Timeline


and

11/17/09:

22:51 Changeset [51102] by darin@chromium.org
  • 2 edits in trunk/WebCore

2009-11-17 Nicolas Weber <thakis@chromium.org>

Reviewed by Darin Fisher.

Fix crash in Chromium/Mac where dropdowns weren't dismissed correctly
after navigation.
https://bugs.webkit.org/show_bug.cgi?id=31609

  • platform/chromium/PopupMenuChromium.cpp: (WebCore::PopupContainer::showExternal): Set parent for external dropdowns, so that |PopupListBox::hidePopup()| can successfully notify its parent's client.
21:19 Changeset [51101] by tkent@chromium.org
  • 4 edits
    3 adds in trunk

2009-11-17 Hayato Ito <hayato@google.com>

Reviewed by Darin Adler.

Avoid infinite mutual recursion when deeply nested tags are loaded
https://bugs.webkit.org/show_bug.cgi?id=30651

  • fast/parser/block-nesting-cap-table-expected.txt: Added.
  • fast/parser/block-nesting-cap-table.html: Added.
  • fast/parser/script-tests/block-nesting-cap-table.js: Added.

2009-11-17 Hayato Ito <hayato@google.com>

Reviewed by Darin Adler.

Avoid infinite mutual recursion when deeply nested tags are loaded
https://bugs.webkit.org/show_bug.cgi?id=30651

Test: fast/parser/block-nesting-cap-table.html

  • html/HTMLParser.cpp: (WebCore::HTMLParser::parseToken): (WebCore::tagPriorityOfNode): (WebCore::HTMLParser::limitBlockDepth): (WebCore::HTMLParser::insertNodeAfterLimitBlockDepth): (WebCore::HTMLParser::insertNode):
  • html/HTMLParser.h:
20:18 Changeset [51100] by jhoneycutt@apple.com
  • 2 edits in trunk/JavaScriptCore

Add JSContextRefPrivate.h to list of copied files.

Reviewed by Mark Rowe.

17:12 Changeset [51099] by mrowe@apple.com
  • 2 edits in trunk/WebKitTools

<http://webkit.org/b/31603> WebSocket server is confused if WebKit tests run from within /tmp on Mac OS X

Reviewed by Alexey Proskuryakov.

  • pywebsocket/mod_pywebsocket/dispatch.py: Use os.path.realpath as it returns the canonical path of a file.

This prevents symlinks from confusing the descendant check.

17:08 Changeset [51098] by bfulgham@webkit.org
  • 2 edits in trunk/WebCore

Rubber-stamped by Alexey Proskuryakov.

Final clean-ups for minor coding standard violations.
https://bugs.webkit.org/show_bug.cgi?id=26102.

  • platform/graphics/cairo/GraphicsContextCairo.cpp:
16:35 Changeset [51097] by dimich@chromium.org
  • 5 edits
    2 adds in trunk

WebCore: In all valid script tags for JavaScript, the event handler in <script...for> should not get executed.
https://bugs.webkit.org/show_bug.cgi?id=31567

Patch by Johnny Ding <jnd@chromium.org> on 2009-11-17
Reviewed by Darin Adler.

  • dom/ScriptElement.cpp:

(WebCore::ScriptElementData::shouldExecuteAsJavaScript):

LayoutTests: In all valid script tags for JavaScript, the event handler in <script...for> should not get executed.
Change the original test to pure js test, and test the situation about script tags which have text or language attribute.
https://bugs.webkit.org/show_bug.cgi?id=31567

Patch by Johnny Ding <jnd@chromium.org> on 2009-11-17
Reviewed by Darin Adler.

  • fast/dom/HTMLScriptElement/script-for-attribute-unexpected-execution-expected.txt:
  • fast/dom/HTMLScriptElement/script-for-attribute-unexpected-execution.html:
  • fast/dom/HTMLScriptElement/script-tests: Added.
  • fast/dom/HTMLScriptElement/script-tests/script-for-attribute-unexpected-execution.js: Added.

(ScriptForAttributeExecute):

16:31 Changeset [51096] by mrowe@apple.com
  • 3 edits
    2 adds in trunk

<http://webkit.org/b/31602> Failing to start the WebSocket server shouldn’t terminate entire test run

Reviewed by Alexey Proskuryakov.

WebKitTools:

If the WebSocket server fails to start have DRT load an error page in place of tests that require the
server to be up rather than having run-webkit-tests abort immediately.

  • Scripts/run-webkit-tests:

LayoutTests:

  • websocket/resources/server-failed-to-start.html: Error page that will be loaded in place of websocket

tests when the server fails to start.

16:24 Changeset [51095] by dimich@chromium.org
  • 2 edits in trunk/WebKit/chromium

Revert r50887, it regressed handling of cmd-left when a text box is focused.
https://bugs.webkit.org/show_bug.cgi?id=31566

Patch by Nicolas Weber <thakis@chromium.org> on 2009-11-17
Reviewed by Darin Fisher, Dmitry Titov.

  • src/mac/WebInputEventFactory.mm:

(WebKit::WebInputEventFactory::keyboardEvent):

16:08 CommitQueue edited by thakis@chromium.org
(diff)
15:38 Changeset [51094] by jhoneycutt@apple.com
  • 2 edits in trunk/WebKit/win

Implement DOMHTMLInputElement::replaceCharactersInRange().

https://bugs.webkit.org/show_bug.cgi?id=31492

Reviewed by Dan Bernstein.

  • DOMHTMLClasses.cpp:

(DOMHTMLInputElement::replaceCharactersInRange):
Get the text of the input field. Replace the given range with the
replacement text, and set this new string as the input element's value.
Select from index to the end of the field. This matches the
implementation in the Obj-C bindings.

15:37 Changeset [51093] by jhoneycutt@apple.com
  • 2 edits in trunk/WebKit/win

DOMHTMLOptionElement is missing some functionality.

https://bugs.webkit.org/show_bug.cgi?id=31491

Reviewed by Dan Bernstein.

  • DOMHTMLClasses.cpp:

(DOMHTMLOptionElement::text):
Cast m_element to an HTMLOptionElement, and call its text() function.
(DOMHTMLOptionElement::label):
Ditto, for label().

15:37 Changeset [51092] by jhoneycutt@apple.com
  • 2 edits in trunk/WebKit/win

DOMHTMLSelectElement is missing some implementation.

https://bugs.webkit.org/show_bug.cgi?id=31489

Reviewed by Dan Bernstein.

  • DOMHTMLClasses.cpp:

(DOMHTMLSelectElement::options):
Cast m_element to an HTMLSelectElement. If it has no options, return
E_FAIL. Otherwise, create a DOMHTMLOptionsCollection to wrap the
options, and return it.
(DOMHTMLSelectElement::activateItemAtIndex):
If the index is out of bounds, return E_FAIL. Otherwise, select the
item.

15:37 Changeset [51091] by jhoneycutt@apple.com
  • 3 edits in trunk/WebKit/win

DOMHTMLOptionsCollection is missing some implementation.

https://bugs.webkit.org/show_bug.cgi?id=31488

Reviewed by Dan Bernstein.

  • DOMHTMLClasses.cpp:

(DOMHTMLOptionsCollection::DOMHTMLOptionsCollection):
Initialize m_collection.
(DOMHTMLOptionsCollection::createInstance):
Create a DOMHTMLOptionsCollection. If we fail to query for
IDOMHTMLOptionsCollection, delete it, and return 0. Otherwise, return
the result.
(DOMHTMLOptionsCollection::length):
(DOMHTMLOptionsCollection::item):
Create a DOMNode for the WebCore Node. If this is 0, return E_FAIL.
(DOMHTMLOptionsCollection::namedItem):
Correct the signature of this function.

  • DOMHTMLClasses.h:

Declare DOMHTMLOptionsCollection::createInstance(). Correct the
signature of namedItem() to match IDOMHTMLOptionsCollection. Add a
member to DOMHTMLOptionsCollection to hold the WebCore object.

15:37 Changeset [51090] by jhoneycutt@apple.com
  • 2 edits in trunk/WebKit/win

DOMHTMLInputElement::rectOnScreen() returns the wrong rect

https://bugs.webkit.org/show_bug.cgi?id=31487

Reviewed by Darin Adler.

  • DOMHTMLClasses.cpp:

(DOMHTMLInputElement::rectOnScreen):
Return the rect on screen, not the rect in the window.

14:07 Changeset [51089] by bweinstein@apple.com
  • 2 edits in trunk/LayoutTests

Rubber-stamped by Pavel Feldman.

Skip inspector tests that are flaky on Windows Release - this is being tracked by:
https://bugs.webkit.org/show_bug.cgi?id=31596.

  • platform/win/Skipped:
13:40 Changeset [51088] by jparent@chromium.org
  • 4 edits in trunk/LayoutTests

Eliminate flakiness in char-encoding tests by removing setTimeout,
that was working around an old issue.
https://bugs.webkit.org/show_bug.cgi?id=31571

Reviewed by Darin Adler.

  • fast/encoding/char-encoding-mac.html:
  • fast/encoding/char-encoding.html:
  • fast/encoding/resources/char-encoding-utils.js:
13:30 Changeset [51087] by kenneth@webkit.org
  • 2 edits in trunk/WebKitTools

Unreviewed buildbot fix.

Revert part of earlier patch and add comment, as it
was causing timeouts on the buildbot.

  • DumpRenderTree/qt/LayoutTestControllerQt.cpp:

(LayoutTestController::notifyDone):

12:38 Changeset [51086] by eric@webkit.org
  • 3 edits in trunk/JavaScriptCore

2009-11-17 Martin Robinson <martin.james.robinson@gmail.com>

Reviewed by Adam Barth.

[GTK] Style cleanup for GOwnPtr
https://bugs.webkit.org/show_bug.cgi?id=31506

Remove forward declaration in GOwnPtr and do some style cleanup.

  • wtf/GOwnPtr.cpp:
  • wtf/GOwnPtr.h: (WTF::GOwnPtr::GOwnPtr): (WTF::GOwnPtr::~GOwnPtr): (WTF::GOwnPtr::get): (WTF::GOwnPtr::release): (WTF::GOwnPtr::outPtr): (WTF::GOwnPtr::set): (WTF::GOwnPtr::clear): (WTF::GOwnPtr::operator*): (WTF::GOwnPtr::operator->): (WTF::GOwnPtr::operator!): (WTF::GOwnPtr::operator UnspecifiedBoolType): (WTF::GOwnPtr::swap): (WTF::swap): (WTF::operator==): (WTF::operator!=): (WTF::getPtr): (WTF::freeOwnedGPtr):
12:11 Changeset [51085] by eric@webkit.org
  • 3 edits in trunk/WebKitTools

2009-11-17 Eric Seidel <eric@webkit.org>

Reviewed by Darin Adler.

commit-queue is failing to set reviewer in ChangeLogs
https://bugs.webkit.org/show_bug.cgi?id=31592

  • Scripts/bugzilla-tool: Clarify the "applying" log message.
  • Scripts/modules/bugzilla.py:
    • Add a new _validate_committer_and_reviewer function as a temporary solution until we can make a real Attachment object which knows how to fill in its committer/reviewer fields automatically.
12:05 Changeset [51084] by kenneth@webkit.org
  • 2 edits in trunk/WebKitTools

Make the timeout 15 sec as for the other DRT's and make
it print out the same output when a test timeout.

Reviewed by Oliver Hunt.

  • DumpRenderTree/qt/LayoutTestControllerQt.cpp:

(LayoutTestController::waitUntilDone):
(LayoutTestController::notifyDone):
(LayoutTestController::timerEvent):

11:56 Changeset [51083] by yael.aharon@nokia.com
  • 2 edits in trunk/WebKit/qt

[Qt] QGLLauncher does not support drag&drop of local files
https://bugs.webkit.org/show_bug.cgi?id=31057

Reviewed by Kenneth Rohde Christiansen.

Enable accepting files in QGraphicsWebView.

  • Api/qgraphicswebview.cpp:

(QGraphicsWebView::QGraphicsWebView):
(QGraphicsWebView::dragEnterEvent):

11:33 Changeset [51082] by bfulgham@webkit.org
  • 2 edits in trunk/WebKit/win

Correct build error in Debug_Cairo target after @49705.

Reviewed by NOBODY - Build Fix.

  • WebKit.vcproj/WebKit.vcproj: Revise JavaScriptCore.lib dependency to have proper "_debug" suffix needed by the Debug_Cairo target.
11:27 Web Inspector edited by timothy@apple.com
(diff)
11:26 WebInspector created by timothy@apple.com
Copied from Web Inspector
11:11 Changeset [51081] by bweinstein@apple.com
  • 1 edit in trunk/WebCore/inspector/InspectorController.cpp

Touch InspectorController to force the Inspector HTML and JS files to be
updated.

10:57 Changeset [51080] by ossy@webkit.org
  • 2 edits in trunk/LayoutTests

Not reviewed, follow up to r51072: rename skipped entries

  • platform/qt/Skipped:
10:52 Changeset [51079] by joepeck@webkit.org
  • 2 edits in trunk/WebKitTools

2009-11-17 Joseph Pecoraro <joepeck@webkit.org>

Reviewed by Timothy Hatcher.

Fixed typos in comments.

  • Scripts/modules/committers.py:
10:42 Changeset [51078] by ossy@webkit.org
  • 2 edits in trunk/LayoutTests

[Qt] Crashing tests after updating to Qt-4.6.0
https://bugs.webkit.org/show_bug.cgi?id=31591

Reviewed by Kenneth Rohde Christiansen.

  • platform/qt/Skipped: Put tests into skiplist temporarily.
10:37 Changeset [51077] by bweinstein@apple.com
  • 4 edits in trunk/WebKit/win

Touch files to try to fix the build.

Reviewed by NOBODY - Build Fix.

  • Interfaces/IWebInspector.idl:
  • Interfaces/WebKit.idl:
  • WebKit.vcproj/Interfaces.vcproj:
09:55 Changeset [51076] by tonikitoo@webkit.org
  • 2 edits in trunk/WebKit/qt

[Qt] better test coverage for ErrorPageExtension
https://bugs.webkit.org/show_bug.cgi?id=31583

Reviewed by Kenneth Christiansen.

Improved the coverage of current ErrorPageExtension tests by
adding autotests involving frameset and iframes.

  • tests/qwebpage/tst_qwebpage.cpp:

(ErrorPage::extension): Make the ErrorPageExtension
to work for all frames, not only the main frame.
(tst_QWebPage::errorPageExtension): Stop using
the 'frameset.html' resouce in this method since
an autotest specific for frameset's is being added.
(tst_QWebPage::errorPageExtensionInIFrames): Added.
(tst_QWebPage::errorPageExtensionInFrameset): Added.

06:14 Changeset [51075] by pfeldman@chromium.org
  • 1 edit
    6 moves
    2 adds in trunk/LayoutTests

2009-11-17 Pavel Feldman <pfeldman@chromium.org>

Not reviewed, tests fix. Follow up to r51072: move
platform-specific results for fast/inspector into
fast/inspector-support.

