Why use a regexp?

use strict;
my $str;

while (<DATA>) {
  chomp;
  my @cols = split(/;/);
  for (my $i=1;$i<=@cols;$i++) {
    $cols[$i-1]="<col$i>$cols[$i-1]</col$i>";
  }
  $str .= '<row>' . join('',@cols) . "</row>\n";
}
print $str;

exit;
__DATA__
a;b;c;d;e
f;g;h;i
j;k
l
m;n;o
[download]

prints:

<row><col1>a</col1><col2>b</col2><col3>c</col3><col4>d</col4><col5>e</
+col5></row>
<row><col1>f</col1><col2>g</col2><col3>h</col3><col4>i</col4></row>
<row><col1>j</col1><col2>k</col2></row>
<row><col1>l</col1></row>
<row><col1>m</col1><col2>n</col2><col3>o</col3></row>
[download]

UPDATE: Doh! of course gjb is right. Text::CSV is a great module. While my code is better than using regexps, Text::CSV is better than using my code.

-pete
"Worry is like a rocking chair. It gives you something to do, but it doesn't get you anywhere."