Bug 146718 – [GTK] Layout test accessibility/aria-hidden-false-works-in-subtrees.html fails after r184890

• Status: RESOLVED
• Resolution: FIXED
• Component: Accessibility

[Carlos Alberto Lopez Perez]

On platform GTK the test accessibility/aria-hidden-false-works-in-subtrees.html fails after r184890 <http://trac.webkit.org/r184890>

The diff is the following:

https://build.webkit.org/results/GTK%20Linux%2064-bit%20Release%20%28Tests%29/r184890%20%288846%29/accessibility/aria-hidden-false-works-in-subtrees-pretty-diff.html

[Radar WebKit Bug Importer]

<rdar://problem/21722487>

[Carlos Garcia Campos]

Created attachment 367115 [details]
Patch

[Michael Catanzaro]

This looks like one for Joanie.

[Joanmarie Diggs]

Comment on attachment 367115 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=367115&action=review

> Source/WebCore/accessibility/atk/WebKitAccessible.cpp:1185
> +        || role == AccessibilityRole::ApplicationGroup;

Fine. BUT this reminds me that we really should be implementing AtkText on most things with text (i.e. I'd think pretty much anything which is not a non-text widget). Not this bug, of course....

[Joanmarie Diggs]

(In reply to Joanmarie Diggs (irc: joanie) from comment #4)
> Comment on attachment 367115 [details]
> Patch
> 
> View in context:
> https://bugs.webkit.org/attachment.cgi?id=367115&action=review
> 
> > Source/WebCore/accessibility/atk/WebKitAccessible.cpp:1185
> > +        || role == AccessibilityRole::ApplicationGroup;
> 
> Fine. BUT this reminds me that we really should be implementing AtkText on
> most things with text (i.e. I'd think pretty much anything which is not a
> non-text widget). Not this bug, of course....

Just filed https://bugs.webkit.org/show_bug.cgi?id=196769.

[Carlos Garcia Campos]

Committed r244187: <https://trac.webkit.org/changeset/244187>