  • platform/mac/fast/inspector-support/matchedrules-expected.checksum: Renamed from LayoutTests/platform/mac/fast/inspector/matchedrules-expected.checksum.
  • platform/mac/fast/inspector-support/matchedrules-expected.png: Renamed from LayoutTests/platform/mac/fast/inspector/matchedrules-expected.png.
  • platform/mac/fast/inspector-support/matchedrules-expected.txt: Renamed from LayoutTests/platform/mac/fast/inspector/matchedrules-expected.txt.
  • platform/mac/fast/inspector-support/style-expected.txt: Renamed from LayoutTests/platform/mac/fast/inspector/style-expected.txt.
  • platform/qt/fast/inspector-support/matchedrules-expected.txt: Renamed from LayoutTests/platform/qt/fast/inspector/matchedrules-expected.txt.
  • platform/qt/fast/inspector-support/style-expected.txt: Renamed from LayoutTests/platform/qt/fast/inspector/style-expected.txt.
05:54 Changeset [51074] by kov@webkit.org
  • 2 edits in trunk/LayoutTests

Actually skip the inspector console tests while the problem
discussed in https://bugs.webkit.org/show_bug.cgi?id=31472 is
fixed.

05:38 Changeset [51073] by pfeldman@chromium.org
  • 5 edits in trunk

2009-11-17 Pavel Feldman <pfeldman@chromium.org>

Not reviewed, build fix. Follow up to r51072: rename skipped entries
for the tests that were moved. Also skipped console tests on GTK
while investigating and touched Interfaces.vcproj to clobber win bot.


  • platform/gtk/Skipped:
  • platform/qt/Skipped:
05:11 Changeset [51072] by pfeldman@chromium.org
  • 37 edits
    10 moves
    2 adds in trunk

2009-11-17 Pavel Feldman <pfeldman@chromium.org>

Reviewed by Timothy Hatcher.

Web Inspector: Make DRT show web inspector for tests in inspector/ folder.

  • Updated DRT to show/close inspector for all tests under /inspector
  • Introduced LayoutTestController::setTimelineProfilingEnabled and WebInspector::setTimelineProfilingEnabled beside setJavaScriptProfilingEnabled
  • Removed reload on each inspector test
  • Renamed fast/inspector to fast/inspector-support in order not to trigger inspector for those.
  • Reimplemented timeline tests in order to get rid of reload there.
  • Moved tests that don't require harness into the fast group.

https://bugs.webkit.org/show_bug.cgi?id=31472

WebCore:

  • WebCore.Inspector.exp:
  • inspector/front-end/TimelinePanel.js: (WebInspector.TimelinePanel.prototype._formatRecord):

WebKitTools:

  • DumpRenderTree/LayoutTestController.cpp: (setTimelineProfilingEnabledCallback): (closeWebInspectorCallback): (LayoutTestController::staticFunctions):
  • DumpRenderTree/LayoutTestController.h:
  • DumpRenderTree/gtk/DumpRenderTree.cpp: (shouldOpenWebInspector): (runTest):
  • DumpRenderTree/gtk/LayoutTestControllerGtk.cpp: (LayoutTestController::setTimelineProfilingEnabled):
  • DumpRenderTree/mac/DumpRenderTree.mm: (shouldOpenWebInspector): (runTest):
  • DumpRenderTree/mac/LayoutTestControllerMac.mm: (LayoutTestController::setTimelineProfilingEnabled):
  • DumpRenderTree/win/DumpRenderTree.cpp: (shouldOpenWebInspector): (runTest):
  • DumpRenderTree/win/LayoutTestControllerWin.cpp: (LayoutTestController::setTimelineProfilingEnabled):

LayoutTests:

  • fast/inspector-support/cssURLQuotes-expected.txt: Renamed from LayoutTests/fast/inspector/cssURLQuotes-expected.txt.
  • fast/inspector-support/cssURLQuotes.html: Renamed from LayoutTests/fast/inspector/cssURLQuotes.html.
  • fast/inspector-support/matchedrules.html: Renamed from LayoutTests/fast/inspector/matchedrules.html.
  • fast/inspector-support/style.html: Renamed from LayoutTests/fast/inspector/style.html.
  • fast/inspector-support/uncaught-dom1-exception-expected.txt: Renamed from LayoutTests/inspector/uncaught-dom1-exception-expected.txt.
  • fast/inspector-support/uncaught-dom1-exception.html: Renamed from LayoutTests/inspector/uncaught-dom1-exception.html.
  • fast/inspector-support/uncaught-dom3-exception-expected.txt: Renamed from LayoutTests/inspector/uncaught-dom3-exception-expected.txt.
  • fast/inspector-support/uncaught-dom3-exception.html: Renamed from LayoutTests/inspector/uncaught-dom3-exception.html.
  • fast/inspector-support/uncaught-dom8-exception-expected.txt: Renamed from LayoutTests/inspector/uncaught-dom8-exception-expected.txt.
  • fast/inspector-support/uncaught-dom8-exception.html: Renamed from LayoutTests/inspector/uncaught-dom8-exception.html.
  • inspector/inspector-test.js: (onload): (evaluateInWebInspector): (notifyDone):
  • inspector/resources/timeline-iframe-data.html: Added.
  • inspector/timeline-layout-expected.txt:
  • inspector/timeline-layout.html:
  • inspector/timeline-mark-timeline.html:
  • inspector/timeline-paint.html:
  • inspector/timeline-parse-html-expected.txt:
  • inspector/timeline-parse-html.html:
  • inspector/timeline-recalculate-styles-expected.txt:
  • inspector/timeline-recalculate-styles.html:
  • inspector/timeline-script-tag-1-expected.txt:
  • inspector/timeline-script-tag-1.html:
  • inspector/timeline-script-tag-2-expected.txt:
  • inspector/timeline-script-tag-2.html:
  • inspector/timeline-script-tag-2.js:
  • inspector/timeline-test.js: (printTimelineRecords): (): (dumpTimelineRecords): (printProps): (frontend_getTimelineResults):

WebKit/mac:

WebKit/win:

  • Interfaces/IWebInspector.idl:
  • WebInspector.cpp: (WebInspector::isTimelineProfilingEnabled): (WebInspector::setTimelineProfilingEnabled):
  • WebInspector.h:

WebKit/gtk:

  • webkit/webkitwebinspector.cpp: (webkit_web_inspector_class_init): (webkit_web_inspector_set_property): (webkit_web_inspector_get_property):
04:40 Changeset [51071] by benm@google.com
  • 1 edit
    4 adds in trunk/WebCore

[Android] Android is missing the implementation of the GeolocationService iface.
https://bugs.webkit.org/show_bug.cgi?id=31554

Patch by Andrei Popescu <andreip@google.com> on 2009-11-17
Reviewed by Dimitri Glazkov.

No new tests required as this is platform specific code.

  • platform/android/GeolocationServiceAndroid.cpp: Added.

(WebCore::GeolocationServiceAndroid::create):
(WebCore::GeolocationServiceAndroid::GeolocationServiceAndroid):
(WebCore::GeolocationServiceAndroid::startUpdating):
(WebCore::GeolocationServiceAndroid::stopUpdating):
(WebCore::GeolocationServiceAndroid::suspend):
(WebCore::GeolocationServiceAndroid::resume):
(WebCore::GeolocationServiceAndroid::newPositionAvailable):
(WebCore::GeolocationServiceAndroid::newErrorAvailable):
(WebCore::GeolocationServiceAndroid::timerFired):
(WebCore::GeolocationServiceAndroid::isPositionMovement):
(WebCore::GeolocationServiceAndroid::isPositionMoreAccurate):
(WebCore::GeolocationServiceAndroid::isPositionMoreTimely):

  • platform/android/GeolocationServiceAndroid.h: Added.

(WebCore::GeolocationServiceAndroid::~GeolocationServiceAndroid):
(WebCore::GeolocationServiceAndroid::lastPosition):
(WebCore::GeolocationServiceAndroid::lastError):

  • platform/android/GeolocationServiceBridge.cpp: Added.

(WebCore::):
(WebCore::GeolocationServiceBridge::GeolocationServiceBridge):
(WebCore::GeolocationServiceBridge::~GeolocationServiceBridge):
(WebCore::GeolocationServiceBridge::start):
(WebCore::GeolocationServiceBridge::stop):
(WebCore::GeolocationServiceBridge::setEnableGps):
(WebCore::GeolocationServiceBridge::newLocationAvailable):
(WebCore::GeolocationServiceBridge::newErrorAvailable):
(WebCore::GeolocationServiceBridge::toGeoposition):
(WebCore::GeolocationServiceBridge::startJavaImplementation):
(WebCore::GeolocationServiceBridge::stopJavaImplementation):

  • platform/android/GeolocationServiceBridge.h: Added.
01:36 Changeset [51070] by hamaji@chromium.org
  • 7 edits
    1 add in trunk/LayoutTests

2009-11-17 Yuta Kitamura <yutak@chromium.org>

Reviewed by Eric Seidel.

Fix http tests so that they do not submit a GET form to about:blank.

Original tests expected that submitting an empty form to <about:blank> navigates
the browser to <about:blank?>, not <about:blank>. However, some other browsers
do not act like Safari, and in fact the layout test runner of Chromium was
failing to run these tests.

Prior to this patch, a test that only checks this behavior was added, so now
we can remove the dependency on it from these tests.

Some HTTP tests submit a GET form to about:blank
https://bugs.webkit.org/show_bug.cgi?id=30423

  • http/tests/navigation/onload-navigation-iframe-2-expected.txt:
  • http/tests/navigation/onload-navigation-iframe-2.html:
  • http/tests/navigation/onload-navigation-iframe-expected.txt:
  • http/tests/navigation/onload-navigation-iframe-timeout-expected.txt:
  • http/tests/navigation/onload-navigation-iframe-timeout.html:
  • http/tests/navigation/onload-navigation-iframe.html:
  • http/tests/navigation/resources/blank.txt: Added.
01:27 Changeset [51069] by hamaji@chromium.org
  • 1 edit
    2 adds in trunk/LayoutTests

2009-11-17 Yuta Kitamura <yutak@chromium.org>

Reviewed by Darin Adler.

Add a test to check the behavior of submitting a GET form to <about:blank>.

Currently Safari redirects the browser to <about:blank?>, while some other
browsers do not (including Firefox and Chromium). The objective of this test is
to express the current behavior of Safari and to become ready to remove
unintentional dependency on it from several tests.

Some HTTP tests submit a GET form to about:blank
https://bugs.webkit.org/show_bug.cgi?id=30423

  • fast/forms/get-forms-to-about-blank-expected.txt: Added.
  • fast/forms/get-forms-to-about-blank.html: Added.
00:41 Changeset [51068] by oliver@apple.com
  • 6 edits in trunk

Incorrect use of JavaScriptCore API in DumpRenderTree
https://bugs.webkit.org/show_bug.cgi?id=31577

Reviewed by Maciej Stachowiak

Return undefined rather than a literal null.

11/16/09:

23:16 Changeset [51067] by zoltan@webkit.org
  • 4 edits in trunk/JavaScriptCore

https://bugs.webkit.org/show_bug.cgi?id=31050

Patch by Zoltan Herczeg <zherczeg@inf.u-szeged.hu> on 2009-11-16
Reviewed by Gavin Barraclough.

Minor fixes for JSVALUE32_64: branchConvertDoubleToInt32
failed on a CortexA8 CPU, but not on a simulator; and
JITCall.cpp modifications was somehow not committed to mainline.

  • assembler/ARMAssembler.h:

(JSC::ARMAssembler::fmrs_r):

  • assembler/MacroAssemblerARM.h:

(JSC::MacroAssemblerARM::branchConvertDoubleToInt32):

  • jit/JITCall.cpp:

(JSC::JIT::compileOpCall):

22:31 Changeset [51066] by tkent@chromium.org
  • 3 edits in trunk/WebCore

2009-11-16 Kent Tamura <tkent@chromium.org>

Unreviewd build fix.

  • Fix typo in WebCore.vcproj.
  • Intlude limits.h for INT_MAX.
  • Enclose with parenthesis to suspress warning.
  • WebCore.vcproj/WebCore.vcproj:
  • html/ISODateTime.cpp: (WebCore::ISODateTime::addDay):
21:50 Changeset [51065] by kevino@webkit.org
  • 6 edits in trunk/WebCore

Reviewed by Kevin Ollivier.

Make sure wx scrollbar drawing code factors in transforms when switching backends,
fix calcs for scrollbar length, and tweak the Mac scrollbar tracking rects.

https://bugs.webkit.org/show_bug.cgi?id=31570

21:39 Changeset [51064] by girish@forwardbias.in
  • 1 edit in trunk/WebKitTools/Scripts/modules/committers.py

Add myself to commiters.py

21:13 Changeset [51063] by tkent@chromium.org
  • 7 edits
    2 adds in trunk/WebCore

2009-11-16 Kent Tamura <tkent@chromium.org>

Reviewed by David Levin.

Introduce WebCore::ISODateTime class.
https://bugs.webkit.org/show_bug.cgi?id=31340

This class represents a value of date/time types of the HTML5 INPUT
element, and has some parsing methods for ISO 8601.

This change has no tests because the class is not used yet.

  • GNUmakefile.am: Add ISODateTime.cpp and ISODateTime.h.
  • WebCore.gypi: ditto.
  • WebCore.pro: ditto.
  • WebCore.vcproj/WebCore.vcproj: ditto.
  • WebCore.xcodeproj/project.pbxproj: ditto.
  • WebCoreSources.bkl: ditto.
  • html/ISODateTime.cpp: Added. Implementation of WebCore::ISODateTime class. (WebCore::isLeapYear): (WebCore::maxDayOfMonth): (WebCore::dayOfWeek): (WebCore::ISODateTime::maxWeekNumberInYear): (WebCore::countDigits): (WebCore::toInt): (WebCore::ISODateTime::parseYear): Private helper for parseDate() and parseWeek(). (WebCore::ISODateTime::addDay): Private helper for parseTimeZone(). (WebCore::ISODateTime::addMinute): ditto. (WebCore::ISODateTime::parseTimeZone): Private helper for parseDateTime(). (WebCore::ISODateTime::parseMonth): Parser for <input type=month>. (WebCore::ISODateTime::parseDate): Parser for <input type=date>. (WebCore::ISODateTime::parseWeek): Parser for <input type=week>. (WebCore::ISODateTime::parseTime): Parser for <input type=time>. (WebCore::ISODateTime::parseDateTimeLocal): Parser for <input type=datetime-local>. (WebCore::ISODateTime::parseDateTime): Parser for <input type=datetime>.
  • html/ISODateTime.h: Added. Declare WebCore::ISODateTime class. (WebCore::ISODateTime::ISODateTime): (WebCore::ISODateTime::millisecond): (WebCore::ISODateTime::second): (WebCore::ISODateTime::minute): (WebCore::ISODateTime::hour): (WebCore::ISODateTime::monthDay): (WebCore::ISODateTime::month): (WebCore::ISODateTime::fullYear): (WebCore::ISODateTime::week):
19:05 BuildingQtOnLinux edited by dbates@webkit.org
Added missing "libsqlite3-dev" dependency for Ubuntu (diff)
16:09 Changeset [51062] by ap@apple.com
  • 2 edits in trunk/WebCore

Fix a typo in previous commit.

