Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!usc!elroy.jpl.nasa.gov!jpl-devvax!lwall
From: lwall@jpl-devvax.JPL.NASA.GOV (Larry Wall)
Newsgroups: comp.unix.wizards
Subject: Re: replacing nulls in a file with sed.
Message-ID: <8872@jpl-devvax.JPL.NASA.GOV>
Date: 25 Jul 90 20:58:23 GMT
References: <1990Jul24.224811.22143@midway.uchicago.edu>
Reply-To: lwall@jpl-devvax.JPL.NASA.GOV (Larry Wall)
Organization: Jet Propulsion Laboratory, Pasadena, CA
Lines: 29

In article <1990Jul24.224811.22143@midway.uchicago.edu> monty@delphi.uchicago.edu (Monty Mullig) writes:
: i have a file that comes with null characters randomly placed in it.
: i'd like to replace those nulls with spaces, but i can't seem to get
: sed to do this.  the command:
: 
: sed 's/\(^@\)/ /g' < infile > outfile
: 
: removes the nulls, but doesn't replace them with spaces.
: in emacs, the null characters show up as ^@ characters and
: global search and replace works (using the control-q to enter
: ^@ as a true control character, rather than a carat and an
: at sign).

Most Unix utilities are not null-proof, because C doesn't make it real
easy to be null-proof.  Chances are the null isn't even making it past your
shell!  Even tr can't handle \000, seemingly.  GNU emacs is a fortunate
exception--so's Perl:

	perl -pe 's/\0/ /g' <infile >outfile

or, a bit more efficiently

	perl -pe 'y/\0/ /' <infile >outfile

Note also that you can specify the null in human-readable form with Perl.
So you can sneak it right past your shell.

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