RE: The aria-hidden attribute not being honored within naming calculation for parent controls?

Brian, did you submit that bug on http://connect.microsoft.com/ie ?  Can you send me a link to it?

Thanks,
Cynthia

From: Steve Faulkner [mailto:faulkner.steve@gmail.com]
Sent: Monday, January 19, 2015 1:54 AM
To: Bryan Garaventa
Cc: W3C WAI Protocols & Formats
Subject: Re: The aria-hidden attribute not being honored within naming calculation for parent controls?

Hi Bryan,
submitted bug on Firefox
https://bugzilla.mozilla.org/show_bug.cgi?id=1123248

bug also needs to be filed on IE
Works correctly in chrome on windows.


--

Regards

SteveF
HTML 5.1<http://www.w3.org/html/wg/drafts/html/master/>

On 17 January 2015 at 00:31, Bryan Garaventa <bryan.garaventa@ssbbartgroup.com<mailto:bryan.garaventa@ssbbartgroup.com>> wrote:
Hi,
I wrote the attached grid sample to identify a bug in JAWS+IE where supporting attributes like aria-sort, aria-readonly, and aria-haspopup, are not being conveyed when Forms Mode is active, and discovered another issue in the accessibility tree in both IE and FF.

The grid consists of column nodes, which include a span tag that contains a Unicode down arrow symbol, marked with aria-hidden=”true” to remove it from the accessibility tree, which doesn’t remove it actually but flags it as hidden.

So, regardless, the naming calculation still sets the accessible name for each columnheader node as the text + the hidden symbol text, even though it should not be included since I am explicitly telling it not to using aria-hidden=”true”. This results in JAWS announcing the hidden symbol text as part of the label despite my attempts to keep this out of the accessible name for that node.

So I guess my question is, shouldn’t nodes marked up with aria-hidden=”true” be omitted from the accessible name calculation within the accessibility tree?

Thanks,
Bryan