| commit | cd2bf8971943379bd7828aefa0a4736efa10e310 | |
| author | Steffen (Daode) Nurpmeso <steffen@sdaoden.eu> | |
| Mon, 5 Sep 2016 14:53:45 +0000 (5 16:53 +0200) | ||
| committer | Steffen (Daode) Nurpmeso <steffen@sdaoden.eu> | |
| Thu, 15 Sep 2016 13:58:26 +0000 (15 15:58 +0200) | ||
| tree | 776755d29a18cfb7853abb89840e5bb79c5a85f0 | treesnapshot (tar.gz zip) |
| parent | 45d488d22c8327eb4dbe40c3da413dee81c7fbdc | commitdiff |
sendout.c: fix HAVE_DEBUG From_ quoting..
Rename __savemail() to a_sendout__savemail() and rewrite it a bit.
If we resend a message then we (at least: yet) don't reevaluate
it's contents for real, so if a user resends a message that was
produced by a MUA which doesn't force MIME encoding when there is
"From " in a message body, then (with *record-resent*) we need to
MBOXO quote the message if we save it in *record*.
Sofar i had a HAVE_DEBUG assertion which did _not_ do it, but the
regular user shipout code always did.
Also there was a resource leak in the error path. The file would
have been cleaned up when the command loop ticks, though.
Rename __savemail() to a_sendout__savemail() and rewrite it a bit.
If we resend a message then we (at least: yet) don't reevaluate
it's contents for real, so if a user resends a message that was
produced by a MUA which doesn't force MIME encoding when there is
"From " in a message body, then (with *record-resent*) we need to
MBOXO quote the message if we save it in *record*.
Sofar i had a HAVE_DEBUG assertion which did _not_ do it, but the
regular user shipout code always did.
Also there was a resource leak in the error path. The file would
have been cleaned up when the command loop ticks, though.
| sendout.c | diffblobblamehistory |