid markup from video-timeupdate-during-playback.html
because that is not the point of that test. It adds a new test with
the same markup to webkit01.dat. New test fails with --html5-parser.

  • html5lib/resources/webkit01.dat:
  • media/video-timeupdate-during-playback.html:
10:12 Changeset [60899] by abarth@webkit.org

2010-06-09 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

LayoutTests needlessly depend on old parser quirks covered by other tests
https://bugs.webkit.org/show_bug.cgi?id=40354

There are two quirks of the old HTMLTokenizer which I'm removing
from various tests in this change.

  • <tag<tag> parsing is a quirk unique to WebKit's old tokenizer and is already tested by fast/invalid/016.html and html5lib/runner.html.
  • <script/> is another quirk unique to WebKit's old tokenizer and is covered by fast/parser/script-tag-with-trailing-slash.html
  • fast/block/float/4145535Crash.html:
    • The test tests a CSS crasher. Changed the html to be explicit about how it should be parsed instead of depending on the <tag<tag> quirk of the old parser.
  • fast/dom/HTMLScriptElement/script-load-events.html:
    • Remove testing of <script/> since that's already covered by fast/parser/script-tag-with-trailing-slash.html
  • fast/dom/HTMLScriptElement/script-set-src-expected.txt:
    • Same as above.
  • fast/dom/HTMLScriptElement/script-set-src.html:
    • Same as above.
  • fast/events/pageshow-pagehide.html:
    • Remove dependence on <tag<tag> quirk.
  • fast/events/resources/pageshow-pagehide-subframe-cachable.html:
    • <tag<tag> quirk.
  • fast/events/resources/pageshow-pagehide-subframe.html:
    • <tag<tag> quirk.
  • fast/frames/sandboxed-iframe-navigation-parent.html:
    • <tag<tag> quirk.
  • fast/lists/w3-css3-list-styles-alphabetic.html:
    • <tag<tag> quirk.
  • fast/multicol/negativeColumnWidth.html:
    • <tag<tag> quirk.
  • fast/ruby/ruby-illegal-2.html:
    • The original fix (https://trac.webkit.org/changeset/53525) was a fix to rendering code. I've simply corrected the test to be explicit about how it should be parsed so that it can test the rendering code path without depending on the <tag<tag> quirk of the old HTMLTokenizer.
  • fast/ruby/ruby-illegal-3.html:
    • Same as above.
  • http/tests/misc/generated-content-inside-table.html:
    • Remove dependence on the <script/> quirk.
  • svg/custom/svg-fonts-in-html-linebreaks.html:
    • <tag<tag> quirk.
10:09 Changeset [60898] by abarth@webkit.org

2010-06-09 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

HTML5 Parser needs to integrate with the XSSAuditor
https://bugs.webkit.org/show_bug.cgi?id=40287

This fixes most of the XSSAuditor tests, except for the ones
which rely on the srcValue "context". The previous HTMLTokenizer
implementation was both Lexer and ScriptRunner and thus could
provide the XSSAuditor with the un-modified attribute source.
This naive implementation will fail the context-sensitive tests
but Adam Barth says he'll just have to find a new way to provide
the required information to the XSSAuditor in a later patch.

Covered by numerous http/tests/security/xssAuditor tests.

  • html/HTML5ScriptRunner.cpp: (WebCore::HTML5ScriptRunner::requestScript):
    • Ask the HTML5ScriptRunner host before running any scripts.
  • html/HTML5ScriptRunnerHost.h:
    • Add a shouldLoadExternalScriptFromSrc declaration.
  • html/HTML5Tokenizer.cpp: (WebCore::HTML5Tokenizer::shouldLoadExternalScriptFromSrc):
    • Ask the XSSAuditor if we're allowed to run the passed script.
  • html/HTML5Tokenizer.h:
10:05 Changeset [60897] by abarth@webkit.org

2010-06-09 Tony Gentilcore <tonyg@chromium.org>

Reviewed by Adam Barth.

Fix script-after-frameset test in HTML5 parser
https://bugs.webkit.org/show_bug.cgi?id=40274

The old HTMLTokenizer enforced this in scriptHandler(). We don't use
that code anymore, so it needs to be implemented. The spec does this
as a part of "insertion mode"->"after frameset", so this adds an
InsertionMode enum instead of a one-off boolean.

No new tests because covered by fast/tokenizer/script-after-frameset.html

  • html/HTML5TreeBuilder.cpp: (WebCore::HTML5TreeBuilder::HTML5TreeBuilder): (WebCore::HTML5TreeBuilder::passTokenToLegacyParser):
  • html/HTML5TreeBuilder.h: (WebCore::HTML5TreeBuilder::): (WebCore::HTML5TreeBuilder::setInsertionMode): (WebCore::HTML5TreeBuilder::insertionMode):
09:58 Changeset [60896] by eric@webkit.org

2010-06-09 Kwang Yul Seo <skyul@company100.net>

Reviewed by Kent Tamura.

[BREWMP] Add EventLoop
https://bugs.webkit.org/show_bug.cgi?id=39401

BREW MP does not have an explicit event loop. Add dummy EventLoop::cycle.

  • platform/brew/EventLoopBrew.cpp: Added. (WebCore::EventLoop::cycle):
09:01 Changeset [60895] by loislo@chromium.org

2010-06-09 Sheriff Bot <webkit.review.bot@gmail.com>

Unreviewed, rolling out r60889.
http://trac.webkit.org/changeset/60889
https://bugs.webkit.org/show_bug.cgi?id=40365

gtk bot has some kind of memory corruption (Requested by
loislo on #webkit).

  • inspector/InspectorClient.h:
  • inspector/InspectorController.cpp: (WebCore::InspectorController::setFrontend):
  • inspector/InspectorController.h:
  • inspector/InspectorFrontend.cpp: (WebCore::InspectorFrontend::InspectorFrontend):
  • inspector/InspectorFrontend.h:
  • inspector/InspectorFrontendClientLocal.cpp: (WebCore::InspectorFrontendClientLocal::frontendLoaded):
  • inspector/InspectorValues.cpp: (WebCore::InspectorObject::writeJSON):
  • inspector/front-end/inspector.js:
  • loader/EmptyClients.h:

2010-06-09 Sheriff Bot <webkit.review.bot@gmail.com>

Unreviewed, rolling out r60889.
http://trac.webkit.org/changeset/60889
https://bugs.webkit.org/show_bug.cgi?id=40365

gtk bot has some kind of memory corruption (Requested by
loislo on #webkit).

  • efl/WebCoreSupport/InspectorClientEfl.cpp:
  • efl/WebCoreSupport/InspectorClientEfl.h:

2010-06-09 Sheriff Bot <webkit.review.bot@gmail.com>

Unreviewed, rolling out r60889.
http://trac.webkit.org/changeset/60889
https://bugs.webkit.org/show_bug.cgi?id=40365

gtk bot has some kind of memory corruption (Requested by
loislo on #webkit).

  • WebCoreSupport/WebInspectorClientCF.cpp: (WebInspectorClient::storeSetting):

2010-06-09 Sheriff Bot <webkit.review.bot@gmail.com>

Unreviewed, rolling out r60889.
http://trac.webkit.org/changeset/60889
https://bugs.webkit.org/show_bug.cgi?id=40365

gtk bot has some kind of memory corruption (Requested by
loislo on #webkit).

  • src/InspectorClientImpl.cpp:
  • src/InspectorClientImpl.h:
  • src/WebDevToolsAgentImpl.cpp: (WebKit::WebDevToolsAgentImpl::setInspectorFrontendProxyToInspectorController):

2010-06-09 Sheriff Bot <webkit.review.bot@gmail.com>

Unreviewed, rolling out r60889.
http://trac.webkit.org/changeset/60889
https://bugs.webkit.org/show_bug.cgi?id=40365

gtk bot has some kind of memory corruption (Requested by
loislo on #webkit).

  • WebCoreSupport/InspectorClientGtk.cpp: (WebKit::InspectorClient::InspectorClient): (WebKit::InspectorClient::openInspectorFrontend): (WebKit::InspectorFrontendClient::InspectorFrontendClient): (WebKit::InspectorFrontendClient::destroyInspectorWindow):
  • WebCoreSupport/InspectorClientGtk.h:

2010-06-09 Sheriff Bot <webkit.review.bot@gmail.com>

Unreviewed, rolling out r60889.
http://trac.webkit.org/changeset/60889
https://bugs.webkit.org/show_bug.cgi?id=40365

gtk bot has some kind of memory corruption (Requested by
loislo on #webkit).

  • WebCoreSupport/InspectorClientHaiku.cpp:
  • WebCoreSupport/InspectorClientHaiku.h:

2010-06-09 Sheriff Bot <webkit.review.bot@gmail.com>

Unreviewed, rolling out r60889.
http://trac.webkit.org/changeset/60889
https://bugs.webkit.org/show_bug.cgi?id=40365

gtk bot has some kind of memory corruption (Requested by
loislo on #webkit).

  • WebCoreSupport/WebInspectorClient.h:
  • WebCoreSupport/WebInspectorClient.mm: (WebInspectorClient::WebInspectorClient): (WebInspectorClient::openInspectorFrontend): (-[WebInspectorWindowController destroyInspectorView]):

2010-06-09 Sheriff Bot <webkit.review.bot@gmail.com>

Unreviewed, rolling out r60889.
http://trac.webkit.org/changeset/60889
https://bugs.webkit.org/show_bug.cgi?id=40365

gtk bot has some kind of memory corruption (Requested by
loislo on #webkit).

  • WebCoreSupport/InspectorClientQt.cpp: (WebCore::InspectorClientQt::InspectorClientQt): (WebCore::InspectorClientQt::openInspectorFrontend): (WebCore::InspectorFrontendClientQt::InspectorFrontendClientQt): (WebCore::InspectorFrontendClientQt::closeWindow):
  • WebCoreSupport/InspectorClientQt.h:

2010-06-09 Sheriff Bot <webkit.review.bot@gmail.com>

Unreviewed, rolling out r60889.
http://trac.webkit.org/changeset/60889
https://bugs.webkit.org/show_bug.cgi?id=40365

gtk bot has some kind of memory corruption (Requested by
loislo on #webkit).

  • WebCoreSupport/WebInspectorClient.cpp: (WebInspectorClient::WebInspectorClient): (WebInspectorClient::~WebInspectorClient): (WebInspectorClient::openInspectorFrontend):
  • WebCoreSupport/WebInspectorClient.h: (WebInspectorClient::frontendClosing):

2010-06-09 Sheriff Bot <webkit.review.bot@gmail.com>

Unreviewed, rolling out r60889.
http://trac.webkit.org/changeset/60889
https://bugs.webkit.org/show_bug.cgi?id=40365

gtk bot has some kind of memory corruption (Requested by
loislo on #webkit).

  • WebKitSupport/InspectorClientWx.cpp:
  • WebKitSupport/InspectorClientWx.h:
08:51 Changeset [60894] by tkent@chromium.org

2010-06-09 Kent Tamura <tkent@chromium.org>

Reviewed by Eric Carlson.

[Windows] Build fix on Japanese Windows
https://bugs.webkit.org/show_bug.cgi?id=40358

cl.exe on Japanese Windows assumes files with invalid CP932
sequences broken and doesn't build them at all.

  • platform/graphics/win/MediaPlayerPrivateFullscreenWindow.cpp: Remove non-ASCII characters
  • platform/graphics/win/MediaPlayerPrivateFullscreenWindow.h: ditto.
  • platform/graphics/win/WKCAImageQueue.cpp: ditto.
  • platform/graphics/win/WKCAImageQueue.h: ditto.
08:00 Changeset [60893] by yurys@chromium.org

2010-06-09 Pavel Podivilov <podivilov@chromium.org>

Reviewed by Yury Semikhatsky.

Web Inspector: Add isDebuggerAlwaysEnabled method to ScriptDebugServer.
This method returns true if debugger should always be enabled when
frontend is attached.
https://bugs.webkit.org/show_bug.cgi?id=40289

  • bindings/js/ScriptDebugServer.cpp: (WebCore::ScriptDebugServer::isDebuggerAlwaysEnabled):
  • bindings/js/ScriptDebugServer.h:
  • bindings/v8/ScriptDebugServer.cpp: (WebCore::ScriptDebugServer::isDebuggerAlwaysEnabled):
  • bindings/v8/ScriptDebugServer.h:
  • inspector/InspectorController.cpp: (WebCore::InspectorController::setFrontend):

2010-06-09 Pavel Podivilov <podivilov@chromium.org>

Reviewed by Yury Semikhatsky.

Web Inspector: retreive debugger script source synchronously when
frontend is attached. We need this to be able to stop on top-level
breakpoints.
https://bugs.webkit.org/show_bug.cgi?id=40289

  • WebKit.gypi:
  • src/DebuggerAgent.h:
  • src/DebuggerAgentImpl.cpp:
  • src/DebuggerAgentImpl.h:
  • src/WebDevToolsAgentImpl.cpp: (WebKit::WebDevToolsAgentImpl::createInspectorFrontendProxy):
  • src/js/DebuggerScript.js:
  • src/js/DevTools.js: (WebInspector.loaded):
  • src/js/DevToolsHostStub.js:
07:49 Changeset [60892] by loislo@chromium.org

2010-06-09 Ilya Tikhonovsky <loislo@chromium.org>

Unreviewed build fix.

  • WebProcess/WebCoreSupport/WebInspectorClient.h:
07:40 Changeset [60891] by loislo@chromium.org

2010-06-09 Ilya Tikhonovsky <loislo@chromium.org>

Unreviewed build fix.

WebInspector: On the way to Remote Debugging we want to transfer dom/timeline/etc
data from inspected page to WebInspector as JSON string via http. The native
serialization to JSON string is supported by InspectorValue's classes. This patch
has the implementation of sendMessageToFrontend function. WebKit version of it still
uses ScriptFunctionCall and will be switched to another transport a little bit later.
https://bugs.webkit.org/show_bug.cgi?id=40134

  • WebProcess/WebCoreSupport/WebInspectorClient.cpp: (WebKit::WebInspectorClient::sendMessageToFrontend):
  • WebProcess/WebCoreSupport/WebInspectorClient.h:
07:33 Changeset [60890] by ossy@webkit.org

Unreviewed.

[Qt] canvas/philip/tests/2d.path.stroke.overlap.html fails with Qt-4.6.2 or higher
https://bugs.webkit.org/show_bug.cgi?id=40362

  • platform/qt/Skipped: canvas/philip/tests/2d.path.stroke.overlap.html skipped until fix.
07:18 Changeset [60889] by loislo@chromium.org

WebInspector: On the way to Remote Debugging we want to transfer dom/timeline/etc
data from inspected page to WebInspector as JSON string via http. The native
serialization to JSON string is supported by InspectorValue's classes. This patch
has the implementation of sendMessageToFrontend function. WebKit version of it still
uses ScriptFunctionCall and will be switched to another transport a little bit later.
https://bugs.webkit.org/show_bug.cgi?id=40134

  • inspector/InspectorClient.h:
  • inspector/InspectorController.cpp: (WebCore::InspectorController::connectFrontend): (WebCore::InspectorController::disconnectFrontend):
  • inspector/InspectorController.h:
  • inspector/InspectorFrontend.cpp: (WebCore::InspectorFrontend::InspectorFrontend):
  • inspector/InspectorFrontend.h:
  • inspector/InspectorFrontendClientLocal.cpp: (WebCore::InspectorFrontendClientLocal::frontendLoaded):
  • inspector/InspectorValues.cpp: (WebCore::InspectorObject::writeJSON):
  • inspector/front-end/inspector.js: (WebInspector.dispatchMessageToFrontend):
  • loader/EmptyClients.h: (WebCore::EmptyInspectorClient::sendMessageToFrontend):
  • efl/WebCoreSupport/InspectorClientEfl.cpp: (WebCore::InspectorClientEfl::sendMessageToFrontend):
  • efl/WebCoreSupport/InspectorClientEfl.h:
  • WebCoreSupport/WebInspectorClientCF.cpp: (WebInspectorClient::releaseFrontendPage): (WebInspectorClient::sendMessageToFrontend):
  • src/InspectorClientImpl.cpp: (WebKit::InspectorClientImpl::sendMessageToFrontend):
  • src/InspectorClientImpl.h:
  • src/WebDevToolsAgentImpl.cpp: (WebKit::WebDevToolsAgentImpl::setInspectorFrontendProxyToInspectorController):
  • WebCoreSupport/InspectorClientGtk.cpp: (WebKit::InspectorClient::releaseFrontendPage): (WebKit::InspectorClient::sendMessageToFrontend):
  • WebCoreSupport/InspectorClientGtk.h:
  • WebCoreSupport/InspectorClientHaiku.cpp: (WebCore::InspectorClientHaiku::sendMessageToFrontend):
  • WebCoreSupport/InspectorClientHaiku.h:
  • WebCoreSupport/WebInspectorClient.h:
  • WebCoreSupport/WebInspectorClient.mm: (WebInspectorClient::WebInspectorClient): (WebInspectorClient::openInspectorFrontend):
  • WebCoreSupport/InspectorClientQt.cpp: (WebCore::InspectorClientQt::InspectorClientQt): (WebCore::InspectorClientQt::openInspectorFrontend): (WebCore::InspectorClientQt::releaseFrontendPage): (WebCore::InspectorClientQt::sendMessageToFrontend):
  • WebCoreSupport/InspectorClientQt.h:
  • WebCoreSupport/WebInspectorClient.cpp: (WebInspectorClient::WebInspectorClient): (WebInspectorClient::~WebInspectorClient): (WebInspectorClient::openInspectorFrontend):
  • WebCoreSupport/WebInspectorClient.h:
  • WebKitSupport/InspectorClientWx.cpp: (WebCore::InspectorClientWx::sendMessageToFrontend):
  • WebKitSupport/InspectorClientWx.h:
07:09 Changeset [60888] by senorblanco@chromium.org

2010-06-09 Stephen White <senorblanco@chromium.org>

Unreviewed; expectations fix.

