Path: utzoo!mnetor!tmsoft!torsqnt!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.lang.perl
Subject: Re: Perl bug? - tr///c
Message-ID: <11597@jpl-devvax.JPL.NASA.GOV>
Date: 26 Feb 91 20:03:24 GMT
References: <EVAN.91Feb26154539@murgon.cs.mu.OZ.AU>
Reply-To: lwall@jpl-devvax.JPL.NASA.GOV (Larry Wall)
Distribution: comp
Organization: Jet Propulsion Laboratory, Pasadena, CA
Lines: 24

In article <EVAN.91Feb26154539@murgon.cs.mu.OZ.AU> evan@mullauna.cs.mu.OZ.AU writes:
: The following code:
: 	$foo = '*foo';
: 	$foo =~ tr/a-zA-Z//c;
: 	print "$foo\n";
: does not produce the output (I) expected.  Changing the * to (say) .
: produces a different (but similar) result.

You need to tell it what you want to translate non-alphabetic characters to.
In the absence of a RHS, it'll simply reuse the LHS (unless you supply a d,
in which case it will delete the specified characters).  So you've
essentially said

	$foo =~ tr/a-zA-Z/a-zA-Z/c;

Now, that said, I should say that there's a bug up through 4.0 beta that
incorrectly replicates the last character of the RHS when complementing,
which is why

	$foo =~ tr/a-zA-Z/_/c;

doesn't work right.  It's fixed in real 4.0 though.

Larry