-$Cambridge: exim/doc/doc-txt/ChangeLog,v 1.97 2005/03/22 15:45:35 ph10 Exp $
+$Cambridge: exim/doc/doc-txt/ChangeLog,v 1.98 2005/03/22 16:44:04 ph10 Exp $
Change log file for Exim from version 4.21
-------------------------------------------
PH/12 If the last host in a fallback_hosts list was multihomed, only the first
of its addresses was ever tried. (Bugzilla bug #2.)
+PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
+ the result incorrectly in the debug output. (It correctly added a newline
+ to what was transported.)
+
A note about Exim versions 4.44 and 4.50
----------------------------------------
-/* $Cambridge: exim/src/src/transport.c,v 1.6 2005/03/08 16:57:28 ph10 Exp $ */
+/* $Cambridge: exim/src/src/transport.c,v 1.7 2005/03/22 16:44:04 ph10 Exp $ */
/*************************************************
* Exim - an Internet mail transport agent *
same alias might share some of them) but we want to output them in the
opposite order. This is a bit tedious, but there shouldn't be very many
of them. We just walk the list twice, reversing the pointers each time,
- but on the second time, write out the items. */
+ but on the second time, write out the items.
+
+ Headers added to an address by a router are guaranteed to end with a newline.
+ */
if (addr != NULL)
{
/* If a string containing additional headers exists, expand it and write
out the result. This is done last so that if it (deliberately or accidentally)
isn't in header format, it won't mess up any other headers. An empty string
- or a forced expansion failure are noops. */
+ or a forced expansion failure are noops. An added header string from a
+ transport may not end with a newline; add one if it does not. */
if (add_headers != NULL)
{
if (s[len-1] != '\n' && !write_chunk(fd, US"\n", 1, use_crlf))
return FALSE;
DEBUG(D_transport)
- debug_printf("added header line(s):\n%s---\n", s);
+ {
+ debug_printf("added header line(s):\n%s", s);
+ if (s[len-1] != '\n') debug_printf("\n");
+ debug_printf("---\n");
+ }
}
}
}