Context Navigation

Fix for radar 4611287, make resizable text fields opt-in rather than
opt-out. Open source users will need to write the pref into their defaults
to see the resizers on textareas now.

Reviewed by mjs and adele

WebCore.xcodeproj/project.pbxproj:
bridge/mac/WebCoreSettings.mm: (-[WebCoreSettings setTextAreasAreResizable:]):
css/cssparser.cpp: (WebCore::CSSParser::parseValue):
css/cssstyleselector.cpp: (WebCore::CSSStyleSelector::applyProperty):
css/html4.css:

4:03 PM Changeset in webkit [15513] by thatcher

12 edits in tags/Safari-521.18

Merging important fixes.

3:56 PM Changeset in webkit [15512] by tomernic

2 edits in trunk/JavaScriptCore

Reviewed by Tim Hatcher.

Removed a misleading comment; we recently added support for the NPNVPluginElementNPObject
variable.

bindings/npapi.h:

3:09 PM Changeset in webkit [15511] by thatcher

77 edits
6 copies
7 deletes in tags/Safari-521.18

Merged urgent fixes to a new tag.

1:43 PM Changeset in webkit [15510] by brmorris

33 edits in S60/trunk/WebKit

2006-07-18 Bradley <bradley.morrison@nokia.com>

Reviewed by Sachin/Zalan.

Various fixes - see ChangeLogs for details.
Part 4 of 4.

1:38 PM Changeset in webkit [15509] by brmorris

34 edits in S60/trunk/WebKit

2006-07-17 Bradley <bradley.morrison@nokia.com>

Reviewed by Sachin/Zalan.

Various fixes - see ChangeLogs for details.
Part 3 of 4.

1:31 PM Changeset in webkit [15508] by brmorris

26 edits in S60/trunk/WebCore

2006-07-17 Bradley <bradley.morrison@nokia.com>

Reviewed by Sachin/Zalan.

Various fixes - see ChangeLogs for details.
Part 2 of 4.

1:22 PM Changeset in webkit [15507] by brmorris

21 edits
5 adds
1 delete in S60/trunk

2006-07-17 Bradley <bradley.morrison@nokia.com>

Reviewed by Sachin/Zalan.

Various fixes - see ChangeLogs for details.
Part 1 of 4.

1:22 PM Changeset in webkit [15506] by andersca

5 edits
2 adds in trunk

WebCore:

2006-07-18 Anders Carlsson <acarlsson@apple.com>

Reviewed by Darin.

http://bugzilla.opendarwin.org/show_bug.cgi?id=9959
REGRESSION: iframes stop rendering after 200th one on successive reloads

html/HTMLFrameElement.cpp: (WebCore::HTMLFrameElement::attach): (WebCore::HTMLFrameElement::close):
html/HTMLIFrameElement.cpp: (WebCore::HTMLIFrameElement::insertedIntoDocument): (WebCore::HTMLIFrameElement::willRemove): Remove calls to incrementFrameCount and decrementFrameCount.

page/Frame.cpp: (WebCore::Frame::Frame): Call incrementFrameCount here.

(WebCore::Frame::~Frame):
Call disconnectOwnerElement.

(WebCore::Frame::disconnectOwnerElement):
Call decrementFrameCount here.

LayoutTests:

2006-07-18 Anders Carlsson <acarlsson@apple.com>

Reviewed by Darin.

http://bugzilla.opendarwin.org/show_bug.cgi?id=9959
REGRESSION: iframes stop rendering after 200th one on successive reloads

fast/frames/frame-limit-expected.txt: Added.
fast/frames/frame-limit.html: Added.

12:56 PM Changeset in webkit [15505] by ap

3 edits
1 add in trunk/WebCore

Reviewed by Darin.

Fix http://bugzilla.opendarwin.org/show_bug.cgi?id=9013
Let correct CSS custom cursor declarations parse

css/cssparser.cpp: (WebCore::CSSParser::parseValue): Ignore any values following a custom CSS cursor URI, instead of requiring that it is the only one in the list (which is actually illegal, according to the spec). This is only a temporary hack, with a real implementation to follow in bug 6001/6002.

manual-tests/custom-cursors.html: Added a test for this issue.
manual-tests/resources/helpCursor.tiff: A cursor image used in the test.