Add new expectations for layout tests post-r60885 ("Implement non-scaling stroke"). Mark one test flaky.

  • platform/chromium-linux/svg/custom/non-scaling-stroke-expected.checksum: Added.
  • platform/chromium-mac/fast/css/getComputedStyle/computed-style-expected.txt:
  • platform/chromium-mac/fast/css/getComputedStyle/computed-style-without-renderer-expected.txt:
  • platform/chromium-mac/svg/css/getComputedStyle-basic-expected.txt:
  • platform/chromium-win/fast/css/getComputedStyle/computed-style-expected.txt:
  • platform/chromium-win/fast/css/getComputedStyle/computed-style-without-renderer-expected.txt:
  • platform/chromium-win/svg/css/getComputedStyle-basic-expected.txt:
  • platform/chromium-win/svg/custom/non-scaling-stroke-expected.checksum: Added.
  • platform/chromium-win/svg/custom/non-scaling-stroke-expected.png: Added.
  • platform/chromium/test_expectations.txt:
03:24 Changeset [60887] by ossy@webkit.org

[Qt] Imperfect dependency for generated SVGNames.cpp
https://bugs.webkit.org/show_bug.cgi?id=40359

Reviewed by Dirk Schulze.

  • WebCore.pri: Missing dependency added.
01:57 Changeset [60886] by yurys@chromium.org

2010-06-09 Yury Semikhatsky <yurys@chromium.org>

Reviewed by Pavel Feldman.

Web Inspector: update stack trace after script source editing
https://bugs.webkit.org/show_bug.cgi?id=40357

  • src/js/DebuggerAgent.js: (devtools.DebuggerAgent.prototype.editScriptSource.this.requestSeqToCallback_.cmd.getSequenceNumber): (devtools.DebuggerAgent.prototype.editScriptSource.requestBacktrace): (devtools.DebuggerAgent.prototype.editScriptSource.handleBacktraceResponse): (devtools.DebuggerAgent.prototype.editScriptSource.reportDidCommitEditing): (devtools.DebuggerAgent.prototype.editScriptSource): (devtools.DebuggerAgent.prototype.requestBacktrace_): (devtools.DebuggerAgent.prototype.handleDebuggerOutput_): (devtools.DebuggerAgent.prototype.doHandleBacktraceResponse_): (devtools.DebuggerAgent.prototype.updateCallFramesFromBacktraceResponse_):
  • src/js/InspectorControllerImpl.js: (.devtools.InspectorBackendImpl.prototype.editScriptSource):
00:24 Changeset [60885] by krit@webkit.org

2010-06-08 Dirk Schulze <krit@webkit.org>

Reviewed by Nikolas Zimmermann.

Implement non-scaling-stroke (from SVG Tiny 1.2, also in Opera)
https://bugs.webkit.org/show_bug.cgi?id=31438


This is the implementation of the SVG property vector-effect according
to the specification of SVG Tiny 1.2.
getScreenCTM() was not able to calculate the transformation matrix in
comparison to the host coordinate system for objects, that get refereced
by a use element. This is fixed and covered by the the new test as well.
This patch is based upon a patch of Jeff Schiller.

Test: svg/custom/non-scaling-stroke.svg

  • css/CSSComputedStyleDeclaration.cpp: Added CSSPropertyVectorEffect. (WebCore::):
  • css/CSSPrimitiveValueMappings.h: (WebCore::CSSPrimitiveValue::CSSPrimitiveValue): (WebCore::CSSPrimitiveValue::operator EVectorEffect):
  • css/SVGCSSComputedStyleDeclaration.cpp: (WebCore::CSSComputedStyleDeclaration::getSVGPropertyCSSValue):
  • css/SVGCSSParser.cpp: (WebCore::CSSParser::parseSVGValue):
  • css/SVGCSSPropertyNames.in:
  • css/SVGCSSStyleSelector.cpp: (WebCore::CSSStyleSelector::applySVGProperty):
  • css/SVGCSSValueKeywords.in: Add vector-effect value non-scaling-stroke.
  • rendering/RenderPath.cpp: Revert transformations to ctm and transform path instead. (WebCore::fillAndStrokePath):
  • rendering/RenderSVGResourceContainer.h: (WebCore::RenderSVGResourceContainer::transformOnNonScalingStroke):
  • rendering/RenderSVGResourceGradient.cpp: Transform the gradient with the screenCTM. (WebCore::RenderSVGResourceGradient::applyResource):
  • rendering/RenderSVGResourcePattern.cpp: Transform the pattern with the screenCTM. (WebCore::RenderSVGResourcePattern::applyResource):
  • rendering/style/SVGRenderStyle.h: Added vector-effect property, not interited. (WebCore::SVGRenderStyle::NonInheritedFlags::): (WebCore::SVGRenderStyle::setBitDefaults):
  • rendering/style/SVGRenderStyleDefs.h: Added enums for vector-effect. (WebCore::):
  • svg/SVGLocatable.cpp: Calculate screenCTM for <use> referenced objects. (WebCore::SVGLocatable::computeCTM):
  • svg/SVGStyledElement.cpp: (WebCore::SVGStyledElement::cssPropertyIdForSVGAttributeName):
  • svg/svgattrs.in:

2010-06-08 Dirk Schulze <krit@webkit.org>

Reviewed by Nikolas Zimmermann.

Implement non-scaling-stroke (from SVG Tiny 1.2, also in Opera)
https://bugs.webkit.org/show_bug.cgi?id=31438


Test for SVG property vector-effect. The stroke-width should not get influenced by
any transformation on the object or its parrent.
Some ports needed platform specific test results for getComputedStyle.

  • platform/gtk/svg/css/getComputedStyle-basic-expected.txt:
  • platform/mac/fast/css/getComputedStyle/computed-style-expected.txt:
  • platform/mac/fast/css/getComputedStyle/computed-style-without-renderer-expected.txt:
  • platform/mac/svg/custom/non-scaling-stroke-expected.checksum: Added.
  • platform/mac/svg/custom/non-scaling-stroke-expected.png: Added.
  • platform/mac/svg/custom/non-scaling-stroke-expected.txt: Added.
  • platform/win/fast/css/getComputedStyle/computed-style-expected.txt:
  • platform/win/fast/css/getComputedStyle/computed-style-without-renderer-expected.txt:
  • svg/css/getComputedStyle-basic-expected.txt:
  • svg/custom/non-scaling-stroke.svg: Added.
00:22 Changeset [60884] by tony@chromium.org

2010-06-08 Tony Chang <tony@chromium.org>

Reviewed by David Levin.

[chromium] compile linux chromium in WebKit/out instead of WebKit/WebKit/chromium/out
https://bugs.webkit.org/show_bug.cgi?id=40285

Take 2 at landing this patch, this time rolling DEPS just a bit
further to pick up some chromium fixes.

  • DEPS
  • gyp_webkit:

2010-06-08 Tony Chang <tony@chromium.org>

Reviewed by David Levin.

[chromium] compile linux chromium in WebKit/out instead of WebKit/WebKit/chromium/out
https://bugs.webkit.org/show_bug.cgi?id=40285

  • Scripts/webkitdirs.pm:
  • Scripts/webkitpy/layout_tests/port/chromium_linux.py:

06/08/10:

23:29 Changeset [60883] by rolandsteiner@chromium.org

2010-06-02 Roland Steiner <rolandsteiner@chromium.org>

Reviewed by Tamura Kent

Bug 40052 - [DRT/Chromium] Upstream test_shell_webthemeengine as WebThemeEngineDRT
https://bugs.webkit.org/show_bug.cgi?id=40052

Add WebThemeEngineDRT and WebThemeControlDRT from Chromium rev. 48907

  • DumpRenderTree/DumpRenderTree.gypi:
  • DumpRenderTree/chromium/TestShellWin.cpp: (platformInit):
  • DumpRenderTree/chromium/WebThemeControlDRT.cpp: Added. (WebKit::): (WebKit::WebThemeControlDRT::WebThemeControlDRT): (WebKit::WebThemeControlDRT::~WebThemeControlDRT): (WebKit::WebThemeControlDRT::box): (WebKit::WebThemeControlDRT::line): (WebKit::WebThemeControlDRT::triangle): (WebKit::WebThemeControlDRT::roundRect): (WebKit::WebThemeControlDRT::oval): (WebKit::WebThemeControlDRT::circle): (WebKit::WebThemeControlDRT::nestedBoxes): (WebKit::WebThemeControlDRT::markState): (WebKit::WebThemeControlDRT::draw): (WebKit::WebThemeControlDRT::drawTextField): (WebKit::WebThemeControlDRT::drawProgressBar):
  • DumpRenderTree/chromium/WebThemeControlDRT.h: Added. (WebKit::WebThemeControlDRT::):
  • DumpRenderTree/chromium/WebThemeEngineDRT.cpp: Added. (WebKit::): (WebKit::WebThemeEngineDRT::paintButton): (WebKit::WebThemeEngineDRT::paintMenuList): (WebKit::WebThemeEngineDRT::paintScrollbarArrow): (WebKit::WebThemeEngineDRT::paintScrollbarThumb): (WebKit::WebThemeEngineDRT::paintScrollbarTrack): (WebKit::WebThemeEngineDRT::paintTextField): (WebKit::WebThemeEngineDRT::paintTrackbar): (WebKit::WebThemeEngineDRT::paintProgressBar):
  • DumpRenderTree/chromium/WebThemeEngineDRT.h: Added. (WebKit::WebThemeEngineDRT::WebThemeEngineDRT):
22:55 Changeset [60882] by mrowe@apple.com

<rdar://problem/8072136> REGRESSION (r56051): Inspect Element context menu does nothing in applications linked against 10.4 SDK

Reviewed by Adele Peterson.

WebCore:

Revert the change to ContextMenu::addInspectElementItem from r56051. It was made without
explanation and broke a reliance that WebKit has on the presence of a separator before the
Inspect Element menu item. This also restores the context menu item to the correct location
at the bottom of the context menu in applications built against the Mac OS X 10.4 SDK.

  • platform/ContextMenu.cpp:

(WebCore::ContextMenu::addInspectElementItem):

WebKit/mac:

  • WebCoreSupport/WebContextMenuClient.mm:

(fixMenusReceivedFromOldClients): Fix up the tag on the Inspect Element menu item. This ensures that even if the
context menu layout doesn't match our expectations that the menu item will continue to trigger the web inspector.

22:23 Changeset [60881] by rolandsteiner@chromium.org

2010-06-08 Roland Steiner <rolandsteiner@chromium.org>

Unreviewed build fix

Bug 38145 - Validate *tex* functions input parameters according to ES2 conformance
https://bugs.webkit.org/show_bug.cgi?id=38145
https://bugs.webkit.org/show_bug.cgi?id=40346

Fix for 38145 broke the build ('log2 not found' on Windows),
the fix from 40346 also didn't catch ('log ambiguous').

2nd quick fix: add explicit casts to double and use double constants.

No new tests.

  • html/canvas/WebGLRenderingContext.cpp: (WebCore::WebGLRenderingContext::validateTexFuncParameters):
21:44 Changeset [60880] by yuzo@google.com

2010-05-18 Yuzo Fujishima <yuzo@google.com>

Reviewed by Shinichiro Hamaji.

Fix for Bug 34529 - [CSSOM] issues with cssText and selectorText
Add U16_LENGTH that is needed to implement CSS character serialization.
https://bugs.webkit.org/show_bug.cgi?id=34529

  • wtf/unicode/qt4/UnicodeQt4.h:
  • wtf/unicode/wince/UnicodeWince.h:

2010-05-18 Yuzo Fujishima <yuzo@google.com>

Reviewed by Shinichiro Hamaji.

Fix for Bug 34529 - [CSSOM] issues with cssText and selectorText
https://bugs.webkit.org/show_bug.cgi?id=34529

  • fast/css/selector-text-escape-expected.txt: Added.
  • fast/css/selector-text-escape.html: Added.

2010-05-18 Yuzo Fujishima <yuzo@google.com>

Reviewed by Shinichiro Hamaji.

Fix for Bug 34529 - [CSSOM] issues with cssText and selectorText
Serialize selector as per http://dev.w3.org/csswg/cssom/#serializing-selectors
Without this patch, for example, selectors containing ':' or '.' are not properly serialized.
https://bugs.webkit.org/show_bug.cgi?id=34529

Test: fast/css/selector-text-escape.html

  • Android.mk:
  • CMakeLists.txt:
  • GNUmakefile.am:
  • WebCore.gypi:
  • WebCore.pro:
  • WebCore.vcproj/WebCore.vcproj:
  • WebCore.xcodeproj/project.pbxproj:
  • css/CSSOMUtils.cpp: Added. (WebCore::appendCharacter): (WebCore::serializeCharacter): (WebCore::serializeCharacterAsCodePoint): (WebCore::serializeIdentifier): (WebCore::serializeString):
  • css/CSSOMUtils.h: Added.
  • css/CSSSelector.cpp: (WebCore::CSSSelector::selectorText):
21:30 Changeset [60879] by tony@chromium.org

2010-06-08 Sheriff Bot <webkit.review.bot@gmail.com>

Unreviewed, rolling out r60878.
http://trac.webkit.org/changeset/60878
https://bugs.webkit.org/show_bug.cgi?id=40349

broke linux build (Requested by tonywork on #webkit).

  • DEPS:
  • gyp_webkit:

2010-06-08 Sheriff Bot <webkit.review.bot@gmail.com>

Unreviewed, rolling out r60878.
http://trac.webkit.org/changeset/60878
https://bugs.webkit.org/show_bug.cgi?id=40349

broke linux build (Requested by tonywork on #webkit).

  • Scripts/webkitdirs.pm:
  • Scripts/webkitpy/layout_tests/port/chromium_linux.py:
21:04 Changeset [60878] by tony@chromium.org

2010-06-08 Tony Chang <tony@chromium.org>

Reviewed by David Levin.

[chromium] compile linux chromium in WebKit/out instead of WebKit/WebKit/chromium/out
https://bugs.webkit.org/show_bug.cgi?id=40285

  • gyp_webkit:

2010-06-08 Tony Chang <tony@chromium.org>

Reviewed by David Levin.

[chromium] compile linux chromium in WebKit/out instead of WebKit/WebKit/chromium/out
https://bugs.webkit.org/show_bug.cgi?id=40285

  • Scripts/webkitdirs.pm:
  • Scripts/webkitpy/layout_tests/port/chromium_linux.py:
20:47 Changeset [60877] by rniwa@webkit.org

InsertListCommand needs cleanup
https://bugs.webkit.org/show_bug.cgi?id=36430

Reviewed by Justin Garcia

Separated code for listifying and unlistifying paragraphs.
Also isolated the dependency on the selection in doApply.

No test is added since this is a cleanup.

  • editing/InsertListCommand.cpp:

(WebCore::InsertListCommand::doApply): isolated code to listify / unlistify paragraphs
(WebCore::InsertListCommand::unlistifyParagraph): ditto
(WebCore::InsertListCommand::listifyParagraph): ditto

  • editing/InsertListCommand.h:
19:11 Changeset [60876] by kbr@google.com

2010-06-08 Kenneth Russell <kbr@google.com>

Unreviewed, build fix.

Fix build breakage from 38145
https://bugs.webkit.org/show_bug.cgi?id=40346

The fix for bug 38145 broke the Chromium Win build because
Microsoft's cmath doesn't define log2. Suggested fix by zmo is to
change log2(x) to log(x) / log(2). Built and ran WebGL layout
tests in Safari on Mac OS X. Changed download mirrors for
python-irclib to working ones.

  • html/canvas/WebGLRenderingContext.cpp: (WebCore::WebGLRenderingContext::validateTexFuncParameters):

2010-06-08 Kenneth Russell <kbr@google.com>

Unreviewed, build fix.

Fix build breakage from 38145
https://bugs.webkit.org/show_bug.cgi?id=40346

The fix for bug 38145 broke the Chromium Win build because
Microsoft's cmath doesn't define log2. Suggested fix by zmo is to
change log2(x) to log(x) / log(2). Built and ran WebGL layout
tests in Safari on Mac OS X. Changed download mirrors for
python-irclib to working ones.

  • Scripts/webkitpy/thirdparty/init.py:
18:02 Changeset [60875] by ukai@chromium.org

2010-06-08 Fumitoshi Ukai <ukai@chromium.org>

Reviewed by Eric Seidel.

old-run-webkit-tests should use ensure-valid-python to check if it can run the websockets tests
https://bugs.webkit.org/show_bug.cgi?id=39058

  • Scripts/old-run-webkit-tests: use sourceDir() to find ensure-valid-python in checkPythonVersion()
17:45 Changeset [60874] by eric@webkit.org

2010-06-08 Zhenyao Mo <zmo@google.com>

Reviewed by Dimitri Glazkov.

Validate *tex* functions input parameters according to ES2 conformance
https://bugs.webkit.org/show_bug.cgi?id=38145

  • fast/canvas/webgl/tex-image-and-sub-image-2d-with-image.html: Internalformat and format need to be the same.
  • fast/canvas/webgl/tex-input-validation-expected.txt: Added.
  • fast/canvas/webgl/tex-input-validation.html: Added.

2010-06-08 Zhenyao Mo <zmo@google.com>

Reviewed by Dimitri Glazkov.

