Context Navigation

Fixes <https://bugs.webkit.org/show_bug.cgi?id=15100>:

XMLHttpRequest::urlMatchesDocumentDomain raises error if port
information does not match exactly

Refactor our security check for XMLHttpRequest into SecurityOrigin so
we can reuse it in other places. This leverages our default port
technology in SecurityOrigin.

I wasn't sure how to write a test for this because the LayoutTests run
on non-default ports.

platform/SecurityOrigin.cpp: (WebCore::SecurityOrigin::canRequest):
platform/SecurityOrigin.h:
xml/XMLHttpRequest.cpp: (WebCore::XMLHttpRequest::send): (WebCore::XMLHttpRequest::willSendRequest):
xml/XMLHttpRequest.h:

10:39 PM Changeset in webkit [34531] by abarth@webkit.org

1 edit
2 adds in trunk/LayoutTests

2008-06-13 Adam Barth <abarth@webkit.org>

Reviewed by Darin Adler.

Test whether we permit XMLHttpRequest once document.domain is set.

http/tests/xmlhttprequest/document-domain-set-expected.txt: Added.
http/tests/xmlhttprequest/document-domain-set.html: Added.

10:39 PM Changeset in webkit [34530] by abarth@webkit.org

3 edits in trunk/WebCore

2008-06-13 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

Removed unnecessary dependencies.

platform/SecurityOrigin.cpp:
platform/SecurityOrigin.h:

10:06 PM Changeset in webkit [34529] by cwzwarich@webkit.org

5 edits in trunk/JavaScriptCore

2008-06-13 Cameron Zwarich <cwzwarich@uwaterloo.ca>

Reviewed by Maciej.

Combine TrueNode and FalseNode to make BooleanNode, and remove the
unused class PlaceholderTrueNode.

kjs/grammar.y:
kjs/nodes.cpp: (KJS::BooleanNode::emitCode):
kjs/nodes.h: (KJS::BooleanNode::): (KJS::BooleanNode::precedence):
kjs/nodes2string.cpp: (KJS::BooleanNode::streamTo):

9:40 PM Changeset in webkit [34528] by cwzwarich@webkit.org

4 edits in trunk/JavaScriptCore

2008-06-13 Cameron Zwarich <cwzwarich@uwaterloo.ca>

Reviewed by Maciej.

Eliminate the use of temporaries to store the left hand side of an
expression when the right hand side is a constant. This slightly
improves the generated bytecode for a few SunSpider tests, but it is
mostly in preparation for fixing

Bug 19484: More instructions needs to use temporary registers
<https://bugs.webkit.org/show_bug.cgi?id=19484>

VM/CodeGenerator.h: (KJS::CodeGenerator::leftHandSideNeedsCopy): (KJS::CodeGenerator::emitNodeForLeftHandSide):
kjs/nodes.cpp: (KJS::BracketAccessorNode::emitCode): (KJS::ReadModifyResolveNode::emitCode): (KJS::AssignDotNode::emitCode): (KJS::ReadModifyDotNode::emitCode): (KJS::AssignBracketNode::emitCode): (KJS::ReadModifyBracketNode::emitCode):
kjs/nodes.h: (KJS::ExpressionNode::): (KJS::FalseNode::): (KJS::TrueNode::): (KJS::NumberNode::): (KJS::StringNode::):

9:12 PM Changeset in webkit [34527] by mjs@apple.com

2 edits in trunk/JavaScriptCore

2008-06-13 Maciej Stachowiak <mjs@apple.com>

Reviewed by Oliver.

prettify opcode stats output

I changed things to be a bit more aligned, also there is a new
section listing most common opcodes and most common sequences that
include them.

VM/Opcode.cpp: (KJS::): (KJS::OpcodeStats::~OpcodeStats):
VM/Opcode.h:

4:35 PM Changeset in webkit [34526] by Darin Adler

6 edits in trunk/WebKit

WebKit/gtk:

2008-06-13 Darin Adler <Darin Adler>

try to fix build

WebCoreSupport/FrameLoaderClientGtk.h: Add missing argument.

WebKit/qt:

2008-06-13 Darin Adler <Darin Adler>

try to fix build

WebCoreSupport/FrameLoaderClientQt.h: Add missing argument.

