I don't understand why I would have to use x-prep profiles, I make the md5 sum of the jid_full(sess->jid) and it works fine. Yet, my problem is to know if [EMAIL PROTECTED]/Gabber and [EMAIL PROTECTED]/Gabber are the same users or not. If they are, my md5 sum is perfect, but if not, I would have to use your x-prep profile.
Am I wrong ??
Thanks, good week-end
Matthias Wimmer wrote:
Hi Remy!
Remy HAREL schrieb am 2004-05-07 14:46:01:
Still in my project to make jabberd 2 an high availibility application, I'm currently adding datas in xmpp packets. These datas are MD5 hashed, and among these data there is the full_jid ( i.e. : [EMAIL PROTECTED]/Gabber for example ).
The main problem is to know which of the three parts of this full_jid are case sensitive, because if case is ignored I could have two differents md5 sums for the same full jid. If i'm not clear, the md5 sums for 'remy' and 'Remy' are differents. For a non sensitive case application it would be the same person if we work on name but two differents persons if we work on names'md5sums.
None of the parts are case-sensitive and non of them are case-insensitive. You have to use different string-prep profiles to check if two are the same. Therefore to hash them you first have to normalize them using stringprep with the nodeprep profile for the user part, the nameprep profile for the domain part and the resourceprep profile for the resource part.
Tot kijk Matthias
------------------------------------------------------------------------
_______________________________________________
jdev mailing list
[EMAIL PROTECTED]
https://jabberstudio.org/mailman/listinfo/jdev
-- Remy Harel - [EMAIL PROTECTED] G.I.C.M - Distributed Systems & IT Linux Registered User #224740
_______________________________________________ jdev mailing list [EMAIL PROTECTED] https://jabberstudio.org/mailman/listinfo/jdev