Validate *tex* functions input parameters according to ES2 conformance
https://bugs.webkit.org/show_bug.cgi?id=38145

Test: fast/canvas/webgl/tex-input-validation.html

  • html/canvas/WebGLFramebuffer.cpp: (WebCore::WebGLFramebuffer::getColorBufferFormat): Get color buffer internalformat.
  • html/canvas/WebGLFramebuffer.h: Ditto.
  • html/canvas/WebGLRenderingContext.cpp: (WebCore::WebGLRenderingContext::WebGLRenderingContext): Deal with max tex units and max texture size. (WebCore::WebGLRenderingContext::copyTexImage2D): Validate input parameters. (WebCore::WebGLRenderingContext::copyTexSubImage2D): Ditto. (WebCore::WebGLRenderingContext::texImage2DBase): Ditto. (WebCore::WebGLRenderingContext::texParameterf): Ditto. (WebCore::WebGLRenderingContext::texParameteri): Ditto. (WebCore::WebGLRenderingContext::texSubImage2DBase): Ditto. (WebCore::WebGLRenderingContext::handleNPOTTextures): Deal with max tex units. (WebCore::WebGLRenderingContext::isTexInternalformatColorBufferCombinationValid): Check whether the texture format and framebuffer color buffer internalformat combination is valid. (WebCore::WebGLRenderingContext::getTextureBinding): get the current bound texture for given target. (WebCore::WebGLRenderingContext::validateTexFuncFormatAndType): Validate input format/type for *tex* functions. (WebCore::WebGLRenderingContext::validateTexFuncParameters): Validate input parameters for *tex* functions. (WebCore::WebGLRenderingContext::texParameter): Base function for texParameteri and texParameterf.
  • html/canvas/WebGLRenderingContext.h: Deal with max tex units.
  • html/canvas/WebGLTexture.cpp: (WebCore::WebGLTexture::WebGLTexture): Cache internalformat.
  • html/canvas/WebGLTexture.h: Ditto. (WebCore::WebGLTexture::setInternalformat): (WebCore::WebGLTexture::getInternalformat):
  • platform/graphics/GraphicsContext3D.cpp: internalformat and format need to be the same. (WebCore::GraphicsContext3D::extractImageData):
17:29 Changeset [60873] by eric@webkit.org

2010-06-08 Erik Arvidsson <arv@chromium.org>

Reviewed by Ojan Vafai.

REGRESSION: Weird focus behavior affects quoting on University of Washington message board system
https://bugs.webkit.org/show_bug.cgi?id=38548

  • editing/pasteboard/drop-link-expected.txt: Mouse down on an element where canStartSelection returns

false no longer clears the selection so we get one less
notification from the editing delegate.

  • editing/selection/click-in-focusable-link-should-not-clear-selection-expected.txt: Added.
  • editing/selection/click-in-focusable-link-should-not-clear-selection.html: Added.
  • editing/selection/script-tests/click-in-focusable-link-should-not-clear-selection.js: Added. (description.getElementCenter): (doubleClickOnElement): (mouseDownOnElement): (selectionShouldBe):

2010-06-08 Erik Arvidsson <arv@chromium.org>

Reviewed by Ojan Vafai.

REGRESSION: Weird focus behavior affects quoting on University of Washington message board system
https://bugs.webkit.org/show_bug.cgi?id=38548

Test: editing/selection/click-in-focusable-link-should-not-clear-selection.html

  • page/FocusController.cpp: (WebCore::clearSelectionIfNeeded): Make sure we do not clear selection when canStartSelection

returns false.

17:05 Changeset [60872] by tony@chromium.org

2010-06-08 Tony Chang <tony@chromium.org>

Not reviewed, build fix.

[chromium] fix linux compile
https://bugs.webkit.org/show_bug.cgi?id=40343

  • DEPS:
16:55 Changeset [60871] by andersca@apple.com

<rdar://problem/8071268> WebKit2 URLs are displayed as 1-character strings in log statements

Reviewed by John Sullivan.

Create a CFString from our WebCore string and then create the CFURL from the CFString.

  • UIProcess/API/C/cf/WKURLCF.cpp:

(WKURLCopyCFURL):

16:26 Changeset [60870] by senorblanco@chromium.org

2010-06-08 Stephen White <senorblanco@chromium.org>

Unreviewed, expectations fix.

[CHROMIUM] Remove expectations for editing/pasteboard/drop-link.html on mac, since the change was reverted. Mark http/tests/multipart/invalid-image-data.html as flaky on linux/dbg.

  • platform/chromium/test_expectations.txt:
16:09 Changeset [60869] by jchaffraix@webkit.org

[Qt] Crashfix on Path::contains
https://bugs.webkit.org/show_bug.cgi?id=40253

Patch by Enrico Ros <eros@codeaurora.org> on 2010-06-08
Reviewed by Ariya Hidayat.

Fix accessing the first item of an empty vector, that happens when
CanvasRenderingContext2D::isPointInPath is invoked over a borderless
path (e.g. a path made of just a MoveTo operation).

  • platform/graphics/qt/PathQt.cpp:

(WebCore::isPointOnPathBorder):

15:03 Changeset [60868] by senorblanco@chromium.org

2010-06-08 Sheriff Bot <webkit.review.bot@gmail.com>

Unreviewed, rolling out r60864.
http://trac.webkit.org/changeset/60864
https://bugs.webkit.org/show_bug.cgi?id=40337

Remove expected results, since r60859 was also reverted
(Requested by senorblanco on #webkit).

  • platform/chromium-win/editing/pasteboard/drop-link-expected.txt: Removed.
14:44 Changeset [60867] by eric@webkit.org

2010-06-08 Sheriff Bot <webkit.review.bot@gmail.com>

Unreviewed, rolling out r60859.
http://trac.webkit.org/changeset/60859
https://bugs.webkit.org/show_bug.cgi?id=40334

Included debugging statements. Caused a test to fail.
(Requested by ojan on #webkit).

  • editing/pasteboard/drop-link-expected.txt:
  • editing/pasteboard/drop-link.html:
  • editing/selection/click-in-focusable-link-should-not-clear-selection-expected.txt: Removed.
  • editing/selection/click-in-focusable-link-should-not-clear-selection.html: Removed.
  • editing/selection/script-tests/click-in-focusable-link-should-not-clear-selection.js: Removed.

2010-06-08 Sheriff Bot <webkit.review.bot@gmail.com>

Unreviewed, rolling out r60859.
http://trac.webkit.org/changeset/60859
https://bugs.webkit.org/show_bug.cgi?id=40334

Included debugging statements. Caused a test to fail.
(Requested by ojan on #webkit).

  • page/FocusController.cpp: (WebCore::clearSelectionIfNeeded):
14:27 Changeset [60866] by senorblanco@chromium.org

2010-06-08 Stephen White <senorblanco@chromium.org>

Unreviewed; expectations fix.

Update chromium expectations for r60865, r60859.

  • platform/chromium/test_expectations.txt:
13:48 Changeset [60865] by mrobinson@webkit.org

2010-06-08 Martin Robinson <mrobinson@igalia.com>

Reviewed by Ojan Vafai.

editingBehavior settings needs to be set back to a reasonable default between tests
https://bugs.webkit.org/show_bug.cgi?id=39433

This is a re-land of r59840 rolled out in r59857 due to DRT limitations.

  • editing/selection/extend-selection-after-double-click-expected.txt: Updated expectations.
  • editing/selection/extend-selection-after-double-click.html: Use the new DRT support for changing editing behavior.
13:28 Changeset [60864] by senorblanco@chromium.org

2010-06-08 Stephen White <senorblanco@chromium.org>

Unreviewed; expectations fix.

Added new chromium-win expectations for editing/pasteboard/drop-link.html, after r60859.

  • platform/chromium-win/editing/pasteboard/drop-link-expected.txt: Added.
13:00 Changeset [60863] by crogers@google.com

Take audio mixer's output gain into account

12:56 Changeset [60862] by krit@webkit.org

2010-06-08 Sheriff Bot <webkit.review.bot@gmail.com>

Unreviewed, rolling out r60858.
http://trac.webkit.org/changeset/60858
https://bugs.webkit.org/show_bug.cgi?id=40325

Qt needs a clean build. Rolling out the patch for now and fix
it tomorrow (Requested by krit on #webkit).

  • css/CSSComputedStyleDeclaration.cpp: (WebCore::):
  • css/CSSPrimitiveValueMappings.h:
  • css/SVGCSSComputedStyleDeclaration.cpp: (WebCore::CSSComputedStyleDeclaration::getSVGPropertyCSSValue):
  • css/SVGCSSParser.cpp: (WebCore::CSSParser::parseSVGValue):
  • css/SVGCSSPropertyNames.in:
  • css/SVGCSSStyleSelector.cpp: (WebCore::CSSStyleSelector::applySVGProperty):
  • css/SVGCSSValueKeywords.in:
  • rendering/RenderPath.cpp: (WebCore::fillAndStrokePath):
  • rendering/RenderSVGResourceContainer.h:
  • rendering/RenderSVGResourceGradient.cpp: (WebCore::RenderSVGResourceGradient::applyResource):
  • rendering/RenderSVGResourcePattern.cpp: (WebCore::RenderSVGResourcePattern::applyResource):
  • rendering/style/SVGRenderStyle.h: (WebCore::SVGRenderStyle::NonInheritedFlags::): (WebCore::SVGRenderStyle::setBitDefaults):
  • rendering/style/SVGRenderStyleDefs.h: (WebCore::):
  • svg/SVGLocatable.cpp: (WebCore::SVGLocatable::computeCTM):
  • svg/SVGStyledElement.cpp: (WebCore::SVGStyledElement::cssPropertyIdForSVGAttributeName):
  • svg/svgattrs.in:

2010-06-08 Sheriff Bot <webkit.review.bot@gmail.com>

Unreviewed, rolling out r60858.
http://trac.webkit.org/changeset/60858
https://bugs.webkit.org/show_bug.cgi?id=40325

Qt needs a clean build. Rolling out the patch for now and fix
it tomorrow (Requested by krit on #webkit).

  • platform/mac/fast/css/getComputedStyle/computed-style-expected.txt:
  • platform/mac/fast/css/getComputedStyle/computed-style-without-renderer-expected.txt:
  • platform/mac/svg/custom/non-scaling-stroke-expected.checksum: Removed.
  • platform/mac/svg/custom/non-scaling-stroke-expected.png: Removed.
  • platform/mac/svg/custom/non-scaling-stroke-expected.txt: Removed.
  • svg/css/getComputedStyle-basic-expected.txt:
  • svg/custom/non-scaling-stroke.svg: Removed.
12:49 Changeset [60861] by dbates@webkit.org

2010-06-08 Daniel Bates <dbates@rim.com>

Unreviewed, fix Xcode project file after change-set 60841.

Xcode is not happy about the entries for files EditingBehavior.h and
EditingBehaviorTypes.h that were added in change-set 60841 (Bug # 39854).
Lets make Xcode happy.

Also, lets place these entries in alphabetic order.

  • WebCore.xcodeproj/project.pbxproj:
12:44 Changeset [60860] by robert@webkit.org

2010-06-08 Robert Hogan <robert@webkit.org>

Rubber-stamped by Kenneth Rohde Christiansen.

[Qt] Put even more structure on Skipped file.

  • platform/qt/Skipped:
12:29 Changeset [60859] by arv@chromium.org

2010-06-08 Erik Arvidsson <arv@chromium.org>

Reviewed by ojan@chromium.org.

REGRESSION: Weird focus behavior affects quoting on University of Washington message board system
https://bugs.webkit.org/show_bug.cgi?id=38548

We should not clear the selection when canStartSelection returns false.

  • editing/pasteboard/drop-link-expected.txt: Mouse down on an element where canStartSelection returns

false no longer clears the selection so we get one less
notification from the editing delegate.

  • editing/selection/click-in-focusable-link-should-not-clear-selection-expected.txt: Added.
  • editing/selection/click-in-focusable-link-should-not-clear-selection.html: Added.
  • editing/selection/script-tests/click-in-focusable-link-should-not-clear-selection.js: Added. (description.getElementCenter): (doubleClickOnElement): (mouseDownOnElement): (selectionShouldBe):

2010-06-08 Erik Arvidsson <arv@chromium.org>

Reviewed by ojan@chromium.org.

REGRESSION: Weird focus behavior affects quoting on University of Washington message board system
https://bugs.webkit.org/show_bug.cgi?id=38548

We should not clear the selection when canStartSelection returns false.

Test: editing/selection/click-in-focusable-link-should-not-clear-selection.html

  • page/FocusController.cpp: (WebCore::clearSelectionIfNeeded): Make sure we do not clear selection when canStartSelection

returns false.

12:11 Changeset [60858] by krit@webkit.org

2010-06-08 Dirk Schulze <krit@webkit.org>

Reviewed by Nikolas Zimmermann.

Implement non-scaling-stroke (from SVG Tiny 1.2, also in Opera)
https://bugs.webkit.org/show_bug.cgi?id=31438


This is the implementation of the SVG property vector-effect according
to the specification of SVG Tiny 1.2.
getScreenCTM() was not able to calculate the transformation matrix in
comparison to the host coordinate system for objects, that get refereced
by a use element. This is fixed and covered by the the new test as well.
This patch is based upon a patch of Jeff Schiller.

Test: svg/custom/non-scaling-stroke.svg

  • css/CSSComputedStyleDeclaration.cpp: Added CSSPropertyVectorEffect. (WebCore::):
  • css/CSSPrimitiveValueMappings.h: (WebCore::CSSPrimitiveValue::CSSPrimitiveValue): (WebCore::CSSPrimitiveValue::operator EVectorEffect):
  • css/SVGCSSComputedStyleDeclaration.cpp: (WebCore::CSSComputedStyleDeclaration::getSVGPropertyCSSValue):
  • css/SVGCSSParser.cpp: (WebCore::CSSParser::parseSVGValue):
  • css/SVGCSSPropertyNames.in:
  • css/SVGCSSStyleSelector.cpp: (WebCore::CSSStyleSelector::applySVGProperty):
  • css/SVGCSSValueKeywords.in: Add vector-effect value non-scaling-stroke.
  • rendering/RenderPath.cpp: Revert transformations to ctm and transform path instead. (WebCore::fillAndStrokePath):
  • rendering/RenderSVGResourceContainer.h: (WebCore::RenderSVGResourceContainer::transformOnNonScalingStroke):
  • rendering/RenderSVGResourceGradient.cpp: Transform the gradient with the screenCTM. (WebCore::RenderSVGResourceGradient::applyResource):
  • rendering/RenderSVGResourcePattern.cpp: Transform the pattern with the screenCTM. (WebCore::RenderSVGResourcePattern::applyResource):
  • rendering/style/SVGRenderStyle.h: Added vector-effect property, not interited. (WebCore::SVGRenderStyle::NonInheritedFlags::): (WebCore::SVGRenderStyle::setBitDefaults):
  • rendering/style/SVGRenderStyleDefs.h: Added enums for vector-effect. (WebCore::):
  • svg/SVGLocatable.cpp: Calculate screenCTM for <use> referenced objects. (WebCore::SVGLocatable::computeCTM):
  • svg/SVGStyledElement.cpp: (WebCore::SVGStyledElement::cssPropertyIdForSVGAttributeName):
  • svg/svgattrs.in:

2010-06-08 Dirk Schulze <krit@webkit.org>

Reviewed by Nikolas Zimmermann.

Implement non-scaling-stroke (from SVG Tiny 1.2, also in Opera)
https://bugs.webkit.org/show_bug.cgi?id=31438


Test for SVG property vector-effect. The stroke-width should not get influenced by
any transformation on the object or its parrent.

  • platform/mac/fast/css/getComputedStyle/computed-style-expected.txt:
  • platform/mac/fast/css/getComputedStyle/computed-style-without-renderer-expected.txt:
  • platform/mac/svg/custom/non-scaling-stroke-expected.checksum: Added.
  • platform/mac/svg/custom/non-scaling-stroke-expected.png: Added.
  • platform/mac/svg/custom/non-scaling-stroke-expected.txt: Added.
  • svg/css/getComputedStyle-basic-expected.txt:
  • svg/custom/non-scaling-stroke.svg: Added.
11:41 Changeset [60857] by sullivan@apple.com
  • mac/WebKit2.exp:

Added _WKPageGetEstimatedProgress and _WKFrameGetPage

Rubber-stamped by Anders Carlsson.

11:16 Changeset [60856] by mrobinson@webkit.org

2010-06-08 Martin Robinson <mrobinson@igalia.com>

Not reviewed.

Adding baselines for new tests.

  • platform/gtk/fast/forms/slider-padding-expected.txt: Added.
  • platform/gtk/svg/custom/fill-opacity-hsl-expected.txt: Copied from LayoutTests/platform/mac/svg/custom/fill-opacity-hsl-expected.txt.
  • platform/gtk/svg/custom/fill-opacity-rgba-expected.txt: Copied from LayoutTests/platform/mac/svg/custom/fill-opacity-rgba-expected.txt.
  • platform/gtk/svg/filters/filter-on-filter-for-text-expected.txt: Added.
11:00 Changeset [60855] by ap@apple.com
  • GNUmakefile.am: Also adding platform/Cursor.cpp to Gtk build.
10:52 Changeset [60854] by ossy@webkit.org

[Qt] Unreviewed buildfix after r60849.

  • WebCore.pro: platform/Cursor.cpp is added to Qt build system.
10:52 Changeset [60853] by andersca@apple.com

2010-06-08 Anders Carlsson <andersca@apple.com>

Reviewed by John Sullivan.

Would like a way to query WKPageRef for the current progress value
https://bugs.webkit.org/show_bug.cgi?id=40310
<rdar://problem/8071299>

Add WKPageGetEstimatedProgress. Remove the progress parameter from the didChangeProgress
loader client callback function.

  • UIProcess/API/C/WKPage.cpp: (WKPageGetEstimatedProgress):
  • UIProcess/API/C/WKPage.h:
  • UIProcess/WebLoaderClient.cpp: (WebKit::WebLoaderClient::didChangeProgress):
  • UIProcess/WebLoaderClient.h:
  • UIProcess/WebPageProxy.cpp: (WebKit::WebPageProxy::WebPageProxy): (WebKit::WebPageProxy::close): (WebKit::WebPageProxy::didStartProgress): (WebKit::WebPageProxy::didChangeProgress): (WebKit::WebPageProxy::didFinishProgress): (WebKit::WebPageProxy::processDidExit):
  • UIProcess/WebPageProxy.h: (WebKit::WebPageProxy::estimatedProgress):
10:37 Changeset [60852] by eric@webkit.org

2010-06-08 Sheriff Bot <webkit.review.bot@gmail.com>

Unreviewed, rolling out r60830.
http://trac.webkit.org/changeset/60830
https://bugs.webkit.org/show_bug.cgi?id=40305

Broke the Windows build (Requested by abarth on #webkit).