  • platform/network/cf/SocketStreamHandleCFNet.cpp: #ifdef, not #if.
16:04 Changeset [51061] by ap@apple.com
  • 2 edits in trunk/WebCore

Fix a typo in previous commit.

  • platform/network/cf/SocketStreamHandleCFNet.cpp:
15:57 Changeset [51060] by ap@apple.com
  • 2 edits in trunk/WebCore

Tiger build fix.

  • platform/network/cf/SocketStreamHandleCFNet.cpp: Define CFN_EXPORT, as this macro was named differently in Tiger CFNetwork.
15:32 Changeset [51059] by ap@apple.com
  • 3 edits in trunk/LayoutTests

Disabling WebSocket tests on Tiger and Windows again, as they still time out.

  • platform/mac-tiger/Skipped:
  • platform/win/Skipped:
15:08 Changeset [51058] by dglazkov@chromium.org
  • 2 edits in trunk/WebCore

2009-11-16 Dimitri Glazkov <dglazkov@chromium.org>

Reviewed by Darin Fisher.

[KURLGoogle] setHostAndPort doesn't handle arguments without port correctly.

Fix a bug in the code that was dormant until http://trac.webkit.org/changeset/50784.

Covered by existing test: LayoutTests/fast/dom/HTMLAnchorElement/set-href-attribute-host.html

  • platform/KURLGoogle.cpp: (WebCore::KURL::setHostAndPort): Added handling of arguments without port specified.
14:58 Changeset [51057] by ap@apple.com
  • 2 edits in trunk/WebCore

Rubber-stamped by Jon Honeycutt.

A better Windows build fix

  • platform/network/cf/SocketStreamHandleCFNet.cpp: (WebCore::SocketStreamHandle::createStreams): Use a macro that inserts declspec(dllimport) when appropriate for CFNetwork imports.
14:57 Changeset [51056] by yael.aharon@nokia.com
  • 5 edits in trunk/WebCore

Rename protocolIsValid to isValidProtocol.
https://bugs.webkit.org/show_bug.cgi?id=31503

Reviewed by Darin Adler.

This name change was suggested in https://bugs.webkit.org/show_bug.cgi?id=29972#c19.

No new tests since no new functionality was introduced.

  • html/HTMLAnchorElement.cpp:

(WebCore::HTMLAnchorElement::setProtocol):

  • platform/KURL.cpp:

(WebCore::isValidProtocol):

  • platform/KURL.h:
  • platform/KURLGoogle.cpp:

(WebCore::isValidProtocol):

14:41 Changeset [51055] by cfleizach@apple.com
  • 8 edits
    2 adds in trunk

AX: aria-labelledby duplicates some of its WAI-ARIA label
https://bugs.webkit.org/show_bug.cgi?id=31565

Reviewed by Beth Dakin.

WebCore:

Test: accessibility/aria-labelledby-overrides-label.html

  • accessibility/AccessibilityRenderObject.cpp:

(WebCore::AccessibilityRenderObject::hasTextAlternative):
(WebCore::AccessibilityRenderObject::exposesTitleUIElement):
(WebCore::AccessibilityRenderObject::accessibilityIsIgnored):

  • accessibility/AccessibilityRenderObject.h:

WebKitTools:

  • DumpRenderTree/AccessibilityUIElement.cpp:

(titleUIElementCallback):
(getIsValidCallback):
(AccessibilityUIElement::getJSClass):

LayoutTests:

  • accessibility/aria-labelledby-overrides-label-expected.txt: Added.
  • accessibility/aria-labelledby-overrides-label.html: Added.
  • platform/gtk/Skipped:
  • platform/win/Skipped:
14:40 Changeset [51054] by japhet@chromium.org
  • 2 edits in trunk/WebCore

2009-11-16 Nate Chapin <japhet@chromium.org>

Reviewed by Darin Fisher.

Handle the case of a null NPObject* in NPN_SetException in
the V8 bindings. This allow out of process plugins calling
NPN_SetException to just send null instead of sending an
NPObject* that would be an address in a different
process's memory space.

https://bugs.webkit.org/show_bug.cgi?id=31561

  • bindings/v8/NPV8Object.cpp: (_NPN_SetException): Allow null NPObject* and just throw a general error.
14:30 Changeset [51053] by ap@apple.com
  • 2 edits in trunk/WebCore

Windows build fix.

  • platform/network/cf/SocketStreamHandleCFNet.cpp: (WebCore::SocketStreamHandle::createStreams): Disable CONNECT proxies on Windows until WebKitSupportLibrary includes support for those.
14:21 Changeset [51052] by yael.aharon@nokia.com
  • 3 edits in trunk/WebCore

[Qt] ASSERT failure while running DRT
https://bugs.webkit.org/show_bug.cgi?id=30978

Reviewed by Kenneth Rohde Christiansen.

Add needed Structure typeInfo flags to QtRuntimeObjectImpl and QtRuntimeMethod.
These flags are needed after r49649, where HasDefaultmark was changed to OverrideMarkChildren.

  • bridge/qt/qt_instance.cpp:

(JSC::Bindings::QtRuntimeObjectImp::createStructure):

  • bridge/qt/qt_runtime.h:

(JSC::Bindings::QtRuntimeMethod::createStructure):

13:51 Changeset [51051] by mrowe@apple.com
  • 1 edit
    1 add in trunk/WebCore

Attempt to fix the build. Land a file that was missing from r51049.

  • bindings/js/JSWebGLArrayHelper.h: Copied from WebCore/html/canvas/WebGLByteArray.idl.

(WebCore::setWebGLArrayFromArray):

13:31 Changeset [51050] by kenneth@webkit.org
  • 3 edits in trunk/WebKitTools

r50942 broke output from created windows. Make the
m_enableTextOutput a member of the DRT and not the
WebPage.

Reviewed by Simon Hausmann.

  • DumpRenderTree/qt/DumpRenderTree.cpp:

(WebCore::WebPage::WebPage):
(WebCore::WebPage::javaScriptAlert):
(WebCore::WebPage::javaScriptConsoleMessage):
(WebCore::WebPage::javaScriptConfirm):
(WebCore::WebPage::javaScriptPrompt):
(WebCore::WebPage::acceptNavigationRequest):
(WebCore::DumpRenderTree::DumpRenderTree):
(WebCore::DumpRenderTree::open):
(WebCore::DumpRenderTree::createWindow):

  • DumpRenderTree/qt/DumpRenderTree.h:

(WebCore::DumpRenderTree::setTextOutputEnabled):
(WebCore::DumpRenderTree::isTextOutputEnabled):
(WebCore::WebPage::shouldInterruptJavaScript):
(WebCore::WebPage::isTextOutputEnabled):
(WebCore::WebPage::setViewGeometry):

13:30 Changeset [51049] by cmarrin@apple.com
  • 53 edits in trunk

Update API of WebGLArray and friends
https://bugs.webkit.org/show_bug.cgi?id=31175

13:28 Changeset [51048] by ap@apple.com
  • 2 edits in trunk/WebCore

Windows build fix.

  • platform/network/cf/SocketStreamHandleCFNet.cpp: Declare constants as extern "C".
13:28 Changeset [51047] by hausmann@webkit.org
  • 4 edits in trunk/WebKit/qt

API documentation fixes.

Reviewed by Kenneth Rohde Christiansen.

  • Api/qgraphicswebview.cpp: Removed duplicate docs.
  • Api/qwebelement.cpp: Added missing docs.
  • Api/qwebsettings.cpp: Ditto.
13:01 Changeset [51046] by bfulgham@webkit.org
  • 2 edits in trunk/WebCore

Enable support for webkit-box-shadow in Cairo builds.
https://bugs.webkit.org/show_bug.cgi?id=26102.

Reviewed by Oliver Hunt.

Covered by existing fast/box-shadow tests.

  • platform/graphics/cairo/GraphicsContextCairo.cpp: Add support for fillRect shadows.
12:36 Changeset [51045] by ap@apple.com
  • 2 edits in trunk/WebCore

Windows build fix.

  • platform/network/cf/SocketStreamHandleCFNet.cpp: Don't try to include a file that's not in WebKitSupportLibrary.
12:05 Changeset [51044] by kenneth@webkit.org
  • 3 edits
    4 moves
    4 adds in trunk/WebKitTools

Moved DumpRenderTree/gtk/TestNetscapePlugin to DumpRenderTree/unix/TestNetscapePlugin
as the implementation is being used by at least Qt and Gtk+.

Reviewed by Gustavo Noronha Silva.

Update buildsystems as well.

  • DumpRenderTree/qt/TestNetscapePlugin/TestNetscapePlugin.pro:
  • DumpRenderTree/unix/TestNetscapePlugin/ForwardingHeaders/WebKit/npapi.h:
  • DumpRenderTree/unix/TestNetscapePlugin/ForwardingHeaders/WebKit/npfunctions.h:
  • DumpRenderTree/unix/TestNetscapePlugin/ForwardingHeaders/WebKit/npruntime.h:
  • DumpRenderTree/unix/TestNetscapePlugin/TestNetscapePlugin.cpp:
  • GNUmakefile.am:
11:08 Changeset [51043] by weinig@apple.com
  • 2 edits in trunk/WebKitSite

Update Security Group Members list.

Reviewed by Anders Carlsson.

  • security/security-group-members.html:
10:58 Changeset [51042] by weinig@apple.com
  • 2 edits in trunk/WebKitSite

Update Security Group Members list.

Reviewed by Adele Peterson.

  • security/security-group-members.html:
10:40 Changeset [51041] by ap@apple.com
  • 5 edits in trunk

WebKitTools:
2009-11-16 Yuzo Fujishima <yuzo@google.com>

Reviewed by Alexey Proskuryakov.

Disable wss until all platforms support pyOpenSSL

https://bugs.webkit.org/show_bug.cgi?id=31479

  • Scripts/run-webkit-tests:

LayoutTests:

Re-enabling WebSocket tests on Tiger and Windows, now that SSL is disabled.

  • platform/mac-tiger/Skipped:
  • platform/win/Skipped:
09:59 Changeset [51040] by ap@apple.com
  • 2 edits in trunk/WebCore

Reviewed by Darin Adler.

https://bugs.webkit.org/show_bug.cgi?id=31494
Add unauthenticated proxy support to SocketStreamHandleCFNet

Cannot be tested in DRT.

  • platform/network/cf/SocketStreamHandleCFNet.cpp: (WebCore::SocketStreamHandle::chooseProxy): Fetch proxy information from OS. (WebCore::SocketStreamHandle::createStreams): Apply it to the newly created streams.
08:54 WebKit Team edited by keishi@webkit.org
(diff)
08:51 Changeset [51039] by keishi@webkit.org
  • 1 edit in trunk/WebKitTools/Scripts/modules/committers.py

Adding myself to committers.py.

07:38 Changeset [51038] by tonikitoo@webkit.org
  • 8 edits
    2 adds in trunk

[Qt] Broken back/forward after using ErrorPageExtension to set error page
https://bugs.webkit.org/show_bug.cgi?id=30573

Reviewed by Antti Koivisto.

WebCore:

Make FrameLoader::checkLoadCompleteForThisFrame method
to check for any working DocumentLoader instance (through
activeDocumentLoader()) instead of only checking for
'm_provisionalDocumentLoader' in order to decide to if
it is going to reset of not the back and forward history.
after an error page has been loaded.

Test: LayoutTests/fast/history/back-forward-reset-after-error-handling.html

  • loader/FrameLoader.cpp:

(WebCore::FrameLoader::checkLoadCompleteForThisFrame):

WebKit/qt:

Implemented autotests for covering the back/forward
reset problem involving error pages.

  • tests/qwebpage/tst_qwebpage.cpp:

(tst_QWebPage::errorPageExtension):

LayoutTests:

Skipped fast/history/back-forward-reset-after-error-handling.html
in mac, gtk and win DRT, because their DRT do not support yet
error pages handling.

  • fast/history/back-forward-reset-after-error-handling.html:
  • platform/gtk/Skipped:
  • platform/mac/Skipped:
  • platform/win/Skipped:
07:37 Changeset [51037] by tonikitoo@webkit.org
  • 5 edits in trunk/WebKitTools

[Qt] Implement load error pages support for Qt's DRT.
https://bugs.webkit.org/show_bug.cgi?id=31509

Reviewed by Kenneth Christiansen.

For now, it will not be a default feature, and layout tests
that want to make use of this have to explicitily call
'handleErrorPages();' for the test source.

Any of the other DumpRenderTree's (mac, win and gtk)
support handling error pages. Qt's will be the first.

  • DumpRenderTree/qt/DumpRenderTree.cpp:

(WebCore::WebPage::supportsExtension):
(WebCore::WebPage::extension):

  • DumpRenderTree/qt/LayoutTestControllerQt.cpp:

(LayoutTestController::reset):

  • DumpRenderTree/qt/LayoutTestControllerQt.h:

(LayoutTestController::shouldHandleErrorPages):
(LayoutTestController::handleErrorPages):

07:05 QtWebKitContrib edited by hausmann@webkit.org
(diff)
06:59 QtWebKitJournal edited by hausmann@webkit.org
(diff)
06:58 QtWebKitJournal edited by hausmann@webkit.org
(diff)
06:30 QtWebKitJournal edited by hausmann@webkit.org
(diff)
06:26 QtWebKitJournal edited by hausmann@webkit.org
(diff)
06:24 QtWebKitJournal edited by hausmann@webkit.org
(diff)
06:18 QtWebKitJournal edited by hausmann@webkit.org
(diff)
06:12 QtWebKit edited by kenneth@webkit.org
(diff)
06:06 QtWebKitJournal edited by hausmann@webkit.org
(diff)
06:05 QtWebKit edited by kenneth@webkit.org
(diff)
06:04 QtWebKit edited by hausmann@webkit.org
(diff)
05:59 QtWebKitJournal edited by hausmann@webkit.org
(diff)
05:47 Changeset [51036] by eric@webkit.org
  • 2 edits in trunk/WebKitTools

2009-11-16 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

Rename the --commit-queue flag on land-* now that the commit-queue needs no special treatment
https://bugs.webkit.org/show_bug.cgi?id=31549

Renamed --commit-queue to --non-interactive in most places
and remove the code in land-patches which is no longer needed.

  • Scripts/bugzilla-tool:
05:43 Changeset [51035] by kenneth@webkit.org
  • 1 edit
    4 adds in trunk/LayoutTests

Rubberstamped by Simon Hausmann.

