🚀 go-pugleaf

In article <6883@xenna.Xylogics.COM> loverso@Xylogics.COM (John Robert LoVerso) writes:
} In article <4177@ogccse.ogc.edu> schaefer@ogccse.UUCP (Barton E. Schaefer) writes:
} >     mush> pipe sed -n 1,/\^\$/p >> file
} ...
} > "pipe" does not write the ignored headers ("ignore" command or "show_hdrs"
} > variable).  If you want to save ALL the headers, you have to do a little
} > more work:
} >     mush> set print_cmd='sed -n 1,/\^\$/p >> file'
} >     mush> lpr
} 
} Not to be picky, but this sounds more like the "lpr" command is unneeded,
} and that "pipe" needs additional work.  I.e., perhaps there should be an
} option to pipe, "-a"(?), which says "even send all headers".  Then
} "lpr" just becomes an alias using "pipe -a", and the mechanism pipe out
} complete messages becomes much cleaner.

This has already been under discussion for some days, originally for
other reasons, but nobody has time to work it out at the moment.

I should mention that there is an error in the "sv_hdrs" cmd that I
included in the original article.  If given a list of several messages,
it will save the headers only of the first, because "lpr" starts only
one "print_cmd" process and sends all the messages to it separated by
form feeds.  (A difference from "pipe" that presently does make the
"lpr" command needed.)

To make sv_hdrs work correctly for a list of messages, substitute the
variable "pager" for "print_cmd" and substitute the command "print"
for "lpr".  You probably want to save the original value of $pager in
another variable and restore it afterwards.
-- 
Bart Schaefer           "And if you believe that, you'll believe anything."
                                                            -- DangerMouse
CSNET / Internet                schaefer@cse.ogc.edu
UUCP                            ...{sequent,tektronix,verdix}!ogccse!schaefer