WebKit/wx:

2008-06-13 Darin Adler <Darin Adler>

try to fix build

WebKitSupport/FrameLoaderClientWx.h: Add missing argument.

3:32 PM Changeset in webkit [34525] by cwzwarich@webkit.org

1 edit
4 adds in trunk/LayoutTests

2008-06-13 Cameron Zwarich <cwzwarich@uwaterloo.ca>

Reviewed by Darin.

Add a test for existing loader behaviour in preparation for:

Bug 19422: Distinct redirects from the same link do not create distinct history items
<https://bugs.webkit.org/show_bug.cgi?id=19422>

http/tests/navigation/redirect-cycle-expected.txt: Added.
http/tests/navigation/redirect-cycle.html: Added.
http/tests/navigation/resources/redirect-cycle-1.pl: Added.
http/tests/navigation/resources/redirect-cycle-2.pl: Added.

3:05 PM Changeset in webkit [34524] by mjs@apple.com

3 edits
13 moves
13 adds in trunk/LayoutTests

2008-06-13 Maciej Stachowiak <mjs@apple.com>

Reviewed by Darin.

enable a bunch of disabled layout tests which now run ok https://bugs.webkit.org/show_bug.cgi?id=19540

I also made a minor change to selftxhtml.js to handle the case
where the body is a frameset properly, and to frame.xhtml to match
the HTML version of that file, and what the tests expect.

dom/html/level2/html/HTMLFormElement10-expected.txt: Added.
dom/html/level2/html/HTMLFormElement10.html: Copied from dom/html/level2/html/HTMLFormElement10.html-disabled.
dom/html/level2/html/HTMLFormElement10.html-disabled: Removed.
dom/xhtml/level2/html/HTMLFormElement10-expected.txt: Added.
dom/xhtml/level2/html/HTMLFormElement10.xhtml: Copied from dom/xhtml/level2/html/HTMLFormElement10.xhtml-disabled.
dom/xhtml/level2/html/HTMLFormElement10.xhtml-disabled: Removed.
dom/xhtml/level2/html/HTMLFrameElement01-expected.txt: Added.
dom/xhtml/level2/html/HTMLFrameElement01.xhtml: Copied from dom/xhtml/level2/html/HTMLFrameElement01.xhtml-disabled.
dom/xhtml/level2/html/HTMLFrameElement01.xhtml-disabled: Removed.
dom/xhtml/level2/html/HTMLFrameElement02-expected.txt: Added.
dom/xhtml/level2/html/HTMLFrameElement02.xhtml: Copied from dom/xhtml/level2/html/HTMLFrameElement02.xhtml-disabled.
dom/xhtml/level2/html/HTMLFrameElement02.xhtml-disabled: Removed.
dom/xhtml/level2/html/HTMLFrameElement03-expected.txt: Added.
dom/xhtml/level2/html/HTMLFrameElement03.xhtml: Copied from dom/xhtml/level2/html/HTMLFrameElement03.xhtml-disabled.
dom/xhtml/level2/html/HTMLFrameElement03.xhtml-disabled: Removed.
dom/xhtml/level2/html/HTMLFrameElement04-expected.txt: Added.
dom/xhtml/level2/html/HTMLFrameElement04.xhtml: Copied from dom/xhtml/level2/html/HTMLFrameElement04.xhtml-disabled.
dom/xhtml/level2/html/HTMLFrameElement04.xhtml-disabled: Removed.
dom/xhtml/level2/html/HTMLFrameElement05-expected.txt: Added.
dom/xhtml/level2/html/HTMLFrameElement05.xhtml: Copied from dom/xhtml/level2/html/HTMLFrameElement05.xhtml-disabled.
dom/xhtml/level2/html/HTMLFrameElement05.xhtml-disabled: Removed.
dom/xhtml/level2/html/HTMLFrameElement06-expected.txt: Added.
dom/xhtml/level2/html/HTMLFrameElement06.xhtml: Copied from dom/xhtml/level2/html/HTMLFrameElement06.xhtml-disabled.
dom/xhtml/level2/html/HTMLFrameElement06.xhtml-disabled: Removed.
dom/xhtml/level2/html/HTMLFrameElement07-expected.txt: Added.
dom/xhtml/level2/html/HTMLFrameElement07.xhtml: Copied from dom/xhtml/level2/html/HTMLFrameElement07.xhtml-disabled.
dom/xhtml/level2/html/HTMLFrameElement07.xhtml-disabled: Removed.
dom/xhtml/level2/html/HTMLFrameElement08-expected.txt: Added.
dom/xhtml/level2/html/HTMLFrameElement08.xhtml: Copied from dom/xhtml/level2/html/HTMLFrameElement08.xhtml-disabled.
dom/xhtml/level2/html/HTMLFrameElement08.xhtml-disabled: Removed.
dom/xhtml/level2/html/HTMLFrameElement09-expected.txt: Added.
dom/xhtml/level2/html/HTMLFrameElement09.xhtml: Copied from dom/xhtml/level2/html/HTMLFrameElement09.xhtml-disabled.
dom/xhtml/level2/html/HTMLFrameElement09.xhtml-disabled: Removed.
dom/xhtml/level2/html/HTMLFrameSetElement01-expected.txt: Added.
dom/xhtml/level2/html/HTMLFrameSetElement01.xhtml: Copied from dom/xhtml/level2/html/HTMLFrameSetElement01.xhtml-disabled.
dom/xhtml/level2/html/HTMLFrameSetElement01.xhtml-disabled: Removed.
dom/xhtml/level2/html/HTMLFrameSetElement02-expected.txt: Added.
dom/xhtml/level2/html/HTMLFrameSetElement02.xhtml: Copied from dom/xhtml/level2/html/HTMLFrameSetElement02.xhtml-disabled.
dom/xhtml/level2/html/HTMLFrameSetElement02.xhtml-disabled: Removed.
dom/xhtml/level2/html/resources/frame.xhtml:
dom/xhtml/level2/html/selfxhtml.js:

