[Webkit-unassigned] [Bug 98357] [GTK] accessibility/aria-readonly.html is failing

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=98357





--- Comment #4 from Joanmarie Diggs (irc: joanie) <jdiggs at igalia.com>  2012-12-10 09:46:30 PST ---
(In reply to comment #3)

> Does this get set from calling AccessibilityObject::isAttributeSettable()? 

ATK_STATE_EDITABLE is set here:
>From http://trac.webkit.org/browser/trunk/Source/WebCore/accessibility/atk/WebKitAccessibleWrapperAtk.cpp#L632:

  // FIXME: isReadOnly does not seem to do the right thing for
  // controls, so check explicitly for them. In addition, because
  // isReadOnly is false for listBoxOptions, we need to add one
  // more check so that we do not present them as being "editable".
  if ((!coreObject->isReadOnly()
      || (coreObject->isControl() && coreObject->canSetValueAttribute()))
      && !isListBoxOption)
      atk_state_set_add_state(stateSet, ATK_STATE_EDITABLE);

> Ideally, you'd like your implementation to match that. I'm not sure editable and read-only are always the same thing,

Agreed. We have a bug to add a new state in ATK specific to readonly:
https://bugzilla.gnome.org/show_bug.cgi?id=665598

> I am not sure what is the right behavior. I'll get James craig from ARIA W3C group to check

Awesome. Thank you!! I also pinged Joseph Scheuhammer (GNOME Accessibility Team guy who I believe also participates in that group). I will feel better having an official verdict. (Though I am still highly uncomfortable having a screen reader report something as read-only/non-editable when in reality it is editable.)

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