procmail
[Top] [All Lists]

Re: digest messages

2000-03-08 18:38:52
On Wed, 8 Mar 2000, Philip Guenther wrote:

pg> "S.Toms" <tomas(_at_)primenet(_dot_)com> writes:
pg> >On Tue, 7 Mar 2000, Philip Guenther wrote:
pg> >
pg> >pg> You almost certainly don't want the 'f' flag on that recipe.  It'll 
cause
pg> >pg> an empty message to be processed by the rest of your rcfile.  You 
should
pg> >pg> instead use the 'w' flag to catch the return code of the formail 
command.
pg> >pg> As for making this nicer, you can give formail header alteration flags
pg> >pg> (-i, -a, etc) when splitting and they will be applied to each message
pg> >pg> that is split out.
pg> >pg> 
pg> >pg>        :0 w:
pg> >pg>        | formail +1 -bfi "Reply-To: blah(_at_)blah(_dot_)com" -ds >> 
mailfolder
pg> >pg> 
pg> >
pg>
pg> Sorry, I should have caught that you need to drop the -f flag.  With it,
pg> formail doesn't add the "From " separator to each message burst from
pg> the digest.  It worked in the two formail version because the splitting
pg> version added them.  Also, the -b flag needs to be dropped, as otherwise
pg> an embedded "From " line won't be escaped, resulting in a bogus split in
pg> the mailbox when your mail client reads it.  The -b and -f flags should
pg> only be used when working with non-mbox mailbox formats or when feeding
pg> the message into programs that know they're dealing with one message and
pg> won't try to split it.
pg> 
pg> So, give this a try:
pg> 
pg>     :0 w:
pg>     | formail +1 -i "Reply-To: blah(_at_)blah(_dot_)com" -ds >> mailfolder
pg> 
pg> 

  Yeah, that's what I ended up with, another user messaged me last night
with pretty much the same response. Although, what did you mean above with
the 'two formail version'? Were you referring to version 2 or did you mean
something else?

pg> Philip Guenther
pg> 

-- 
        S.Toms - tomas(_at_)primenet(_dot_)com - www.primenet.com/~tomas
                   SuSE Linux v6.3+ - Kernel 2.2.14

In the long run, every program becomes rococo, and then rubble.
                -- Alan Perlis

<Prev in Thread] Current Thread [Next in Thread>