[jdev] stringprep

Remy HAREL remy.harel at GICM.FR
Tue May 11 09:24:14 CDT 2004

In fact, I add datas in xmpp stanzas to know the sender of a stanza and 
a number to assume ordering between any two entities... I would like to 
have something like Seq:matthias at jabber.gicm.net/Gabber:911 , but in 
order to reduce the size, I hash with MD5 implementation in jabber to 
obtain something like Seq:a3e4bbf6381f89ff3ebf22332268becb:911.

 I have currently two problems :

  first, unicode is use, so your jid could be japanese characters. I use 
stringnameprep which seems to be ok.
  second, matthias at jabber.gicm.net/Gabber and 
Matthias at jabber.gicm.net/Gabber  are the same users where as their md5 
sums are different. so I have to test it. Once I've tested it, I've seen 
that string prep has convert Matthias to matthias, so it's allright.

  So, since I use stringprep_nameprep, I think it's ok and my md5 sums 
are rights, and if i have two differents md5 sums, that's probably 
because that's 2 differents full jids. I have just to make sure and not 
"probably". Do you understand matthias what I mean ?

One more time, thx

Matthias Wimmer wrote:

>Hi Remy!
>Remy HAREL schrieb am 2004-05-11 15:47:43:
>>does anyone knows how works stringprep and profiles ? I'm sturying it 
>>but if someone has already used it it would be graceful to help me...
>Install the libidn documentation and execute "info libidn". You will get
>described how it works in detail.
>For your work on jabberd2 ... I don't know if you need to care about it
>at all. It depends on what you are exactly doing. It might be that you
>are working on already normalized strings.
>Tot kijk
>    Matthias
>jdev mailing list
>jdev at jabber.org

Remy Harel - remy.harel at gicm.fr
G.I.C.M - Distributed Systems & IT
Linux Registered User #224740

More information about the JDev mailing list