"Simple" but non-portable == useless.

[Chip Salzenberg]

According to tchrist at convex.COM (Tom Christiansen):
>I think he was talking about my script, not Karl's.

He certainly was; sorry for the misunderstanding.

>Chip's \0 stuff is sure the right way to go if you're that concerned.

After I posted my script, I realized that _trailing_ \0 is easier to
deal with, since the simplest output format is "s/\0/\n/".  But I went
back to the leading \0 when I changed my output format.

>I'm not sure how easy it is to discern files with \n's in them from
>separate entries in Chip's output.

A directory that ends in newline could go undetected.  Obviously I
could octalize all the names before printing them, but that takes the
output loop from grep() to for(), upping the line count by three. :-)

>Here are 7 lines (although I only see 5 lines of code), and it does run
>pretty fast.

But you cheated by omitting closedir(DIR).

>Notice I don't count files linked together.

Ah, that feature is useful.  But your implementation skips the useful
heuristic of checking st_nlinks.

Here's my latest version, still only ten lines long (with closedir()),
and with a better output format ("command dir1 dir2"):

===============================================================================
for $dir (split(/:/, $ENV{'PATH'})) {
	opendir(DIR, length($dir) ? $dir : ".") || next;
	while (defined($file = readdir(DIR))) {
		$_ = "$dir/$file";
		$D{$file} .= "\0$dir" unless m#/\.\.?$# || !-f $_ || !-x _
		    || (stat(_))[3] > 1 && $F{$file,(stat(_))[0,1]}++;
	}
	closedir(DIR);
}
grep(((($x = $D{$_}) =~ s/\0/ /g) > 1 && print "$_\t$x\n"), sort keys(%D));
===============================================================================

Try getting that output format from a shell script...  :-)
-- 
Chip Salzenberg at Teltronics/TCT     <chip at tct.uucp>, <uunet!pdn!tct!chip>
 "I want to mention that my opinions whether real or not are MY opinions."
             -- the inevitable William "Billy" Steinmetz