From: Philip Hazel <ph10@???>
> On Fri, 24 Sep 1999, Paul Makepeace wrote:
> > (autoreply.c has been minisculey hacked to enable this.)
>
> To enable what, exactly?
To enable me to use the combination of headers= and text= in the autoreply
transport to send message/rfc822 MIME encapsulated bounce messages. You can
see this (and the LDAP name resolution/suggestion system) in effect if you
mail the non-existent account macgregor@???
What I did was remove the lines which exim puts in at the front of the body
since these need to be MIME message delimiter paragraphs.
--- src/transports/autoreply.c Mon Aug 2 10:43:09 1999
+++ ../exim-3.03p1/src/transports/autoreply.c Wed Sep 8 17:13:32 1999
@@ -432,6 +432,12 @@
/* Copy the original message if required, observing the return size
limit. */
+
+/* Paul Makepeace, 23 June 1999: experimentally removing the
+ message about the copying of the message.
+ Action: commented out both headers.
+ */
+
if (return_message)
{
if (return_size_limit > 0)
@@ -447,11 +453,11 @@
"------ The body of the message is %d characters long; only the first\n"
"------ %d or so are included here.\n\n", size, (max/1000)*1000);
}
- else fprintf(f, "\n"
-"------ This is a copy of the message, including all the
headers. ------\n\n");
+ else ; /* fprintf(f, "\n"
+"------ This is a copy of the message, including all the
headers. ------\n\n");*/
}
- else fprintf(f, "\n"
-"------ This is a copy of the message, including all the
headers. ------\n\n");
+ else ; /* fprintf(f, "\n"
+"------ This is a copy of the message, including all the
headers. ------\n\n");*/
fflush(f);
transport_write_message(addr, fileno(f),
> The reason for checking for printing characters is to ensure that the
> format of the header lines conforms to RFC 822. It also seemed sensible
> to me that the text of the message should conform in this case.
I'm sure you're right: I just noticed them occurring with marked regularity
from people with non-English sounding surnames!
I mentioned this before and I think it's even on the wishlist but having the
ability to create MIME encapsulated bounce messages would be I think very
useful. And that's not just because our (paying as opposed to email) client
requires it or that it's on the IETF Standards Track. I'd ultimately like to
see exim heading towards full Disposition Notification `a la RFC2298
http://www.ietf.org/rfc/rfc2298.txt itself based on RFC1894 about DSNs ("A
DSN can be used to notify the sender of a message of any of several
conditions: failed delivery, delayed delivery, successful delivery, ...")
> > Is there any way of having exim ignore these errors?
>
> You can set print_topbitchars, but that is a global switch.
Thanks, I'll take a look at that.
Paul
>
> --
> Philip Hazel University of Cambridge Computing Service,
> ph10@??? Cambridge, England. Phone: +44 1223 334714.
