Toen wij Dallman Ross kietelden, kwam er dit uit:
Ruud:
one sed-call can do it.
Yup, I saved it before when you posted it. Very nice! Unfortunately,
you did it too late in the day to spare me a couple of hours of
unplanned screwing around. :-) The exercise is still useful
if only as a tutorial (partly to myself).
One could also store the full headers in their original folded state
in a variable first, and work with that.
# is there a cheaper way to isolate the (non-unfolded) header-blob?
:0w
{ my_HEADERS = `sed '/^$/q'` }
NL = "
" SPC = " "
TAB = " "
b = "[$SPC$TAB]"
S = "[^$SPC$TAB]"
my_FIELD = "Received"
my_REGEX = "by [a-z]+check"
# Enjoy the way to get rid of the last line
:0
*$ my_HEADERS ?? ^^\/(.+$)+$\my_FIELD:.*(^$b.*)*$my_REGEX
*$ MATCH ?? ^^\/.*(^.*)*^$S.*
* MATCH ?? ^^\/(.+$)+.
* MATCH ?? ^^\/(.*$)+..+$
{ my_HEAD = "$MATCH" }
my_TAIL
:0
*$ my_HEADERS ??
^$\my_FIELD:.*(^$b.*)*.*$my_REGEX.*(^$b.*)*^\/$S.*(^.*)*^^
{ my_TAIL = "$MATCH" }
:0 h fw
| echo "$my_HEAD$my_TAIL$NL"
Nice going, Ruud.
You too. And sed-user Gudermez rules.
http://groups.yahoo.com/group/sed-users/messages/4696
(if you have a yahoogroups-login)
--
Grtz, Ruud
____________________________________________________________
procmail mailing list Procmail homepage: http://www.procmail.org/
procmail(_at_)lists(_dot_)RWTH-Aachen(_dot_)DE
http://MailMan.RWTH-Aachen.DE/mailman/listinfo/procmail