Also adding the pycurl scraper for Westminster, just in case it is useful to remind us how to do stuff later.