From: Charles Lindsey (chl@clw.cs.man.ac.uk)
Date: Wed Oct 20 1999 - 04:34:31 CDT
In <1dzxhbw.1w4b7bl72487jM%phenix@interpath.com> phenix@interpath.com (John Moreno) writes:
Do I interpret that as a mild preference for #1?
>I think either will *do, but have a couple of quibbles -- the MUST seems,
>well unusual, if someone wants their program to crash let them, and since
>it is a warning that they WILL encounter messages with more than 21 id's
>why doesn't the second warning (that they WILL encounter things in the
>header which are NOT id's) have the same status? Particularly since it's
>going to happen more often?
> Trimming SHOULD be done by removing sufficient identifiers starting
> with the second so as to bring the total down to 21. Non-compliant
> programs will sometimes include more than 21 id's and/or text that is
> not id's.
I am not convinced that Reference counts over 21 are necessarily
non-compliant. The syntax does not say so, and neither does anything
written in section 4. I agree they should be discouraged, but trimming is
currently only a SHOULD (and it will be a long time before everybody is
routinely doing it).
How about:
However, it would be wrong to assume that References headers containing
more than 21 message identifiers will not occur.
-- Charles H. Lindsey ---------At Home, doing my own thing------------------------ Email: chl@clw.cs.man.ac.uk Web: http://www.cs.man.ac.uk/~chl Voice/Fax: +44 161 437 4506 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