[Standards] Reg MUC Roomname

Peter Saint-Andre stpeter at stpeter.im
Tue Jul 1 22:10:59 UTC 2008


Michal 'vorner' Vaner wrote:
> Hello
> 
> On Mon, Jun 30, 2008 at 09:29:03PM +0530, Thanik wrote:
>> Can someone clarify whether the roomname in MUC is case sensitive or not?

The concept of "case sensitive" applies only to US-ASCII -- or at least 
it does not apply to all UTF-8 characters. Stringprep uses the concept 
of "case folding" instead.

>> Also clarify whether the following attributes are case sensitive or not?
>> 1) Transaction ID (TID)
> 
> Where's that thing from?

Maybe the 'id' attribute?

>> 2) Username in JID
> 
> Case insensitive (see nodeprep profile). Or, it is close-to case
> insensitive, it also considers other similar-looking characters as the
> same (like roman number four and IV).

Right, but even stringprep doesn't help you with that AFAIK.

>> 3) Room User Nickname
> 
> Case sensitive, as it is JID resource, it goes trough resourceprep.

Correct. Some people find that confusing, though, because "vorner" is a 
different person than "Vorner" (etc.).

>> 4) Room Name
> 
> Case insensitive ‒ it is node part (username) of JID.

Correct.

>> 5) PrivacyList Name

Unspecified. :(

I think we'd apply resourceprep to determine matching. But that needs to 
be added to XEP-0016.

>> 6) Group Name of a Roster Item.

According to section 2.3.2 of draft-saintandre-rfc3921bis, duplicate 
groups are determined using the Resourceprep profile of stringprep.

I think that could be called out more prominently in the spec, so I'll 
fix that in the next version.

> I don't know, try looking which stringpreps and rules apply to them (in
> XMPP-IM).
>> In specification they have not mentioned anything about this.
> 
> They do, it is just well hidden ;-)

:)

Peter



More information about the Standards mailing list