  • JavaScriptCore.vcproj/WTF/WTF.vcproj:
  • wtf/OwnPtrCommon.h:
  • wtf/brew/OwnPtrBrew.h: Removed.
  • wtf/win/OwnPtrWin.h: Removed.
10:34 Changeset [60851] by kenneth@webkit.org

Unreviewed Buildbot fix.

Reset the Qt TextBreakIterator when reusing it.

  • platform/text/qt/TextBreakIteratorQt.cpp:

(WebCore::setUpIterator):

10:32 Changeset [60850] by andersca@apple.com

2010-06-08 Anders Carlsson <andersca@apple.com>

Reviewed by John Sullivan.

Would like a way to tell which WKPageRef a WKFrameRef is part of
https://bugs.webkit.org/show_bug.cgi?id=40308
<rdar://problem/8071251>

Add and implement WKFrameGetPage.

  • UIProcess/API/C/WKFrame.cpp: (WKFrameGetPage):
  • UIProcess/API/C/WKFrame.h:
  • UIProcess/WebFrameProxy.h: (WebKit::WebFrameProxy::page):
10:29 Changeset [60849] by ap@apple.com

Reviewed by John Sullivan.

https://bugs.webkit.org/show_bug.cgi?id=15779
<rdar://problem/8002964> Custom CSS .cur cursor anchor point is ignored

Test: manual-tests/cur-hotspot.html

Currently, this patch only has effect in Safari 5 on Windows. Other platforms will need
to implement getting hot spot information from .cur files.

  • WebCore.vcproj/WebCore.vcproj:
  • WebCore.xcodeproj/project.pbxproj: Added Cursor.cpp. Looks like TextWrangler fixed line endings in vcproj file.
  • css/CSSCursorImageValue.cpp: (WebCore::CSSCursorImageValue::CSSCursorImageValue): Renamed "hotspot" to "hotSpot" to match other code. (WebCore::CSSCursorImageValue::updateIfSVGCursorIsUsed): Ditto. Added a FIXME about possibly incorrect code for resolving differences with CSS specified hotspot.
  • css/CSSCursorImageValue.h: (WebCore::CSSCursorImageValue::create): (WebCore::CSSCursorImageValue::hotSpot): Renamed "hotspot" to "hotSpot" to match other code.
  • css/CSSParser.cpp: (WebCore::CSSParser::parseValue): Start with an invalid value for hot spot, so that we can differentiate between missing and (0, 0) CSS hot spots. Missing and invalid (-1, -1) are treated identically by Firefox, so we don't need to differentiate these.
  • css/CSSStyleSelector.cpp: (WebCore::CSSStyleSelector::applyProperty): Renamed "hotspot" to "hotSpot" to match other code.
  • page/EventHandler.cpp: (WebCore::EventHandler::selectCursor): Don't ignore cursors with invalid CSS hot spots. Firefox ignores only the hot spot, not the whole cursor in this case (and IE doesn't support CSS3 cursor declarations).
  • platform/Cursor.cpp: Added. (WebCore::determineHotSpot): Added a function that decides where hot spot is, provided an image and a possible out of band coordinate specification.
  • platform/Cursor.h: Added determineHotSpot().
  • platform/graphics/BitmapImage.cpp: (WebCore::BitmapImage::getHotSpot):
  • platform/graphics/BitmapImage.h:
  • platform/graphics/Image.h: (WebCore::Image::getHotSpot):
  • platform/graphics/ImageSource.cpp: (WebCore::ImageSource::getHotSpot):
  • platform/graphics/ImageSource.h: Added plumbing to get hot spot data from an image provider.
  • platform/graphics/cg/ImageSourceCG.cpp: (WebCore::ImageSource::getHotSpot): CG implemantation.
  • platform/mac/CursorMac.mm: (WebCore::createCustomCursor): Call determineHotSpot() to determine where hot spot should be. (WebCore::Cursor::Cursor): Renamed "hotspot" to "hotSpot" to match other code.
  • platform/win/CursorWin.cpp: (WebCore::Cursor::Cursor): Call determineHotSpot() to determine where hot spot should be.
  • platform/gtk/CursorGtk.cpp: (WebCore::Cursor::Cursor): Ditto. This must be done despite this platform not having an implementation for ImageSource::getHotSpot(), because we no longer check for out of bounds hot spots in EventHandler::selectCursor().
  • platform/qt/CursorQt.cpp: (WebCore::Cursor::Cursor): Ditto.
10:25 Changeset [60848] by sullivan@apple.com
  • mac/WebKit2.exp:

Added a few more symbols needed by Mac clients.

Rubber-stamped by Mark Rowe.

10:21 QtWebKitTriageRoster edited by laszlo.1.gombos@nokia.com
(diff)
10:04 Changeset [60847] by kenneth@webkit.org

[Qt] TextBreakIterator Qt performance
https://bugs.webkit.org/show_bug.cgi?id=39958

Reviewed by Antti Koivisto.

Rework TextBreakIteratorQt to be more in line with the ICU version.

We now reuse iterators where ever possible. The string data is compared
with memcmp, which should be faster than using a hash, as you need
to traverse the full buffer in the case the strings don't match,
where as the compare would fail quickly.

  • platform/text/qt/TextBreakIteratorQt.cpp:

(WebCore::TextBreakIterator::TextBreakIterator):
(WebCore::setUpIterator):
(WebCore::wordBreakIterator):
(WebCore::characterBreakIterator):
(WebCore::lineBreakIterator):
(WebCore::sentenceBreakIterator):

08:48 Changeset [60846] by abecsi@webkit.org

Unreviewed build fix after r60785.

[GTK] Add missing GtkVersioning.h include to fix the build
on older GTK+ versions.

No new tests needed.

  • platform/gtk/WidgetGtk.cpp:
08:24 Changeset [60845] by abecsi@webkit.org

Unreviewed buildfix after r60479.

[GTK] Move the declaration of getRootCoords higher up to fix the build on GTK
versions other than 2.17.3.

  • DumpRenderTree/gtk/EventSender.cpp:

(getRootCoords):

08:04 Changeset [60844] by senorblanco@chromium.org

2010-06-08 Stephen White <senorblanco@chromium.org>

Unreviewed; expectations fix.

Restore expectations for media/video-duration-known-after-eos.html, since it's still failing.

  • platform/chromium/test_expectations.txt:
07:52 Changeset [60843] by yurys@chromium.org

2010-06-08 Yury Semikhatsky <yurys@chromium.org>

Reviewed by Pavel Feldman.

Web Inspector: extend ScriptDebugServer to support script source editing
https://bugs.webkit.org/show_bug.cgi?id=40297

  • bindings/js/ScriptDebugServer.cpp: (WebCore::ScriptDebugServer::editScriptSource):
  • bindings/js/ScriptDebugServer.h:
  • bindings/v8/ScriptDebugServer.cpp: (WebCore::ScriptDebugServer::editScriptSource):
  • bindings/v8/ScriptDebugServer.h:
  • inspector/InspectorBackend.cpp: (WebCore::InspectorBackend::editScriptSource):
  • inspector/InspectorBackend.h:
  • inspector/InspectorBackend.idl:
  • inspector/InspectorController.cpp: (WebCore::InspectorController::InspectorController): (WebCore::InspectorController::disableDebugger): (WebCore::InspectorController::editScriptSource): (WebCore::InspectorController::currentCallFrames): (WebCore::InspectorController::didPause): (WebCore::InspectorController::didContinue):
  • inspector/InspectorController.h:
  • inspector/InspectorFrontend.cpp: (WebCore::InspectorFrontend::didEditScriptSource):
  • inspector/InspectorFrontend.h:
  • inspector/front-end/Script.js:
  • inspector/front-end/ScriptsPanel.js: (WebInspector.ScriptsPanel.prototype.canEditScripts): (WebInspector.ScriptsPanel.prototype.editScriptSource.mycallback): (WebInspector.ScriptsPanel.prototype.editScriptSource):
  • inspector/front-end/Settings.js:
  • inspector/front-end/SourceView.js: (WebInspector.SourceView.prototype._editLine):
  • inspector/front-end/TextViewer.js: (WebInspector.TextViewer.prototype._handleDoubleClick): (WebInspector.TextViewer.prototype._commitEditingLine):
  • inspector/front-end/inspector.css: (#scripts-files option.extension-script):

2010-06-08 Yury Semikhatsky <yurys@chromium.org>

Reviewed by Pavel Feldman.

Web Inspector: extend ScriptDebugServer to support script source editing
https://bugs.webkit.org/show_bug.cgi?id=40297

  • src/js/DebuggerAgent.js: (devtools.ScriptInfo.prototype.worldType):
  • src/js/DebuggerScript.js: (debuggerScriptConstructor.DebuggerScript.editScriptSource):
  • src/js/DevTools.js: (WebInspector.loaded):
  • src/js/InspectorControllerImpl.js: (devtools.InspectorBackendImpl):
07:42 Changeset [60842] by xan@webkit.org

2010-06-08 Xan Lopez <xlopez@igalia.com>

Reviewed by Gustavo Noronha.

[GTK] Unit testing for WebKitDOMNode hierarchy walk
https://bugs.webkit.org/show_bug.cgi?id=40171

Add DOM node test to the build system.

  • GNUmakefile.am:

WebKit/gtk:

2010-06-08 Xan Lopez <xlopez@igalia.com>

Reviewed by Gustavo Noronha.

[GTK] Unit testing for WebKitDOMNode hierarchy walk
https://bugs.webkit.org/show_bug.cgi?id=40171

Add some DOM node unit testing.

  • tests/testdomnode.c: Added. (finish_loading): (dom_node_fixture_setup): (dom_node_fixture_teardown): (test_dom_node_hierarchy_navigation): (main):
06:18 Changeset [60841] by tonikitoo@webkit.org

Refactor platform dependent editing behavior code out of Settings
https://bugs.webkit.org/show_bug.cgi?id=39854

[ Re-landing of r60816 with the m_editor initialized before m_selectionController in Frame constructor. ]

Reviewed by Ojan Vafai and Darin Adler.
Patch by Antonio Gomes <tonikitoo@webkit.org>

WebCore:

As per discussion in bug 36627, the following was suggested by Darin Adler: "before the addition of any
new EditingBehavior, I think we need to add functions to translate the overall behavior policy into specific
behaviors rather than directly saying 'behavior == Mac' or 'behavior == Windows'. Helper functions that
expression the different rules in plain language.".

Patch is a first step of an incremental approach to address this request. It does not yet replace all
direct use of Editing{Mac,Windows}Behaviors, but adds infra for that and replaces some of them.

No behavior change, so no new tests.

  • GNUmakefile.am:
  • WebCore.pro:
  • WebCore/WebCore.gypi
  • WebCore/WebCore.vcproj/WebCore.vcproj
  • WebCore/WebCore.xcodeproj/project.pbxproj
  • page/Settings.cpp:

(WebCore::Settings::Settings):

  • page/Settings.h: Moved out EditingBehavior enum to WebCore/editing/EditingBehaviorTypes.h,

and renamed to EditingBehaviorTypes.
(WebCore::Settings::setEditingBehaviorType): Renamed from setEditingBehavior.
(WebCore::Settings::editingBehaviorType): Renamed from editingBehavior.

  • editing/EditingBehavior.h: Added.

(WebCore::EditingBehavior::EditingBehavior): Class to work as a central point for
editing behavior that might have to be handled differently in the different platforms/ports.
Class should be not used or instantiated directly, but rather implicitly constructed
through the Editor class.
(WebCore::EditingBehavior::shouldMoveCaretToHorizontalBoundaryWhenPastTopOrBottom):
(WebCore::EditingBehavior::shouldConsiderSelectionAsDirectional):
(WebCore::EditingBehavior::shouldCenterAlignWhenSelectionIsRevealed):

  • editing/EditingBehaviorTypes.h: Added.

(WebCore::):

  • editing/Editor.cpp:

(WebCore::Editor::behavior): Getter for EditingBehavior class.

  • editing/Editor.h:
  • editing/EditorCommand.cpp:

(WebCore::executeToggleStyle):
(WebCore::SelectionController::setIsDirectional):
(WebCore::SelectionController::positionForPlatform):
(WebCore::SelectionController::modify):

  • page/EventHandler.cpp:

(WebCore::EventHandler::handleMousePressEventSingleClick):

  • rendering/RenderBlock.cpp:

(WebCore::RenderBlock::positionForPointWithInlineChildren):

  • Call sites will look like:

if (frame->editor()->behavior().shouldThisIfThat()))

do something

WebKit:

EditingBehavior enum was renamed to EditingBehaviorTypes and moved out from Settings.h to
EditingBehaviorTypes.h . Call sites in WebKit/ adjusted accordingly.

WebKit/chromium:

  • src/WebSettingsImpl.cpp:

(WebKit::WebSettingsImpl::setEditingBehavior):

WebKit/gtk:

  • webkit/webkitprivate.cpp:

(WebKit::core):

  • webkit/webkitprivate.h:
  • webkit/webkitwebview.cpp:

(webkit_web_view_update_settings):
(webkit_web_view_settings_notify):

WebKit/mac:

  • WebView/WebFrame.mm:

(core):

  • WebView/WebFrameInternal.h:
  • WebView/WebView.mm:

(-[WebView _preferencesChangedNotification:]):

WebKit/qt:

  • WebCoreSupport/DumpRenderTreeSupportQt.cpp:

(DumpRenderTreeSupportQt::setEditingBehavior):

WebKit/win:

  • WebView.cpp:

(WebView::notifyPreferencesChanged):

05:36 QtWebKit edited by jarkko.j.sakkinen@gmail.com
(diff)
05:29 Changeset [60840] by steveblock@google.com

Prevent Geolocation making callbacks to a ScriptExecutionContext that no longer exists
https://bugs.webkit.org/show_bug.cgi?id=40162

Reviewed by Jeremy Orlow.

WebCore:

Before making callbacks, we check that the relevant ScriptExecutionContext still exists.
To achieve this, the callbacks inherit from ActiveDOMObject.

The ScriptExecutionContext is ref'ed from script, so may not be GC'ed for some time after
it is disconnected from its frame. Making the callback currently involves accessing the
Frame, so an additional check for the Frame is required.

This change also prevents the V8 bindings from incorrectly holding a reference to the Frame.

Test: fast/dom/Geolocation/callback-to-deleted-context.html

  • bindings/js/JSCallbackData.cpp:

(WebCore::JSCallbackData::invokeCallback):

  • bindings/js/JSCustomPositionCallback.cpp:

(WebCore::JSCustomPositionCallback::JSCustomPositionCallback):
(WebCore::JSCustomPositionCallback::handleEvent):

  • bindings/js/JSCustomPositionErrorCallback.cpp:

(WebCore::JSCustomPositionErrorCallback::JSCustomPositionErrorCallback):
(WebCore::JSCustomPositionErrorCallback::handleEvent):

  • bindings/v8/custom/V8CustomPositionCallback.cpp:

(WebCore::V8CustomPositionCallback::V8CustomPositionCallback):
(WebCore::V8CustomPositionCallback::handleEvent):

  • bindings/v8/custom/V8CustomPositionCallback.h:

(WebCore::V8CustomPositionCallback::create):

  • bindings/v8/custom/V8CustomPositionErrorCallback.cpp:

(WebCore::V8CustomPositionErrorCallback::V8CustomPositionErrorCallback):
(WebCore::V8CustomPositionErrorCallback::handleEvent):

  • bindings/v8/custom/V8CustomPositionErrorCallback.h:

(WebCore::V8CustomPositionErrorCallback::create):

  • bindings/v8/custom/V8GeolocationCustom.cpp:

(WebCore::createPositionCallback):
(WebCore::createPositionErrorCallback):

  • page/PositionCallback.h:

(WebCore::PositionCallback::PositionCallback):

  • page/PositionErrorCallback.h:

(WebCore::PositionErrorCallback::PositionErrorCallback):

LayoutTests:

  • fast/dom/Geolocation/callback-to-deleted-context-expected.txt: Added.
  • fast/dom/Geolocation/callback-to-deleted-context.html: Added.
  • fast/dom/Geolocation/resources/callback-to-deleted-context-inner1.html: Added.
  • fast/dom/Geolocation/resources/callback-to-deleted-context-inner2.html: Added.
  • fast/dom/Geolocation/script-tests/callback-to-deleted-context.js: Added.
  • platform/gtk/Skipped: Modified.
05:17 Changeset [60839] by steveblock@google.com

Geolocation should have a LayoutTest for reentrant calls that cause a PERMISSION_DENIED error
https://bugs.webkit.org/show_bug.cgi?id=40174

Reviewed by Alexey Proskuryakov.

Tests for reentrant calls with success and POSITION_UNAVAILABLE responses already exist,
but the code path for PERMISSION_DENIED is different.

  • fast/dom/Geolocation/reentrant-permission-denied-expected.txt: Added.
  • fast/dom/Geolocation/reentrant-permission-denied.html: Added.
  • fast/dom/Geolocation/script-tests/reentrant-permission-denied.js: Added.
  • platform/gtk/Skipped: Modified.
05:05 Changeset [60838] by bulach@chromium.org

Unreviewed: adding bulach@chromium.org to commiters list.

04:39 Changeset [60837] by hamaji@chromium.org

2010-06-08 Shinichiro Hamaji <hamaji@chromium.org>

Unreviewed. fast/js/large-expressions is failing only in debug mode.

  • platform/chromium/test_expectations.txt:
04:32 Changeset [60836] by hamaji@chromium.org

2010-06-08 Shinichiro Hamaji <hamaji@chromium.org>

Unreviewed. Remove passing tests from expectations.txt.

  • platform/chromium/test_expectations.txt:
04:29 Changeset [60835] by xan@webkit.org

2010-06-08 Xan Lopez <xlopez@igalia.com>

Reviewed by Gustavo Noronha.

