AX: Support @scope in HTML tables

• Component: Accessibility
• Status: RESOLVED FIXED

[chris fleizach]

We need to support scope in tables for AX

http://www.w3.org/TR/html401/struct/tables.html#adef-scope

<rdar://problem/6237955>

[chris fleizach]

Created attachment 222338 [details]
patch

[chris fleizach]

Created attachment 222340 [details]
patch

[chris fleizach]

Created attachment 222387 [details]
patch

[Mario Sanchez Prada]

Comment on attachment 222387 [details]
patch 

View in context: https://bugs.webkit.org/attachment.cgi?id=222387&action=review

Assuming that the Objective-C bits are Ok (which do seem Ok to me, with the little knowledge I have about the language), the patch looks good to me.

I would only ask you for uploading a new version (no need to review it again) of the patch removing the uiElementArrayAttributeValue() empty stub from WKTR's AccessibilityUIElementAtk.cpp before landing it, so we make sure that such a change is enough to make EWS bots happy.

> Source/WebCore/accessibility/mac/WebAccessibilityObjectWrapperMac.mm:-93
> -#ifndef NSAccessibilityRowHeaderUIElementsAttribute
> -#define NSAccessibilityRowHeaderUIElementsAttribute @"AXRowHeaderUIElements"
> -#endif
> -

I guess this is intentional? Just double checking

> Tools/WebKitTestRunner/InjectedBundle/atk/AccessibilityUIElementAtk.cpp:828
> +JSValueRef AccessibilityUIElement::uiElementArrayAttributeValue(JSStringRef attribute) const
> +{
> +    // FIXME: implement
> +    return nullptr;
> +}
> +    

You don't need to add this empty method, it's already there (and the reason why it's failing to build in GTK and EFL)

[chris fleizach]

(In reply to comment #4)
> (From update of attachment 222387 [details])
> View in context: https://bugs.webkit.org/attachment.cgi?id=222387&action=review
> 
> Assuming that the Objective-C bits are Ok (which do seem Ok to me, with the little knowledge I have about the language), the patch looks good to me.
> 
> I would only ask you for uploading a new version (no need to review it again) of the patch removing the uiElementArrayAttributeValue() empty stub from WKTR's AccessibilityUIElementAtk.cpp before landing it, so we make sure that such a change is enough to make EWS bots happy.
> 

Yep of course. I would have fixed these already but I'm no longer getting emails when builds fail

> > Source/WebCore/accessibility/mac/WebAccessibilityObjectWrapperMac.mm:-93
> > -#ifndef NSAccessibilityRowHeaderUIElementsAttribute
> > -#define NSAccessibilityRowHeaderUIElementsAttribute @"AXRowHeaderUIElements"
> > -#endif
> > -
> 
> I guess this is intentional? Just double checking

Yep, this name has now been defined for 3 releases.

> 
> > Tools/WebKitTestRunner/InjectedBundle/atk/AccessibilityUIElementAtk.cpp:828
> > +JSValueRef AccessibilityUIElement::uiElementArrayAttributeValue(JSStringRef attribute) const
> > +{
> > +    // FIXME: implement
> > +    return nullptr;
> > +}
> > +    
> 
> You don't need to add this empty method, it's already there (and the reason why it's failing to build in GTK and EFL)

Thanks!

[chris fleizach]

Created attachment 222449 [details]
patch

[WebKit Commit Bot]

Attachment 222449 [details] did not pass style-queue:


ERROR: Source/WebCore/accessibility/AccessibilityTableCell.cpp:33:  Alphabetical sorting problem.  [build/include_order] [4]
Total errors found: 1 in 28 files


If any of these errors are false positives, please file a bug against check-webkit-style.

[chris fleizach]

Created attachment 222451 [details]
patch

[chris fleizach]

http://trac.webkit.org/changeset/162986

[David Farler]

This breaks the iOS build - filed https://bugs.webkit.org/show_bug.cgi?id=127917 for implementation of methods in AccessibilityUIElement for iOS.