[Tickets #11128] Re: Infinite loop expanding containers
    bugs at horde.org 
    bugs at horde.org
       
    Thu Apr  5 17:14:59 UTC 2012
    
    
  
DO NOT REPLY TO THIS MESSAGE. THIS EMAIL ADDRESS IS NOT MONITORED.
Ticket URL: http://bugs.horde.org/ticket/11128
------------------------------------------------------------------------------
  Ticket             | 11128
  Updated By         | Michael Slusarz <slusarz at horde.org>
  Summary            | Infinite loop expanding containers
  Queue              | IMP
  Version            | Git master
  Type               | Bug
  State              | Feedback
  Priority           | 2. Medium
  Milestone          |
  Patch              |
  Owners             | Michael Slusarz
------------------------------------------------------------------------------
Michael Slusarz <slusarz at horde.org> (2012-04-05 11:14) wrote:
>> Is "sent" really a container?  If so, it should NOT appear in dynamic
>> view since it doesn't have any non-special children.
>
> Good point. It has unsubscribed folders, but it indeed only has  
> subscribed special folders. I agree it shouldn't show up at all in  
> the first place.
There's still some tweakiness when it comes to these "empty"  
containers (meaning containers that only contain special mailboxes).   
For example, if you subsequently add a mailbox to this container, it  
won't appear until the tree is rebuilt.  And if you delete this  
mailbox, the container won't be destroyed.
Doing the latter, I can end up in the situation where we have just the  
container.  But opening the container does not cause an infinite loop  
- it only calls listMailboxes once.
I don't think having an unsubscribed mailbox should change things,  
since the Tree code is not normally aware this mailbox ever exists.
>> Sure enough, if I create a container with only special mailboxes, the
>> container doesn't appear in dynamic view.
>
> I'll see if I can track further down why it's added to the tree.
This whole special-mailboxes container code is difficult to implement  
correctly.  I might have to go back and give it a second look.
    
    
More information about the bugs
mailing list