[GTK] Avoid duplicated signals in DOM bindings
https://bugs.webkit.org/show_bug.cgi?id=40170

Do not create duplicated event signals between a class and its
ancestor.

  • bindings/scripts/CodeGeneratorGObject.pm:
04:25 Changeset [60834] by xan@webkit.org

2010-06-08 Xan Lopez <xlopez@igalia.com>

Reviewed by Gustavo Noronha.

[GTK] Add inner-node property to WebKitHitTestResult
https://bugs.webkit.org/show_bug.cgi?id=40131

Add a 'inner-node' property to WebKitHitTestResult, carrying the
DOM node where the hit test happened.

  • tests/testhittestresult.c: (load_status_cb):
  • webkit/webkithittestresult.cpp: (webkit_hit_test_result_get_property): (webkit_hit_test_result_set_property): (webkit_hit_test_result_class_init):
  • webkit/webkitprivate.cpp: (WebKit::kit):
03:58 Changeset [60833] by tkent@chromium.org

Unreviewed, build box fix.
Touch DOMWindow.idl to reflect ENABLE_METER_TAG change on Mac.

03:52 Changeset [60832] by morrita@google.com

2010-06-08 MORITA Hajime <morrita@google.com>

Unreviewed. An attempt to fix test break.

  • Configurations/FeatureDefines.xcconfig:

2010-06-08 MORITA Hajime <morrita@google.com>

Unreviewed. An attempt to fix test break.

  • Configurations/FeatureDefines.xcconfig:
02:54 Changeset [60831] by morrita@google.com

2010-06-08 MORITA Hajime <morrita@google.com>

Unreviewd. Update expectations to follow r60820, r60821

  • fast/dom/Window/window-properties-expected.txt:
  • fast/dom/Window/window-property-descriptors-expected.txt:
  • fast/dom/prototype-inheritance-2-expected.txt:
  • fast/dom/prototype-inheritance-expected.txt:
  • fast/js/global-constructors-expected.txt:

2010-06-08 MORITA Hajime <morrita@google.com>

Unreviewd. Touched the file to force rebuild.

  • html/HTMLMeterElement.idl:
01:35 Changeset [60830] by eric@webkit.org

2010-06-08 Kwang Yul Seo <skyul@company100.net>

Reviewed by Adam Barth.

Change OwnPtrCommon to include platform-specific headers
https://bugs.webkit.org/show_bug.cgi?id=40279

Adding new type to OwnPtrCommon needlessly causes all ports to do full rebuilds.
Change OwnPtrCommon to include platform-specific headers to avoid all ports rebuilds.

  • JavaScriptCore.vcproj/WTF/WTF.vcproj:
  • wtf/OwnPtrCommon.h:
  • wtf/brew/OwnPtrBrew.h: Added.
  • wtf/win/OwnPtrWin.h: Added.
01:32 QtWebKitWebGL edited by jarkko.j.sakkinen@gmail.com
(diff)
01:30 QtWebKitWebGL edited by jarkko.j.sakkinen@gmail.com
(diff)
01:29 QtWebKitWebGL edited by jarkko.j.sakkinen@gmail.com
(diff)
01:28 QtWebKit edited by jarkko.j.sakkinen@gmail.com
(diff)
01:26 QtWebKitWebGL created by jarkko.j.sakkinen@gmail.com
01:16 Changeset [60829] by morrita@google.com

2010-06-08 MORITA Hajime <morrita@google.com>

Unreviewd. Updated test_expectations.txt to follow r60822

  • platform/chromium/test_expectations.txt:
01:08 Changeset [60828] by eric@webkit.org

2010-06-08 Vangelis Kokkevis <vangelis@chromium.org>

Reviewed by Dimitri Glazkov.

[Chromium] Fix the math for transform matrices of composited layers. There
were problems both with how content layers were setting their position and
how the overal matrix stack was computed when the layer anchor isn't at the
center of the layer.
https://bugs.webkit.org/post_bug.cgi

  • platform/graphics/chromium/GraphicsLayerChromium.cpp: (WebCore::GraphicsLayerChromium::updateContentsRect):
  • platform/graphics/chromium/LayerRendererChromium.cpp: (WebCore::LayerRendererChromium::updateLayersRecursive): (WebCore::LayerRendererChromium::drawLayer):
00:20 Changeset [60827] by eric@webkit.org

2010-06-07 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

HTML5Lexer does not handle <div FOO ><img><img></div> correctly
https://bugs.webkit.org/show_bug.cgi?id=40283

  • html5lib/resources/webkit01.dat:
    • Add another test for this failure case.

2010-06-07 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

HTML5Lexer does not handle <div FOO ><img><img></div> correctly
https://bugs.webkit.org/show_bug.cgi?id=40283

Fix a typo in the AfterAttributeNameState.
The compiler really should have caught this since the typo
resulted in unreachable code.

Tests:

10 tables/mozilla/marvin/colgroup* tests.
Also added a new sub-test in html5lib/resources/webkit01.dat

  • html/HTML5Lexer.cpp: (WebCore::HTML5Lexer::nextToken):
    • Fix typo of = instead of >
00:15 Changeset [60826] by abarth@webkit.org

2010-06-08 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

Implement HTML5's forceQuirks flag
https://bugs.webkit.org/show_bug.cgi?id=40284

I didn't implement this before because I didn't know how to test it,
but now I do.


Tests:

  • fast/doctypes/doctype-parsing.html.

When we actually implement the quirks mode handling in the tree
builder, we'll probably want to add to this test.

  • html/HTML5Lexer.cpp: (WebCore::HTML5Lexer::nextToken):
  • html/HTML5Token.h: (WebCore::HTML5Token::forceQuirks): (WebCore::HTML5Token::setForceQuirks):
  • html/HTML5TreeBuilder.cpp: (WebCore::HTML5TreeBuilder::passTokenToLegacyParser):
  • html/HTMLParser.cpp: (WebCore::HTMLParser::parseDoctypeToken):
  • html/HTMLTokenizer.h: (WebCore::DoctypeToken::reset):
00:10 Changeset [60825] by abarth@webkit.org

2010-06-08 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

Remove ASSERT in ~HTML5Tokenizer
https://bugs.webkit.org/show_bug.cgi?id=40282

We thought this ASSERT would be a good idea, but the problem is that
this object is deleted without warning in some circumstances, so
there's no way to know whether it has reached the end of its state
machine.

  • html/HTML5Tokenizer.cpp: (WebCore::HTML5Tokenizer::~HTML5Tokenizer):
00:08 Changeset [60824] by abarth@webkit.org

2010-06-08 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

Remove another unterminated <title> element
https://bugs.webkit.org/show_bug.cgi?id=40281

We're already testing untermianted <title> element behavior in specific
tests for that beheavior. We don't need to test it redudantly here.

  • http/tests/security/frame-loading-via-document-write.html:

06/07/10:

23:36 Changeset [60823] by tony@chromium.org

2010-06-07 Tony Chang <tony@chromium.org>

Reviewed by Kent Tamura.

[chromium] roll chromium DEPS
https://bugs.webkit.org/show_bug.cgi?id=40277

  • DEPS:
23:26 Changeset [60822] by morrita@google.com

2010-06-07 MORITA Hajime <morrita@google.com>

Reviewed by Kent Tamura.

[Mac] <meter> elements should be rendered as level indicators.
https://bugs.webkit.org/show_bug.cgi?id=40217

  • fast/dom/HTMLMeterElement/meter-appearances-capacity.html: Added.
  • fast/dom/HTMLMeterElement/meter-appearances-rating-relevancy.html: Added.
  • fast/dom/HTMLMeterElement/meter-optimums.html: Added.
  • platform/chromium/test_expectations.txt:
  • platform/gtk/Skipped:
  • platform/mac/Skipped: removed <meter> related tests.
  • platform/mac/fast/dom/HTMLMeterElement/meter-appearances-capacity-expected.checksum: Added.
  • platform/mac/fast/dom/HTMLMeterElement/meter-appearances-capacity-expected.png: Added.
  • platform/mac/fast/dom/HTMLMeterElement/meter-appearances-capacity-expected.txt: Added.
  • platform/mac/fast/dom/HTMLMeterElement/meter-appearances-rating-relevancy-expected.checksum: Added.
  • platform/mac/fast/dom/HTMLMeterElement/meter-appearances-rating-relevancy-expected.png: Added.
  • platform/mac/fast/dom/HTMLMeterElement/meter-appearances-rating-relevancy-expected.txt: Added.
  • platform/mac/fast/dom/HTMLMeterElement/meter-element-expected.checksum: Added.
  • platform/mac/fast/dom/HTMLMeterElement/meter-element-expected.png: Added.
  • platform/mac/fast/dom/HTMLMeterElement/meter-element-expected.txt: Copied from LayoutTests/platform/qt/fast/dom/HTMLMeterElement/meter-element-expected.txt.
  • platform/mac/fast/dom/HTMLMeterElement/meter-optimums-expected.checksum: Added.
  • platform/mac/fast/dom/HTMLMeterElement/meter-optimums-expected.png: Added.
  • platform/mac/fast/dom/HTMLMeterElement/meter-optimums-expected.txt: Added.
  • platform/win/Skipped:

2010-06-07 MORITA Hajime <morrita@google.com>

Reviewed by Kent Tamura.

[Mac] <meter> elements should be rendered as level indicators.
https://bugs.webkit.org/show_bug.cgi?id=40217

Implemented RenderThemeMac::paintMeter() using NSLevelIndicatorCell.

This change also added new -webkit-appearance values to select the
style of level indicators:

  • relevancy-level-indicator,
  • continuous-capacity-level-indicator,
  • discrete-capacity-level-indicator, and
  • rating-level-indicator.

The size of elements are adjusted based on the bounds of the indicator
to make room for painting whole indicator cells.

Tests: fast/dom/HTMLMeterElement/meter-appearances-capacity.html

fast/dom/HTMLMeterElement/meter-appearances-rating-relevancy.html
fast/dom/HTMLMeterElement/meter-optimums.html

  • css/CSSPrimitiveValueMappings.h: (WebCore::CSSPrimitiveValue::CSSPrimitiveValue):
  • css/CSSValueKeywords.in:
  • html/HTMLMeterElement.cpp: (WebCore::HTMLMeterElement::gaugeRegion):
  • html/HTMLMeterElement.h: (WebCore::HTMLMeterElement::):
  • platform/ThemeTypes.h: (WebCore::):
  • rendering/RenderMeter.cpp: (WebCore::RenderMeter::calcWidth): (WebCore::RenderMeter::calcHeight):
  • rendering/RenderMeter.h:
  • rendering/RenderTheme.cpp: (WebCore::RenderTheme::adjustStyle): (WebCore::RenderTheme::paint): (WebCore::RenderTheme::paintBorderOnly): (WebCore::RenderTheme::paintDecorations): (WebCore::RenderTheme::meterSizeForBounds):
  • rendering/RenderTheme.h:
  • rendering/RenderThemeMac.h:
  • rendering/RenderThemeMac.mm: (WebCore::RenderThemeMac::meterSizeForBounds): (WebCore::RenderThemeMac::paintMeter): (WebCore::RenderThemeMac::levelIndicatorStyleFor): (WebCore::RenderThemeMac::levelIndicatorFor):
22:23 Changeset [60821] by morrita@google.com

2010-06-06 MORITA Hajime <morrita@google.com>

Unreviewd, follow up to r60820

https://bugs.webkit.org/show_bug.cgi?id=40219
[Mac] ENABLE_METER_TAG should be enabled

Added ENABLE_METER_TAG.

  • Configurations/FeatureDefines.xcconfig:

2010-06-06 MORITA Hajime <morrita@google.com>

Unreviewd, follow up to r60820

https://bugs.webkit.org/show_bug.cgi?id=40219
[Mac] ENABLE_METER_TAG should be enabled

  • Configurations/FeatureDefines.xcconfig:
22:02 Changeset [60820] by morrita@google.com

2010-06-07 MORITA Hajime <morrita@google.com>

Reviewed by Kent Tamura.

https://bugs.webkit.org/show_bug.cgi?id=40219
[Mac] ENABLE_METER_TAG should be enabled

Added ENABLE_METER_TAG.

  • Configurations/FeatureDefines.xcconfig:

2010-06-07 MORITA Hajime <morrita@google.com>

Reviewed by Kent Tamura.

[Mac] ENABLE_METER_TAG should be enabled.
https://bugs.webkit.org/show_bug.cgi?id=40219

  • Configurations/FeatureDefines.xcconfig:
  • WebCore.xcodeproj/project.pbxproj:
21:09 Changeset [60819] by tonikitoo@webkit.org

Revert "Refactor platform dependent editing behavior code out of Settings"

This reverts commit r60816: Crashed all bots.

21:08 Changeset [60818] by tonikitoo@webkit.org

Revert "2010-06-07 Antonio Gomes <tonikitoo@webkit.org>"

This reverts r60817: Unsuccessful attept to fix the bots.

21:00 Changeset [60817] by tonikitoo@webkit.org

2010-06-07 Antonio Gomes <tonikitoo@webkit.org>

Unreviewed attempt to fix crashy bots (Mac,Win,Qt,Gtk) after r60816.

  • editing/Editor.cpp: (WebCore::Editor::behavior):
20:08 Changeset [60816] by tonikitoo@webkit.org

Refactor platform dependent editing behavior code out of Settings
https://bugs.webkit.org/show_bug.cgi?id=39854

Reviewed by Ojan Vafai and Darin Adler.
Patch by Antonio Gomes <tonikitoo@webkit.org>

WebCore:

As per discussion in bug 36627, the following was suggested by Darin Adler: "before the addition of any
new EditingBehavior, I think we need to add functions to translate the overall behavior policy into specific
behaviors rather than directly saying 'behavior == Mac' or 'behavior == Windows'. Helper functions that
expression the different rules in plain language.".

Patch is a first step of an incremental approach to address this request. It does not yet replace all
direct use of Editing{Mac,Windows}Behaviors, but adds infra for that and replaces some of them.

No behavior change, so no new tests.

  • GNUmakefile.am:
  • WebCore.pro:
  • WebCore/WebCore.gypi
  • WebCore/WebCore.vcproj/WebCore.vcproj
  • WebCore/WebCore.xcodeproj/project.pbxproj
  • page/Settings.cpp:

(WebCore::Settings::Settings):

  • page/Settings.h: Moved out EditingBehavior enum to WebCore/editing/EditingBehaviorTypes.h,

and renamed to EditingBehaviorTypes.
(WebCore::Settings::setEditingBehaviorType): Renamed from setEditingBehavior.
(WebCore::Settings::editingBehaviorType): Renamed from editingBehavior.

  • editing/EditingBehavior.h: Added.

(WebCore::EditingBehavior::EditingBehavior): Class to work as a central point for
editing behavior that might have to be handled differently in the different platforms/ports.
Class should be not used or instantiated directly, but rather implicitly constructed
through the Editor class.
(WebCore::EditingBehavior::shouldMoveCaretToHorizontalBoundaryWhenPastTopOrBottom):
(WebCore::EditingBehavior::shouldConsiderSelectionAsDirectional):
(WebCore::EditingBehavior::shouldCenterAlignWhenSelectionIsRevealed):

  • editing/EditingBehaviorTypes.h: Added.

(WebCore::):

  • editing/Editor.cpp:

(WebCore::Editor::behavior): Getter for EditingBehavior class.

  • editing/Editor.h:
  • editing/EditorCommand.cpp:

(WebCore::executeToggleStyle):
(WebCore::SelectionController::setIsDirectional):
(WebCore::SelectionController::positionForPlatform):
(WebCore::SelectionController::modify):

  • page/EventHandler.cpp:

(WebCore::EventHandler::handleMousePressEventSingleClick):

  • rendering/RenderBlock.cpp:

(WebCore::RenderBlock::positionForPointWithInlineChildren):

  • Call sites will look like:

if (frame->editor()->behavior().shouldThisIfThat()))

do something

WebKit:

EditingBehavior enum was renamed to EditingBehaviorTypes and moved out from Settings.h to
EditingBehaviorTypes.h . Call sites in WebKit/ adjusted accordingly.

WebKit/chromium:

  • src/WebSettingsImpl.cpp:

(WebKit::WebSettingsImpl::setEditingBehavior):

WebKit/gtk:

  • webkit/webkitprivate.cpp:

(WebKit::core):

  • webkit/webkitprivate.h:
  • webkit/webkitwebview.cpp:

(webkit_web_view_update_settings):
(webkit_web_view_settings_notify):

WebKit/mac:

  • WebView/WebFrame.mm:

(core):

  • WebView/WebFrameInternal.h:
  • WebView/WebView.mm:

(-[WebView _preferencesChangedNotification:]):

WebKit/qt:

  • WebCoreSupport/DumpRenderTreeSupportQt.cpp:

(DumpRenderTreeSupportQt::setEditingBehavior):

WebKit/win:

  • WebView.cpp:

(WebView::notifyPreferencesChanged):

19:33 Changeset [60815] by eric@webkit.org

2010-06-07 Eric Seidel <eric@webkit.org>

Unreviewed. Just commit an edit XCode keeps making automatically.

It appears someone hand-edited the file, XCode no-like.

  • WebCore.xcodeproj/project.pbxproj:
19:25 Changeset [60814] by eric@webkit.org

2010-06-07 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

HTML5 Parser fails script-tests which use document.write
https://bugs.webkit.org/show_bug.cgi?id=40276

We were leaving the TreeBuilder paused when executing the scripts
resulting in document.write() calls being ignored.

I don't see a good way to ASSERT this correct behavior, since
the HTML5ScriptRunner and the HTML5TreeBuilder do not know about
each other. We should never have document.write() called while
the HTML5TreeBuilder is paused, however both document.write and
the network call the same HTML5Tokenizer::write method. It's OK
to be paused when the network writes, but during document.write
(or any other script execution).

Tested by all of fast/url and numerous other tests.