12:28 PM Changeset in webkit [15504] by thatcher

2 edits in trunk/WebCore

Reviewed by Anders.

<rdar://problem/4636216> NetNewsWire fails to launch on 9A224 - missing symbols

WebCore needs to add -sub_library libobjc to maintain
backwards compatibility with binaries linked with WebKit
before JavaScriptCore moved out of WebKit.

WebCore.xcodeproj/project.pbxproj:

10:56 AM Changeset in webkit [15503] by thatcher

1 copy in tags/Safari-521.18

New tag.

10:03 AM Changeset in webkit [15502] by darin

5 edits
4 deletes in trunk/LayoutTests

Reviewed by Alice.

convert a couple of tests to "dump as text" tests

fast/dom/css-selectorText-expected.checksum: Removed.
fast/dom/css-selectorText-expected.png: Removed.
fast/dom/css-selectorText-expected.txt: Replaced render tree results with text results.
fast/dom/css-selectorText.html: Added a call to layoutTestController.dumpAsText.

fast/dom/css-shortHands-expected.checksum: Removed.
fast/dom/css-shortHands-expected.png: Removed.
fast/dom/css-shortHands-expected.txt: Replaced render tree results with text results.
fast/dom/css-shortHands.html: Added a call to layoutTestController.dumpAsText.

9:01 AM Changeset in webkit [15501] by thatcher

2 edits in trunk/JavaScriptCore

Made the following headers public:

JavaScriptCore.h
JSBase.h
JSContextRef.h
JSObjectRef.h
JSStringRef.h
JSValueRef.h

JavaScriptCore.xcodeproj/project.pbxproj:

3:55 AM Changeset in webkit [15500] by ddkilzer

3 edits
4 adds in trunk

LayoutTests:

Reviewed by Justin. Patch by Mitz.

test for http://bugzilla.opendarwin.org/show_bug.cgi?id=9978 REGRESSION (r12949-r12988): Clicking the first letter on a line places the caret at the end of the previous line

editing/selection/click-start-of-line-expected.checksum: Added.
editing/selection/click-start-of-line-expected.png: Added.
editing/selection/click-start-of-line-expected.txt: Added.
editing/selection/click-start-of-line.html: Added.

WebCore:

Reviewed by Justin. Patch by Mitz.

fix http://bugzilla.opendarwin.org/show_bug.cgi?id=9978 REGRESSION (r12949-r12988): Clicking the first letter on a line places the caret at the end of the previous line

Test: editing/selection/click-start-of-line.html

rendering/RenderText.cpp: (WebCore::RenderText::positionForCoordinates): Changed to return downstream affinity if the x coordinate is to the left of the middle of the first character in the text box.

2:24 AM Changeset in webkit [15499] by andersca

3 edits
2 adds in trunk

WebCore:

2006-07-18 Anders Carlsson <acarlsson@apple.com>

Reviewed by Darin.

http://bugzilla.opendarwin.org/show_bug.cgi?id=9695
<rdar://problem/4614085>
TOT REGRESSION: NativeTextArea: Text area does not respond to Cmd-UpArrow/Cmd-DownArrow (9695)

editing/SelectionController.cpp: (WebCore::SelectionController::modifyExtendingRightForward): (WebCore::SelectionController::modifyMovingRightForward): (WebCore::SelectionController::modifyExtendingLeftBackward): (WebCore::SelectionController::modifyMovingLeftBackward): If the caret is inside an editable region, the beginning/end of the document should actually be the beginning/end of the editable region.

LayoutTests:

2006-07-18 Anders Carlsson <acarlsson@apple.com>

Reviewed by Darin.

http://bugzilla.opendarwin.org/show_bug.cgi?id=9695
<rdar://problem/4614085>
TOT REGRESSION: NativeTextArea: Text area does not respond to Cmd-UpArrow/Cmd-DownArrow (9695)

editing/selection/move-begin-end-expected.txt: Added.
editing/selection/move-begin-end.html: Added.

Jul 17, 2006:

9:55 PM Changeset in webkit [15498] by tomernic

12 edits in trunk

WebCore:

Reviewed by Maciej.

