Here's an example that I use in something where I want to filter out lines of a file that contain words in an array.

sub filter_line {
  my ($i, $line);
  $line = shift;    
  foreach $i (@_) { 
    if ($line =~ /\b\Q$i\E\b/) { return 1 } 
  }
  return 0;
}
[download]

Hope this helps..
Rich

Addition: Read perlre and look at what constitutes a word boundry. In your example, it will match radio and not radiohead, but it will also match radio-head or radio.head, etc. But it will consider the underbar ( _ ) as part of the word.. so it won't match radio_head.

Use |s enclosed in parens to specify alternatives.

Try something like below:

#!/usr/bin/perl -w


my @sentence_list = (
            'radiohead',
            'turn off your radio',
            'jill has a radio that is black',
            'the stereo is broken',
            'radio for happiness',
           );

foreach $sentence (@sentence_list){
    if ($sentence =~ /(^|\W)radio(\W|$)/){
        print "Found\n";
    } else {
        print "Not found\n";
    }
}
[download]

This gives results:

Not found
Found
Found
Not found
Found
[download]

# Andy Lester  http://www.petdance.com  AIM:petdance
%_=split';','.; Perl ;@;st a;m;ker;p;not;o;hac;t;her;y;ju';
print map $_{$_}, split //,
'andy@petdance.com'

Try:

     m/\Wradio\W|^radio\W|\Wradio$/
[download]

I wanted to really say something like:

     m/[^\W]radio[\W$]/
[download]

Excpet that means something radically different than what I would be trying say which "Match anthing that either starts at the begining of a line or starts on a word boundery that has the text "radio" and ends on a word boundery or the end of a line." That is not what the thing above means though, so you have to settle for the first (unless someone else has a more simplified way of doing it).

Cheers...james