I often struggle to extract information from crappy HTML. My default approach typically involves using grep or sed while issuing a steady stream of expletives. This library makes it easy to extricate useful content from a veritable soup of insidiously crafted HTML.