<rdar://problem/4612079> need a way to prevent pages from scrolling to reveal elements that are focused
by script

bridge/mac/WebCoreFrameBridge.h:
bridge/mac/WebCoreFrameBridge.mm: (-[WebCoreFrameBridge setProhibitsScrolling:]):
page/Frame.cpp: (WebCore::Frame::prohibitsScrolling): (WebCore::Frame::setProhibitsScrolling):
page/Frame.h:
page/FramePrivate.h: (WebCore::FramePrivate::FramePrivate):
page/FrameView.cpp: (WebCore::FrameView::scrollPointRecursively): (WebCore::FrameView::setContentsPos):
page/FrameView.h:
platform/ScrollView.h:

WebKit:

Reviewed by Maciej.

<rdar://problem/4612079> need a way to prevent pages from scrolling to reveal elements that are focused
by script

WebView/WebViewPrivate.h:
WebView/WebView.m: (-[WebView setProhibitsMainFrameScrolling:]): New method. Prohibits scrolling in the WebView's main frame. Used to "lock" a WebView to a specific scroll position.

9:33 PM Changeset in webkit [15497] by ggaren

19 edits in trunk/JavaScriptCore

Reviewed by Maciej.

Added automatic prototype creation for classes.

A class stores a weak reference to a prototype, which is cleared when
the prototype is garbage collected, to avoid a reference cycle.

We now have an attributes field in JSClassDefinition, that currently is
used only to override automatic prototype creation when you want to manage your
own prototypes, but can be extended in the future for other nefarious purposes.

Similarly, we have JSObjectMake and JSObjectMakeWithPrototype, the latter
allowing you to manage your own prototypes.

JSObjectMakeConstructor is more interesting now, able to make a constructor
on your behalf if you just give it a class.

Removed bogus old code from minidom.js.

Tweaked the headerdocs.

Added more GC testing, which caught some leaks, and tested more funny edge cases in lookup, which caught a lookup bug. Removed some testing we used to do with MyObject because it was redundant with the new, cool stuff.

While fixing the lookup bug I retracted this change:

"If a static setProperty callback returns 'false', to indicate that the
property was not set, we no longer forward the set request up the class
chain, because that's almost certainly not what the programmer expected."

Returning false when setting a static property is a little silly, but you can see
it being useful when shadowing a base class's static properties, and, regardless
of usefullness, this is the defined behavior of the setProperty callback.

Plus a little ASCII art, for the kids.

9:32 PM Changeset in webkit [15496] by mjs

3 edits
4 adds in trunk

LayoutTests:

Test cases not reviewed but the corresponding fix was.

test cases for <rdar://problem/4618089> Blot crashes when I paste in all the contents of http://www.apple.com/startpage/

editing/pasteboard/nested-blocks-with-text-area-expected.txt: Added.
editing/pasteboard/nested-blocks-with-text-area.html: Added.
editing/pasteboard/nested-blocks-with-text-field-expected.txt: Added.
editing/pasteboard/nested-blocks-with-text-field.html: Added.

WebCore:

Reviewed by Dave Harrison.

fixed <rdar://problem/4618089> Blot crashes when I paste in all the contents of http://www.apple.com/startpage/

editing/htmlediting.cpp: (WebCore::editingIgnoresContent):

6:18 PM Changeset in webkit [15495] by justing

2 edits in trunk/WebCore

Reviewed by levi

<rdar://problem/4621728>
REGRESSION: Selecting by dragging down creates selection in wrong direction, with certain steps
<rdar://problem/4604932>
REGRESSION: Dragging too far left on a line will select all lines above it.

No layout test because of 9980.

rendering/RenderBlock.cpp: (WebCore::RenderBlock::positionForCoordinates): Don't return positions inside editable roots for coordinates outside those roots, except for coordinates outside a document that is entirely editable.

5:55 PM Changeset in webkit [15494] by thatcher

2 edits in trunk/WebKit

Reviewed by Darin.

<rdar://problem/4635311> REGRESSION: WebKit should call windowScriptObjectAvailable before attaching the script debugger

WebCoreSupport/WebFrameBridge.m: (-[WebFrameBridge windowObjectCleared]):

5:48 PM Changeset in webkit [15493] by thatcher

3 edits in trunk/WebCore

Reviewed by Darin.

