[jdev] Re: SOCKS5 Bytestream File Transfer problem

Anthony Ortiz anthonypaulo at gmail.com
Fri Feb 25 09:30:00 CST 2005


"I SEE!!" said the blind man to the deaf man as he pissed into the
wind... "It's all coming back to me!"

I get it... so what you're saying is that given :

venkata at 68.101.84.233/Psi

I should do :

Nodeprep("venkata") + "@" + Nameprep("68.101.84.233") + "/" +
Resourceprep("Psi")

Well, I've implemented it and it seems that it didn't do a thing to
the JID's... they're all the same, which explains why your hash came
out right. Well, now mine matches yours and so now I guess we're on
the same page as far as the SHA1 goes, but it still doesn't work,
still gives me the "Unknown Host" exception. :(

Thanks for the assist on the hash Magnus!

As for the SOCKS5 connection problem... anyone have any ideas?

Anthony

On Fri, 25 Feb 2005 15:28:24 +0100, Magnus Henoch <mange at freemail.hu> wrote:
> Anthony Ortiz <anthonypaulo at gmail.com> writes:
> 
> > Thanks for the quick reply, but there's no problem with the SHA-1 due to :
> >
> > 1) The difference in our hash results is due to your applying SHA1(sid
> > + from + to) rather than SHA1(sid + Stringprep.nameprep(from + to)),
> > which in this particular case simply transforms the from and to
> > variables to all lowercase. So your hash is based on a string
> > containing uppercase characters while mine was based on lowercase
> > characters, and the protocol specifically states that the JID's need
> > to be Stringprepped, so I assume my hash is correct.
> 
> Ah... I didn't think about that.
> 
> But it seems that I accidentally generated the correct hash.  The
> resources should be stringprepped according to the Resourceprep
> profile, which doesn't fold case, even though both the username (by
> Nodeprep) and the server (by Nameprep) do.  So you need to split the
> JIDs and stringprep the parts separately.
> 
> Magnus
> 
> _______________________________________________
> jdev mailing list
> jdev at jabber.org
> http://mail.jabber.org/mailman/listinfo/jdev
>



More information about the JDev mailing list