"... Regexps. John tried a beautifulSoup parser, but it turned out to be more trouble than it was worth ...'

Took me a bit of mucking around to get it working. My trick was just utilising the FOX HTML Validator to find the page structure, then CUT+PASTE the text into IDLE and call up BS in IDLE, then write the BS expression to oarse the string till I had the right data.

"... I don't mind sharing the code if PG doesn't mind the potential flood of screen scrapers. ..."

A better suggestion might be supply the raw data you collected as a service and let others do what they want with it. Most just want the data and are not particularly interested in the HOW. If they are interested in the HOW then it's better to point them to some scrapers to play with.