<rdar://problem/4635281> JSCanvasRenderingContext2D::drawImage needs to initialize the exception code to zero

bindings/js/JSCanvasRenderingContext2DCustom.cpp: (WebCore::JSCanvasRenderingContext2D::drawImage): initialize ec to zero
html/CanvasRenderingContext2D.cpp: (WebCore::CanvasRenderingContext2D::drawImage): initialize ec before earlier return

5:27 PM Changeset in webkit [15492] by thatcher

20 edits
1 copy
2 moves
1 add in trunk

JavaScriptCore:

Reviewed by Maciej.

<rdar://problem/4634874> WebScriptObject and WebUndefined are no longer defined by WebKit

Moves WebScriptObject and WebUndefined up to WebCore.
This change does create an upwards-dependancy on WebScriptObject existing
in the loaded process, but this code path in JavaScriptCore does not get used
unless it is through WebKit/WebCore. Moving all of the binding code out of
JavaScriptCore might make sense in the future.

JavaScriptCore.exp:
JavaScriptCore.xcodeproj/project.pbxproj:
bindings/objc/WebScriptObject.h: Replaced.
bindings/objc/WebScriptObject.mm: Removed.
bindings/objc/WebScriptObjectPrivate.h: Removed.
bindings/objc/objc_class.h:
bindings/objc/objc_instance.h:
bindings/objc/objc_instance.mm: (ObjcInstance::~ObjcInstance):
bindings/objc/objc_runtime.h:
bindings/objc/objc_runtime.mm: (convertValueToObjcObject):
bindings/objc/objc_utility.mm: (KJS::Bindings::convertValueToObjcValue): (KJS::Bindings::convertObjcValueToValue): (KJS::Bindings::createObjcInstanceForValue):

WebCore:

Reviewed by Maciej.

<rdar://problem/4634874> WebScriptObject and WebUndefined are no longer defined by WebKit

Moves WebScriptObject and WebUndefined to WebCore.

WebCore.exp:
WebCore.xcodeproj/project.pbxproj:
bindings/objc/DOM.mm:
bindings/objc/DOMCore.h:
bindings/objc/DOMInternal.mm:
bindings/objc/DOMUtility.mm:
bindings/objc/WebScriptObject.mm: Added. (+[WebUndefined allocWithZone:]): (-[WebUndefined initWithCoder:]): (-[WebUndefined encodeWithCoder:]): (-[WebUndefined copyWithZone:]):
bindings/objc/WebScriptObjectPrivate.h: Added.
bridge/mac/FrameMac.mm:
bridge/mac/WebCoreScriptDebugger.mm:

WebKit:

Reviewed by Maciej.

<rdar://problem/4634874> WebScriptObject and WebUndefined are no longer defined by WebKit

Copy WebScriptObject.h from WebCore's private headers, not JavaScriptCore.

WebKit.xcodeproj/project.pbxproj:

3:29 PM Changeset in webkit [15491] by andersca

1 edit
1 add in trunk/LayoutTests

2006-07-17 Anders Carlsson <acarlsson@apple.com>

Reviewed by Darin.

Add Apache 2 config file.

http/conf/apache2-httpd.conf: Added.

3:18 PM Changeset in webkit [15490] by hyatt

2 edits in trunk/WebCore

Back out the fix for 5564, since it turns out font:x-small; is a pretty
prominent IE-specific CSS hack. Because Web sites rely on IE's incorrect
font parsing as a means of also correcting IE's incorrect font size rules.

This fixes Yahoo.com.

Reviewed by darin

css/cssparser.cpp: (WebCore::CSSParser::parseFont):

3:05 PM Changeset in webkit [15489] by sullivan

2 edits in trunk/WebKit

Reviewed by Tim Omernick.

fixed <rdar://problem/4604366> Orange Find highlight displays text in wrong size on PDF pages if they're not at "actual size"

To match WebHTMLView, I made the methods that return attributed strings take the view's scale
factor into account.

