in reply to Extract text from HTML

For this sort of thing I usually recommend HTML::TokeParser or Ovid's slightly more intuitive HTML::TokeParser::Simple.

-- vek --