powered by Jive Software

Psi vCards

The Psi client prompts a user with a vCard every time they log in until they fill out at least one field and publish it. I asked on the Psi forum if there was a way to not prompt a user with this after first login or something to that degree. There really is no way around it without populating some information to it. They suggested I check here to find out if it’‘s possible to populate a few basic pieces of information into the vCard automatically. We are using LDAP connecting to our AD for user authentication. Is there a way we could fill out a few fields for the vCard via LDAP automatically? We’'d rather not leave it up to the users to put information into the vCard and have to deal with that popup if possible.

What’'s the scoop on vCards and auto-populating them with LDAP info?

You can follow the progress of this issue (and vote for it!) here: JM-121. I tested the code attached to that JIRA issue but uncovered a problem in the way JM’'s vCard IQ handler (even without the ldap-vcard code applied). You can see me talking about it in this thread:

http://www.jivesoftware.org/forums/thread.jspa?threadID=14905&tstart=15

I’‘ve emailed a patch to Matt for the core problem, but haven’'t seen it in CVS yet. With my patch and the code Robert posted to the JIRA issue, LDAP vCard works for me (with Active Directory and Pandion).

Hey Cameron,

Sorry for not getting back to you before about your patch. Before Matt sent me your patch I was adding support for vCards in Smack (implemented by a community member) and doing some tests with Jive Messenger. While doing those tests I found that we are having an important problem with vCards in the server. The main problem is that we are mixing information and even losing information in the vCards. I checked your patch and if I’‘m not wrong it is fixing something else. So I decided to review your patch when I’'m going to fix the vCard problem we are having.

Regards,

– Gato

Gato,

Thanks for the response. It sounds like you may have found a separate issue. The second half of the dev thread does a pretty good job of detailing the problem I was trying to fix. Let me know if you have any questions or comments. Thanks

Cameron