[Webkit-unassigned] [Bug 52148] [Gtk] Implement support for Embedded Objects

bugzilla-daemon at webkit.org bugzilla-daemon at webkit.org
Thu Jan 13 03:05:01 PST 2011


https://bugs.webkit.org/show_bug.cgi?id=52148





--- Comment #3 from Mario Sanchez Prada <msanchez at igalia.com>  2011-01-13 03:05:01 PST ---
(In reply to comment #0)
> Created an attachment (id=78400)
 --> (https://bugs.webkit.org/attachment.cgi?id=78400&action=review) [details]
> test case
> 
> WebKitGtk currently does not have support for embedded objects. In particular, given the attached test case:
> 
> 1. The AtkText implementation for each paragraph should contain an embedded object character for each embedded object:

You mean some kind of special character like those found in ff, right? 
Something like this:

 In [2]: t.getText(0, -1)
 Out[2]: "Here's our logo: \xef\xbf\xbc It reminds me of airport security
          screenings. "

> * Here's our logo: <eoc> It reminds me of airport security screenings.
> 
> * Choose: <eoc> <eoc> (pick one or more)
> 
> * Choose: <eoc> <eoc> (just pick one)
> 
> * Choose: <eoc> <eoc> (just pick one)

My main question is "how that special character helps to screen readers"? 
I guess the answer is that ORCA would use it to know that it should dive in the object to extract more info (actually I recall you talking to me about this in Seville, not sure if I recall correctly, though :-)), but I wonder whether it shouldn't be better to provide a textual representation for them.

Of course, I'm talking from my vast ignorance on the issue :-), and I realize that then we'd need to agree on a specific text representation for each kind of object (e.g. 'checked/unchecked' for check boxes, 'selected/unselected' for radio buttons...) but asking is not harmful either.


> 2. Each embedded object should implement AtkHyperlink.

Is this really always needed for each embedded object of any kind? 

I can see it for the image, but don't know how this interface would be implemented for other objects, such as the check boxes or the radio buttons (e.g. what should they return in the atk_hyperlink_get_uri() method?).

Taking a look at the following document (look for the text "Check one or more" to easily find the part I'm talking about)...

http://accessibility.linuxfoundation.org/~gk4/a11y/ddc06/guide/atkguideddc06.html

...it seems they suggest using AthHyperText/Link stuff for the case of the image, but not for check buttons and their associated text, which would just implement AtkAction and AtkText, respectively.

But of course, we could decide to do something else. Just pointing out that link as a reference trying to confirm things.


> 3. Each paragraph should implement AtkHypertext.

Makes sense.


> (I realize this is likely at least three bugs/rfes. But I will let Mario triage and make this an embedded object metabug if and how he sees fit. :-) )

At the moment, things are not perfectly clear so I'll keep this one as reference, but as soon as this translates into actual code and patches you can be sure I'll probably file more specific bugs depending on this one.

Thanks again, and sorry for the delay from my side on this topic.

-- 
Configure bugmail: https://bugs.webkit.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.



More information about the webkit-unassigned mailing list