2:47 PM Changeset in webkit [34523] by Darin Adler

21 edits in trunk

WebCore:

2008-06-13 Darin Adler <Darin Adler>

Reviewed by John Sullivan.

added FormState argument to action policy functions

loader/FrameLoader.cpp: (WebCore::FrameLoader::checkNewWindowPolicy): (WebCore::FrameLoader::checkNavigationPolicy):
loader/FrameLoaderClient.h:
svg/graphics/SVGImageEmptyClients.h: (WebCore::SVGEmptyFrameLoaderClient::dispatchDecidePolicyForNewWindowAction): (WebCore::SVGEmptyFrameLoaderClient::dispatchDecidePolicyForNavigationAction):

WebKit/gtk:

2008-06-13 Darin Adler <Darin Adler>

Reviewed by John Sullivan.

updated for addition of FormState argument to action policy functions

WebCoreSupport/FrameLoaderClientGtk.cpp: (WebKit::FrameLoaderClient::dispatchDecidePolicyForNewWindowAction): (WebKit::FrameLoaderClient::dispatchDecidePolicyForNavigationAction):

WebKit/mac:

2008-06-13 Darin Adler <Darin Adler>

Reviewed by John Sullivan.

updated for addition of FormState argument to action policy functions
added WebActionFormKey

WebCoreSupport/WebFrameLoaderClient.h:
WebCoreSupport/WebFrameLoaderClient.mm: (WebFrameLoaderClient::dispatchDecidePolicyForNewWindowAction): (WebFrameLoaderClient::dispatchDecidePolicyForNavigationAction): (WebFrameLoaderClient::actionDictionary):
WebKit.exp:
WebView/WebPolicyDelegate.mm:
WebView/WebPolicyDelegatePrivate.h:

WebKit/qt:

2008-06-13 Darin Adler <Darin Adler>

Reviewed by John Sullivan.

updated for addition of FormState argument to action policy functions

WebCoreSupport/FrameLoaderClientQt.cpp: (WebCore::FrameLoaderClientQt::dispatchDecidePolicyForNewWindowAction): (WebCore::FrameLoaderClientQt::dispatchDecidePolicyForNavigationAction):

WebKit/win:

2008-06-13 Darin Adler <Darin Adler>

Reviewed by John Sullivan.

updated for addition of FormState argument to action policy functions
added WebActionFormKey