Added new Qt expected results for animation tests.

  • platform/qt/animations/change-transform-in-end-event-expected.txt: Added.
  • platform/qt/animations/state-at-end-event-expected.txt: Added.
  • platform/qt/animations/state-at-end-event-transform-expected.txt: Added.
05:34 Changeset [51034] by eric@webkit.org
  • 2 edits in trunk/WebKitTools

2009-11-16 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

The commit-queue should use land-attachment
https://bugs.webkit.org/show_bug.cgi?id=31548

  • Scripts/bugzilla-tool:
05:25 Changeset [51033] by eric@webkit.org
  • 2 edits in trunk/WebKitTools

2009-11-16 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

Convert CommitQueue over to PatchCollection
https://bugs.webkit.org/show_bug.cgi?id=31547

Also fixes a bug in workqueue and adds a test!

  • Scripts/bugzilla-tool:
  • Scripts/modules/workqueue.py:
  • Scripts/modules/workqueue_unittest.py:
05:20 Changeset [51032] by hausmann@webkit.org
  • 3 edits in trunk/JavaScriptCore

Fix Qt build on Windows CE 6.

Patch by Joerg Bornemann <joerg.bornemann@trolltech.com> on 2009-11-16
Reviewed by Simon Hausmann.

  • JavaScriptCore.pri: Add missing include path.
  • wtf/Platform.h: Include ce_time.h for Windows CE 6.
05:05 Changeset [51031] by abarth@webkit.org
  • 4 edits in trunk/WebKitTools

2009-11-16 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

Convert CommitQueue over to PatchCollection
https://bugs.webkit.org/show_bug.cgi?id=31547

Also fixes a bug in workqueue and adds a test!

  • Scripts/bugzilla-tool:
  • Scripts/modules/workqueue.py:
  • Scripts/modules/workqueue_unittest.py:
04:29 Changeset [51030] by abarth@webkit.org
  • 4 edits in trunk/WebKitTools

2009-11-16 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

Move StyleQueue over to using PatchCollection
https://bugs.webkit.org/show_bug.cgi?id=31544

That's what the class it's for.

  • Scripts/bugzilla-tool:
  • Scripts/modules/patchcollection.py:
  • Scripts/modules/patchcollection_unittest.py:
04:11 Changeset [51029] by eric@webkit.org
  • 2 edits in trunk/WebKitTools

2009-11-16 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

Move more patch-landing code into WebKitLandingScripts in preparation for land-attachment
https://bugs.webkit.org/show_bug.cgi?id=31543

Just moving code and updating the one caller to use WebKitLandingScripts instead of 'self'.

  • Scripts/bugzilla-tool:
04:04 Changeset [51028] by eric@webkit.org
  • 2 edits in trunk/WebKitTools

2009-11-16 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

Disable the style queue from posting to the commit queue status page.

  • Scripts/bugzilla-tool:
03:54 Changeset [51027] by eric@webkit.org
  • 2 edits in trunk/WebKitTools

2009-11-16 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

Implement PatchCollection
https://bugs.webkit.org/show_bug.cgi?id=31541

This class holds a set of patches and lets clients iterate through
them. Optionally, clients can install a filter.

  • Scripts/modules/patchcollection.py: Added.
  • Scripts/modules/patchcollection_unittest.py: Added.
  • Scripts/run-webkit-unittests:
03:47 Changeset [51026] by abarth@webkit.org
  • 2 edits
    1 copy
    1 add in trunk/WebKitTools

2009-11-16 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

Implement PatchCollection
https://bugs.webkit.org/show_bug.cgi?id=31541

This class holds a set of patches and lets clients iterate through
them. Optionally, clients can install a filter.

  • Scripts/modules/patchcollection.py: Added.
  • Scripts/modules/patchcollection_unittest.py: Added.
  • Scripts/run-webkit-unittests:
03:31 Changeset [51025] by eric@webkit.org
  • 2 edits in trunk/WebKitTools

2009-11-16 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

bugzilla-tool check-style should work with attachment ids instead of bug ids
https://bugs.webkit.org/show_bug.cgi?id=31540

  • Scripts/bugzilla-tool:
03:04 Changeset [51024] by eric@webkit.org
  • 4 edits in trunk/WebKitTools

2009-11-16 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

WorkQueue is the only place that should know about special exit codes
https://bugs.webkit.org/show_bug.cgi?id=31534

Move LandPatchesFromBugs.handled_error to WorkQueue.exit_after_handled_error
and add tests for handling exit codes.
I also cleaned up workqueue_unittest.py more.

  • Scripts/bugzilla-tool:
  • Scripts/modules/workqueue.py:
  • Scripts/modules/workqueue_unittest.py:
03:02 Changeset [51023] by eric@webkit.org
  • 2 edits in trunk/WebKitTools

2009-11-16 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

Re-factor workqueue_unittest to allow for more than one test.
https://bugs.webkit.org/show_bug.cgi?id=31535

  • Scripts/modules/workqueue_unittest.py:
03:00 Changeset [51022] by eric@webkit.org
  • 2 edits in trunk/WebKitTools

2009-11-16 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

land-patches needs to be re-factored into smaller chunks
https://bugs.webkit.org/show_bug.cgi?id=31532

  • Scripts/bugzilla-tool: fix a couple obvious typos.
02:50 Changeset [51021] by eric@webkit.org
  • 2 edits in trunk/WebKitTools

2009-11-16 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

land-patches needs to be re-factored into smaller chunks
https://bugs.webkit.org/show_bug.cgi?id=31532

The next patch will move these methods into WebKitLandingScripts.

  • Scripts/bugzilla-tool:
    • Split out _land_patch and _close_bug_if_no_active_patches.
02:45 Changeset [51020] by abarth@webkit.org
  • 2 edits in trunk/WebKitTools

2009-11-16 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

Fix silly copy-and-paste code. I am a terrible coder.

  • Scripts/modules/bugzilla.py:
02:34 Changeset [51019] by abarth@webkit.org
  • 3 edits in trunk/WebKitTools

2009-11-16 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

Implement a StyleQueue
https://bugs.webkit.org/show_bug.cgi?id=31537

The first iteration of the style queue only produces output locally.
There is also a limit of 10 patches because it's not that useful to
iterate through the entire review queue at this point. We can remove
the limit later.

  • Scripts/bugzilla-tool:
  • Scripts/modules/bugzilla.py:
01:25 Changeset [51018] by abarth@webkit.org
  • 5 edits
    1 add in trunk/WebKitTools

2009-11-16 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

Unit test WorkQueue
https://bugs.webkit.org/show_bug.cgi?id=31531

Adds basic unit testing for WorkQueue. Just runs through one cycle.

  • Scripts/bugzilla-tool:
  • Scripts/modules/statusbot.py:
  • Scripts/modules/workqueue.py:
  • Scripts/modules/workqueue_unittest.py: Added.
  • Scripts/run-webkit-unittests:
00:49 Changeset [51017] by eric@webkit.org
  • 2 edits in trunk/WebKitTools

2009-11-16 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

bugzilla-tool land-diff should know how to parse bug ids out of ChangeLogs
https://bugs.webkit.org/show_bug.cgi?id=31530

  • Scripts/bugzilla-tool:
00:39 Changeset [51016] by eric@webkit.org
  • 4 edits in trunk/WebKitTools

2009-11-16 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

bugzilla-tool needs apply-attachment
https://bugs.webkit.org/show_bug.cgi?id=31528

  • Scripts/bugzilla-tool:
    • Add ApplyAttachment command.
    • Abstract applying code into WebKitApplyingScripts.
    • Rename setup_for_landing to prepare_clean_working_directory and make local_commit checking optional.
  • Scripts/modules/bugzilla.py:
    • Add fetch_attachment and bug_id_for_attachment_id.
  • Scripts/modules/bugzilla_unittest.py:
    • Add test for new parsing.
    • Fix previous parsing test which broke with Adam's check-style patch (bug 31515).
00:38 Changeset [51015] by eric@webkit.org
  • 3 edits in trunk/WebKitTools

2009-11-16 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

Document check-style's use of force_clean.

  • Scripts/bugzilla-tool:
  • Scripts/modules/scm.py:
00:28 Changeset [51014] by abarth@webkit.org
  • 3 edits
    1 add in trunk/WebKitTools

2009-11-16 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

Move WorkQueue to its own file
https://bugs.webkit.org/show_bug.cgi?id=31529

WorkQueue and WorkQueueDelegate are separate concerns from
bugzilla-tool. Also added a missing include to logging.py.

  • Scripts/bugzilla-tool:
  • Scripts/modules/logging.py:
  • Scripts/modules/workqueue.py:
00:11 Changeset [51013] by abarth@webkit.org
  • 3 edits in trunk/WebKitTools

2009-11-16 Adam Barth <abarth@webkit.org>

Reviewed by Eric Siedel.

Move OutputTee to logging.py.

  • Scripts/bugzilla-tool:
  • Scripts/modules/logging.py:
00:07 Changeset [51012] by abarth@webkit.org
  • 2 edits in trunk/WebKitTools

2009-11-15 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

Refactor bugzilla-tool to allow for multiple queues
https://bugs.webkit.org/show_bug.cgi?id=31513

Divide the commit queue class into three class to make creating
additional queues easier.

  • Scripts/bugzilla-tool:

11/15/09:

22:19 Changeset [51011] by cfleizach@apple.com
  • 3 edits in trunk/LayoutTests

Fixing broken layout test.

Need to implement ARIA role="directory"
https://bugs.webkit.org/show_bug.cgi?id=31516

Lists are not supported on Leopard or Tiger, so the test for this
bug needs to be skipped.

  • platform/mac-leopard/Skipped:
  • platform/mac-tiger/Skipped:
22:02 Changeset [51010] by hamaji@chromium.org
  • 3 edits in trunk/WebKitTools

2009-11-15 Shinichiro Hamaji <hamaji@chromium.org>

Reviewed by Eric Seidel.

svn-apply may not handle git patches created by bugzilla-tool
https://bugs.webkit.org/show_bug.cgi?id=31457

  • Scripts/modules/scm.py:
  • Scripts/modules/scm_unittest.py:
20:44 Changeset [51009] by dglazkov@chromium.org
  • 2 edits in trunk/WebKit/chromium

2009-11-15 Dimitri Glazkov <dglazkov@chromium.org>

Unreviewed, build fix.

Correct an error in the previous commit

  • public/WebAccessibilityRole.h: It should be WebAccessibilityRoleDirectory, not WebAccessibilityDirectoryRole.
19:48 Changeset [51008] by dglazkov@chromium.org
  • 3 edits in trunk/WebKit/chromium

2009-11-15 Dimitri Glazkov <dglazkov@chromium.org>

Unreviewed, build fix.

Bring WebAccessibilityRole up to sync after
http://trac.webkit.org/changeset/51007.

  • public/WebAccessibilityRole.h: Added WebAccessibilityRoleDirectory.
  • src/AssertMatchingEnums.cpp: Added a match between DirectoryRole and WebAccessibilityRoleDirectory.
18:05 Changeset [51007] by cfleizach@apple.com
  • 10 edits
    2 adds in trunk

Need to implement ARIA role="directory"
https://bugs.webkit.org/show_bug.cgi?id=31516

Reviewed by Darin Adler.

WebCore:

Test: platform/mac/accessibility/aria-directory.html

  • accessibility/AXObjectCache.cpp:

(WebCore::AXObjectCache::getOrCreate):

  • accessibility/AccessibilityList.cpp:

(WebCore::AccessibilityList::isOrderedList):

  • accessibility/AccessibilityRenderObject.cpp:

(WebCore::createARIARoleMap):

WebKit/mac:

Fix a spelling error in the comment of a localized (accessibility) string.

  • WebCoreSupport/WebViewFactory.mm:

(-[WebViewFactory AXARIAContentGroupText:]):

LayoutTests:

  • platform/mac/accessibility/aria-directory-expected.txt: Added.
  • platform/mac/accessibility/aria-directory.html: Added.
15:34 Changeset [51006] by abarth@webkit.org
  • 2 edits in trunk/WebCore

2009-11-15 Dave Tapuska <dtapuska@rim.com>

Reviewed by George Staikos.


Compare UChars single unit at a time as opposed to the uint32_t
approach as casting to unaligned addresses may cause a bus failure
on ARMv5 and below. This change replicates the same defines that
exists in AtomicString.cpp

https://bugs.webkit.org/show_bug.cgi?id=31475


  • platform/text/StringHash.h: (WebCore::StringHash::equal):
15:32 Changeset [51005] by abarth@webkit.org
  • 3 edits in trunk/WebKitTools

2009-11-15 Shinichiro Hamaji <hamaji@chromium.org>

Reviewed by Eric Seidel.

bugzilla-tool should post git binary diff
https://bugs.webkit.org/show_bug.cgi?id=31458

Add --binary option to Git.create_patch.

  • Scripts/modules/scm.py:
  • Scripts/modules/scm_unittest.py:
15:29 Changeset [51004] by abarth@webkit.org
  • 2 edits in trunk/WebCore

2009-11-15 Evan Martin <evan@chromium.org>

Reviewed by Adam Barth.

Wrap some SVG code in V8DOMWrapper with an ENABLE(SVG) test.

https://bugs.webkit.org/show_bug.cgi?id=31490

  • bindings/v8/V8DOMWrapper.cpp:
15:28 Changeset [51003] by abarth@webkit.org
  • 2 edits in trunk/WebCore

2009-11-15 Maxime Simon <simon.maxime@gmail.com>

Reviewed by Adam Barth.

[Haiku] Build fix. The FileChooser constructor doesn't need to be redefined.

  • platform/haiku/FileChooserHaiku.cpp:
15:26 Changeset [51002] by abarth@webkit.org
  • 2 edits in trunk/WebCore

2009-11-15 Maxime Simon <simon.maxime@gmail.com>

Reviewed by Adam Barth.

[Haiku] Build fix. ColorSpace name had a wrong CamelCase.

  • platform/graphics/haiku/GraphicsContextHaiku.cpp: (WebCore::GraphicsContext::setPlatformStrokeColor):
15:24 Changeset [51001] by abarth@webkit.org
  • 4 edits in trunk/WebKitTools

2009-11-15 Adam Barth <abarth@webkit.org>

Reviewed by Darin Adler.

Add bugzilla-tool check-style
https://bugs.webkit.org/show_bug.cgi?id=31515

  • Scripts/bugzilla-tool:
  • Scripts/modules/bugzilla.py:
15:15 Changeset [51000] by dbates@webkit.org
  • 8 edits in trunk/WebCore

2009-11-15 Daniel Bates <dbates@webkit.org>

No review, rolling out r50999.
http://trac.webkit.org/changeset/50999

Need to fix some issues in the Windows build. Missed some places where
RenderTextControl::isEdited is called.

