In article <Pine.SOL.4.44.0203050849320.18501-100000@???>,
Philip Hazel <ph10@???> wrote:
>If I recall correctly, you had two addresses in the message which both
>alias to "you". Is that right?
Yes, that is correct.
>So you have
>
>address1 -> you -> .forward -> pipe
>
>and then Exim hits
>
>address2 -> you
Yes.
>At this point it says "that's a duplicate address", so it discards it.
>Consequently, it never knows that the second address would also end up
>at the pipe if delivered on its own.
>At least, I think this is what's happening.
Oh dear. But can't it at that point add the duplicate to the
list of addresses that will end up in the envelope-to header
anyway? Perhaps with a flag set "don't really deliver" so that
the rest of the code ignores it ?
That's the way I implemented it in sendmail some time ago, btw
But the patch never got accepted, so sendmail still doesn't
have (X-)Envelope-to support.
>I'm afraid this is pretty fundamental to the way Exim works, and it is
>the same in Exim 4.
Ouch. I just noticed that by switching from procmail to an exim
filter, the same thing happens. So that doesn't help - it's the
.forward file processing that causes it.
>I note the problem, but I don't think there's anything I can do about it
>in a hurry. Sorry.
Do you think it can eventually be fixed, longer term ?
Anyway, if I can make some time I'll dive into the source myself,
since I'll be needing this.
Mike.
--
Computers are useless, they only give answers. --Pablo Picasso
