On Mon, 10 Jun 1996 08:11:56 -0700, "Alan Stebbens
<stebbens(_at_)anywhere(_dot_)engr(_dot_)sgi(_dot_)com>"
<stebbens(_at_)anywhere(_dot_)engr(_dot_)sgi(_dot_)com> said:
# This recipe filters FROM through the badaddrs file, if egrep succeeds,
# then we don't want this mail, and we'll bounce it
:0E
* ? echo "$FROM" | egrep -f $BADADDRS
{ BOUNCE=1 }
I usually get the error "egrep: regular expression too long" when I
run the above command. This is on SunOS 4.1. Here's how to reproduce
it:
--- file junkfile.data: ---
u1(_at_)alpha(_dot_)beta(_dot_)gamma
u2(_at_)delta(_dot_)epsilon
u3(_at_)foo(_dot_)bar(_dot_)baz
u4(_at_)plugh(_dot_)plover
u5(_at_)xyzzy(_dot_)com
u6(_at_)frobozz(_dot_)magic-wells(_dot_)com
mrose.stsci.edu
--- end file ---
The command is:
echo "mrose(_at_)stsci(_dot_)edu" |egrep -f junkfile.data
If I change the last line of the file from "mrose.stsci.edu" to
"mrose(_at_)stsci(_dot_)edu", then the command works correctly. That's not a
solution, just a boundary.
Does anyone have any suggestions how to make this work? It would be
very convenient to be able to put a bunch of regexp's in a file and
match for junk mail (and other things) against each line in the file.
Thanks,
Mike Rose