From: greg andruk (meowing@banet.net)
Date: Fri Nov 05 1999 - 13:04:48 CST
On Fri, 5 Nov 1999, Charles Lindsey wrote:
> But hold on a minute. Our draft currently forbids duplicated headers,
> except where explicitly allowed. So what INN is doing is currently
> absolutely correct. So I think you should leave it alone unless we
> deliberately decide to allow them.
innd (nnrpd is more convoluted at the moment) only checks for duplication
of the six required headers, and SFAIK that's all it ever done. Its
notion of "standard headers" will need to be revamped a bit to be able
to find duplicates of the optional set, though.
> Does anybody think that we should?
Anything that looks like it wants a duplicate header ought to go into
header parameters or the body instead. Otherwise too much ambiguity
creeps in.