Path: utzoo!utgpu!watserv1!watmath!att!occrsh!uokmax!munnari.oz.au!samsung!uakari.primate.wisc.edu!sdd.hp.com!elroy.jpl.nasa.gov!jpl-devvax!lwall
From: lwall@jpl-devvax.JPL.NASA.GOV (Larry Wall)
Newsgroups: comp.unix.questions
Subject: Re: Selective translation
Message-ID: <8677@jpl-devvax.JPL.NASA.GOV>
Date: 11 Jul 90 01:00:46 GMT
References: <1553@dfsun1.electro.swri.edu>
Reply-To: lwall@jpl-devvax.JPL.NASA.GOV (Larry Wall)
Organization: Jet Propulsion Laboratory, Pasadena, CA
Lines: 51

In article <1553@dfsun1.electro.swri.edu> jackson@dfsun1.electro.swri.edu (Keith Jackson) writes:
: I was trying to filter a file by making the first word lowercase and
: leaving the rest as is.  My solution:
: 
:     % awk -f filt1.awk foo | tr A-Z a-z > stage1.b
:     % pr -m -t -s -l1 stage1.a stage1.b > final
: 
: where filt1.awk contains:
: {
:     print $1;
:     for (i = 2; i < NF; i++)
: 	printf("%s ", $i) >> "stage1.a";
:     if (NF > 1)
: 	print $NF >> "stage1.a";
: }
: 
: One flaw to this solution is that pr(1) ends up adding an extra
: line with the separation character (tab) on it.  The question is, how
: does one do this more easily?

Well, there's

  perl -pe 's/\S+/($tmp = $&) =~ y:A-Z:a-z:, $tmp/e' foo >final

or

  perl -pe '/\S+/ && substr($_,length($`),length($&)) =~ y/A-Z/a-z/' foo >final

or

  perl -pe '($x)=/(\S+)/ && $x =~ y/A-Z/a-z/ && s//$x/' foo >final

or

  perl -ne '@x = split(/(\S+)/); $x[1] =~ y/A-Z/a-z/; print @x' foo >final

or even

  sed -e 'h' \
	-e 's/^\([ ]*[^ ][^ ]*\).*/\1/' \
	-e 'y/ABCDEFGHIJKLMNOPQRSTUVWXYZ/abcdefghijklmnopqrstuvwxyz/' \
	-e 'G' \
	-e 's/\n[ ]*[^ ][^ ]*//'  foo >final

    where [ ] and [^ ] can be expanded to have a tab as well.

Take your pick.  All of these will preserve the whitespace around the
first word on each line, which you can't do very easily with awk.

Larry Wall
lwall@jpl-devvax.jpl.nasa.gov