* lothar wrote: As already said by Georg, regexes are the wrong tool for such tasks, but anyway...
> give an re to find every innermost "table" element: <table(?:\s[^>]*)?>[^<]*(?:<(?!/table>|table(?:\s[^>]*)?>)[^<]*)*</table> > give an re to find every "pre" element directly followed by an "a" > element: <pre(?:\s[^>]*)?>[^<]*(?:<(?!/pre>|pre(?:\s[^>]*)?>)[^<]*)*</pre>(?=<a[\s>]) The are written more common than needed for your samples. Depending on the data to be expected, they can be written a bit shorter, but this is left as an exercise for the reader. HTH, nd -- http://mail.python.org/mailman/listinfo/python-list
