[Webkit-unassigned] [Bug 141862] AX: Expose caret browsing preference to accessibility API

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=141862

--- Comment #12 from chris fleizach <cfleizach at apple.com> ---
(In reply to comment #4)
> (In reply to comment #3)
> > Comment on attachment 247025 [details]
> > Patch
> > 
> > View in context:
> > https://bugs.webkit.org/attachment.cgi?id=247025&action=review
> > 
> > > Source/WebCore/accessibility/AccessibilityObject.h:978
> > > +    virtual bool caretBrowsingEnabled() const { return false; }
> > 
> > I would just put these implementations into AccessibilityObject and remove
> > from AccessibilityRenderObject. It doesn't seem like there's a benefit to
> > making those virtual
> 
> The methods are virtual because AccessibilityObject doesn't know about the
> Frame that caretBrowsingEnabled is ultimately set on.
> 
> FrameView* AccessibilityRenderObject::documentFrameView() const
> 

documentFrameView exists on AccessibilityObject.h

virtual FrameView* documentFrameView() const;

it seems like this can be

> > 
> > > Tools/DumpRenderTree/AccessibilityUIElement.cpp:1404
> > > +    return 0;
> > 
> > return nullptr
> 
> Wil do
> 
> > 
> > > LayoutTests/platform/mac/accessibility/caret-browsing-attribute.html:13
> > > +    var webArea = accessibilityController.focusedElement;
> > 
> > the body focus() idiom i think might have some issues on GTK. I usually
> > prefer accessibilityController.rootElement.childAtIndex(0) I think. Or maybe
> > the rootElement is the web area
> 
> Wil do
> 
> > 
> > > LayoutTests/platform/mac/accessibility/caret-browsing-tab-selection.html:22
> > > +            shouldBe("selectedElement(webArea)", "");
> > 
> > can you (in general) put some new lines between tests units, to make it a
> > bit easier parse
> 
> Wil do
> 
> > 
> > > LayoutTests/platform/mac/accessibility/resources/accessibility-helper.js:11
> > > +function selectedElement(webArea) {
> > 
> > can you name this a little better so its clear its related to text marker
> > selection
> 
> Does
> 
> function elementAtStartMarkerOfSelectedTextMarkerRange(webArea)
> 
> work as far as clarity?

sounds good

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.webkit.org/pipermail/webkit-unassigned/attachments/20150222/65be4a6e/attachment-0002.html>