Date: 7 Apr 1998 17:43:26 -0000
Message-ID: <19980407174326.15367.qmail@cr.yp.to>
From: "D. J. Bernstein" <djb@cr.yp.to>
To: drums@cs.utk.edu
Subject: Re: Bare CRs and LFs
Mail-Followup-To: drums@cs.utk.edu
References: <19980407050912.11393.qmail@cr.yp.to> <7029.891940885@aussie.cs.mu.OZ.AU>

Robert Elz writes:
> And if a MAC user says the same about bare LFs, and starts sending you
> messages containing bare LFs in lines and insisting that you must be able
> to deal with them? (that is, not treat them as line breaks).

Thanks for asking. I do not, in fact, treat them as line breaks. I don't
accept them in the first place.

Perhaps you'll have an easier time grasping the following example. Which
of these strategies is a better way to handle queue-size limitations?

   (1) Disallow messages over 64K.

   (2) Allow arbitrarily long messages. It's up to each MTA to decide
       what it can handle, and to bounce messages it can't handle.

---Dan
Smaller, faster, safer than inetd+tcpd. http://pobox.com/~djb/ucspi-tcp.html