On Mon, Jan 05, 2004 at 10:13:11PM +0000, David Woodhouse wrote:
>
> 'Another tricky aspect of this feature is that an early delivery happens
> before the local_scan() function and the system filter are run.'
>
> Why so? I was thinking of doing it after DATA...\r\n.\r\n is received by
> the relaying host, but before returning a response to the DATA command.
> That's when local_scan() is run already, isn't it? Not sure about the
> system filter.
In order for early delivery to work as an ACL modifier you have to run
it there and then in order for the result to be passed to the ACL.
The reason I think it should be an ACL modifier is to provide enough
flexibility in the admin's error handling options.
> Just how difficult _is_ this to achieve, architecturally? That's a
> genuine question not a rhetorical one -- my knowledge of the internals
> of Exim isn't that great.
I don't think a patch to implement it would be particularly big or
complicated, unless it ends up causing repercussions. Things that
could make it interesting include the need to write the -H file
earlier than usual, and having two Exim processes owning the same
message (both the receiver and the deliverer).
Tony.
--
f.a.n.finch <dot@???>
http://dotat.at/
NORTHWEST FITZROY WEST SOLE: SOUTHWEST BACKING SOUTH 4 INCREASING 6 TO GALE 8,
PERHAPS SEVERE GALE 9 LATER. RAIN OR DRIZZLE. MODERATE OR POOR.
