[Webkit-unassigned] [Bug 95286] New: buildAccessibilityTree() needs to be removed from individual layout tests

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=95286

           Summary: buildAccessibilityTree() needs to be removed from
                    individual layout tests
           Product: WebKit
           Version: 528+ (Nightly build)
          Platform: Unspecified
        OS/Version: Unspecified
            Status: UNCONFIRMED
          Severity: Normal
          Priority: P2
         Component: Accessibility
        AssignedTo: webkit-unassigned at lists.webkit.org
        ReportedBy: jdiggs at igalia.com
                CC: cfleizach at apple.com, apinheiro at igalia.com


I stumbled across a nifty function that seems like it might be an excellent way to have more WebCore-wide a11y Layout tests and fewer mysterious failures and ignores. It's the buildAccessibilityTree() function found here:

  http://trac.webkit.org/browser/trunk/LayoutTests/accessibility/adjacent-continuations-cause-assertion-failure.html

and here:

  http://trac.webkit.org/browser/trunk/LayoutTests/accessibility/div-within-anchors-causes-crash.html

and here:

  http://trac.webkit.org/browser/trunk/LayoutTests/accessibility/deleting-iframe-destroys-axcache.html

It would also solve the problem found here:

  http://trac.webkit.org/browser/trunk/LayoutTests/accessibility/legend.html

Because at least two platforms (Mac and Gtk) have very different a11y hierarchies (not just roles, but objects included and numbers of children), the expected results will often be quite different. But the fundamental things being tested often are not.

So.... Why is this method not <somewhere else> where we can stop duplicating it and start using for all our tests? Assuming the answer is "We just haven't gotten to it yet", where should it go?

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