  • bindings/objc/DOMHTML.mm: (-[DOMHTMLInputElement _isEdited]): (-[DOMHTMLTextAreaElement _isEdited]):
  • dom/Document.cpp: (WebCore::Document::setFocusedNode):
  • html/HTMLInputElement.cpp: (WebCore::HTMLInputElement::defaultEventHandler):
  • rendering/RenderTextControl.cpp: (WebCore::RenderTextControl::RenderTextControl): (WebCore::RenderTextControl::setInnerTextValue): (WebCore::RenderTextControl::setUserEdited): (WebCore::RenderTextControl::subtreeHasChanged):
  • rendering/RenderTextControl.h: (WebCore::RenderTextControl::isEdited): (WebCore::RenderTextControl::setEdited): (WebCore::RenderTextControl::isUserEdited):
  • rendering/RenderTextControlSingleLine.cpp: (WebCore::RenderTextControlSingleLine::subtreeHasChanged):
  • wml/WMLInputElement.cpp: (WebCore::WMLInputElement::defaultEventHandler):
14:55 Changeset [50999] by dbates@webkit.org
  • 8 edits in trunk/WebCore

2009-11-15 Daniel Bates <dbates@webkit.org>

Reviewed by Darin Adler.

https://bugs.webkit.org/show_bug.cgi?id=31186

Renames RenderTextControl::m_edited and RenderTextControl::m_userEdited to
m_wasChangedSinceLastChangeEvent and m_lastChangeWasUserEdit, respectively.
These are more descriptive names so as to clear an ambiguity surrounding
their usage. Also, renames associated setters and getters so that they
coincide with the renamed fields.

No functionality was changed. So, no tests were included.

  • bindings/objc/DOMHTML.mm: (-[DOMHTMLInputElement _isEdited]): (-[DOMHTMLTextAreaElement _isEdited]):
  • dom/Document.cpp: (WebCore::Document::setFocusedNode):
  • html/HTMLInputElement.cpp: (WebCore::HTMLInputElement::defaultEventHandler):
  • rendering/RenderTextControl.cpp: (WebCore::RenderTextControl::RenderTextControl): (WebCore::RenderTextControl::setInnerTextValue): (WebCore::RenderTextControl::setLastChangeWasUserEdit): Formerly named setUserEdited. (WebCore::RenderTextControl::subtreeHasChanged):
  • rendering/RenderTextControl.h: (WebCore::RenderTextControl::wasChangedSinceLastChangeEvent): Formerly named isEdited. (WebCore::RenderTextControl::setChangedSinceLastChangeEvent): Formerly named setEdited. (WebCore::RenderTextControl::lastChangeWasUserEdit): Formerly named setUserEdited.
  • rendering/RenderTextControlSingleLine.cpp: (WebCore::RenderTextControlSingleLine::subtreeHasChanged):
  • wml/WMLInputElement.cpp: (WebCore::WMLInputElement::defaultEventHandler):

11/14/09:

12:55 Changeset [50998] by adele@apple.com
  • 11 edits
    5 adds in trunk

WebCore: Fix for <rdar://problem/6946165> Would like to be able to specify the number of visible lines when using -webkit-line-clamp

Reviewed by Dan Bernstein.

Test: fast/overflow/line-clamp.html

  • css/CSSComputedStyleDeclaration.cpp: (WebCore::CSSComputedStyleDeclaration::getPropertyCSSValue): Updated to handle different types of values.
  • css/CSSParser.cpp: (WebCore::CSSParser::parseValue): ditto.
  • css/CSSStyleSelector.cpp: (WebCore::CSSStyleSelector::applyProperty): ditto.
  • rendering/RenderFlexibleBox.cpp: (WebCore::RenderFlexibleBox::layoutVerticalBox):

Use the line count value if available. Otherwise, convert the percentage to the line count, as we did before. Also,
if there is anchor as the last child box, still allow adding the ellipsis.

  • rendering/RenderLayer.cpp:

(WebCore::RenderLayer::scrollByRecursively): Call isNone on the LineClampValue to see if the line-clamp property has been set.
(WebCore::RenderLayer::scrollRectToVisible): ditto.

  • WebCore.xcodeproj/project.pbxproj: Added LineClampValue.h
  • rendering/style/LineClampValue.h: Added.

(WebCore::LineClampValue::LineClampValue):
(WebCore::LineClampValue::value):
(WebCore::LineClampValue::isPercentage):
(WebCore::LineClampValue::isNone):
(WebCore::LineClampValue::operator==):
(WebCore::LineClampValue::operator!=):

  • rendering/style/RenderStyleConstants.h: (WebCore::): Define ELineClampType enum for percentage or line count.
  • rendering/style/RenderStyle.h: Use LineClampValue.

(WebCore::InheritedFlags::lineClamp):
(WebCore::InheritedFlags::setLineClamp):
(WebCore::InheritedFlags::initialLineClamp):

  • rendering/style/StyleRareNonInheritedData.h:

LayoutTests: Test for <rdar://problem/6946165> Would like to be able to specify the number of visible lines when using -webkit-line-clamp

Reviewed by Dan Bernstein.

  • fast/overflow/line-clamp.html: Added.
  • platform/mac/fast/overflow/line-clamp-expected.checksum: Added.
  • platform/mac/fast/overflow/line-clamp-expected.png: Added.
  • platform/mac/fast/overflow/line-clamp-expected.txt: Added.
12:24 QtBackLog edited by zecke@selfish.org
Mention individual commits (diff)
09:53 Changeset [50997] by eric.carlson@apple.com
  • 8 edits in trunk

2009-11-14 Eric Carlson <eric.carlson@apple.com>

Reviewed by Oliver Hunt.

<rdar://problem/7287487>
Do not use QuickTime version to detect media controller theme

  • WebCore.base.exp: Export wkMediaControllerThemeAvailable
  • platform/mac/WebCoreSystemInterface.h: Ditto.
  • platform/mac/WebCoreSystemInterface.mm: Ditto.
  • rendering/RenderThemeMac.mm: (WebCore::mediaControllerTheme): Use wkMediaControllerThemeAvailable instead of the QuickTime version to see if it is possible to use MediaControllerThemeQuickTime.

2009-11-14 Eric Carlson <eric.carlson@apple.com>

Reviewed by Oliver Hunt.

<rdar://problem/7287487>
Do not use QuickTime version to detect media controller theme

  • WebCoreSupport/WebSystemInterface.m: (InitWebCoreSystemInterface): Initialize wkMediaControllerThemeAvailable.
07:21 Changeset [50996] by tkent@chromium.org
  • 4 edits
    3 adds in trunk

2009-11-14 Kent Tamura <tkent@chromium.org>

Reviewed by Darin Adler.

Add tests for reading/writing .type property of HTMLInputElement.
https://bugs.webkit.org/show_bug.cgi?id=29004

  • fast/forms/input-type-change3-expected.txt: Added.
  • fast/forms/input-type-change3.html: Added.
  • fast/forms/script-tests/input-type-change3.js: Added.

2009-11-14 Kent Tamura <tkent@chromium.org>

Reviewed by Darin Adler.

  • Recognizes date/datetime/datetime-local/month/time/week types of INPUT element. They have no dedicated UI and no type validation for now.
  • Clean up setInputType() and formControlType() of HTMLInputElement. https://bugs.webkit.org/show_bug.cgi?id=29004

Test: fast/forms/input-type-change3.html

  • html/HTMLInputElement.cpp: (WebCore::HTMLInputElement::valueMissing): (WebCore::HTMLInputElement::patternMismatch): (WebCore::HTMLInputElement::tooLong): (WebCore::createTypeMap): Initializer for a static local variable. (WebCore::HTMLInputElement::setInputType): (WebCore::createFormControlTypes): Initializer for a static local variable. (WebCore::HTMLInputElement::formControlType): (WebCore::HTMLInputElement::saveFormControlState): (WebCore::HTMLInputElement::restoreFormControlState): (WebCore::HTMLInputElement::accessKeyAction): (WebCore::HTMLInputElement::rendererIsNeeded): (WebCore::HTMLInputElement::createRenderer): (WebCore::HTMLInputElement::appendFormData): (WebCore::HTMLInputElement::isTextField): (WebCore::HTMLInputElement::valueWithDefault): (WebCore::HTMLInputElement::storesValueSeparateFromAttribute): (WebCore::HTMLInputElement::defaultEventHandler): (WebCore::HTMLInputElement::isRequiredFormControl): (WebCore::HTMLInputElement::dataList):
  • html/HTMLInputElement.h: Add new values to InputType enum, and define numberOfTypes.
06:08 Changeset [50995] by kov@webkit.org
  • 2 edits in trunk/LayoutTests

Skip new test that needs new DRT infrastructure, with bug filled.

11/13/09:

22:24 Changeset [50994] by cfleizach@apple.com
  • 5 edits
    2 adds in trunk

WAI-ARIA: checkbox does not determine its label from text content
https://bugs.webkit.org/show_bug.cgi?id=31456

Reviewed by Darin Adler.

WebCore:

Test: accessibility/aria-checkbox-text.html

  • accessibility/AccessibilityRenderObject.cpp:

(WebCore::AccessibilityRenderObject::title):

LayoutTests:

  • accessibility/aria-checkbox-text.html: Added.
  • platform/gtk/Skipped:
  • platform/mac/accessibility/aria-checkbox-text-expected.txt: Added.
  • platform/win/Skipped:
20:54 Changeset [50993] by dglazkov@chromium.org
  • 7 edits in trunk

WebCore:

2009-11-13 Dimitri Glazkov <dglazkov@chromium.org>

Unreviewed, build fix.

[Chromium] Fix build to catch up with http://trac.webkit.org/changeset/50973.
This is just enough changes to unbreak the port.

  • bindings/v8/ScriptController.cpp: (WebCore::mainThreadNormalWorld):
  • bindings/v8/V8Proxy.cpp: (WebCore::V8Proxy::initContextIfNeeded):
  • loader/FrameLoaderClient.h:

WebKit/chromium:

2009-11-13 Dimitri Glazkov <dglazkov@chromium.org>

Unreviewed, build fix.

[Chromium] Rename windowObjectCleared to dispatchDidClearWindowObjectInWorld
to catch up http://trac.webkit.org/changeset/50973.

  • src/FrameLoaderClientImpl.cpp: (WebKit::FrameLoaderClientImpl::dispatchDidClearWindowObjectInWorld):
  • src/FrameLoaderClientImpl.h:
19:34 Changeset [50992] by aa@chromium.org
  • 2 edits in trunk/WebCore

2009-11-13 Aaron Boodman <aa@chromium.org>

Unreviewed fix for Chromium build.

  • loader/FrameLoaderClient.h: (WebCore::FrameLoaderClient::dispatchDidClearWindowObjectInWorld): Provide an empty implementation of this method because I don't know what it is supposed to do on Chromium.
19:11 Changeset [50991] by aa@chromium.org
  • 2 edits in trunk/WebCore

2009-11-13 Aaron Boodman <aa@chromium.org>

Unreviewed fix for Chromium build.

  • loader/FrameLoader.h: Make dispatchDidClearWindowObjectsInAllWorlds() public, as Chromium's V8Proxy calls it.
18:58 Changeset [50990] by aa@chromium.org
  • 2 edits in trunk/WebCore

2009-11-13 Aaron Boodman <aa@chromium.org>

Unreviewed fix to Chromium build.

  • bindings/v8/ScriptController.cpp: (WebCore::ScriptController::getAllWorlds):
18:45 Changeset [50989] by aa@chromium.org
  • 2 edits in trunk/WebCore

2009-11-13 Aaron Boodman <aa@chromium.org>

Unreviewed fix for Chromium build.

  • platform/text/TextBoundaries.cpp: Use longer path to refer to Unicode.h.
18:29 Changeset [50988] by abarth@webkit.org
  • 4 edits in trunk/WebCore

2009-11-13 Adam Barth <abarth@webkit.org>

Unreviewed partial build fix for Chromium. Should fix failure #4.

  • bindings/v8/ScriptController.cpp: (WebCore::ScriptController::getAllWorlds):
  • bindings/v8/ScriptController.h:
  • bindings/v8/V8Proxy.cpp: (WebCore::V8Proxy::initContextIfNeeded):
18:18 Changeset [50987] by abarth@webkit.org
  • 2 edits in trunk/WebCore

2009-11-13 Adam Barth <abarth@webkit.org>

Unreviewed partial build fix for Chromium.

  • bindings/v8/ScriptController.h: (WebCore::ScriptController::getAllWorlds):
17:47 Changeset [50986] by ap@apple.com
  • 2 edits in trunk/LayoutTests

Disable websocket tests on Tiger, as they time out.

Filed <https://bugs.webkit.org/show_bug.cgi?id=31501> to investigate this issue.

  • platform/mac-tiger/Skipped:
17:40 Changeset [50985] by eric@webkit.org
  • 2 edits in trunk/WebKit/chromium

2009-11-13 Yaar Schnitman <yaar@chromium.org>

Reviewed by Dimitri Glazkov.

Chromium WebKit Port to compile as dynamic lib

https://bugs.webkit.org/show_bug.cgi?id=31478

  • WebKit.gyp:
17:33 Changeset [50984] by kevino@webkit.org
  • 2 edits in trunk/WebKit/wx

wx build fix after r50923 change to externalRepresentation.

17:22 Changeset [50983] by aa@chromium.org
  • 4 edits in trunk/WebKit/chromium

woo

17:07 Changeset [50982] by zoltan@webkit.org
  • 3 edits in trunk/JavaScriptCore

https://bugs.webkit.org/show_bug.cgi?id=31050

Patch by Zoltan Herczeg <zherczeg@inf.u-szeged.hu> on 2009-11-13
Reviewed by Gavin Barraclough.

Adding optimization support for mode JSVALUE32_64
on ARM systems.

  • jit/JIT.h:
  • jit/JITCall.cpp:

(JSC::JIT::compileOpCall):

  • jit/JITPropertyAccess.cpp:

(JSC::JIT::emit_op_method_check):
(JSC::JIT::compileGetByIdHotPath):
(JSC::JIT::compileGetByIdSlowCase):
(JSC::JIT::emit_op_put_by_id):

16:44 Changeset [50981] by zoltan@webkit.org
  • 8 edits in trunk/JavaScriptCore

https://bugs.webkit.org/show_bug.cgi?id=31050

Patch by Zoltan Herczeg <zherczeg@inf.u-szeged.hu> on 2009-11-14
Reviewed by Gavin Barraclough.

Adding JSVALUE32_64 support for ARM (but not turning it
on by default). All optimizations must be disabled, since
this patch is only the first of a series of patches.

