After an evaluation, GNOME has moved from Bugzilla to GitLab. Learn more about GitLab.
No new issues can be reported in GNOME Bugzilla anymore.
To report an issue in a GNOME project, go to GNOME GitLab.
Do not go to GNOME Gitlab for: Bluefish, Doxygen, GnuCash, GStreamer, java-gnome, LDTP, NetworkManager, Tomboy.
Bug 407647 - Indicate the tree level, expanded state, and child nodes in the Pidgin buddy list.
Indicate the tree level, expanded state, and child nodes in the Pidgin buddy ...
Status: RESOLVED FIXED
Product: orca
Classification: Applications
Component: general
2.17.x
Other All
: Normal minor
: 2.22.0
Assigned To: Joanmarie Diggs (IRC: joanie)
Orca Maintainers
Depends on: 395781 474504
Blocks:
 
 
Reported: 2007-02-13 22:04 UTC by Rich Burridge
Modified: 2008-07-22 19:32 UTC
See Also:
GNOME target: ---
GNOME version: 2.17/2.18


Attachments
take 1 (9.09 KB, patch)
2007-11-15 21:31 UTC, Joanmarie Diggs (IRC: joanie)
none Details | Review
Take 2 (14.38 KB, patch)
2007-11-15 23:56 UTC, Joanmarie Diggs (IRC: joanie)
none Details | Review
minor adjustment in the custom whereAmI (14.39 KB, patch)
2007-11-16 00:08 UTC, Joanmarie Diggs (IRC: joanie)
committed Details | Review

Description Rich Burridge 2007-02-13 22:04:20 UTC
In Gaim version 2.0.0 beta 4, the structure of the buddy list was changed,
necessitating a corresponding change in Orca (see bug #376791).

While we now have access to the buddy list in these later versions of Gaim,
Orca does not indicate the tree level of items as you navigate.  As a result no
distinction is made between groups and individual "buddies".  If possible, Orca
should indicate the tree level of items in the (post 2.0 beta 3.1) buddy list.

This is a tracking bug for atk/gail bug #395781.
See that bug for more details.
Comment 1 Willie Walker 2007-05-25 16:27:35 UTC
Removing target milestone from [blocked] bugs.  We have little control over them, so we're better off letting priority and severity be our guide for poking the related components.
Comment 2 Joanmarie Diggs (IRC: joanie) 2007-11-15 21:31:33 UTC
Created attachment 99169 [details] [review]
take 1

Li posted a patch to Gail.  Having it is necessary for my patch to work.  Li's patch can be found here: http://bugzilla.gnome.org/show_bug.cgi?id=474504#c7

The combined result of both patches is (hopefully) the following:

1. We indicate the tree level (but wait, there's more!)

2. When arrowing among items, we indicate the expandedness of the node.  If the node is expanded and we're using the verbose level of speech, it should also indicate the number of children.

3. When expanding/collapsing the current node, it should indicate the change in state.

I think I got all of the issues.  Will, thoughts?
Comment 3 Joanmarie Diggs (IRC: joanie) 2007-11-15 21:50:50 UTC
Take 2 will include defensive code because we can't count on everyone have the latest gail.  (I'll look at that later this evening).
Comment 4 Joanmarie Diggs (IRC: joanie) 2007-11-15 21:58:59 UTC
Take 2 will also address the whereAmI-related issues.
Comment 5 Joanmarie Diggs (IRC: joanie) 2007-11-15 23:56:10 UTC
Created attachment 99180 [details] [review]
Take 2

This version does what the previous version did plus:

1. Does some checking before we attempt to use the new object provided to us by the Gail patch.

2. Adds a custom whereAmI to gaim.py

Will before I asked for your thoughts.  Then I immediately found new issues.  I don't want to jinx myself, but.... Thoughts? :-)
Comment 6 Joanmarie Diggs (IRC: joanie) 2007-11-16 00:08:00 UTC
Created attachment 99182 [details] [review]
minor adjustment in the custom whereAmI
Comment 7 Willie Walker 2007-11-16 14:16:31 UTC
Looks like a nice workaround to me.  Thanks!  At some point in the future, perhaps so distant that it doesn't event warrant a bug or RFE, we might want to investigate what it would mean for the client to give non-negative column numbers to hidden columns.  Joanie and I discussed this yesterday, though, and realize it might expose problems in scripts that make an assumption about the visibility of columns with non-negative index numbers (e.g., Evolution.py). Again, just something to keep in the depths of our minds and not something to perseverate on right now.  Thanks Joanie!
Comment 8 Joanmarie Diggs (IRC: joanie) 2007-11-16 18:13:40 UTC
Thanks Will!

If it's alright with you, I would feel better committing this patch after it's gone through some more testing, both without the Gail patch and with it.  Mike and Rich, mind giving it a go?  Without the Gail patch, but with my patch you should see no difference.  With the Gail patch and my patch, you should see the reported bug fixed. :-)

I think I'd also feel better committing this patch once Li has committed his.  If for some reason the decision is made to do things differently on the Gail side, this patch may no longer be applicable.

Changing the status to [testing required] and adding Li's bug as a depends.  Optimistically setting the target milestone to the next release. :-)
Comment 9 Joanmarie Diggs (IRC: joanie) 2007-11-19 15:17:39 UTC
Li's patch has been checked in so there's no longer a need to patch Gail in order to test my patch; merely get the latest Gail.  Thanks!
Comment 10 Mike Pedersen 2007-11-20 20:56:29 UTC
I've just tested this one with the latest gail and it looks good to me. 
Comment 11 Joanmarie Diggs (IRC: joanie) 2007-11-30 04:19:47 UTC
I just re-tested this "one more time to be sure" (or so I thought) with the intent of checking it in.  Alas, something seems to have changed somewhere as I'm now getting sporadic announcements of the tree levels for buddies even though the level has not changed. :-(  

I'll look at this tomorrow.
Comment 12 Joanmarie Diggs (IRC: joanie) 2007-11-30 14:17:24 UTC
Now I can't reproduce the problem and the patch seems to be doing the right thing.  Go figure.... Patch committed, moving to pending.  Please be on the lookout for the sporadic level announcements.
Comment 13 Mike Pedersen 2007-11-30 17:55:15 UTC
This one is now working fine for me.  
Comment 14 Joanmarie Diggs (IRC: joanie) 2007-12-01 00:53:53 UTC
Thanks Mike.  Closing as FIXED.