Interfaces/IWebPolicyDelegate.idl:
WebActionPropertyBag.cpp: (WebActionPropertyBag::WebActionPropertyBag): (WebActionPropertyBag::createInstance): (WebActionPropertyBag::AddRef): (WebActionPropertyBag::Release): (WebActionPropertyBag::Read):
WebActionPropertyBag.h:
WebFrame.cpp: (WebFrame::dispatchDecidePolicyForNewWindowAction): (WebFrame::dispatchDecidePolicyForNavigationAction):

WebKit/wx:

2008-06-13 Darin Adler <Darin Adler>

Reviewed by John Sullivan.

updated for addition of FormState argument to action policy functions

WebKitSupport/FrameLoaderClientWx.cpp: (WebCore::FrameLoaderClientWx::dispatchDecidePolicyForNewWindowAction): (WebCore::FrameLoaderClientWx::dispatchDecidePolicyForNavigationAction):

11:23 AM Changeset in webkit [34522] by Darin Adler

2 edits in trunk/WebKitTools

2008-06-13 Darin Adler <Darin Adler>

Scripts/make-js-test-wrappers: Added three more exceptions.

10:45 AM Changeset in webkit [34521] by kmccullough@apple.com

6 edits in trunk/JavaScriptCore

2008-06-13 Kevin McCullough <kmccullough@apple.com>

Reviewed by Geoff.

<rdar://problem/5969992> JSProfiler: Remove the recursion limit in the
profiler.

Remove recursion from exclude(). This leaves only focus() to fix.

JavaScriptCore.exp: Change the signatures of the exported functions.
profiler/Profile.cpp: (KJS::Profile::forEach): I added a traverseNextNodePreOrder() function and so needed to distinguish the other function by labeling it traverseNextNodePostOrder(). (KJS::Profile::exclude): All new exclude that iteratively walks the tree
profiler/Profile.h: (KJS::Profile::focus): Add a null check for m_head.
profiler/ProfileNode.cpp: (KJS::ProfileNode::traverseNextNodePostOrder): Renamed (KJS::ProfileNode::traverseNextNodePreOrder): Walks the tree in pre- order, where the parent is processed before the children. (KJS::ProfileNode::setTreeVisible): Iterate over the sub-tree and set all of the nodes visible value. This changes another function that used recursion. (KJS::ProfileNode::exclude): Remove recursion from this function. Because we now check for m_visible and we are walking the tree in pre- order we do not need to check if an excluded node is in an excluded sub-tree.
profiler/ProfileNode.h: Added specific selfTime functions to facilitate exclude(). (KJS::ProfileNode::setSelfTime): (KJS::ProfileNode::setActualSelfTime): (KJS::ProfileNode::setVisibleSelfTime):

10:42 AM Changeset in webkit [34520] by justin.garcia@apple.com

2 edits in trunk/WebCore

2008-06-13 Justin Garcia <justin.garcia@apple.com>

Reviewed by Jess.

<rdar://problem/5702248> Crash on Copy Image from the contextual menu on a broken image in Mail

platform/mac/PasteboardMac.mm: (WebCore::Pasteboard::writeImage): Check to see the image loaded successfully before trying to put data for it on the Pasteboard. Mail shouldn't really include the Copy Image menu item, but if it does and the user selects it, we shouldn't crash.

5:51 AM Changeset in webkit [34519] by Simon Hausmann

2 edits in trunk/WebCore

2008-06-11 Marc Ordinas i Llopis <marc.ordinasillopis@collabora.co.uk>

With help from Siraj Razick <siraj.razick@collabora.co.uk>

Reviewed by Simon.

https://bugs.webkit.org/show_bug.cgi?id=19367
[Qt] Plugins are drawn over the scrollbars

1:22 AM WebDevelopers edited by cwzwarich@webkit.org

(diff)

Jun 12, 2008:

9:53 PM Changeset in webkit [34518] by Darin Adler

12 edits in trunk/JavaScriptCore

2008-06-12 Darin Adler <Darin Adler>

Reviewed by Maciej.

https://bugs.webkit.org/show_bug.cgi?id=19434 speed up SunSpider by avoiding some string boxing

Speeds up SunSpider by 1.1%.

Optimized code path for getting built-in properties from strings -- avoid
boxing with a string object in that case. We can make further changes to avoid
even more boxing, but this change alone is a win.