During the work, a lot of x86 specific code revealed and
made platform independent.
See revisions: 50531 50541 50593 50594 50595

  • assembler/ARMAssembler.h:

(JSC::ARMAssembler::):
(JSC::ARMAssembler::fdivd_r):

  • assembler/MacroAssemblerARM.h:

(JSC::MacroAssemblerARM::lshift32):
(JSC::MacroAssemblerARM::neg32):
(JSC::MacroAssemblerARM::rshift32):
(JSC::MacroAssemblerARM::branchOr32):
(JSC::MacroAssemblerARM::set8):
(JSC::MacroAssemblerARM::setTest8):
(JSC::MacroAssemblerARM::loadDouble):
(JSC::MacroAssemblerARM::divDouble):
(JSC::MacroAssemblerARM::convertInt32ToDouble):
(JSC::MacroAssemblerARM::zeroDouble):

  • jit/JIT.cpp:
  • jit/JIT.h:
  • jit/JITOpcodes.cpp:

(JSC::JIT::privateCompileCTIMachineTrampolines):

  • jit/JITStubs.cpp:
  • wtf/StdLibExtras.h:
16:39 Changeset [50980] by eric@webkit.org
  • 3 edits in trunk/WebCore

2009-11-13 Eric Seidel <eric@webkit.org>

No review, build fix only.

Fix Windows and Chromium builds after http://trac.webkit.org/changeset/50977.

Unify TextBoundaries implementations by only relying on WTF Unicode abstractions
https://bugs.webkit.org/show_bug.cgi?id=31468

  • WebCore.gypi: Rename TextBoundariesICU -> TextBoundaries
  • WebCore.vcproj/WebCore.vcproj: Rename TextBoundariesICU -> TextBoundaries
16:27 Changeset [50979] by jorlow@chromium.org
  • 5 edits in trunk

2009-11-12 Jeremy Orlow <jorlow@chromium.org>

Reviewed by Dmitry Titov.

LocalStorage quota should include key sizes in its count
https://bugs.webkit.org/show_bug.cgi?id=31451

  • storage/StorageMap.cpp: (WebCore::StorageMap::setItem):

Count keys in the quota when adding a new item.

(WebCore::StorageMap::removeItem):

Remove the key's length from the quota if we're removing the item.

(WebCore::StorageMap::importItem):

Assume that we're adding things for the first time.
Count keys in the quota.

2009-11-12 Jeremy Orlow <jorlow@chromium.org>

Reviewed by Dmitry Titov.

Now that we're tracking key size in the quota, we can't fit as much in.
https://bugs.webkit.org/show_bug.cgi?id=31451

  • storage/domstorage/quota-expected.txt:
  • storage/domstorage/script-tests/quota.js: (testQuota):
16:25 Changeset [50978] by eric@webkit.org
  • 2 edits in trunk/WebKitSite

2009-11-13 Chris Jerdonek <chris.jerdonek@gmail.com>

Reviewed by Darin Adler.

Added to the WebKit Coding Style Guidelines not to use
"using namespace" statements in header files.

  • coding/coding-style.html:
16:09 Changeset [50977] by eric@webkit.org
  • 8 edits
    1 move in trunk

2009-11-13 Dominik Röttsches <dominik.roettsches@access-company.com>

Reviewed by Eric Seidel.

Unify TextBoundaries implementations by only relying on WTF Unicode abstractions
https://bugs.webkit.org/show_bug.cgi?id=31468

Adding isAlphanumeric abstraction, required
by TextBoundaries.cpp.

  • wtf/unicode/glib/UnicodeGLib.h: (WTF::Unicode::isAlphanumeric):
  • wtf/unicode/icu/UnicodeIcu.h: (WTF::Unicode::isAlphanumeric):

2009-11-13 Dominik Röttsches <dominik.roettsches@access-company.com>

Reviewed by Eric Seidel.

Unify TextBoundaries implementations by only relying on WTF Unicode abstractions
https://bugs.webkit.org/show_bug.cgi?id=31468

Moving TextBoundariesICU.cpp to TextBoundaries.cpp
by removing the direct ICU dependency and replacing it
with WTF functions and WebCore's own TextBreakIterator
abstractions.

  • GNUmakefile.am:
  • platform/graphics/gtk/SimpleFontDataGtk.cpp:
  • platform/text/TextBoundaries.cpp: Added. (WebCore::findNextWordFromIndex): (WebCore::findWordBoundary):
  • platform/text/TextBoundariesICU.cpp: Removed.
  • platform/text/TextBreakIterator.h:
  • platform/text/TextBreakIteratorICU.cpp: (WebCore::textBreakLast): (WebCore::textBreakPrevious):
16:07 Changeset [50976] by hamaji@chromium.org
  • 4 edits in trunk

2009-11-13 Shinichiro Hamaji <hamaji@chromium.org>

Reviewed by Darin Adler.

WebCore::externalRepresentation should update layout before getting render object
https://bugs.webkit.org/show_bug.cgi?id=31459

Before this change, the render tree output for this test was wrong.
Note that #FF0000 (red) becomes #008000 (green) now.

  • platform/mac/svg/custom/gradient-stop-style-change-expected.txt:

2009-11-13 Shinichiro Hamaji <hamaji@chromium.org>

Reviewed by Darin Adler.

WebCore::externalRepresentation should update layout before getting render object
https://bugs.webkit.org/show_bug.cgi?id=31459

  • rendering/RenderTreeAsText.cpp: (WebCore::externalRepresentation):
16:06 Changeset [50975] by andersca@apple.com
  • 5 edits in trunk/WebKitLibraries

Update WebKitSystemInterface.

Reviewed by Dan Bernstein.

  • WebKitSystemInterface.h:
  • libWebKitSystemInterfaceLeopard.a:
  • libWebKitSystemInterfaceSnowLeopard.a:
  • libWebKitSystemInterfaceTiger.a:
15:56 Changeset [50974] by aroben@apple.com
  • 2 edits in trunk/WebKit/win

Build fix

  • Interfaces/WebKit.idl: Touch this to force interfaces to rebuild.
15:49 Changeset [50973] by aroben@apple.com
  • 45 edits
    2 adds in trunk

Tell the WebFrameLoadDelegate when window objects in isolated worlds are cleared

Fixes <http://webkit.org/b/31124>.

Reviewed by Dave Hyatt.

WebCore:

Tell FrameLoaderClient when window objects in isolated worlds are
cleared

Test: http/tests/security/isolatedWorld/didClearWindowObject.html

  • bindings/js/JSDOMBinding.h:

(WebCore::WebCoreJSClientData::getAllWorlds): Added. Copies all the
worlds in m_worldSet to the passed-in Vector.

  • bindings/js/ScriptController.cpp:

(WebCore::ScriptController::getAllWorlds): Added. Calls through to
WebCoreJSClientData.
(WebCore::ScriptController::initScript): Changed to call
FrameLoader::dispatchDidClearWindowObjectInWorld.

  • bindings/js/ScriptController.h: Added getAllWorlds.
  • loader/EmptyClients.h:

(WebCore::EmptyFrameLoaderClient::dispatchDidClearWindowObjectInWorld):
Updated for FrameLoaderClient change.

  • loader/FrameLoader.cpp:

(WebCore::FrameLoader::receivedFirstData):
(WebCore::FrameLoader::begin):
Changed to call dispatchDidClearWindowObjectsInAllWorlds.

(WebCore::FrameLoader::dispatchDidClearWindowObjectsInAllWorlds):
Added. Retrieves all the worlds, then calls through to
dispatchDidClearWindowObjectInWorld for each one.
(WebCore::FrameLoader::dispatchDidClearWindowObjectInWorld): Replaces
dispatchWindowObjectAvailable. Calls up to the client, then, if the
world is the mainThreadNormalWorld(), tells the Inspector about it,
too.

  • loader/FrameLoader.h: Replaced dispatchWindowObjectAvailable with

dispatchDidClearWindowObjectInWorld. Added
dispatchDidClearWindowObjectsInAllWorlds.

  • loader/FrameLoaderClient.h: Replaced windowObjectCleared with

dispatchDidClearWindowObjectForWorld.

WebKit:

Add WebFrameLoadDelegatePrivat.h to WebKit.xcodeproj

  • WebKit.xcodeproj/project.pbxproj: Added

WebFrameLoadDelegatePrivate.h, which already existed.

WebKit/gtk:

Update for changes to FrameLoaderClient

  • WebCoreSupport/FrameLoaderClientGtk.cpp:

(WebKit::FrameLoaderClient::dispatchDidClearWindowObjectInWorld):

  • WebCoreSupport/FrameLoaderClientGtk.h:

Replaced windowObjectCleared with this function. Does nothing if the
passed-in world is not the mainThreadNormalWorld().

WebKit/haiku:

Update for changes to FrameLoaderClient

  • WebCoreSupport/FrameLoaderClientHaiku.cpp:

(WebCore::FrameLoaderClientHaiku::dispatchDidClearWindowObjectInWorld):

  • WebCoreSupport/FrameLoaderClientHaiku.h:

Replaced windowObjectCleared with this function. Does nothing if the
passed-in world is not the mainThreadNormalWorld().

WebKit/mac:

  • WebCoreSupport/WebFrameLoaderClient.h:
  • WebCoreSupport/WebFrameLoaderClient.mm:

(WebFrameLoaderClient::dispatchDidClearWindowObjectInWorld):
Replaced windowObjectCleared with this function. If the delegate
implements it, call
-webView:didClearWindowObjectForFrame:inScriptWorld:. Otherwise, if
the passed-in world is the mainThreadNormalWorld(), call
-webView:didClearWindowObject:forFrame:.

  • WebView/WebDelegateImplementationCaching.h: Added a new entry in the

frame load delegate implementation cache for the new delegate method.

  • WebView/WebFrameLoadDelegatePrivate.h:
  • WebView/WebScriptWorld.mm:

(allWorlds): Added. Returns a HashMap of all the WebScriptWorlds in
existence.
(-[WebScriptWorld initWithWorld:]): Add ourselves to allWorlds().
(-[WebScriptWorld dealloc]): Remove ourselves from allWorlds().
(+[WebScriptWorld findOrCreateWorld:]): Returns the existing
WebScriptWorld for this DOMWrapperWorld, or a new one if one doesn't
already exist.

  • WebView/WebScriptWorldInternal.h: Declared +findOrCreateWorld:.
  • WebView/WebView.mm:

(-[WebView _cacheFrameLoadDelegateImplementations]): Cache the
implementation of the new frame load delegate method.

WebKit/qt:

Update for changes to FrameLoaderClient

  • WebCoreSupport/FrameLoaderClientQt.cpp:

(WebCore::FrameLoaderClientQt::dispatchDidClearWindowObjectInWorld):

  • WebCoreSupport/FrameLoaderClientQt.h:

Replaced windowObjectCleared with this function. Does nothing if the
passed-in world is not the mainThreadNormalWorld().

WebKit/win:

  • Interfaces/IWebFrameLoadDelegatePrivate2.idl: Added

didClearWindowObjectForFrameInScriptWorld.

  • WebFrame.cpp:

(WebFrame::dispatchDidClearWindowObjectInWorld):

  • WebFrame.h:

Replaced windowObjectCleared with this function. If the delegate
implements IWebFrameLoadDelegatePrivate2, call
didClearWindowObjectForFrameInScriptWorld. Otherwise, if the passed-in
world is the mainThreadNormalWorld(), call
didClearWindowObjectForFrame.

  • WebScriptWorld.cpp:

(allWorlds): Added. Returns a HashMap of all the WebScriptWorlds in
existence.
(WebScriptWorld::WebScriptWorld): Add ourselves to allWorlds().
(WebScriptWorld::~WebScriptWorld): Remove ourselves from allWorlds().
(WebScriptWorld::standardWorld): Added this non-COM getter for the
standard world, which the COM getter now calls through to.
(WebScriptWorld::findOrCreateWorld): Added. Returns the existing
WebScriptWorld for this DOMWrapperWorld, or a new one if one doesn't
already exist.

  • WebScriptWorld.h: Added new standardWorld overload, made one

overload of createInstance private, and added findOrCreateWorld.

WebKit/wx:

Update for changes to FrameLoaderClient

  • WebKitSupport/FrameLoaderClientWx.cpp:

(WebCore::FrameLoaderClientWx::dispatchDidClearWindowObjectInWorld):

  • WebKitSupport/FrameLoaderClientWx.h:

Replaced windowObjectCleared with this function. Does nothing if the
passed-in world is not the mainThreadNormalWorld().

WebKitTools:

Add support for
http/tests/security/isolatedWorld/didClearWindowObject.html

  • DumpRenderTree/mac/DumpRenderTreeMac.h: Declared worldIDForWorld.
  • DumpRenderTree/mac/FrameLoadDelegate.mm:

(-[FrameLoadDelegate didClearWindowObjectInStandardWorldForFrame:]):
Added. Moved code here from -webView:didClearWindowObject:forFrame:.
(-[FrameLoadDelegate didClearWindowObjectForFrame:inIsolatedWorld:]):
Added. Sets a worldID property on the global object whose value is
the ID of this world.
(-[FrameLoadDelegate webView:didClearWindowObjectForFrame:inScriptWorld:]):
Respond to this new delegate callback by calling through to one of the
above new methods.

  • DumpRenderTree/mac/LayoutTestControllerMac.mm:

(worldMap): Added. Returns a HashMap containing all the worlds we've
created.
(worldIDForWorld): Added. Returns the ID for this world, or 0 if we
haven't kept track of this world.
(LayoutTestController::evaluateScriptInIsolatedWorld): Use worldMap()
instead of declaring our own.

  • DumpRenderTree/win/DumpRenderTreeWin.h: Declared worldIDForWorld.
  • DumpRenderTree/win/FrameLoadDelegate.cpp:

(FrameLoadDelegate::didClearWindowObject): Moved code from hear to
didClearWindowObjectForFrameInStandardWorld.
(FrameLoadDelegate::didClearWindowObjectForFrameInScriptWorld):
Respond to this new delegate callback by calling through to one of the
below new methods.
(FrameLoadDelegate::didClearWindowObjectForFrameInIsolatedWorld):
Added. Sets a worldID property on the gobal object whose value is
the ID of this world.
(FrameLoadDelegate::didClearWindowObjectForFrameInStandardWorld):
Added. Moved code here from didClearWindowObject.

  • DumpRenderTree/win/FrameLoadDelegate.h: Added the

didClearWindowObjectForFrame* functions.

  • DumpRenderTree/win/LayoutTestControllerWin.cpp:

(worldMap): Added. Returns a HashMap containing all the worlds we've
created.
(worldIDForWorld): Added. Returns the ID of this world, or 0 if we
haven't kept track of this world.
(LayoutTestController::evaluateScriptInIsolatedWorld): Use worldMap()
instead of declaring our own.

LayoutTests:

Test that delegates can modify global objects in isolated worlds

  • http/tests/security/isolatedWorld/didClearWindowObject-expected.txt: Added.
  • http/tests/security/isolatedWorld/didClearWindowObject.html: Added.

Tests that the global object in the normal world does not have a
worldID property, and that the global object in isolated worlds has
a
worldID property whose value is equal to the world's ID.

15:46 Changeset [50972] by dglazkov@chromium.org
  • 2 edits in trunk/WebCore

2009-11-13 Vitaly Repeshko <vitalyr@chromium.org>

Reviewed by Dimitri Glazkov.

