Open
Bug 1056334
Opened 11 years ago
Updated 2 years ago
Expose accessible "showlongdesc" action and context menu for empty images
Categories
(Core :: Disability Access APIs, defect)
Tracking
()
NEW
People
(Reporter: jdiggs, Unassigned)
References
(Blocks 1 open bug)
Details
Given the test case below from the W3C, Gecko:
* Does not include the "View Description" right-click menu item
* Does exposes the empty image as ATK_ROLE_TEXT rather than ATK_ROLE_IMAGE
* Fails to expose the accessible action "showlongdesc"
As a result, it is not possible for Orca to provide any support for longdesc if the image is "empty."
=============================
<!DOCTYPE html>
<html>
<head>
<title>Longdesc test: Empty image, data: URI description</title>
</head>
<body>
<img src="" alt="the image" longdesc="data:text/html;charset=utf-8,%3C%21DOCTYPE%20html%3E%0D%0A%3Chtml%3E%0D%0A%20%3Chead%3E%0D%0A%20%20%3Ctitle%3ELongdesc%20test%20Pass%20page%3C%2Ftitle%3E%0D%0A%20%3C%2Fhead%3E%0D%0A%20%3Cbody%3E%0D%0A%20%20%3Ch1%3EPass%3C%2Fh1%3E%0D%0A%20%3C%2Fbody%3E%0D%0A%3C%2Fhtml%3E">
</body>
</html>
Comment 1•11 years ago
|
||
(In reply to Joanmarie Diggs from comment #0)
> Given the test case below from the W3C, Gecko:
which is invalid per http://www.whatwg.org/specs/web-apps/current-work/multipage/embedded-content.html#attr-source-src you need to have a src attribute on a <img> for it to be valid and it needs to contain a URl. Since this image doesn't point at a valid image I expect gecko just displays the alt attribute as text which IMO is the reasonable thing to do.
> * Does not include the "View Description" right-click menu item
I'll grant that this is arguably a bug probably caused by us rendering the image element as text not an image.
> * Does exposes the empty image as ATK_ROLE_TEXT rather than ATK_ROLE_IMAGE
given that there's no image, instead there's text on the screen that seems like the correct behavior to me.
> * Fails to expose the accessible action "showlongdesc"
basically the same as issue 1.
> As a result, it is not possible for Orca to provide any support for longdesc
> if the image is "empty."
parts 1 and three seem like they could reasonably be considered problems, but I hope you agree it'll be a very nice day when they're the most important things to work on.
Comment 2•10 years ago
|
||
I agree that's on the propose: no physical image no a11y image. It's not addressed by HTML a11y spec though. Where does the testcase go from?
Reporter | ||
Comment 3•10 years ago
|
||
It's the second test in table 1 on: http://w3c.github.io/test-results/html-longdesc/cr-report.html#mfresults-discovery
I don't care so much about the role. But how can Orca provide access to the associated longdesc? The table pointed to above suggests Orca should be able to do this.
Comment 4•10 years ago
|
||
(In reply to alexander :surkov from comment #2)
> I agree that's on the propose: no physical image no a11y image. It's not
> addressed by HTML a11y spec though. Where does the testcase go from?
Well, IMHO its addressed by the HTML spec saying its invalid as per the link in my previous comment.
> I don't care so much about the role. But how can Orca provide access to the
> associated longdesc? The table pointed to above suggests Orca should be able
> to do this.
I don't see how it can, but I'm not sure that's an issue given the test is clearly bogus given the spec for the image element.
Updated•2 years ago
|
Severity: normal → S3
You need to log in
before you can comment on or make changes to this bug.
Description
•