in reply to One line assigment statement with regex match

I tend to use something like this; 

  my $re = qr/(?:${\(
                 join "|", map { qr/\Q$_\E/ } @terms
              )})/;

  ...

  my ($match) = ($line =~ m/($re)/);

[download]

This is especially important if your @terms contain metacharacters such as parantheses, asterix, period, etc.

Also, it is usually faster because you're only doing one regular expression match per line, not many. 

$h=$ENV{HOME};my@q=split/\n\n/,`cat $h/.quotes`;$s="$h/."
."signature";$t=`cat $s`;print$t,"\n",$q[rand($#q)],"\n";

[download]

Replies are listed 'Best First'.
Re^2: One line assigment statement with regex match
by ketema (Scribe) on Jun 23, 2005 at 15:37 UTC
    I get search pattern not terminated when I try this: 
     my $re = qr/(?:${\(
                 join "|", map { qr/\Q$_\E/ } @terms
              )})/;

    [download]
[reply]
[d/l]

      Terribly sorry, I didn't test it.

      Yes, the double use of "/" to delimit regular expressions is tripping up the parser. Much better to use braces; 

      @terms=qw(foo bar b.*az);
my $re = qr{(?:${\(
                 join "|", map { qr{\Q$_\E} } @terms
              )})};
print "$re\n";

      [download]

      The above will print "(?-xism:(?:(?-xism:foo)|(?-xism:bar)|(?-xism:b\.\*az)))" 

      $h=$ENV{HOME};my@q=split/\n\n/,`cat $h/.quotes`;$s="$h/."
."signature";$t=`cat $s`;print$t,"\n",$q[rand($#q)],"\n";

      [download]
[reply]
[d/l]
[select]

In Section Seekers of Perl Wisdom