WebView/WebPDFView.m: (-[WebPDFView _scaledAttributedString:]): new helper method, takes an attributed string and returns one that's scaled by the view's current scale factor (-[WebPDFView attributedString]): pass result through _scaledAttributedString: (-[WebPDFView selectedAttributedString]): ditto

2:54 PM Changeset in webkit [15488] by justing

3 edits
4 adds in trunk

LayoutTests:

Reviewed by levi

<rdar://problem/4618389> REGRESSION: After applying a link to the last word of a sentence, a new selection is created at the start of the sentence

editing/selection/replace-selection-1-expected.checksum: Added.
editing/selection/replace-selection-1-expected.png: Added.
editing/selection/replace-selection-1-expected.txt: Added.
editing/selection/replace-selection-1.html: Added.

WebCore:

Reviewed by levi

<rdar://problem/4618389> REGRESSION: After applying a link to the last word of a sentence, a new selection is created at the start of the sentence

dom/Document.cpp: (WebCore::Document::updateSelection):

12:38 PM Changeset in webkit [15487] by justing

4 edits in trunk

LayoutTests:

Reviewed by levi

editing/selection/drag-to-contenteditable-iframe-expected.txt:
editing/selection/subframe-with-selection-expected.txt:

WebKit:

Reviewed by levi

Rolled the first fix for:
<http://bugzilla.opendarwin.org/show_bug.cgi?id=9642>
GMail Editor: Operations that use drop down menus blow away the selection
back in and removed the call to _clearSelectionInOtherFrames from
-[WebHTMLView becomeFirstResponder] to fix the bug.

WebView/WebHTMLView.m: (-[NSArray maintainsInactiveSelection]): (-[NSArray becomeFirstResponder]):
WebView/WebView.m: (-[WebView maintainsInactiveSelection]):

8:19 AM Changeset in webkit [15486] by darin

3 edits in trunk/JavaScriptCore

API/JSBase.h: Fix comment formatting where things used to be lined up but are now ragged. Got rid of spaces that attempted to line things up.
API/JSObjectRef.h: Ditto. Also add missing periods for a couple of comments.

4:50 AM Changeset in webkit [15485] by mjs

4 edits
2 adds in trunk

LayoutTests:

Reviewed by Beth.

<rdar://problem/4604946> REGRESSION: 'checkboxRef.checked = true' fires onchange event in Leopard

fast/events/programmatic-check-no-change-event-expected.txt: Added.
fast/events/programmatic-check-no-change-event.html: Added.

WebCore:

Reviewed by Beth.

<rdar://problem/4604946> REGRESSION: 'checkboxRef.checked = true' fires onchange event in Leopard

html/HTMLInputElement.cpp: (WebCore::HTMLInputElement::setChecked): (WebCore::HTMLInputElement::preDispatchEventHandler):
html/HTMLInputElement.h:

3:49 AM Changeset in webkit [15484] by ggaren

12 edits in trunk/JavaScriptCore

Reviewed by Maciej.

Removed the exception parameter from the initialize callback and, by extension, JSObjectMake. We have never had a need for exceptions when iniitializing, so the parameter seemed likely to "get in the way."

Also, an exception in JavaScript is thrown in response to input --
"invalid URL", "index not a number", etc., so it's the job of the
constructor function, not the initialize method, to throw.

If initialize *really* wants to throw, it can communicate the throw to
the constructor through the constructed object's private data (e.g., set
it to NULL, signaling to the consntructor that initialization failed.)

Added JSObjectMakeWithData, which enables a constructor to set private data on an object *before* it has been initialized. That way, the initialize methods can properly operate on the data.

API/JSNode.c: Moved ref into the initialize method, for better encapsulation, now that it's possible.
API/JSNodeList.c: ditto
API/minidom.c: (main): Do more aggressive garbage collection to test ref/deref and initialize/finalize.
API/minidom.js: store childNodes in a temporary so it doesn't get re-created like a thousand times. This makes debugging ref/deref easier

2:06 AM Changeset in webkit [15483] by ggaren

8 edits in trunk/JavaScriptCore

Reviewed by Maciej.

Changed the initialize callback to run from least derived class (parent class) to most derived class. This enables C++ style initialization, and derived class overriding of member data.

