On Thu, 02 Dec 1999 09:16:56 -0500, "W. Mark Herrick, Jr."
<markh(_at_)va(_dot_)rr(_dot_)com> wrote:
I *think* it has something to do with the pipe. The error that I was
getting in my log file was "Error while writing to cat".
That's extremely much different from cat not being found. You need an
i flag on any recipe which (on purpose) fails to read the input that
Procmail pipes to it. (But adding a dash to cat's arguments fixes that
by reading the entire pipe, of course, so your fixed recipe should
certainly not have an i flag.)
Here's a suggestion for handling both that and empty incoming
subjects, and saving a process by not calling formail there at
all:
SUBJ='(no subject)' # default value
:0h # if there is a non-blank subject, use that instead
* ^Subject:[ ]*\/[^ ].*
{ SUBJ=$MATCH }
That's cool. So, then later on I would still call $SUBJ?
[and not MATCH]?
That's the idea, yes. The tricky case is coping with messages which
have an empty Subject: header. I was sort of under the impression that
you didn't want to cope with those at all, but maybe I misread one of
your earlier messages.
:0 fhw
| formail -rtI"From: $FROMSIG" \
-I"Reply-To: please_do_not_reply(_at_)rr(_dot_)com" \
-I"Precedence: junk" \
-I"Subject: [$ACKM] $SUBJ " \
-I"X-Loop: $MY_ADDR" \
-I"References:"
If you don't want people to reply, you might also want to set the
envelope sender to something invalid, or to something that forwards to
/dev/null. (I might prefer an invalid address. Those who generate
double bounces deserve to have them delivered to their postmaster,
dammit.)
I don't understand why you zap References: here. Some threading mail
readers rely on References: too (at least as a secondary when
In-Reply-To: is in a silly format, which it won't be if formail
generated it, fortunately).
/* era */
--
Too much to say to fit into this .signature anyway: <http://www.iki.fi/era/>
Fight spam in Europe: <http://www.euro.cauce.org/> * Sign the EU petition