Subject: Re: [libssh2] Bug 1703472

Re: [libssh2] Bug 1703472

From: Sara Golemon <>
Date: Sat, 16 Jun 2007 23:27:52 -0700 (PDT)

> Eberhard Mattes has filed a bug report about
> the libssh2 use of 0xFF instead of 1 for a TRUE
> boolean value. He is correct about the RFC,
> , section 5.
> It does look like we are in the wrong, and have
> been since the beginning. My question is why might
> 0xFF been used. He says it fails with one of the
> commercial servers, but would 0x01 fail on more?
> Is it really safe to do the change?
> Should there be an option to use 0xFF?
Well, we're in the wrong, but the servers that aren't respecting non-zero
values as truth are wrong too. That said, we should change it so that
we're at least obeying the rules.

In defense of both libssh2's behavior and the behavior of that server
implementation, the earlier SSH2 drafts didn't contain this clarification
(at least, not that I can recall...

Anyway, let's make the change. If it breaks compatability with anyone,
it's that server's fault and we can introduce a compapability fallback in
that case, but there's no use trying to jump in front of a problem that
probably doesn't exist.


This email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
libssh2-devel mailing list
Received on 2007-06-17