Re: Limiting extraction with \/

On Wed, 6 Jan 1999 22:52:43 -0700 (MST), Bill McClatchie
<wmcclatc(_at_)primenet(_dot_)com> wrote:
> I am re-doing most of my procmail filters; and have been using the
> following for most of my mailing lists.  How can I eliminate the sed
> command?
> :0:
> * ^From[     ]owner-\/.*@
> {
>         LIST=`echo $MATCH|sed -e 's/@/\.gz/'`
>
>         :0 w:
>         |gzip -fc >> $LIST
> }
This can be very hard in the general case, but assuming we can assume
[heh] that there is only going to be one @ in the string (unless you
subscribe to some rather unusually set up lists, nothing should
break), just look for everything non-@ and grab that into MATCH
instead:

    :0w:
    * ^From[    ]+owner-\/[^(_at_)]+
    | gzip -fc >>$MATCH

If you want to verify that there +is+ an @ in the string, and/or
specifically need the $LIST variable to be set after this recipe, you
can do something fairly close to what you already had:

    :0  # extraneous locallockfile elided -- you can't lock on a brace
    * ^From[    ]+owner-\/[^(_at_)]+@
    {
        :0  # extract stuff up to trailing @ into corrected MATCH 
        * MATCH ?? ^^[^(_at_)]+
        { LIST=$MATCH }

        :0w:
        | gzip -fc >>$LIST
    }

For just the @ verification, having two overlapping conditions is
probably simpler:

    :0w:
    * ^From[    ]+owner-[^(_at_)]+@
    * ^From[    ]+owner-\/[^(_at_)]+
    | gzip -fc >>$MATCH

Hope this helps,

/* era */

-- 
.obBotBait: It shouldn't even matter whether    <http://www.iki.fi/~era/>
I am a resident of the state of Washington. <http://members.xoom.com/procmail/>