Bug 100204 – AX: WebKit exposes abstract ARIA role range as AXSlider; should be generic AXGroup

• Component: Accessibility
• Status: RESOLVED FIXED

[chris fleizach]

WebKit exposes abstract ARIA role range as AXSlider; should be generic AXGroup

1. https://dvcs.w3.org/hg/pfwg/raw-file/tip/ARIA/1.0/tests/test-files/roles-plain-abstract/roles-plain-abstract-range.html

The <div role="range">  element is exposed as the *abstract* ARIA role range which is the parent role for both slider and spinbutton.

Note: the HTML element type is "range" (<input type="range") and is equivalent to ARIA slider (<div role="slider") so this is probably where the confusion started. Both of these should remain as AXSlider.

It's only the ROLE of "range" that is exposed incorrectly. Since it's abstract, it should be exposed as a generic AXGroup.

[chris fleizach]

rdar://12085957

[chris fleizach]

Created attachment 170324 [details]
patch

[Dominic Mazzoni]

Unofficial review - looks good. Can you move the test to the main directory rather than platform/mac, though? It should pass on Chromium and GTK now, especially with the latest GTK refactoring.

[chris fleizach]

Created attachment 170403 [details]
patch

Moved the test to /accessibility

[chris fleizach]

http://trac.webkit.org/changeset/132381
Moved that test to the top level accessibility folder too

[Roger Fong]

Is accessibility/aria-slider-value.html  another test that should be skipped on Windows or get Windows specific results?

[chris fleizach]

(In reply to comment #6)
> Is accessibility/aria-slider-value.html  another test that should be skipped on Windows or get Windows specific results?

that's a good question. has the windows platform mapped "slider" to the appropriate role? I don't know. My guess is that that test should be skipped on windows

[Roger Fong]

Ah yes, I'm remembering now from previous test failures that none of the "role stuff" works on windows, I'll skip it.