  • html/HTML5Tokenizer.cpp: (WebCore::HTML5Tokenizer::notifyFinished):
    • Unpause the treebuilder before executing scripts.

(WebCore::HTML5Tokenizer::executeScriptsWaitingForStylesheets):

  • Unpause the treebuilder before executing scripts.
17:57 Changeset [60813] by eric@webkit.org

2010-06-07 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

HTML5 Parser hits ASSERT in fast/events/stop-load-in-unload-handler-using-document-write.html
https://bugs.webkit.org/show_bug.cgi?id=40268

Teach HTML5Tokenizer about m_parserStopped.

While tracking down how m_parserStopped was used, I found
several useless implementations of stopParsing() which I removed.

I also found a comment in HTMLTokenizer which still talks about the
"part" (an old class now split into Frame, Page, FrameLoader, etc.)
and an nearly empty FrameLoader::tokenizerProcessedData which I just
inlined into its one caller.

Tests:

fast/events/stop-load-in-unload-handler-using-document-write.html
http/tests/security/xssAuditor/full-block-base-href.html

  • html/HTML5Tokenizer.cpp: (WebCore::HTML5Tokenizer::pumpLexerIfPossible):
    • New method used instead of direct calls to pumpLexer for when callers would like to pump the lexer if possible, but don't want to check if the parser is stopped, etc.

(WebCore::HTML5Tokenizer::pumpLexer):

  • Added an ASSERT that the parser is not stopped.
  • Stop pumping if the parser is stopped.

(WebCore::HTML5Tokenizer::write):

  • Ignore the write if the parser is already stopped.

(WebCore::HTML5Tokenizer::end):
(WebCore::HTML5Tokenizer::attemptToEnd):

  • Move comment back to finish() which was incorrectly moved here.

(WebCore::HTML5Tokenizer::endIfDelayed):
(WebCore::HTML5Tokenizer::finish):

  • Move comment back here from attemptToEnd()

(WebCore::HTML5Tokenizer::resumeParsingAfterScriptExecution):

  • Remove a bogus ASSERT. The lexer will stop when it needs more data, which may not necessarily mean m_source.isEmpty().
  • html/HTML5Tokenizer.h:
    • Add pumpLexerIfPossible().
  • html/HTMLTokenizer.cpp: (WebCore::HTMLTokenizer::stopParsing):
    • Add a FIXME about if we really should call checkCompleted() here.
    • Inline FrameLoader::tokenizerProcessedData() since this was the only caller.
  • loader/FrameLoader.cpp:
    • Remove tokenizerProcessedData()
  • loader/FrameLoader.h:
    • Remove tokenizerProcessedData()
  • loader/MediaDocument.cpp:
    • Remove unneeded Tokenizer::stopParsing() override.
  • loader/PluginDocument.cpp:
    • Remove unneeded Tokenizer::stopParsing() override.
  • loader/SinkDocument.cpp:
    • Remove unneeded Tokenizer::stopParsing() override.
16:42 Changeset [60812] by crogers@google.com

Better support for HRIR composite generation

16:42 Changeset [60811] by crogers@google.com

Fix audio file writer code to properly close the file

16:42 Changeset [60810] by crogers@google.com

Add composite HRIR impulse responses

16:41 Changeset [60809] by crogers@google.com

Still need AudioFileWriter.h header

16:41 Changeset [60808] by crogers@google.com

Fix order of initialization in FFTFrame copy constructor

16:41 Changeset [60807] by crogers@google.com

Move AudioFileWriter to mac/AudioFileWriterMac

  • for testing purposes only
  • simplify implementation
16:25 Changeset [60806] by eric@webkit.org

2010-06-07 Anders Bakken <agbakken@gmail.com>

Reviewed by David Levin.

[Qt] NotificationPresenterClientQt.cpp has coding-style errors
https://bugs.webkit.org/show_bug.cgi?id=40256

  • WebCoreSupport/NotificationPresenterClientQt.cpp: (WebCore::NotificationPresenterClientQt::show):
15:45 Changeset [60805] by mrobinson@webkit.org

2010-06-07 Martin Robinson <mrobinson@igalia.com>

Reviewed by Oliver Hunt.

[GTK] HTML5 input type=range is not drawn
https://bugs.webkit.org/show_bug.cgi?id=39561

Uncomment and add baselines for slider tests which are now passing. Group
remaining failing slider tests by their type of failure.

  • platform/gtk/Skipped:
  • platform/gtk/fast/forms/range-thumb-height-percentage-expected.txt: Added.
  • platform/gtk/fast/forms/slider-thumb-shared-style-expected.txt: Added.
  • platform/gtk/fast/forms/slider-thumb-stylability-expected.txt: Added.
  • platform/gtk/fast/forms/thumbslider-crash-expected.txt: Added.
  • platform/gtk/fast/forms/thumbslider-no-parent-slider-expected.txt: Added.

2010-06-07 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

Fix XFrameOptions and xssAuditor crashes in HTML5 parser
https://bugs.webkit.org/show_bug.cgi?id=40265

We're not supposed to call end() while the tokenizer's write() method.
This causes a bunch of LayoutTests to crash. In particular, this patch
fixes crashes in the following tests:

Tests:

  • http/tests/security/XFrameOptions/x-frame-options-deny-meta-tag-in-body.html
  • http/tests/security/XFrameOptions/x-frame-options-deny-meta-tag-parent-same-origin-deny.html
  • http/tests/security/XFrameOptions/x-frame-options-deny-meta-tag.html
  • http/tests/security/xssAuditor/full-block-base-href.html
  • http/tests/security/xssAuditor/full-block-get-from-iframe.html
  • http/tests/security/xssAuditor/full-block-iframe-javascript-url.html
  • http/tests/security/xssAuditor/full-block-link-onclick.html
  • http/tests/security/xssAuditor/full-block-post-from-iframe.html
  • http/tests/security/xssAuditor/full-block-script-tag.html
  • http/tests/security/xssAuditor/xss-protection-parsing-01.html
  • html/HTML5Tokenizer.cpp: (WebCore::): (WebCore::HTML5Tokenizer::HTML5Tokenizer): (WebCore::HTML5Tokenizer::write): (WebCore::HTML5Tokenizer::attemptToEnd): (WebCore::HTML5Tokenizer::endIfDelayed): (WebCore::HTML5Tokenizer::finish): (WebCore::HTML5Tokenizer::resumeParsingAfterScriptExecution):
  • html/HTML5Tokenizer.h: (WebCore::HTML5Tokenizer::inWrite):
15:32 Changeset [60804] by mrobinson@webkit.org

2010-06-07 Martin Robinson <mrobinson@igalia.com>

Unreviewed.

[GTK] r60803 broke media/controls-drag-timebar.html
https://bugs.webkit.org/show_bug.cgi?id=40269

  • platform/gtk/Skipped: Skip test which is now failing.
15:13 Changeset [60803] by mrobinson@webkit.org

2010-06-07 Martin Robinson <mrobinson@igalia.com>

Reviewed by Oliver Hunt.

[GTK] Mouse movement should not trigger a replay of saved events in the event sender
https://bugs.webkit.org/show_bug.cgi?id=40267

When a mouse button is down, automatically queue mouse motion events and
do not play them back until the mouse button is released. This matches the
behavior of other ports' EventSenders.

  • DumpRenderTree/gtk/EventSender.cpp: (mouseMoveToCallback): Don't automatically replaySavedEvents for mouse motion. (sendOrQueueEvent): Add a shouldReplaySavedEvents argument defaulting to true.
14:53 Changeset [60802] by abarth@webkit.org

2010-06-07 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

Fix XFrameOptions and xssAuditor crashes in HTML5 parser
https://bugs.webkit.org/show_bug.cgi?id=40265

We're not supposed to call end() while the tokenizer's write() method.
This causes a bunch of LayoutTests to crash. In particular, this patch
fixes crashes in the following tests:

Tests:

  • http/tests/security/XFrameOptions/x-frame-options-deny-meta-tag-in-body.html
  • http/tests/security/XFrameOptions/x-frame-options-deny-meta-tag-parent-same-origin-deny.html
  • http/tests/security/XFrameOptions/x-frame-options-deny-meta-tag.html
  • http/tests/security/xssAuditor/full-block-base-href.html
  • http/tests/security/xssAuditor/full-block-get-from-iframe.html
  • http/tests/security/xssAuditor/full-block-iframe-javascript-url.html
  • http/tests/security/xssAuditor/full-block-link-onclick.html
  • http/tests/security/xssAuditor/full-block-post-from-iframe.html
  • http/tests/security/xssAuditor/full-block-script-tag.html
  • http/tests/security/xssAuditor/xss-protection-parsing-01.html
  • html/HTML5Tokenizer.cpp: (WebCore::): (WebCore::HTML5Tokenizer::HTML5Tokenizer): (WebCore::HTML5Tokenizer::write): (WebCore::HTML5Tokenizer::attemptToEnd): (WebCore::HTML5Tokenizer::endIfDelayed): (WebCore::HTML5Tokenizer::finish): (WebCore::HTML5Tokenizer::resumeParsingAfterScriptExecution):
  • html/HTML5Tokenizer.h: (WebCore::HTML5Tokenizer::inWrite):
14:40 Changeset [60801] by eric@webkit.org

2010-06-07 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

HTML5 parser hits ASSERT in http/tests/misc/window-dot-stop.html
https://bugs.webkit.org/show_bug.cgi?id=40264

  • html/HTML5Tokenizer.cpp: (WebCore::HTML5Tokenizer::executeScript):
14:13 Changeset [60800] by kinuko@chromium.org

2010-06-07 Kinuko Yasuda <kinuko@chromium.org>

Unreviewed.

Avoid "warning: no previous prototype for" in Snow Leopard compilation.

  • platform/BlobItem.cpp: Added static to getFileSnapshotModificationTime().
13:55 Changeset [60799] by kinuko@chromium.org

2010-06-07 Kinuko Yasuda <kinuko@chromium.org>

Reviewed by Jian Li.

Refactor FormData and Blob for better support of Blobs synthesized by BlobBuilder.
https://bugs.webkit.org/show_bug.cgi?id=39083

Replace FormDataList::Item list with BlobItemList to get it compiled
with the refactoring in FormDataList.

  • src/WebSearchableFormData.cpp: (WebCore::HasSuitableTextElement):

2010-06-07 Kinuko Yasuda <kinuko@chromium.org>

Reviewed by Jian Li.

Refactor FormData and Blob for better support of Blobs synthesized by BlobBuilder.
https://bugs.webkit.org/show_bug.cgi?id=39083

  • Introduces a new class BlobItem as a basic component of Blob and FormDataList.
  • File would become a special type of Blob that contains only one FileBlobItem.
  • Fix the dependency violation in FormData so that the files under platform/ do not include any html/ files.

The patch doesn't support the latest File API changes (e.g. type,
url and slice's type parameters) and should not change any of its
existing behaviors.

The existing tests should be able to be used for regression.

  • CMakeLists.txt:
  • GNUmakefile.am:
  • WebCore.gypi:
  • WebCore.pro:
  • WebCore.vcproj/WebCore.vcproj:
  • WebCore.xcodeproj/project.pbxproj:
  • html/Blob.cpp: (WebCore::Blob::Blob): (WebCore::Blob::size): (WebCore::Blob::path): (WebCore::Blob::append): (WebCore::Blob::slice):
  • html/Blob.h: (WebCore::Blob::create): (WebCore::Blob::type): (WebCore::Blob::items): (WebCore::Blob::Blob):
  • html/File.cpp: (WebCore::File::File): (WebCore::File::name):
  • html/File.h: (WebCore::File::fileName):
  • html/FileReader.cpp: (WebCore::FileReader::readAsBinaryString): (WebCore::FileReader::readAsText):
  • html/FileStream.cpp: (WebCore::FileStream::openForRead):
  • html/FormDataList.cpp: (WebCore::FormDataList::appendString): Moved the line ending fix logic to StringBlobItem::convertToCString. (WebCore::FormDataList::appendBlob):
  • html/FormDataList.h: (WebCore::FormDataList::items):
  • html/HTMLFormElement.cpp: (WebCore::HTMLFormElement::createFormData):
  • html/HTMLProgressElement.cpp:
  • platform/BlobItem.cpp: Added.
  • platform/BlobItem.h: Added.
  • platform/network/FormData.cpp: (WebCore::FormDataElement::FormDataElement): (WebCore::FormData::create): (WebCore::FormData::createMultiPart): (WebCore::FormData::deepCopy): (WebCore::FormData::appendData): (WebCore::FormData::appendString): (WebCore::FormData::appendFile): (WebCore::FormData::appendFileRange): (WebCore::FormData::appendItems): (WebCore::FormData::appendItem): (WebCore::FormData::appendKeyValuePairItems):
  • platform/network/FormData.h: (WebCore::operator!=):
  • platform/network/mac/FormDataStreamMac.mm: (WebCore::closeCurrentStream): (WebCore::advanceCurrentStream): (WebCore::formCreate): (WebCore::formRead): (WebCore::setHTTPBody):
  • xml/XMLHttpRequest.cpp: (WebCore::XMLHttpRequest::send):
13:14 Changeset [60798] by eric@webkit.org

2010-06-07 Vangelis Kokkevis <vangelis@chromium.org>

Reviewed by Dimitri Glazkov.

[chromium] Decoupled transform matrix update of composited layers from
drawing and interposed a sorting step so that the layers are now composited
back to front to get transparency working correctly. Also added missing code
for the preserves3D CSS property.

  • platform/graphics/chromium/GraphicsLayerChromium.cpp: (WebCore::GraphicsLayerChromium::updateLayerPreserves3D):
  • platform/graphics/chromium/LayerChromium.h: (WebCore::LayerChromium::setDrawTransform): (WebCore::LayerChromium::drawTransform): (WebCore::LayerChromium::setDrawOpacity): (WebCore::LayerChromium::drawOpacity): (WebCore::LayerChromium::preserves3D):
  • platform/graphics/chromium/LayerRendererChromium.cpp: (WebCore::compareLayerZ): (WebCore::LayerRendererChromium::drawLayers): (WebCore::LayerRendererChromium::updateLayersRecursive): (WebCore::LayerRendererChromium::drawLayer):
  • platform/graphics/chromium/LayerRendererChromium.h:
13:01 Changeset [60797] by eric@webkit.org

2010-06-07 Anton Muhin <antonm@chromium.org>

Reviewed by Nate Chapin.

[Chromium] convert a key to AtomicString in V8NodeList::namedPropertyGetter
https://bugs.webkit.org/show_bug.cgi?id=40238

Comparison to "length" below could be performed faster in this (major) case
as it would only require direct pointer comparison (cf. to full text
comparison if key is converted to WebCore::String.)

  • bindings/v8/custom/V8NodeListCustom.cpp: (WebCore::V8NodeList::namedPropertyGetter):
12:27 Changeset [60796] by eric@webkit.org

2010-06-07 Ariya Hidayat <ariya.hidayat@codeaurora.org>

Reviewed by Kenneth Rohde Christiansen.

[Qt] Fix GraphicsContext::clipOut to take the transformation matrix into account.

  • platform/graphics/qt/GraphicsContextQt.cpp: (WebCore::GraphicsContext::clipOut):
12:22 Changeset [60795] by kov@webkit.org

2010-06-07 Gustavo Noronha Silva <gustavo.noronha@collabora.co.uk>

Unreviewed.

[gtk] web fonts not loaded properly in scribd html5 reader
https://bugs.webkit.org/show_bug.cgi?id=38758

Also reset the expected result for the dynamic test for
backslash/yen sign.

  • platform/gtk/fast/text/backslash-to-yen-sign-dynamic-expected.txt:
12:08 Changeset [60794] by kov@webkit.org

Adding files I forgot last commit.

12:03 Changeset [60793] by kov@webkit.org

WebCore

2010-06-07 Gustavo Noronha Silva <gustavo.noronha@collabora.co.uk>

Reviewed by Xan Lopez.

[gtk] web fonts not loaded properly in scribd html5 reader
https://bugs.webkit.org/show_bug.cgi?id=38758

Avoid always giving a positive reply in createFontPlatformData. We
do not want every non-existing family name to be considered a hit,
or weird usage of local sources in @font-face rules will make
WebKitGTK+ never consider the online fonts.

  • GNUmakefile.am:
  • platform/graphics/cairo/FontCacheCairo.cpp: (WebCore::isWellKnownFontName): (WebCore::FontCache::createFontPlatformData):
  • platform/graphics/cairo/GOwnPtrCairo.cpp: Added. (WTF::FcPattern): (WTF::FcObjectSet): (WTF::FcFontSet):
  • platform/graphics/cairo/GOwnPtrCairo.h: Added.

LayoutTest

2010-06-07 Gustavo Noronha Silva <gustavo.noronha@collabora.co.uk>

Reviewed by Xan Lopez.

[gtk] web fonts not loaded properly in scribd html5 reader
https://bugs.webkit.org/show_bug.cgi?id=38758

Reset expected result for test that uses fonts we do not really
support, and are not "web fonts".

  • platform/gtk/fast/text/backslash-to-yen-sign-expected.txt:
11:54 Changeset [60792] by crogers@google.com

More windows porting work - get rid of usleep() - courtesy Ian Lewis

11:51 Changeset [60791] by abarth@webkit.org

2010-06-07 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

"Fix" fast/dom/stripNullFromTextNodes.html in HTML5 parser
https://bugs.webkit.org/show_bug.cgi?id=40200

The proper handling of null characters is to replace them with U+FFFD.
This patch "fixes" fast/dom/stripNullFromTextNodes.html in the sense
that it makes it give the proper output w.r.t. HTML5. However, that
doesn't match the behavior of the old parser, which just stripped the
nulls.

  • html/HTML5Lexer.h: (WebCore::HTML5Lexer::InputStreamPreprocessor::peek):
11:34 Changeset [60790] by abarth@webkit.org

2010-06-07 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

HTML5 parser should normalize line endings
https://bugs.webkit.org/show_bug.cgi?id=40199

When reading characters from the source, we need to normalize line
endings, which means treating \r as \n and folding \r\n sequences into
a single \n. Doing this requires a slightly tighter pattern for
advancing the source (to make sure we update our state machine
correctly).

