Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.1 6/24/83 (MC840302); site diku.UUCP
Path: utzoo!watmath!clyde!bonnie!akgua!whuxlm!harpo!decvax!genrad!panda!talcott!harvard!seismo!mcvax!diku!storm
From: storm@diku.UUCP (Kim Fabricius Storm)
Newsgroups: net.sources
Subject: Re: Re: shortshort way of reversing the lines in a file
Message-ID: <1168@diku.UUCP>
Date: Sun, 8-Sep-85 03:42:16 EDT
Article-I.D.: diku.1168
Posted: Sun Sep  8 03:42:16 1985
Date-Received: Tue, 10-Sep-85 03:03:40 EDT
References: <822@lll-crg.UUCP> <1124@mtgzz.UUCP>
Organization: DIKU, U of Copenhagen, DK
Lines: 17

In article <1124@mtgzz.UUCP> ecl@mtgzz.UUCP (e.c.leeper) writes:
>What's wrong with:
>	pr -n -t file | sort -nr | sed -e 's/[ 0-9]*	//'

Nothing except that for large files (some uucp logs are > Mbyte) it
this takes time.  I am very greatful for the FAST revfile program, which
I have been missing for a long time. 
  Revfile solves a REAL problem, while your (and others) shortways are
pretty, but also pretty useless.  I also prefer a VAX running UNIX to
a Turing machine even though the latter is all that is really needed
to do my calculations.  And C is better (or at least more structured)
than assembler, or what?
  I welcome any program, that solves my problems faster that the
standard tools.  My time is more valuable than your principles!

--
Kim F. Storm, Institute of Datalogy, U of Copenhagen, Denmark.