718990 - Atk-based plugins to slot into our a11y tree stucks orca in some flash web pages.

Status: RESOLVED FIXED

(Core :: Disability Access APIs, defect)

Description

[Jignesh Kakadiya [:jhk]]

By Looking at Bug 665820 comment 3. It observed that implementation of atk-based plugins in Bug 480317 leads orca to stuck in some of the flash based web pages.

Comment 1

[Jignesh Kakadiya [:jhk]]

Attachment: Patch

Comment 2

[Ginn Chen]

I've no access to Bug 665820, but it looks like you're simply backout Bug 480317.

Comment 3

[Jignesh Kakadiya [:jhk]]

(In reply to Ginn Chen from comment #2)
> I've no access to Bug 665820, but it looks like you're simply backout Bug
> 480317.

Sorry! here is the link. https://bugzilla.gnome.org/show_bug.cgi?id=665820. 
Yes this is not final patch. It may require some debugging.

Comment 4

[alexander :surkov (:asurkov)]

please cc us to bug 665820, I don't have an access too.

Comment 5

[alexander :surkov (:asurkov)]

(In reply to alexander :surkov from comment #4)
> please cc us to bug 665820, I don't have an access too.

I missed that was supposed to be gnome bug, ignore my last comment.

Comment 6

[alexander :surkov (:asurkov)]

(In reply to Ginn Chen from comment #2)
> but it looks like you're simply backout Bug
> 480317.

yep, that's what Trevor asked for in https://bugzilla.gnome.org/show_bug.cgi?id=665820 and it shouldn't be considered as proposed patch.

Comment 7

[Trevor Saunders (:tbsaunde)]

(In reply to alexander :surkov from comment #6)
> (In reply to Ginn Chen from comment #2)
> > but it looks like you're simply backout Bug
> > 480317.
> 
> yep, that's what Trevor asked for in
> https://bugzilla.gnome.org/show_bug.cgi?id=665820 and it shouldn't be
> considered as proposed patch.

yeah, I asked for it as an experiment not as a fix.

Comment 8

[Trevor Saunders (:tbsaunde)]

Attachment: patch

noticed this while reviewing other stuff, it doesn't really make sense to treat "" as a valid atk plug id

Comment 9

[alexander :surkov (:asurkov)]

Comment on attachment 591712 [details] [diff] [review]
patch

r=me for accessible module change, since don't see anything bad here but when this case happens?

Comment 10

[Trevor Saunders (:tbsaunde)]

(In reply to alexander :surkov from comment #9)
> Comment on attachment 591712 [details] [diff] [review]
> patch
> 
> r=me for accessible module change, since don't see anything bad here but
> when this case happens?

when the plugin  isn't accessible.  I'm not sure why  the plug id is "" but if I load a youtube video with a breakpoint in the constructor for AtkSocketAccessible I see that plugId its arg has mData of "".

Comment 11

[Trevor Saunders (:tbsaunde)]

landed https://hg.mozilla.org/integration/mozilla-inbound/rev/e4e85a8b32c7

Comment 12

[Matt Brubeck (:mbrubeck)]

https://hg.mozilla.org/mozilla-central/rev/e4e85a8b32c7

Comment 13

[David Bolter [:davidb] (NeedInfo me for attention)]

Comment on attachment 591712 [details] [diff] [review]
patch

Nice.