Fixes fast/css/first-child-pseudo-class.html and hundreds of other
LayoutTests in the HTML5 parser.

  • html/HTML5Lexer.cpp: (WebCore::HTML5Lexer::nextToken):
  • html/HTML5Lexer.h: (WebCore::HTML5Lexer::InputStreamPreprocessor::LineEndingNormalizer):
    • Introduced InputStreamPreprocessor to keep track of our line endings state machine. This class will grow to eventually do all the input stream preprocessing required by the HTML5 spec.

(WebCore::HTML5Lexer::InputStreamPreprocessor::nextInputCharacter):

  • We need to be careful not to read *source more than once because we need to know whether we've transformed the character. Keeping the next input character in this state machine requires some careful coding, but seems advantageous.

(WebCore::HTML5Lexer::InputStreamPreprocessor::peek):
(WebCore::HTML5Lexer::InputStreamPreprocessor::advance):

10:53 Changeset [60789] by andreip@google.com

2010-06-07 Andrei Popescu <andreip@google.com>

Reviewed by Jeremy Orlow.

run-bindings-tests results broken by Changeset 60776
https://bugs.webkit.org/show_bug.cgi?id=40230

The change to CodeGeneratorJS.pm in
http://trac.webkit.org/changeset/60776/trunk/WebCore/bindings/scripts/CodeGeneratorJS.pm
added the possibility to use [CallWith=ScriptExecutionContext] in idl files.

Before the above change was submitted, CodeGeneratorJS.pm was modified so that native
functions return EncodedJSValue instead of JSValues. This was done in
http://trac.webkit.org/changeset/60631/trunk/WebCore/bindings/scripts/CodeGeneratorJS.pm

This CL updates the CodeGeneratorJS.pm to correctly return an EncodedJSValue in the case
where the ScriptExecutionContext pointer is 0 at the time when the native function is invoked.

No new tests, just fixing run-bindings-tests.

  • bindings/scripts/CodeGeneratorJS.pm:
  • bindings/scripts/test/JS/JSTestObj.cpp: (WebCore::jsTestObjPrototypeFunctionWithScriptExecutionContext):
  • bindings/scripts/test/JS/JSTestObj.h:
10:50 Changeset [60788] by eric@webkit.org

2010-06-07 Kwang Yul Seo <skyul@company100.net>

Reviewed by Eric Seidel.

[BREWMP] Add dummy SearchPopupMenu implementation
https://bugs.webkit.org/show_bug.cgi?id=40224

Add dummy implementation of SearchPopupMenu.

  • platform/brew/SearchPopupMenuBrew.cpp: Added. (WebCore::SearchPopupMenu::saveRecentSearches): (WebCore::SearchPopupMenu::loadRecentSearches): (WebCore::SearchPopupMenu::SearchPopupMenu): (WebCore::SearchPopupMenu::enabled):
10:34 Changeset [60787] by eric@webkit.org

2010-06-07 Kwang Yul Seo <skyul@company100.net>

Reviewed by Eric Seidel.

[BREWMP] Add more types to OwnPtr
https://bugs.webkit.org/show_bug.cgi?id=39667

Add ISSL and ISocket to the list of OwnPtr-ed type.

  • wtf/OwnPtrCommon.h:
  • wtf/brew/OwnPtrBrew.cpp: (WTF::deleteOwnedPtr):
08:26 Changeset [60786] by jchaffraix@webkit.org

Unreviewed, reverting my HTML comment parsing change as it
breaks Gtk and Qt.

08:02 Changeset [60785] by mrobinson@webkit.org

2010-06-07 Martin Robinson <mrobinson@igalia.com>

Reviewed by Xan Lopez.

[GTK] gtk_widget_get_window should replace widget->window
https://bugs.webkit.org/show_bug.cgi?id=40180

Replace all uses of widget->window with gtk_widget_get_window. For older
GTK+ versions, #define gtk_widget_get_window in GtkVersioning.h.

No tests necessary as functionality has not changed.

  • platform/gtk/GtkPluginWidget.cpp: (WebCore::GtkPluginWidget::invalidateRect): Replace widget->window use.
  • platform/gtk/GtkVersioning.h: Add gtk_widget_get_window for old GTK+ versions.
  • platform/gtk/PlatformScreenGtk.cpp: (WebCore::getVisual): Replace widget->window use. (WebCore::screenRect): Ditto.
  • platform/gtk/PopupMenuGtk.cpp: (WebCore::PopupMenu::show): Ditto.
  • platform/gtk/WidgetGtk.cpp: (WebCore::gdkDrawable): Ditto. (WebCore::Widget::setCursor): Ditto.

2010-06-07 Martin Robinson <mrobinson@igalia.com>

Reviewed by Xan Lopez.

[GTK] gtk_widget_get_window should replace widget->window
https://bugs.webkit.org/show_bug.cgi?id=40180

Replace uses of widget->window with gtk_widget_get_window.

  • WebCoreSupport/ChromeClientGtk.cpp: (WebKit::ChromeClient::invalidateContentsAndWindow): Replace widget->window uses. (WebKit::ChromeClient::scroll): Ditto. (WebKit::widgetScreenPosition): Ditto.
  • WebCoreSupport/DragClientGtk.cpp: Move gtk_widget_get_window define to GtkVersioning.h.
  • webkit/webkitwebview.cpp: (webkit_web_view_popup_menu_handler): Replace widget->window uses. (webkit_web_view_button_press_event): Ditto.
07:47 Changeset [60784] by zimmermann@webkit.org

2010-06-07 Nikolas Zimmermann <nzimmermann@rim.com>

Reviewed by Sam Weinig.

Add CPP bindings generator
https://bugs.webkit.org/show_bug.cgi?id=38279

Add CPP bindings generator used in Olympia platform.
We can succesfully generate cpp bindings for all IDLs except:

  • DOMWindow.idl / EventTarget.idl (filtered out in the IDL list in DerivedSources.make for non-jsc/v8 generators anyways, so no problem at all)
  • inspector/ (not needed for us at all)
  • svg/ (needs custom handling, related to animated attributes)

All generated files are free of style issues: "Total errors found: 0 in 510 files"
Not integrated in any build system so far, though I've successfully tested on Mac and inside Olympia, of course.

  • bindings/cpp: Added.
  • bindings/cpp/WebDOMCString.cpp: Added. (WebDOMCString::reset): (WebDOMCString::assign): (WebDOMCString::length): (WebDOMCString::data): (WebDOMCString::utf16): (WebDOMCString::fromUTF16): (WebDOMCString::WebDOMCString): (WebDOMCString::operator=): (WebDOMCString::operator WTF::CString):
  • bindings/cpp/WebDOMCString.h: Added. (WebDOMCString::~WebDOMCString): (WebDOMCString::WebDOMCString): (WebDOMCString::operator=): (WebDOMCString::isEmpty): (WebDOMCString::isNull):
  • bindings/cpp/WebDOMEventListenerCustom.cpp: Added. (WebDOMEventListener::handleEvent): (toWebKit):
  • bindings/cpp/WebDOMEventTarget.cpp: Added. (WebDOMEventTarget::WebDOMEventTargetPrivate::WebDOMEventTargetPrivate): (WebDOMEventTarget::WebDOMEventTarget): (WebDOMEventTarget::~WebDOMEventTarget): (WebDOMEventTarget::impl): (toWebCore): (toWebKit):
  • bindings/cpp/WebDOMEventTarget.h: Added.
  • bindings/cpp/WebDOMHTMLDocumentCustom.cpp: Added. (documentWrite): (WebDOMHTMLDocument::write): (WebDOMHTMLDocument::writeln):
  • bindings/cpp/WebDOMNodeCustom.cpp: Added. (WebDOMNode::insertBefore): (WebDOMNode::replaceChild): (WebDOMNode::removeChild): (WebDOMNode::appendChild): (WebDOMNode::addEventListener): (WebDOMNode::removeEventListener):
  • bindings/cpp/WebDOMObject.h: Added. (WebDOMObject::WebDOMObject): (WebDOMObject::~WebDOMObject):
  • bindings/cpp/WebDOMString.cpp: Added. (WebDOMString::reset): (WebDOMString::assign): (WebDOMString::length): (WebDOMString::data): (WebDOMString::utf8): (WebDOMString::fromUTF8): (WebDOMString::WebDOMString): (WebDOMString::operator=): (WebDOMString::operator WebCore::String): (WebDOMString::operator WebCore::AtomicString): (WebDOMString::equals):
  • bindings/cpp/WebDOMString.h: Added. (WebDOMString::~WebDOMString): (WebDOMString::WebDOMString): (WebDOMString::operator=): (WebDOMString::isEmpty): (WebDOMString::isNull):
  • bindings/cpp/WebExceptionHandler.cpp: Added. (globalExceptionHandler): (webInstallExceptionHandler): (webRaiseDOMException):
  • bindings/cpp/WebExceptionHandler.h: Added. (webDOMRaiseError):
  • bindings/cpp/WebNativeEventListener.cpp: Added. (WebNativeEventListener::WebNativeEventListener): (WebNativeEventListener::~WebNativeEventListener): (WebNativeEventListener::handleEvent): (WebNativeEventListener::reportError): (WebNativeEventListener::operator==):
  • bindings/cpp/WebNativeEventListener.h: Added. (WebNativeEventListener::create): (WebNativeEventListener::cast):
  • bindings/scripts/CodeGeneratorCPP.pm: Added.
  • bindings/scripts/test/CPP: Added.
  • bindings/scripts/test/CPP/WebDOMTestCallback.cpp: Added. (WebDOMTestCallback::WebDOMTestCallbackPrivate::WebDOMTestCallbackPrivate): (WebDOMTestCallback::WebDOMTestCallback): (WebDOMTestCallback::impl): (WebDOMTestCallback::~WebDOMTestCallback): (WebDOMTestCallback::callbackWithClass1Param): (WebDOMTestCallback::callbackWithClass2Param): (WebDOMTestCallback::callbackWithNonBoolReturnType): (toWebCore): (toWebKit):
  • bindings/scripts/test/CPP/WebDOMTestCallback.h: Added.
  • bindings/scripts/test/CPP/WebDOMTestInterface.cpp: Added. (WebDOMTestInterface::WebDOMTestInterfacePrivate::WebDOMTestInterfacePrivate): (WebDOMTestInterface::WebDOMTestInterface): (WebDOMTestInterface::impl): (WebDOMTestInterface::~WebDOMTestInterface): (toWebCore): (toWebKit):
  • bindings/scripts/test/CPP/WebDOMTestInterface.h: Added.
  • bindings/scripts/test/CPP/WebDOMTestObj.cpp: Added. (WebDOMTestObj::WebDOMTestObjPrivate::WebDOMTestObjPrivate): (WebDOMTestObj::WebDOMTestObj): (WebDOMTestObj::impl): (WebDOMTestObj::~WebDOMTestObj): (WebDOMTestObj::readOnlyIntAttr): (WebDOMTestObj::readOnlyStringAttr): (WebDOMTestObj::readOnlyTestObjAttr): (WebDOMTestObj::intAttr): (WebDOMTestObj::setIntAttr): (WebDOMTestObj::longLongAttr): (WebDOMTestObj::setLongLongAttr): (WebDOMTestObj::unsignedLongLongAttr): (WebDOMTestObj::setUnsignedLongLongAttr): (WebDOMTestObj::stringAttr): (WebDOMTestObj::setStringAttr): (WebDOMTestObj::testObjAttr): (WebDOMTestObj::setTestObjAttr): (WebDOMTestObj::attrWithException): (WebDOMTestObj::setAttrWithException): (WebDOMTestObj::attrWithSetterException): (WebDOMTestObj::setAttrWithSetterException): (WebDOMTestObj::attrWithGetterException): (WebDOMTestObj::setAttrWithGetterException): (WebDOMTestObj::scriptStringAttr): (WebDOMTestObj::voidMethod): (WebDOMTestObj::voidMethodWithArgs): (WebDOMTestObj::intMethod): (WebDOMTestObj::intMethodWithArgs): (WebDOMTestObj::objMethod): (WebDOMTestObj::objMethodWithArgs): (WebDOMTestObj::methodThatRequiresAllArgs): (WebDOMTestObj::methodThatRequiresAllArgsAndThrows): (WebDOMTestObj::serializedValue): (WebDOMTestObj::methodWithException): (WebDOMTestObj::addEventListener): (WebDOMTestObj::removeEventListener): (WebDOMTestObj::withDynamicFrame): (WebDOMTestObj::withDynamicFrameAndArg): (WebDOMTestObj::withDynamicFrameAndOptionalArg): (WebDOMTestObj::withScriptStateVoid): (WebDOMTestObj::withScriptStateObj): (WebDOMTestObj::withScriptStateVoidException): (WebDOMTestObj::withScriptStateObjException): (WebDOMTestObj::methodWithOptionalArg): (WebDOMTestObj::methodWithNonOptionalArgAndOptionalArg): (WebDOMTestObj::methodWithNonOptionalArgAndTwoOptionalArgs): (toWebCore): (toWebKit):
  • bindings/scripts/test/CPP/WebDOMTestObj.h: Added.
  • css/StyleSheet.idl: Export helper method isCSSStyleSheet() for cpp bindings only.
  • dom/CustomEvent.idl: Disabled for cpp bindings, needs arbitary input parameter handling support ('DOMObject' type).
  • dom/Document.idl: Ignore getCSSCanvasContext() for cpp bindings and export helper method isHTMLDocument() (cpp bindings only).
  • dom/Event.idl: Export isMutationEvent()/isMouseEvent()/isUIEvent() helpers for cpp bindings only.
  • dom/EventListener.h: Add CPPEventListenerType.
  • dom/Node.idl: addEventListener/removeEventListener need [Custom] flag for cpp bindings.
  • dom/PopStateEvent.idl: Disabled just like CustomEvent, for the same reason ("any" argument type handling).
  • html/HTMLCanvasElement.idl: Disable getContext() method for cpp bindings, as it's done for objc bindings.
  • html/HTMLDocument.idl: Add [NoCPPCustom] marker for open(), as we're just autogenerating it.
  • html/HTMLInputElement.idl: Disable valueAsDate() method for cpp bindings, as we're missing native 'Date' type support.
  • html/TextMetrics.h: Add missing PassRefPtr.h include.
  • html/canvas/CanvasRenderingContext2D.idl: Disable methods (setLineCap, etc..) for cpp bindings whose names clash with property setters.
  • page/AbstractView.idl: Rename 'Media' type to 'StyleMedia' - this should have been done before.
  • page/Location.idl: Disable several methods which would need a custom implementation for the cpp bindings.
  • page/WebKitPoint.h: Add missing PassRefPtr.h include.
  • storage/SQLResultSet.idl: Correct type from 'long' to 'long long', for cpp bindings only though as it wouldn't build on Chromium/V8.
  • workers/DedicatedWorkerContext.idl: Disable postMessage() for now when using cpp bindings, would need custom code for us.
  • workers/WorkerContext.idl: Hide constructors from cpp bindings.

2010-06-07 Nikolas Zimmermann <nzimmermann@rim.com>

Reviewed by Sam Weinig.

Add CPP bindings generator
https://bugs.webkit.org/show_bug.cgi?id=38279

Integrate CPP bindings in run-bindings-tests.

  • Scripts/run-bindings-tests:
07:43 Changeset [60783] by eric@webkit.org

2010-06-07 Jeremy Moskovich <jeremy@chromium.org>

Reviewed by Darin Adler.

Add documentation on the various assertion macros in WebKit.

https://bugs.webkit.org/show_bug.cgi?id=39950

  • coding/assertion-guidelines.html: Added.
07:43 Changeset [60782] by robert@webkit.org

2010-06-07 Robert Hogan <robert@webkit.org>

Reviewed by Kenneth Rohde Christiansen.

[Qt] Put some structure on the Qt Skipped file

https://bugs.webkit.org/show_bug.cgi?id=38818

  • platform/qt/Skipped:
07:41 Changeset [60781] by jchaffraix@webkit.org

WebCore: Space should be allowed between -- and > in comment end
https://bugs.webkit.org/show_bug.cgi?id=21945

Reviewed by Alexey Proskuryakov.

Implemented the HTML comment parsing algorithm so that we match HTML5 and
FF when parsing comments. Missing from this patch is
the parser errors, which will be added in a follow up patch.

Added tests cases for broken comments.

Tests: fast/parser/broken-comment-1.html

fast/parser/broken-comment-2.html
fast/parser/broken-comment-3.html
fast/parser/broken-comment-4.html
fast/parser/broken-comment-5.html
fast/parser/broken-comment-6.html
fast/parser/broken-comment-in-head-1.html
fast/parser/broken-comment-in-head-2.html
fast/parser/broken-comment-in-head-3.html
fast/parser/broken-comment-in-head-4.html
fast/parser/broken-comment-in-head-5.html

  • html/HTMLTokenizer.cpp:

(WebCore::HTMLTokenizer::parseComment): Now we use a state machine
that matches the HTML5 specification.
(WebCore::HTMLTokenizer::emitCommentToken): This emits the comment token
and keep some of the original behavior as parseComment is called in wrong
context (inside a <title> for example). Added a more explicit comment about
this.
(WebCore::HTMLTokenizer::parseTag): Remove the handling of <!--> in quirks mode.
HTML5 specifies that we should accept this in strict mode too.

  • html/HTMLTokenizer.h:

(WebCore::HTMLTokenizer::): Added a new enum for the comment parsing state.

LayoutTests: Tests for:
Space should be allowed between -- and > in comment end
https://bugs.webkit.org/show_bug.cgi?id=21945

Reviewed by Alexey Proskuryakov.

  • fast/css/word-space-extra.html: Tweaked the comment as -- > is now

a valid comment end.

