On Oct 24, 2007 at 3:47 PM Peter Saint-Andre wrote:
> Mark Doliner wrote:
> > Oh Oct 24, 2007 at 3:38 PM Peter Saint-Andre wrote:
> >> Mark Doliner wrote:
> >>> On Oct 24, 2007 at 1:50 PM Peter Saint-Andre wrote:
> >>>> 6. XEP-0175: Best Practices for Use of SASL ANONYMOUS
> >>>>
> >>>> Approve version 1.1pre1?
> >>>>
> >>>> All Council members in attendance voted +1. Ralph to vote on the
> list.
> >>>> Dave raised issue about possible confusion regarding trace data.
> >> Council
> >>>> wordsmithed an acceptable solution during the meeting:
> >>>>
> >>>> http://svn.xmpp.org:18080/browse/XMPP/trunk/extensions/xep-
> >>>> 0175.xml?r1=1297&r2=1310
> >>> The revised text says, "although RFC 4505 allows the initiating entity
> >> (client) to provide so-called 'trace data' when authenticating via SASL
> >> ANONYMOUS, it is NOT RECOMMENDED to include trace data as the XML
> >> character data of the <auth/> element (instead, the <auth/> element
> SHOULD
> >> be empty). However, if trace data is included, the server MUST NOT use
> it
> >> for any purpose other than tracing (e.g., in server logs)."
> >>> However, this seems to conflict with section 7.5.5 of draft-
> saintandre-
> >> rfc3920bis-04 [1] which shows a "malformed-request" failure when this
> data
> >> is provided for SASL ANONYMOUS.
> >>
> >> I think I already fixed that:
> >>
> >> http://svn.xmpp.org:18080/browse/XMPP/trunk/internet-drafts/draft-
> >> saintandre-rfc3920bis-05.xml?r1=1294&r2=1296
> >
> > Oh, er, is "[ ... extremely-long-token ... ]" intended to imply that the
> token is unusably long, like a million characters or something absurd?
> 
> It's intended to mean that it violates the rule from RFC 4505:
> 
>       token       = 1*255TCHAR

I see, thanks.  For what it's worth that wasn't immediately obviously to me 
from reading the spec, but I'm not used to reading specs.

-Mark

Reply via email to