Michael Janssen wrote: > this is important but not enough. re.split(r'\b', 'word boundary') is > yet infunctional. I've looked through the sources to find out why. re.findall(r'\w+', ...) should do what is intended. Neil