  • fast/parser/broken-comment-1-expected.txt: Added.
  • fast/parser/broken-comment-1.html: Added.
  • fast/parser/broken-comment-2-expected.txt: Added.
  • fast/parser/broken-comment-2.html: Added.
  • fast/parser/broken-comment-3-expected.txt: Added.
  • fast/parser/broken-comment-3.html: Added.
  • fast/parser/broken-comment-4-expected.txt: Added.
  • fast/parser/broken-comment-4.html: Added.
  • fast/parser/broken-comment-5-expected.txt: Added.
  • fast/parser/broken-comment-5.html: Added.
  • fast/parser/broken-comment-6-expected.txt: Added.
  • fast/parser/broken-comment-6.html: Added.

Those test checks that broken comments in body are correctly parsed.

  • fast/parser/broken-comment-in-head-1-expected.txt: Added.
  • fast/parser/broken-comment-in-head-1.html: Added.
  • fast/parser/broken-comment-in-head-2-expected.txt: Added.
  • fast/parser/broken-comment-in-head-2.html: Added.
  • fast/parser/broken-comment-in-head-3-expected.txt: Added.
  • fast/parser/broken-comment-in-head-3.html: Added.
  • fast/parser/broken-comment-in-head-4-expected.txt: Added.
  • fast/parser/broken-comment-in-head-4.html: Added.
  • fast/parser/broken-comment-in-head-5-expected.txt: Added.
  • fast/parser/broken-comment-in-head-5.html: Added.

Those test checks that broken comments in head are correctly parsed.

  • fast/parser/broken-comments-vs-parsing-mode.html: Needed to wrap the broken comment

in a title as we match FF here and would not recover from it.

  • fast/parser/comments-expected.txt:
  • fast/parser/comments.html:

Added more tests for comment parsing. Made sure that we close empty comments
<!--> and <!---> as is specified in HTML5.

  • html5lib/runner-expected.txt: Removed the test that we now pass.
07:29 Changeset [60780] by eric@webkit.org

2010-06-07 Mahesh Kulakrni <mahesh.kulkarni@nokia.com>

Reviewed by Simon Hausmann.

[QT] QT_BEARER flag is not enabled on S60 properly
https://bugs.webkit.org/show_bug.cgi?id=39357

enable QT_BEARER for all platform based on qtmobility +
bearer module availability or for qt 4.7+

  • WebCore.pri:
06:59 Changeset [60779] by andreip@google.com

Unreviewed, add Andrei Popescu to committers list.

06:56 Changeset [60778] by tonikitoo@webkit.org

Expose the editing behavior setting in DRT to test all editing code paths https://bugs.webkit.org/show_bug.cgi?id=39680

Patch by Antonio Gomes <tonikitoo@webkit.org>, Laszlo Gombos <laszlo.1.gombos@nokia.com> on 2010-06-06
Reviewed by Kenneth Christiansen, Eric Seidel.

Make setEditingBehavior() a no-operation if the editingBehavior
argument is not recognized to avoid using an uninitialized variable.

  • WebCoreSupport/DumpRenderTreeSupportQt.cpp:

(DumpRenderTreeSupportQt::setEditingBehavior):

06:40 WebKit Team edited by andreip@google.com
Added myself to list of Committers. (diff)
05:32 Changeset [60777] by eric@webkit.org

2010-06-07 Leon Clarke <leonclarke@google.com>

Reviewed by Jeremy Orlow.

Fix indexeddb idls in android makefile
https://bugs.webkit.org/show_bug.cgi?id=40169

No new tests. Android-specific build fix.

  • Android.derived.v8bindings.mk:
05:01 Changeset [60776] by andreip@google.com

2010-06-07 Andrei Popescu <andreip@google.com>

Reviewed by Jeremy Orlow.

[indexedDB] It is impossible to create object stores
https://bugs.webkit.org/show_bug.cgi?id=40054

No new tests. Layout tests will be added separately.

  • Android.mk:
  • CMakeLists.txt:
  • GNUmakefile.am:
  • WebCore.gypi:
  • WebCore.pro:
  • WebCore.vcproj/WebCore.vcproj:
  • WebCore.xcodeproj/project.pbxproj:
  • bindings/scripts/CodeGeneratorJS.pm:
  • bindings/scripts/CodeGeneratorV8.pm:
  • bindings/scripts/test/GObject/WebKitDOMTestObj.cpp: (webkit_dom_test_obj_with_script_execution_context):
  • bindings/scripts/test/GObject/WebKitDOMTestObj.h:
  • bindings/scripts/test/JS/JSTestObj.cpp: (WebCore::): (WebCore::jsTestObjPrototypeFunctionWithScriptExecutionContext):
  • bindings/scripts/test/JS/JSTestObj.h:
  • bindings/scripts/test/ObjC/DOMTestObj.h:
  • bindings/scripts/test/ObjC/DOMTestObj.mm: (-[DOMTestObj withScriptExecutionContext]):
  • bindings/scripts/test/TestObj.idl:
  • bindings/scripts/test/V8/V8TestObj.cpp: (WebCore::TestObjInternal::withScriptExecutionContextCallback): (WebCore::):
  • page/DOMWindow.cpp: (WebCore::DOMWindow::clear): (WebCore::DOMWindow::indexedDB):
  • storage/IDBCallbacks.h:
  • storage/IDBDatabase.h:
  • storage/IDBDatabaseImpl.cpp: (WebCore::IDBDatabaseImpl::createObjectStore):
  • storage/IDBDatabaseImpl.h:
  • storage/IDBDatabaseRequest.cpp: (WebCore::IDBDatabaseRequest::IDBDatabaseRequest): (WebCore::IDBDatabaseRequest::createObjectStore):
  • storage/IDBDatabaseRequest.h:
  • storage/IDBDatabaseRequest.idl:
  • storage/IDBObjectStore.cpp: Removed.
  • storage/IDBObjectStore.h: (WebCore::IDBObjectStore::~IDBObjectStore):
  • storage/IDBObjectStoreImpl.cpp: Added. (WebCore::IDBObjectStoreImpl::~IDBObjectStoreImpl): (WebCore::IDBObjectStoreImpl::IDBObjectStoreImpl): (WebCore::IDBObjectStoreImpl::indexNames): (WebCore::IDBObjectStoreImpl::createIndex): (WebCore::IDBObjectStoreImpl::index): (WebCore::IDBObjectStoreImpl::removeIndex):
  • storage/IDBObjectStoreImpl.h: Added. (WebCore::IDBObjectStoreImpl::create): (WebCore::IDBObjectStoreImpl::name): (WebCore::IDBObjectStoreImpl::keyPath):
  • storage/IDBObjectStoreRequest.cpp: (WebCore::IDBObjectStoreRequest::IDBObjectStoreRequest):
  • storage/IDBObjectStoreRequest.h: (WebCore::IDBObjectStoreRequest::create):
  • storage/IDBRequest.cpp: (WebCore::IDBRequest::onSuccess):
  • storage/IDBRequest.h:
  • storage/IndexedDatabaseRequest.cpp: (WebCore::IndexedDatabaseRequest::IndexedDatabaseRequest): (WebCore::IndexedDatabaseRequest::open):
  • storage/IndexedDatabaseRequest.h: (WebCore::IndexedDatabaseRequest::create):
  • storage/IndexedDatabaseRequest.idl:

2010-06-07 Andrei Popescu <andreip@google.com>

Reviewed by Jeremy Orlow.

[indexedDB] It is impossible to create object stores
https://bugs.webkit.org/show_bug.cgi?id=40054

  • WebKit.gyp:
  • public/WebIDBCallbacks.h: (WebKit::WebIDBCallbacks::onSuccess):
  • public/WebIDBDatabase.h: (WebKit::WebIDBDatabase::createObjectStore):
  • public/WebIDBObjectStore.h: Added. (WebKit::WebIDBObjectStore::~WebIDBObjectStore): (WebKit::WebIDBObjectStore::name): (WebKit::WebIDBObjectStore::keyPath):
  • src/IDBCallbacksProxy.cpp: (WebCore::IDBCallbacksProxy::onSuccess):
  • src/IDBCallbacksProxy.h:
  • src/IDBDatabaseProxy.cpp: (WebCore::IDBDatabaseProxy::createObjectStore):
  • src/IDBDatabaseProxy.h:
  • src/IDBObjectStoreProxy.cpp: Added. (WebCore::IDBObjectStoreProxy::create): (WebCore::IDBObjectStoreProxy::IDBObjectStoreProxy): (WebCore::IDBObjectStoreProxy::~IDBObjectStoreProxy): (WebCore::IDBObjectStoreProxy::name): (WebCore::IDBObjectStoreProxy::keyPath): (WebCore::IDBObjectStoreProxy::indexNames): (WebCore::IDBObjectStoreProxy::createIndex): (WebCore::IDBObjectStoreProxy::index): (WebCore::IDBObjectStoreProxy::removeIndex):
  • src/IDBObjectStoreProxy.h: Added.
  • src/IndexedDatabaseProxy.cpp: (WebCore::IndexedDatabaseProxy::open):
  • src/WebIDBCallbacksImpl.cpp: (WebCore::WebIDBCallbacksImpl::onSuccess):
  • src/WebIDBCallbacksImpl.h:
  • src/WebIDBDatabaseImpl.cpp: (WebKit::WebIDBDatabaseImpl::createObjectStore):
  • src/WebIDBDatabaseImpl.h:
  • src/WebIDBObjectStoreImpl.cpp: Added. (WebKit::WebIDBObjectStoreImpl::WebIDBObjectStoreImpl): (WebKit::WebIDBObjectStoreImpl::~WebIDBObjectStoreImpl): (WebKit::WebIDBObjectStoreImpl::name): (WebKit::WebIDBObjectStoreImpl::keyPath):
  • src/WebIDBObjectStoreImpl.h: Added.
  • src/WebIndexedDatabaseImpl.cpp: (WebKit::WebIndexedDatabaseImpl::open):
04:57 QtWebKitTodo edited by hausmann@webkit.org
(diff)
04:48 Changeset [60775] by zimmermann@webkit.org

2010-06-07 Nikolas Zimmermann <nzimmermann@rim.com>

Not reviewed. Updated run-bindings-tests results.

  • bindings/scripts/test/JS/JSTestInterface.cpp: (WebCore::JSTestInterfaceConstructor::constructTestInterface):
  • bindings/scripts/test/JS/JSTestObj.cpp: (WebCore::jsTestObjPrototypeFunctionVoidMethod): (WebCore::jsTestObjPrototypeFunctionVoidMethodWithArgs): (WebCore::jsTestObjPrototypeFunctionIntMethod): (WebCore::jsTestObjPrototypeFunctionIntMethodWithArgs): (WebCore::jsTestObjPrototypeFunctionObjMethod): (WebCore::jsTestObjPrototypeFunctionObjMethodWithArgs): (WebCore::jsTestObjPrototypeFunctionMethodThatRequiresAllArgs): (WebCore::jsTestObjPrototypeFunctionMethodThatRequiresAllArgsAndThrows): (WebCore::jsTestObjPrototypeFunctionSerializedValue): (WebCore::jsTestObjPrototypeFunctionMethodWithException): (WebCore::jsTestObjPrototypeFunctionCustomMethod): (WebCore::jsTestObjPrototypeFunctionCustomMethodWithArgs): (WebCore::jsTestObjPrototypeFunctionCustomArgsAndException): (WebCore::jsTestObjPrototypeFunctionAddEventListener): (WebCore::jsTestObjPrototypeFunctionRemoveEventListener): (WebCore::jsTestObjPrototypeFunctionWithDynamicFrame): (WebCore::jsTestObjPrototypeFunctionWithDynamicFrameAndArg): (WebCore::jsTestObjPrototypeFunctionWithDynamicFrameAndOptionalArg): (WebCore::jsTestObjPrototypeFunctionWithDynamicFrameAndUserGesture): (WebCore::jsTestObjPrototypeFunctionWithDynamicFrameAndUserGestureASAD): (WebCore::jsTestObjPrototypeFunctionWithScriptStateVoid): (WebCore::jsTestObjPrototypeFunctionWithScriptStateObj): (WebCore::jsTestObjPrototypeFunctionWithScriptStateVoidException): (WebCore::jsTestObjPrototypeFunctionWithScriptStateObjException): (WebCore::jsTestObjPrototypeFunctionMethodWithOptionalArg): (WebCore::jsTestObjPrototypeFunctionMethodWithNonOptionalArgAndOptionalArg): (WebCore::jsTestObjPrototypeFunctionMethodWithNonOptionalArgAndTwoOptionalArgs): (WebCore::jsTestObjPrototypeFunctionOverloadedMethod1): (WebCore::jsTestObjPrototypeFunctionOverloadedMethod2): (WebCore::jsTestObjPrototypeFunctionOverloadedMethod3): (WebCore::jsTestObjPrototypeFunctionOverloadedMethod4): (WebCore::jsTestObjPrototypeFunctionOverloadedMethod):
  • bindings/scripts/test/JS/JSTestObj.h:
04:45 QtWebKitTodo edited by hausmann@webkit.org
(diff)
04:40 QtWebKitTodo edited by hausmann@webkit.org
(diff)
04:33 Changeset [60774] by eric@webkit.org

2010-06-07 Pavel Podivilov <podivilov@chromium.org>

Reviewed by Pavel Feldman.

Web Inspector: Implement JSON parsing for InspectorValue.
https://bugs.webkit.org/show_bug.cgi?id=40064

  • inspector/InspectorValues.cpp: (WebCore::): (WebCore::InspectorValue::asBool): (WebCore::InspectorValue::asNumber): (WebCore::InspectorValue::asString): (WebCore::InspectorValue::asObject): (WebCore::InspectorValue::asArray): (WebCore::InspectorValue::readJSON): (WebCore::InspectorValue::writeJSON): (WebCore::InspectorBasicValue::asBool): (WebCore::InspectorBasicValue::asNumber): (WebCore::InspectorBasicValue::writeJSON): (WebCore::InspectorString::asString): (WebCore::InspectorObject::asObject): (WebCore::InspectorObject::getBool): (WebCore::InspectorObject::getNumber): (WebCore::InspectorObject::getString): (WebCore::InspectorObject::getObject): (WebCore::InspectorObject::getArray): (WebCore::InspectorObject::get): (WebCore::InspectorArray::asArray):
  • inspector/InspectorValues.h: (WebCore::InspectorObject::begin): (WebCore::InspectorObject::end):
04:24 Changeset [60773] by hausmann@webkit.org

[Qt] Crash when compiling on Snow Leopard and running on Leopard
https://bugs.webkit.org/show_bug.cgi?id=31403

Patch by Benjamin Poulain <benjamin.poulain@nokia.com> on 2010-06-07
Reviewed by Simon Hausmann.

Disable the use of pthread_setname_np and other symbols
when targetting Leopard.

Use the defines TARGETING_XX instead of BUILDING_ON_XX
for features that cannot be used before Snow Leopard.

  • wtf/Platform.h:
03:36 Changeset [60772] by jocelyn.turcotte@nokia.com

[Qt] Fix text selection drawing.
https://bugs.webkit.org/show_bug.cgi?id=40221

Reviewed by Simon Hausmann.

The regression was introduced in r60169.

  • platform/graphics/qt/FontQt.cpp:

(WebCore::drawTextCommon):

02:57 Changeset [60771] by yurys@chromium.org

2010-06-07 Yury Semikhatsky <yurys@chromium.org>

Reviewed by Pavel Feldman.

Web Inspector: should be possible to distinguish extension scripts from main world scripts
https://bugs.webkit.org/show_bug.cgi?id=40220

  • bindings/js/ScriptDebugServer.cpp: remove global listeners set which is not used anymore. (WebCore::ScriptDebugServer::dispatchDidParseSource): pass script wrold type to the listeners. (WebCore::currentWorldType): (WebCore::ScriptDebugServer::sourceParsed):
  • bindings/js/ScriptDebugServer.h:
  • bindings/v8/ScriptDebugServer.cpp: (WebCore::ScriptDebugServer::dispatchDidParseSource):
  • inspector/InspectorController.cpp: (WebCore::InspectorController::didParseSource):
  • inspector/InspectorController.h:
  • inspector/InspectorFrontend.cpp: (WebCore::InspectorFrontend::parsedScriptSource):
  • inspector/InspectorFrontend.h:
  • inspector/ScriptDebugListener.h: pass type of the isolated world where the script was compiled to didParseSource. (WebCore::):
  • inspector/front-end/InjectedScript.js: (injectedScriptConstructor):
  • inspector/front-end/Script.js: (WebInspector.Script):
  • inspector/front-end/ScriptsPanel.js: (WebInspector.ScriptsPanel.prototype.addScript): (WebInspector.ScriptsPanel.prototype._addScriptToFilesMenu): use different style to highlight content scripts.
  • inspector/front-end/inspector.css: (#scripts-files option.extension-script):
  • inspector/front-end/inspector.js: (WebInspector.parsedScriptSource):

2010-06-07 Yury Semikhatsky <yurys@chromium.org>

Reviewed by Pavel Feldman.

Web Inspector: should be possible to distinguish extension scripts from main world scripts
https://bugs.webkit.org/show_bug.cgi?id=40220

  • src/js/DebuggerAgent.js: (devtools.DebuggerAgent.prototype.initUI): (devtools.DebuggerAgent.prototype.addScriptInfo_): (devtools.DebuggerAgent.prototype.formatCallFrame_): (devtools.ScriptInfo.prototype.worldType):
  • src/js/DebuggerScript.js: (debuggerScriptConstructor.DebuggerScript.getScripts): use only context id without context type when filtering scripts so that extension scripts are also included. (debuggerScriptConstructor.DebuggerScript._formatScript):
  • src/js/devTools.css:
01:58 Changeset [60770] by loki@webkit.org
  • jit/JITOpcodes32_64.cpp:

(JSC::JIT::privateCompileCTINativeCall):

Reviewed by NOBODY (JSVALUE32_64 build fix).

Note: See TracTimeline for information about the timeline view.