From: Charles Lindsey (chl@clw.cs.man.ac.uk)
Date: Mon Mar 04 2002 - 07:32:18 CST
In <fgPG++CWs7f8QAZC@pillar.turnpike.com> Paul Overell <paulo@turnpike.com> writes:
>In message <200202281918.TAA05694@clw.cs.man.ac.uk>, Charles Lindsey=20
><chl@clw.cs.man.ac.uk> writes
>>Here, as requested by Paul Overell, is the alternative header syntax
>>without schemata or templates. I have shown the basic format from
>>section 4, plus the headers in section 5. If this is considered OK, I
>>will apply it throughout the system.
>>
>Yes please! I like the use of =/
OK, so I shall press on.
>>
>>[Observe that other-parameters are omitted from headers whuch are
>>"unstructured".]
>>
>I have raised this in the past, but I am still very unhappy about the=20
>use of other-parameters on headers we are using from RFC2822. They are=20
>not ours to change.
Yes, I realised that this now opens the possibility to omit parameters
from other selected cases, so this issue can be discussed.
I can see it makes sense in headers such as From, Reply-To and Sender (and
especially those headers where the syntax allows an address-list which can
contain a ';' for other reasons). But I would be reluctant to do it for
those headers which are really "ours", even though they are defined,
somewhat reluctantly, in RFC 2822. Message-ID and References spring to
mind.
But in fact it makes little practical difference, because all these cases
are "SHOULD NOT generate", "MUST accept" for "our" headers, and "SHOULD
accept" for all headers (even the mail ones).
-- Charles H. Lindsey ---------At Home, doing my own thing------------------------ Tel: +44 161 436 6131 Fax: +44 161 436 6133 Web: http://www.cs.man.ac.uk/~chl Email: chl@clw.cs.man.ac.uk Snail: 5 Clerewood Ave, CHEADLE, SK8 3JU, U.K. PGP: 2C15F1A9 Fingerprint: 73 6D C2 51 93 A0 01 E7 65 E8 64 7E 14 A4 AB A5