[V8] Fix SVG context assignment for pod types.
https://bugs.webkit.org/show_bug.cgi?id=31497

I broke this in r50958.

Tested by svg/custom/viewport-update2.svg.

  • bindings/scripts/CodeGeneratorV8.pm:
15:38 Changeset [50971] by dimich@chromium.org
  • 12 edits in trunk/WebCore

Bring the platform/android files inline with Android 2.0
https://bugs.webkit.org/show_bug.cgi?id=31423

Patch by Andrei Popescu <andreip@google.com> on 2009-11-13
Reviewed by Dmitry Titov.

No new tests required: these are all Android-specific files.

  • platform/android/ClipboardAndroid.cpp:

(WebCore::ClipboardAndroid::files):

  • platform/android/ClipboardAndroid.h:
  • platform/android/FileChooserAndroid.cpp:

(WebCore::FileChooser::basenameForWidth):
(WebCore::fileButtonChooseFileLabel):

  • platform/android/KeyEventAndroid.cpp:

(WebCore::PlatformKeyboardEvent::PlatformKeyboardEvent):

  • platform/android/LocalizedStringsAndroid.cpp:

(WebCore::contextMenuItemTagOpenLinkInNewWindow):
(WebCore::contextMenuItemTagDownloadLinkToDisk):
(WebCore::contextMenuItemTagCopyLinkToClipboard):
(WebCore::contextMenuItemTagOpenImageInNewWindow):
(WebCore::contextMenuItemTagDownloadImageToDisk):
(WebCore::contextMenuItemTagCopyImageToClipboard):
(WebCore::contextMenuItemTagOpenFrameInNewWindow):
(WebCore::contextMenuItemTagCopy):
(WebCore::contextMenuItemTagGoBack):
(WebCore::contextMenuItemTagGoForward):
(WebCore::contextMenuItemTagStop):
(WebCore::contextMenuItemTagReload):
(WebCore::contextMenuItemTagCut):
(WebCore::contextMenuItemTagPaste):
(WebCore::contextMenuItemTagNoGuessesFound):
(WebCore::contextMenuItemTagIgnoreSpelling):
(WebCore::contextMenuItemTagLearnSpelling):
(WebCore::contextMenuItemTagSearchWeb):
(WebCore::contextMenuItemTagLookUpInDictionary):
(WebCore::contextMenuItemTagOpenLink):
(WebCore::contextMenuItemTagIgnoreGrammar):
(WebCore::contextMenuItemTagSpellingMenu):
(WebCore::contextMenuItemTagShowSpellingPanel):
(WebCore::contextMenuItemTagCheckSpelling):
(WebCore::contextMenuItemTagCheckSpellingWhileTyping):
(WebCore::contextMenuItemTagCheckGrammarWithSpelling):
(WebCore::contextMenuItemTagFontMenu):
(WebCore::contextMenuItemTagBold):
(WebCore::contextMenuItemTagItalic):
(WebCore::contextMenuItemTagUnderline):
(WebCore::contextMenuItemTagOutline):
(WebCore::contextMenuItemTagWritingDirectionMenu):
(WebCore::contextMenuItemTagTextDirectionMenu):
(WebCore::contextMenuItemTagDefaultDirection):
(WebCore::contextMenuItemTagLeftToRight):
(WebCore::contextMenuItemTagRightToLeft):
(WebCore::mediaElementLoadingStateText):
(WebCore::mediaElementLiveBroadcastStateText):
(WebCore::searchableIndexIntroduction):
(WebCore::resetButtonDefaultLabel):
(WebCore::submitButtonDefaultLabel):
(WebCore::inputElementAltText):

  • platform/android/RenderThemeAndroid.cpp:

(WebCore::RenderTheme::themeForPage):
(WebCore::RenderThemeAndroid::baselinePosition):
(WebCore::RenderThemeAndroid::paintButton):
(WebCore::adjustMenuListStyleCommon):
(WebCore::RenderThemeAndroid::paintCombo):

  • platform/android/ScreenAndroid.cpp:
  • platform/android/ScrollViewAndroid.cpp:

(WebCore::ScrollView::platformOffscreenContentRectangle):

  • platform/android/TemporaryLinkStubs.cpp:

(JSC::Bindings::dispatchJNICall):

  • platform/android/WidgetAndroid.cpp:

(WebCore::Widget::setFrameRect):

15:33 UsingGitWithWebKit edited by eric@webkit.org
(diff)
15:21 Changeset [50970] by eric@webkit.org
  • 4 edits in trunk

2009-11-13 Norbert Leser <norbert.leser&nokia.com>

Reviewed by Eric Seidel.

Added macros for USERINCLUDE paths within symbian blocks
to guarantee inclusion of respective header files from local path
first (to avoid clashes with same names of header files in system include path).

2009-11-13 Norbert Leser <norbert.leser&nokia.com>

Reviewed by Eric Seidel.

Added macros for USERINCLUDE paths within symbian blocks
to guarantee inclusion of respective header files from local path
first (to avoid clashes with same names of header files in system include path).

  • WebCore.pro:
15:20 Changeset [50969] by dumi@chromium.org
  • 2 edits in trunk/LayoutTests

Remove the unnecessary call to dumpDatabaseCallbacks().

Reviewed by Eric Seidel.

https://bugs.webkit.org/show_bug.cgi?id=31495

  • storage/hash-change-with-xhr.html:
13:51 Changeset [50968] by dimich@chromium.org
  • 23 edits
    5 adds in trunk

Implement composition events introduced in DOM Level 3.
This change adds a new IDL which defines the composition events, adds a class which
implements the composition events, and sends the composition events according to
the specification.
https://bugs.webkit.org/show_bug.cgi?id=26310

Patch by Hironori Bono <hbono@chromium.org> on 2009-11-13
Reviewed by Oliver Hunt.

WebCore:

Test: fast/events/ime-composition-events-001.html

  • DerivedSources.make: Added CompositionEvent so we can compile "CompositionEvent.idl".
  • GNUmakefile.am: Added "CompositionEvent.{cpp,h,idl}".
  • WebCore.gypi: ditto.
  • WebCore.pro: ditto
  • WebCore.vcproj/WebCore.vcproj: ditto.
  • WebCore.xcodeproj/project.pbxproj: Added "CompositionEvent.{cpp,h,idl}" and "JSCompositionEvent.{cpp,h}".
  • WebCoreSources.bkl: Added "JSCompositionEvent.{cpp,h}".
  • bindings/js/JSEventCustom.cpp:

(WebCore::toJS): Call isCompositionEvent() to create the CompositionEvent wrapper.

  • bindings/v8/DOMObjectsInclude.h: Added "JSCompositionEvent.h".
  • bindings/v8/DerivedSourcesAllInOne.cpp: Added "JSCompositionEvent.cpp".
  • bindings/v8/V8DOMWrapper.cpp: Call isCompositionEvent() to identify an event as a CompositionEvent.

(WebCore::V8DOMWrapper::convertEventToV8Object):

  • bindings/v8/V8Index.cpp: Added "V8CompositionEvent.h".
  • bindings/v8/V8Index.h: Added V8Index::COMPOSITIONEVENT.
  • dom/CompositionEvent.cpp: Implements the CompositionEvent class.

(WebCore::CompositionEvent::CompositionEvent):
(WebCore::CompositionEvent::~CompositionEvent):
(WebCore::CompositionEvent::initCompositionEvent):
(WebCore::CompositionEvent::isCompositionEvent):

  • dom/CompositionEvent.h: Declares the CompositionEvent class.

(WebCore::CompositionEvent::create):
(WebCore::CompositionEvent::data):

  • dom/CompositionEvent.idl: Added the IDL of DOM CompositionEvent.
  • dom/Event.cpp:

(WebCore::Event::isCompositionEvent): Added a method to identify an event is a CompositionEvent.

  • dom/Event.h:
  • dom/EventNames.h: Added composition{start,update,end} to eventNames.
  • editing/Editor.cpp:

(WebCore::Editor::confirmComposition): Sent a CompositionEnd event.
(WebCore::Editor::setComposition): Sent a Composition{Start,Update,End} event.

LayoutTests:

  • fast/events/ime-composition-events-001-expected.txt: Added.
  • fast/events/ime-composition-events-001.html: Added.
  • platform/gtk/Skipped: Skip this test because TextInputController is not implemented.
  • platform/qt/Skipped: ditto.
  • platform/win/Skipped: ditto.
13:26 Changeset [50967] by snej@chromium.org
  • 2 edits in trunk/WebCore

Fix a link error in unofficial GCC 4.4 builds on Linux.
https://bugs.webkit.org/show_bug.cgi?id=31477

Reviewed by Dimitri Glazkov.

  • bindings/v8/V8Binding.cpp: Add explicit instantiations of v8StringToWebCoreString template.
12:55 Changeset [50966] by eric@webkit.org
  • 3 edits in trunk/WebCore

2009-11-13 Eric Seidel <eric@webkit.org>

No review, build fix only.

Fix Debug build after http://trac.webkit.org/changeset/50960.

The CounterNode class does not support all methods necessary to efficiently update the counter tree as needed per CSS2.1
https://bugs.webkit.org/show_bug.cgi?id=31213

  • rendering/CounterNode.cpp: (WebCore::showTreeAndMark):
  • rendering/RenderCounter.cpp: (WebCore::destroyCounterNodeChildren):
12:50 Changeset [50965] by kenneth@webkit.org
  • 3 edits
    1 add in trunk

WebKitTools: Make the Qt Linux only --valgrind feature, suppress errors
defined in the SuppressedValgrindErrors file.

Reviewed by Simon Hausmann.

  • Scripts/run-webkit-tests:

LayoutTests: Add a file for suppressing invalid valgrind errors (Qt / Linux only)

Reviewed by Simon Hausmann.

  • platform/qt/SuppressedValgrindErrors: Added.
12:42 Changeset [50964] by oliver@apple.com
  • 4 edits in trunk/JavaScriptCore

JSValueProtect and JSValueUnprotect don't protect API wrapper values
https://bugs.webkit.org/show_bug.cgi?id=31485

Reviewed by Geoff Garen.

Make JSValueProtect/Unprotect use a new 'toJS' function, 'toJSForGC' that
does not attempt to to strip out API wrapper objects.

12:35 Changeset [50963] by krit@webkit.org
  • 2 edits in trunk/WebCore

2009-11-13 Dirk Schulze <krit@webkit.org>

Reviewed by Gustavo Noronha.

[CAIRO] shadow support for Canvas and SVG
https://bugs.webkit.org/show_bug.cgi?id=30960

We currently fill a path with solid color instead of filling
a clipping path. This causes problems on some composite operators,
since Cairo modifies the area outside the path.
This fixes the behavior of WebKitGtk on fast/canvas/canvas-composite-alpha.html

Thanks to Benjamin Otte for tracking the bug down.

  • platform/graphics/cairo/GraphicsContextCairo.cpp: (WebCore::setPlatformFill):
12:29 Changeset [50962] by eric@webkit.org
  • 2 edits in trunk/JavaScriptCore

2009-11-13 İsmail Dönmez <ismail@namtrac.org>

Reviewed by Antti Koivisto.

Fix typo, ce_time.cpp should be ce_time.c

12:25 Changeset [50961] by dumi@chromium.org
  • 11 edits
    1 add in trunk

Adding Chromium's DatabaseTracker implementation.

Reviewed by Dimitri Glazkov.

https://bugs.webkit.org/show_bug.cgi?id=31440

WebCore:

  • WebCore.gyp/WebCore.gyp:
  • WebCore.gypi:
  • storage/DatabaseTracker.h:
  • storage/chromium/DatabaseTrackerChromium.cpp:

(WebCore::DatabaseTracker::fullPathForDatabase):
(WebCore::DatabaseTracker::getMaxSizeForDatabase):

  • storage/chromium/QuotaTracker.cpp:

(WebCore::QuotaTracker::updateDatabaseSize):

  • storage/chromium/QuotaTracker.h:

WebKit/chromium:

  • WebKit.gyp:
  • src/ChromeClientImpl.cpp:

(WebKit::ChromeClientImpl::exceededDatabaseQuota):

  • src/DatabaseObserver.cpp: Added.

(WebCore::DatabaseObserver::databaseOpened):
(WebCore::DatabaseObserver::databaseModified):
(WebCore::DatabaseObserver::databaseClosed):

  • src/WebDatabase.cpp:

(WebKit::WebDatabase::name):
(WebKit::WebDatabase::displayName):
(WebKit::WebDatabase::estimatedSize):
(WebKit::WebDatabase::securityOrigin):
(WebKit::WebDatabase::updateDatabaseSize):

  • src/WebSecurityOrigin.cpp:

(WebKit::WebSecurityOrigin::toString):
(WebKit::WebSecurityOrigin::databaseIdentifier):

12:21 Changeset [50960] by eric@webkit.org
  • 7 edits in trunk/WebCore

2009-11-13 Carol Szabo <carol.szabo@nokia.com>

Reviewed by Darin Adler.

The CounterNode class is missing some basic tree navigation methods common in other WebKit trees such as the rendering tree
https://bugs.webkit.org/show_bug.cgi?id=31213
Added tree navigation methods that permit full implementation of CSS2.1
counter feature without using recursion proportional to the counter
tree depth.
No new tests because I did not find any bug that is fixed by this
commit yet, this just reduces the size of the patch for 11031 and
helps respond to some concerns regarding that patch.

  • rendering/CounterNode.cpp: (WebCore::CounterNode::CounterNode):

(WebCore::CounterNode::nextInPreOrderAfterChildren):
(WebCore::CounterNode::nextInPreOrder):
Added to support non-recursive tree traversal necessary for
efficient full implementation of CSS2.1 counters.

(WebCore::CounterNode::lastDescendant):
(WebCore::CounterNode::previousInPreOrder):
Moved this methods such that they occupy a place similar to that of
identically named methods on the render tree. This allows for their
broader use needed in full implementation of CSS2.1 counters.

(WebCore::CounterNode::resetRenderer):
(WebCore::CounterNode::resetRenderers):
(WebCore::CounterNode::recount):
(WebCore::CounterNode::insertAfter):
(WebCore::CounterNode::removeChild):
Changed such that insertion/removal of a counter, triggers not only
recalculation of PrefixWidths, but also reassesment of values in
counter nodes. This is the basis full implementation of CSS2.1
counters. It does not change current behavior by much because of
changes needed to the recalculation algorithm, but those are comming
in the patch for 11031.
(WebCore::showTreeAndMark):

  • rendering/CounterNode.h:
  • rendering/RenderCounter.cpp: (WebCore::counter): Only changed argument type to prepare for implementation of Darin Adler's recommendation for the patch to 11031.