Added excpetion propopgation to JSObjectMake, to support initialize exceptions, and generally round out our policy of making function signatures as long as possible.

API/JSCallbackObject.h: Use ExecState instead of ContextRef, cuz we're in C++ land now.

1:20 AM Changeset in webkit [15482] by ggaren

16 edits in trunk/JavaScriptCore

Reviewed by Maciej.

Changed JSObjectMakeConstructor to JSObjectMakeConstructorWithCallback, to match JSObjectMakeFunctionWithCallback.

Added prototype parameter, so the generated constructor automatically works with hasInstance / instanceof

Moved hasInstance implementation from InternalFunctionImp to JSObject so that subclasses can inherit it without inheriting function-related baggage. More refactoring here would be good, but this seems like a good short-term solution.

(KJS::JSCallbackFunction::implementsHasInstance): override and return false,
because callback functions aren't constructors.

1:14 AM Changeset in webkit [15481] by mjs

15 edits in trunk/JavaScriptCore

Reviewed by Geoff.

add a JSContextRef parameter to all JSValueRef, JSObjectRef, and JSContextRef operations; except JSObject{Get,Set}PrivateData which can be assumed to be simple pure accessors.

Also renamed the parameter "context" to "ctx" because it makes the code read better with this pervasive
but usually uninteresting parameter.

API/JSBase.cpp: (JSEvaluateScript): (JSCheckScriptSyntax): (JSGarbageCollect):
API/JSBase.h:
API/JSCallbackObject.cpp: (KJS::JSCallbackObject::JSCallbackObject): (KJS::JSCallbackObject::init): (KJS::JSCallbackObject::getOwnPropertySlot): (KJS::JSCallbackObject::put): (KJS::JSCallbackObject::deleteProperty): (KJS::JSCallbackObject::toNumber): (KJS::JSCallbackObject::toString):
API/JSContextRef.cpp: (JSGlobalContextCreate): (JSGlobalContextRetain): (JSGlobalContextRelease): (JSContextGetGlobalObject):
API/JSContextRef.h:
API/JSNode.c: (JSNodePrototype_appendChild): (JSNodePrototype_removeChild): (JSNodePrototype_replaceChild): (JSNode_getNodeType): (JSNode_getFirstChild): (JSNode_prototype):
API/JSNodeList.c: (JSNodeListPrototype_item): (JSNodeList_length): (JSNodeList_getProperty): (JSNodeList_prototype):
API/JSObjectRef.cpp: (JSObjectMake): (JSObjectMakeFunctionWithCallback): (JSObjectMakeConstructor): (JSObjectMakeFunction): (JSObjectGetPrototype): (JSObjectSetPrototype): (JSObjectHasProperty): (JSObjectGetProperty): (JSObjectSetProperty): (JSObjectGetPropertyAtIndex): (JSObjectSetPropertyAtIndex): (JSObjectDeleteProperty): (JSObjectIsFunction): (JSObjectCallAsFunction): (JSObjectIsConstructor): (JSObjectCallAsConstructor): (JSObjectCopyPropertyNames):
API/JSObjectRef.h:
API/JSStringRef.cpp:
API/JSValueRef.cpp: (JSValueGetType): (JSValueIsUndefined): (JSValueIsNull): (JSValueIsBoolean): (JSValueIsNumber): (JSValueIsString): (JSValueIsObject): (JSValueIsObjectOfClass): (JSValueIsEqual): (JSValueIsStrictEqual): (JSValueIsInstanceOfConstructor): (JSValueMakeUndefined): (JSValueMakeNull): (JSValueMakeBoolean): (JSValueMakeNumber): (JSValueMakeString): (JSValueToBoolean): (JSValueToNumber): (JSValueToStringCopy): (JSValueToObject): (JSValueProtect): (JSValueUnprotect):
API/JSValueRef.h:
API/minidom.c: (print):
API/testapi.c: (MyObject_getProperty): (MyObject_deleteProperty): (MyObject_callAsFunction): (MyObject_callAsConstructor): (MyObject_convertToType): (print_callAsFunction): (main):

Note: See TracTimeline for information about the timeline view.

Download in other formats:

RSS Feed