Fixed Inconsistent Conversation Participant Count

TimWF

Active member
There seems to be a bug in the count of users participating in a conversation that is inconsistent between what is listed on the conversation listing page (ex: http://xenforo.com/community/conversations/ ) and the count that is shown on the actual conversation page (ex: http://xenforo.com/community/conversations/post-title-example.123/ ). This isn't a consistent bug at all from what I have seen, but it certainly appears that the participant count in the two locations above is somehow calculated differently causing inconsistency in some cases.
 
I can't say I've ever seen this and looking through my list, I don't notice anything out of order. Do you have wrong counts here on XF.com?
 
None here, though I only have a handful of conversations going on here.

Also something that might be of use, in one of the cases that I have seen there are three people listed on the conversation list view, two listed in the conversation view and in looking at the actual participants one of the users that was in the conversation was deleted.

That isn't the only way I have seen this happening though, in another case I saw the list showed 18 participants and the conversation showed 17, though none of the participants had been deleted (and as far as the users knew there were no participants that left the conversation.)
 
It looks normal to me if I read your explination.

In list: you are one of the participants
In conversation: you are not

;)
 
I haven't encountered anything out of the ordinary with this myself. If you could work out the steps that need to be followed in order to reproduce the issue we would have more of a chance of being able to comment on it.
 
The one case that I have an idea of why it happened was when a user was deleted that was a participant in a conversations it seems their conversations would have this problem (for others involved in those). That said I have seen examples where there was not a deleted user involved as well, so that is either unrelated or only one thing that can cause it.
 
Is this anything to do with...?

Test:
In this case, to set up a conversation marker thread (or holding space) for more people to be invited later.

1. Start a new conversation with a person (so 2 are in the list)
2. The invited person leaves, permanently, which should leave 1
3. Conversation list still shows them
4. Invite more people
5. The initial invitee is still present in the list even though they've left
6. Inviting the 1st person back, does not enable them to re-attend nor see the conversation (as they've left permanently) but yet are still on the list.
 
That particular behavior is intentional - when you leave the conversation, it doesn't let people know. It assumes that if you leave the conversation permanently, you're doing it because the conversation is abusive/annoying and notifying the other parties would defeat the point. (Allowing them to be invited back would also be a problem, though we could allow that if you didn't leave permanently I suppose.)
 
That particular behavior is intentional - when you leave the conversation, it doesn't let people know. It assumes that if you leave the conversation permanently, you're doing it because the conversation is abusive/annoying and notifying the other parties would defeat the point. (Allowing them to be invited back would also be a problem, though we could allow that if you didn't leave permanently I suppose.)
Gotcha. Thanks Mike. (y)
Reading the OPs question, does my test there above pertain to the perceived problem, that actually isn't a problem?

Last thought, if someone were to leave temporarily as in your final words there Mike, then the tool is doing its job properly and "as described".
Seems all clear to me, unless there's an angle I'm missing.
Just to say I'm not experiencing any difficulties with the below option...so that's all well and good.

Future Message Handling:

  • Accept future messages
    Should this conversation receive further responses in the future, this conversation will be restored to your inbox.
 
Reading the OPs question, does my test there above pertain to the perceived problem, that actually isn't a problem?

Part of the problem relates to deleted users, but we don't know if there's another reason for the problem.

Generally speaking, users leaving shouldn't have an effect on the count. We never make an indication that a user has left the conversation, so it shouldn't appear to change.
 
I actually noticed this a few weeks ago, this conversation hasn't had anyone removed, the only people involved were the original 4 shown in the bottom right panel. Should I rebuild the counters to get the correct figures showing?

convo.webp
 
Top Bottom