(WebCore::RenderCounter::invalidate):
(WebCore::destroyCounterNodeChildren):
(WebCore::RenderCounter::destroyCounterNodes):

  • rendering/RenderCounter.h:
  • rendering/RenderObjectChildList.cpp: (WebCore::invalidateCountersInContainer): (WebCore::RenderObjectChildList::invalidateCounters):
  • rendering/RenderObjectChildList.h: Added the ability to restrict invalidation to counters with a given identifier. Also invalidated counters that are on the child container itself which were missed by the previous algorithm, but were a valid case.
12:04 Changeset [50959] by kenneth@webkit.org
  • 2 edits in trunk/WebKitTools

Set the locale to C and not to the current one in use on the system.

Reviewed by Simon Hausmann.

  • DumpRenderTree/qt/DumpRenderTree.cpp:

(WebCore::DumpRenderTree::resetToConsistentStateBeforeTesting):

11:56 Changeset [50958] by eric@webkit.org
  • 3 edits in trunk/WebCore

2009-11-13 Vitaly Repeshko <vitalyr@chromium.org>

Reviewed by Dimitri Glazkov.

[V8] Protect SVG animated properties from destruction in bindings.
https://bugs.webkit.org/show_bug.cgi?id=31474

See http://crbug.com/26719.

Tested by LayoutTests/svg/custom/js-update-transform-addition.svg
under Valgrind.

Made sure we keep a reference to SVG properties while setting a
context:

  • bindings/scripts/CodeGeneratorV8.pm:
  • bindings/v8/V8Proxy.h: (WebCore::V8Proxy::withSVGContext):
11:30 Changeset [50957] by ap@apple.com
  • 2 edits in trunk/LayoutTests

Disable websocket tests on Windows, as the server cannot start on buildbot.

Filed <https://bugs.webkit.org/show_bug.cgi?id=31479> to investigate this issue.

  • platform/win/Skipped:
11:21 Changeset [50956] by bfulgham@webkit.org
  • 2 edits in trunk/WebCore

[CAIRO] shadow support for Canvas and SVG.
https://bugs.webkit.org/show_bug.cgi?id=30960

Reviewed by Alexey Proskuryakov.

Incorporate Benjamin Otte's recommendations to avoid
a buffer overrun, and small performance improvement.

  • platform/graphics/cairo/GraphicsContextCairo.cpp:

(WebCore::copyContextProperties): Correctly size output

storage for cairo_get_dash to avoid buffer overrun.

(WebCore::drawPathShadow): Prefer cairo_fill_extents

to slower cairo_stroke_extents when not drawing shadows.

11:19 Changeset [50955] by ap@apple.com
  • 1 edit in trunk/WebCore/ChangeLog

Restoring parst of ChangeLog that were somehow deleted with my latest commit.

11:14 Changeset [50954] by dumi@chromium.org
  • 4 edits in trunk/WebCore

Do not register Chromium's HTML5 DB VFS as the default
VFS. Otherwise, other sqlite DB users in the same process will
stop working.

Reviewed by Dimitri Glazkov.

https://bugs.webkit.org/show_bug.cgi?id=31462

  • platform/sql/chromium/SQLiteFileSystemChromium.cpp:

(WebCore::SQLiteFileSystem::openDatabase):

  • platform/sql/chromium/SQLiteFileSystemChromiumPosix.cpp:

(WebCore::SQLiteFileSystem::registerSQLiteVFS):

  • platform/sql/chromium/SQLiteFileSystemChromiumWin.cpp:

(WebCore::SQLiteFileSystem::registerSQLiteVFS):

10:51 Changeset [50953] by ap@apple.com
  • 2 edits in trunk/WebCore

Windows build fix

  • platform/network/cf/SocketStreamHandleCFNet.cpp: (WebCore::SocketStreamHandle::SocketStreamHandle): Explicitly cast "-1" to CFOptionFlags, avoiding a sign mismatch warning.
10:47 Changeset [50952] by ap@apple.com
  • 2 edits in trunk/WebCore

Release build fix.

  • platform/network/cf/SocketStreamHandleCFNet.cpp: (WebCore::SocketStreamHandle::readStreamCallback): (WebCore::SocketStreamHandle::writeStreamCallback): Use ASSERT_UNUSED for unused stream parameter.
10:14 Changeset [50951] by ap@apple.com
  • 7 edits in trunk

Reviewed by Darin Adler.

https://bugs.webkit.org/show_bug.cgi?id=31441
Implement SocketStreamHandleCFNet

Existing WebSocket tests now pass on Mac. No proxy support yet.

  • platform/network/SocketStreamHandleClient.h:
  • platform/network/cf/SocketStreamHandle.h: (WebCore::SocketStreamHandle::shouldUseSSL): (WebCore::SocketStreamHandle::refAuthenticationClient): (WebCore::SocketStreamHandle::derefAuthenticationClient): (WebCore::SocketStreamHandle::):
  • platform/network/cf/SocketStreamHandleCFNet.cpp: (WebCore::SocketStreamHandle::SocketStreamHandle): (WebCore::SocketStreamHandle::chooseProxy): (WebCore::SocketStreamHandle::createStreams): (WebCore::SocketStreamHandle::copyCFStreamDescription): (WebCore::SocketStreamHandle::readStreamCallback): (WebCore::SocketStreamHandle::writeStreamCallback): (WebCore::SocketStreamHandle::~SocketStreamHandle): (WebCore::SocketStreamHandle::platformSend): (WebCore::SocketStreamHandle::platformClose): (WebCore::SocketStreamHandle::receivedCredential): (WebCore::SocketStreamHandle::receivedRequestToContinueWithoutCredential): (WebCore::SocketStreamHandle::receivedCancellation):
10:07 Changeset [50950] by dimich@chromium.org
  • 2 edits in trunk/LayoutTests

Fix worker-timeout.html layout test that fails in Chromium harness.
https://bugs.webkit.org/show_bug.cgi?id=31452

Reviewed by Darin Adler.

  • fast/workers/worker-timeout.html:

Stop logging timer messages at the same time as calling NotifyDone() because
in test harness Chromium uses for workers it does not capture the output immediately.

09:51 Changeset [50949] by bfulgham@webkit.org
  • 2 edits in trunk/WebKitLibraries

Build correction. No review.

The WinCairo build requires ENABLE_FILTERS to work properly.

  • win/tools/vsprops/WinCairo.vsprops:
09:50 Changeset [50948] by eric@webkit.org
  • 2 edits in trunk/WebKit/chromium

2009-11-13 Yaar Schnitman <yaar@chromium.org>

Reviewed by Dimitri Glazkov.

Rolling Chromium dependencies to match Chromium revision 31834.

https://bugs.webkit.org/show_bug.cgi?id=31445

  • DEPS:
09:26 Changeset [50947] by pfeldman@chromium.org
  • 2 edits in trunk/WebCore

2009-11-12 Pavel Feldman <pfeldman@chromium.org>

Reviewed by Timothy Hatcher.

Web Inspector: preload status bar button glyphs
in order to prevent them from flickering (take 2).

https://bugs.webkit.org/show_bug.cgi?id=31439

  • inspector/front-end/inspector.js: (preloadImages):
09:10 Changeset [50946] by pfeldman@chromium.org
  • 3 edits
    2 deletes in trunk

2009-11-13 Pavel Feldman <pfeldman@chromium.org>

Reviewed by Adam Barth.

Chromium: [REGRESSION] Crash while stopping on a breakpoint.
Rolling back r50890.

https://bugs.webkit.org/show_bug.cgi?id=31467

  • bindings/v8/V8Proxy.cpp: (WebCore::V8Proxy::canAccessPrivate):
09:10 Changeset [50945] by aroben@apple.com
  • 2 edits in trunk/WebKit/win

Build fix

  • Interfaces/WebKit.idl: Un-sort the #includes again. Apparently they

are order-dependent!

09:05 Changeset [50944] by aroben@apple.com
  • 2 edits in trunk/WebKit/win

Touch WebKit.idl to fix the Windows build

  • Interfaces/WebKit.idl: Sorted #includes.
08:58 Changeset [50943] by aroben@apple.com
  • 10 edits in trunk

Finish replacing worldIDs with world objects

The only remaining use of worldIDs was in a method only used by DRT
for the isolated worlds tests.

Fixes <http://webkit.org/b/31414> Replace worldIDs with world objects

Reviewed by Mark Rowe.

WebKit/mac:

  • WebView/WebFrame.mm:

(-[WebFrame _stringByEvaluatingJavaScriptFromString:withGlobalObject:inScriptWorld:]):

  • WebView/WebFramePrivate.h:

Renamed from
_stringByEvaluatingJavaScriptInIsolatedWorld:WithGobalObject:FromString:.
Now takes a WebScriptWorld instead of a worldID, so we don't need to
maintain a map of worldID -> world anymore.

WebKit/win:

  • Interfaces/IWebFramePrivate.idl:
  • WebFrame.cpp:

(WebFrame::stringByEvaluatingJavaScriptInScriptWorld):

  • WebFrame.h:

Renamed from stringByEvaluatingJavaScriptInIsolatedWorld. Now takes an
IWebScriptWorld instead of a worldID, so we don't need to maintain a
map of worldID -> world anymore.

WebKitTools:

  • DumpRenderTree/mac/LayoutTestControllerMac.mm:

(LayoutTestController::evaluateScriptInIsolatedWorld):

  • DumpRenderTree/win/LayoutTestControllerWin.cpp:

(LayoutTestController::evaluateScriptInIsolatedWorld):
Updated for changes to WebFrame. Now holds the map of worldID -> world
at this level instead of making WebKit do it.

08:43 Changeset [50942] by vestbo@webkit.org
  • 3 edits in trunk/WebKitTools

[Qt] Ensure the DRT does not output anything until first test is run

Reviewed by Simon Hausmann.

  • DumpRenderTree/qt/DumpRenderTree.cpp:
  • DumpRenderTree/qt/DumpRenderTree.h:
08:11 Changeset [50941] by kenneth@webkit.org
  • 2 edits in trunk/WebKitTools

Clear the undo stack in before each new test run.

Reviewed by Tor Arne Vestbø.

Locally fixes 3 incorrect layouts, 9 timeouts and 7 crashes.

  • DumpRenderTree/qt/DumpRenderTree.cpp:

(WebCore::DumpRenderTree::resetToConsistentStateBeforeTesting):

07:06 Changeset [50940] by vestbo@webkit.org
  • 6 edits in trunk/WebKit/qt

Update documentation for the Qt API

Reviewed by Simon Hausmann.

  • Api/qgraphicswebview.cpp:
  • Api/qwebelement.cpp:
  • Api/qwebframe.cpp:
  • Api/qwebsettings.cpp:
  • Api/qwebview.cpp:
06:57 Changeset [50939] by krit@webkit.org
  • 4 edits in trunk

2009-11-13 Dirk Schulze <krit@webkit.org>

Reviewed by Gustavo Noronha.

[CAIRO] shadow support for Canvas and SVG
https://bugs.webkit.org/show_bug.cgi?id=30960


This is the fix of a regression, caused by the shadow patch
from the bug above. Reinserted the save and restore calls
that were accidently removed by the previous patch.

  • platform/graphics/cairo/GraphicsContextCairo.cpp: (WebCore::setPlatformFill): (WebCore::setPlatformStroke):

Skip fast/canvas/canvas-shadow.html we need to enable filters
on cairo to pass this test.

  • platform/gtk/Skipped:
05:49 Changeset [50938] by eric@webkit.org
  • 5 edits in trunk

2009-11-13 Andras Becsi <becsi.andras@stud.u-szeged.hu>

Reviewed by Kenneth Rohde Christiansen.

[Qt] Enable Page Cache if setMaximumPagesInCache needs it.
This fixes https://bugs.webkit.org/show_bug.cgi?id=31266.

  • Api/qwebsettings.cpp: (QWebSettingsPrivate::apply): (QWebSettings::setMaximumPagesInCache):

2009-11-13 Andras Becsi <becsi.andras@stud.u-szeged.hu>

Reviewed by Kenneth Rohde Christiansen.

[Qt] Add WebKitUsesPageCachePreferenceKey overriding support to DRT's LayoutTestController.

  • DumpRenderTree/qt/DumpRenderTree.cpp: (WebCore::WebPage::resetSettings):
  • DumpRenderTree/qt/LayoutTestControllerQt.cpp: (LayoutTestController::overridePreference):
05:21 Changeset [50937] by eric@webkit.org
  • 2 edits in trunk/WebKit/qt

2009-11-13 Jocelyn Turcotte <jocelyn.turcotte@nokia.com>

Reviewed by Kenneth Rohde Christiansen.

[Qt] Fix initial QWebView focus behavior.

focusController->setFocused(true) was not always called.
https://bugs.webkit.org/show_bug.cgi?id=31466

  • Api/qwebpage.cpp: (QWebPagePrivate::focusInEvent):
05:12 Changeset [50936] by vestbo@webkit.org
  • 2 edits in trunk/WebKit/qt

Use correct install-path when running qdoc3

Reviewed by Simon Hausmann.

  • docs/docs.pri:
02:45 Changeset [50935] by pfeldman@chromium.org
  • 2 edits in trunk/WebCore

2009-11-13 Pavel Feldman <pfeldman@chromium.org>

Not reviewed. Reverting r50908 since it makes inspector tests
time out.

https://bugs.webkit.org/show_bug.cgi?id=31439

  • inspector/front-end/inspector.js:
02:21 Changeset [50934] by kov@webkit.org
  • 2 edits in trunk/LayoutTests

Skip one new test.

01:51 Changeset [50933] by pfeldman@chromium.org
  • 5 edits
    1 add in trunk/WebCore

2009-11-13 Mikhail Naganov <mnaganov@chromium.org>

Reviewed by Timothy Hatcher.

Enable 'console.profile()' and 'console.profileEnd()'
regardless of JAVASCRIPT_DEBUGGER.

https://bugs.webkit.org/show_bug.cgi?id=31293

  • WebCore.gypi:
  • bindings/js/JSConsoleCustom.cpp: (WebCore::JSConsole::profile): (WebCore::JSConsole::profileEnd):
  • bindings/v8/custom/V8ConsoleCustom.cpp: Added. (WebCore::CALLBACK_FUNC_DECL):
  • bindings/v8/custom/V8CustomBinding.h:
  • page/Console.idl:
01:34 Changeset [50932] by levin@chromium.org
  • 2 edits in trunk/WebCore

Chromium build fix.

Reviewed by NOBODY.

  • bindings/v8/ScriptController.h:

(WebCore::ScriptController::evaluateInWorld): Add a dummy

method which isn't called in chromium to make things compile.

00:11 Changeset [50931] by eric@webkit.org
  • 2 edits in trunk/WebKitTools

2009-11-13 Eric Seidel <eric@webkit.org>

Reviewed by Adam Barth.

Update committers.py based on svn records
https://bugs.webkit.org/show_bug.cgi?id=31366

This list was generated using validate-committer-lists from
https://bugs.webkit.org/show_bug.cgi?id=30970
and makes committers.py current for committers who have
committed in the last 3 years.

  • Scripts/modules/committers.py:
Note: See TracTimeline for information about the timeline view.