[Webkit-unassigned] [Bug 158107] New: [GTK][Stable]: Merge r201072 and r201216 (and some misc gardening of accessibility tests)

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=158107

            Bug ID: 158107
           Summary: [GTK][Stable]: Merge r201072 and r201216 (and some
                    misc gardening of accessibility tests)
    Classification: Unclassified
           Product: WebKit
           Version: Other
          Hardware: Unspecified
                OS: Linux
            Status: NEW
          Severity: Normal
          Priority: P2
         Component: WebKit Gtk
          Assignee: webkit-unassigned at lists.webkit.org
          Reporter: jdiggs at igalia.com
                CC: bugs-noreply at webkitgtk.org

Created attachment 279859
  --> https://bugs.webkit.org/attachment.cgi?id=279859&action=review
Proposed patch

1. ​https://trac.webkit.org/changeset/201072 AX: [ATK] Use WebCore Accessibility's AccessibilityText for AtkObject name and description 

This change brings us into better compliance with a couple of W3C specs and fixes several bugs in which the accessible name of objects was missing or broken. Because Orca doesn't always know when/if the accessible name is broken, having this fix is desirable.

As part of that change, quite a few related layout tests were modified in a sad-hack interim step so that we could verify the correctness of the changes. The sad test hacks were removed as part of the next item.

2. https://trac.webkit.org/changeset/201216 AX: Layout tests related to text alternative computation need to be done differently

In addition to undoing the aforementioned sad test hacks, some testing utilities were added (just in a JavaScript resources file) and a bunch of tests modified to use those utilities. While this would not normally be something to include in a stable release, I would like us to include it if there are no objections. The reason why is that I'm finding quite a few accessibility bugs that impact Orca users. Whenever possible, I would like to backport them to the stable branch. And if we are going to cherry-pick patches for stable, I would like to be able to use our Layout Tests to ensure that we're cherry-picking everything needed and not breaking the stable release.

3. Some unrelated gardening that was done in master and would be helpful to include for reason described above.

Given that 2 is a continuation of 1, and limited to test changes only, I'm attaching a single patch.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.webkit.org/pipermail/webkit-unassigned/attachments/20160526/055692a3/attachment.html>