Re: [Exim] -t and Resent- header lines

On Thu, May 08, 2003 at 02:47:59PM -0400, Greg A. Woods wrote:
> > Subject: Re: [Exim] -t and Resent- header lines
> > RFC2822 needs to be clearer, and should require a seperator between
> > blocks as it does for Trace fields.
>
> Ineed, but that's far from the correct solution. Anything that
> introduces required (i.e. "MUST") order to headers is totally bogus and
> broken by definition.
> The best solution is for the MTA to simply add another "resent-" prefix
> to all existing "resent-*" headers when it is resending a message so
> that now most sane MTAs will simply ignore these new "resent-resent-*"
> headers. Note too that this is existing practice, though I forget just
> exactly where. (maybe even in the MUA I use, ViewMail! ;-)

So, should the MTA be responsible for adding Resent- headers, or the
MUA? The MTA might be the "right" place, I don't know where it currently
happens from.
Sounds like RMS wants the MTA to handle it, in which case I thought the
result should be easy: ignore all previous Resent- headers, and add your
own :-)