API/JSCallbackObjectFunctions.h: (KJS::JSCallbackObject::staticValueGetter): Use isObject instead of inherits in asssert, since the type of slotBase() is now JSValue, not JSObject. (KJS::JSCallbackObject::staticFunctionGetter): Ditto. (KJS::JSCallbackObject::callbackGetter): Ditto.

kjs/internal.cpp: (KJS::StringImp::getPrimitiveNumber): Updated for change of data member name. (KJS::StringImp::toBoolean): Ditto. (KJS::StringImp::toNumber): Ditto. (KJS::StringImp::toString): Ditto. (KJS::StringInstance::create): Added; avoids a bit of cut and paste code. (KJS::StringImp::toObject): Use StringInstance::create. (KJS::StringImp::toThisObject): Ditto. (KJS::StringImp::lengthGetter): Added. Replaces the getter that used to live in the StringInstance class. (KJS::StringImp::indexGetter): Ditto. (KJS::StringImp::indexNumericPropertyGetter): Ditto. (KJS::StringImp::getOwnPropertySlot): Added. Deals with built in properties of the string class without creating a StringInstance.

kjs/internal.h: (KJS::StringImp::getStringPropertySlot): Added. To be used by both the string and string object getOwnPropertySlot function.

kjs/lookup.h: (KJS::staticFunctionGetter): Updated since slotBase() is now a JSValue rather than a JSObject.

kjs/object.h: Removed PropertySlot::slotBase() function, which can now move back into property_slot.h where it belongs since it doesn't have to cast to JSObject*.

kjs/property_slot.cpp: (KJS::PropertySlot::functionGetter): Updated since slot.slotBase() is now a JSValue* instead of JSObject*. setGetterSlot still guarantees the base is a JSObject*.
kjs/property_slot.h: (KJS::PropertySlot::PropertySlot): Changed base to JSValue* intead of JSCell*. (KJS::PropertySlot::setStaticEntry): Ditto. (KJS::PropertySlot::setCustom): Ditto. (KJS::PropertySlot::setCustomIndex): Ditto. (KJS::PropertySlot::setCustomNumeric): Ditto. (KJS::PropertySlot::slotBase): Moved inline here since it no longer involves a downcast to JSObject*. (KJS::PropertySlot::setBase): Changed to JSValue*.

kjs/string_object.cpp: (KJS::StringInstance::getOwnPropertySlot): Changed to use getStringPropertySlot instead of coding the properties here. This allows sharing the code with StringImp.

kjs/string_object.h: Removed inlineGetOwnPropertySlot, lengthGetter, and indexGetter. Made one of the constructors protected.

kjs/value.h: Made getOwnPropertySlot private in the JSCell class -- this is better since it's not the real JSObject getOwnPropertySlot semantic and most callers shouldn't use it.

9:19 PM Changeset in webkit [34517] by sullivan@apple.com

3 edits in trunk/WebKit/mac

2008-06-12 John Sullivan <sullivan@apple.com>

Reviewed by Dan and Darin

Clear up the confusion about _close (older private method) vs -close (newer public method).

WebView/WebView.mm: (-[WebView _isClosed]): simplified style (-[WebView _close]): added a comment about how clients and subclasses should use -close instead (-[WebView dealloc]): call -close instead of _close, so subclasses that override the public method will have the intended behavior (-[WebView close]): added a comment (-[WebView _windowWillClose:]): call -close instead of _close, so subclasses that override the public method will have the intended behavior

WebView/WebViewPrivate.h: added a comment about how clients and subclasses should use -close instead

8:28 PM Changeset in webkit [34516] by Darin Adler

2 edits in trunk/WebCore

2008-06-12 Darin Adler <Darin Adler>

try to fix no-SVG build

css/CSSComputedStyleDeclaration.cpp: (WebCore::CSSComputedStyleDeclaration::getPropertyCSSValue): Added a case for CSSPropertyWebkitMask.

8:10 PM Changeset in webkit [34515] by Darin Adler

2 edits in trunk/WebCore

2008-06-12 Darin Adler <Darin Adler>

try to fix Windows build

dom/Element.idl: Turn off ElementTraversal functions when generating COM. Not sure why they're failing, but turning them off for now seems OK.

7:34 AM Changeset in webkit [34514] by ap@webkit.org

6 edits in trunk/JavaScriptCore

Reviewed by Maciej.

