[jdev] BOM

[Jonathan Dickinson]

Much obliged. As a case of interopability, maybe something like: entities MUST NOT send byte order marks, however, they MUST tolerate them.

I am sure there are one or two (possible no longer maintained) libraries out there that ignorantly write them out. 2 years ago agsxmpp did (qualified with a quick paroosal, so don't quote me on that) so I wouldn't be surprised at all if some others did.

A little OT, is there any list of specific ammendments to 3920, I am pretty familiar with 3920 so I might miss some of the changes/clarifications.

Sorry for not conforming to the list standards, I am on my phone.

-----Original Message-----
From: Waqas Hussain <waqas20 at gmail.com>
Sent: 06 November 2008 09:07 PM
To: Jabber/XMPP software development list <jdev at jabber.org>
Subject: Re: [jdev] BOM


On Thu, Nov 6, 2008 at 4:00 PM, Jonathan Dickinson
<jonathan.dickinson at k2.com> wrote:
> Hi All,
>
>
>
> What's the official stance on BOM (Byte order marks)? I have had to
> specifically remove them because most client (e.g. Cocinnella) sporadically
> break if they are there (strangely, first time you try to log in, after that
> it's fine).
>
>
>
> The client says unexpected char "[]" (where the block is an 'unprintable'),
> so I assume it's the BOM.
>
>
>
> -- Jonathan
>
> _______________________________________________
> JDev mailing list
> FAQ: http://www.jabber.org/discussion-lists/jdev-faq
> Forum: http://www.jabberforum.org/forumdisplay.php?f=20
> Info: http://mail.jabber.org/mailman/listinfo/jdev
> Unsubscribe: JDev-unsubscribe at jabber.org
> _______________________________________________
>
>

While RFC 3920 [XMPP] doesn't say anything about byte order marks, RFC
3629 [UTF-8] does suggest:

   o  A protocol SHOULD forbid use of U+FEFF as a signature for those
      textual protocol elements that the protocol mandates to be always
      UTF-8, the signature function being totally useless in those
      cases.

I don't think the use of the byte order marks in an XMPP stream was
ever intended by the protocol designers. This looks like an oversight
and should be explicitly specified in the XMPP RFC.

--
Waqas Hussain
_______________________________________________
JDev mailing list
FAQ: http://www.jabber.org/discussion-lists/jdev-faq
Forum: http://www.jabberforum.org/forumdisplay.php?f=20
Info: http://mail.jabber.org/mailman/listinfo/jdev
Unsubscribe: JDev-unsubscribe at jabber.org
_______________________________________________