Instead of using the "\w" which is short hand for word characters, just insert your own charactar ranges you consider valid word characters.
\w is does not always include the same character sets across all the different implementations of regular expressions. Hope that helps... probably good to get a good primer article on regular expressions, they are powerful... yet until you understand all the rules and options you have they are the most confusing things. --Scott --- In flexcoders@yahoogroups.com, "jmfillman" <[EMAIL PROTECTED]> wrote: > > I'm using the following pattern to analyze simple text: /\b\w+/gi > > This seems to work well, except for hyphenated words, like "pre- > school", and contractions, like "isn't". > > Please help!! >