Preparation to making JavaScript heap per-thread.

kjs/collector.cpp: (KJS::Collector::collect):
kjs/collector.h: (KJS::Collector::markListSet): The collector now holds the list of protected lists itself, to be made per-instance.

kjs/list.h: Changed to hold a pointer to a mark set this list is in, if any. (KJS::List::List): Explicitly initialize m_size with zero, as m_vector.size() is guaranteed to be such anyway. (KJS::List::append): Changed the fast case to only be executed as long as inline buffer is used, because otherwise, we now do more expensive checks.

kjs/list.cpp: (KJS::List::markLists): Renamed from markProtectedListsSlowCase, made it take the list set as a parameter. (KJS::List::slowAppend): If a non-immediate value is appended, the list needs to be added to an appropriate Heap's protected list. For now, a static Collector::markListSet() is used, but the code is layed out in preparation to making the switch to multiple heaps.

JavaScriptCore.exp: Updated export list.

5:52 AM Disk Cache edited by ben@meyerhome.net

(diff)

3:40 AM Changeset in webkit [34513] by abarth@webkit.org

4 edits
2 copies in trunk

WebCore:

2008-06-12 Adam Barth <abarth@webkit.org>

Rubberstamped by Maciej.

Roll 34504 back in.

Test: http/tests/xmlhttprequest/access-control-basic-allow-access-control-origin-header-data-url.html

platform/SecurityOrigin.cpp: (WebCore::SecurityOrigin::toString):

LayoutTests:

2008-06-12 Adam Barth <abarth@webkit.org>

Rubberstamped by Maciej.

Add the data url test back and set more agressive Cache-Control headers.

http/tests/xmlhttprequest/access-control-basic-allow-access-control-origin-header-data-url-expected.txt: Copied from LayoutTests/http/tests/xmlhttprequest/access-control-basic-allow-access-control-origin-header-data-url-expected.txt.
http/tests/xmlhttprequest/access-control-basic-allow-access-control-origin-header-data-url.html: Copied from LayoutTests/http/tests/xmlhttprequest/access-control-basic-allow-access-control-origin-header-data-url.html.
http/tests/xmlhttprequest/resources/access-control-basic-allow-access-control-origin-header.cgi:

3:02 AM Changeset in webkit [34512] by abarth@webkit.org

2 deletes in trunk/LayoutTests/http/tests/xmlhttprequest

Continue backout of 34504.

2:35 AM Changeset in webkit [34511] by abarth@webkit.org

1 edit in trunk/WebCore/platform/SecurityOrigin.cpp

Partial backout of 34504. I know this will regress the *-data-url.html test, but I'd like to see if it fixes the broken test.

2:03 AM Changeset in webkit [34510] by cwzwarich@webkit.org

3 edits
4 adds in trunk

2008-06-12 Cameron Zwarich <cwzwarich@uwaterloo.ca>

Reviewed by Maciej.

Bug 19510: CodeBlock::needsFullScopeChain not always set for global code
<https://bugs.webkit.org/show_bug.cgi?id=19510>

This fixes the symptoms by using CodeGenerator::m_codeType to determine
when to use temporaries instead of CodeBlock::needsFullScopeChain, but
it does not fix the problem itself.

JavaScriptCore:

VM/CodeGenerator.h: (KJS::CodeGenerator::leftHandSideNeedsCopy):

LayoutTests:

fast/js/codegen-temporaries-multiple-global-blocks-expected.txt: Added.
fast/js/codegen-temporaries-multiple-global-blocks.html: Added.
fast/js/resources/codegen-temporaries-multiple-global-blocks-1.js: Added.
fast/js/resources/codegen-temporaries-multiple-global-blocks-2.js: Added.

2:00 AM Changeset in webkit [34509] by abarth@webkit.org

2 edits in trunk/LayoutTests

2008-06-12 Adam Barth <abarth@webkit.org>

Reviewed by Maciej.

Prevent caching of this resource so we always see the up-to-date value
of the Access-Control-Origin header.

http/tests/xmlhttprequest/resources/access-control-basic-allow-access-control-origin-header.cgi:

Note: See TracTimeline for information about the timeline view.

Download in other formats:

RSS Feed