Even in the third decade of the Web, scraping web sites remains a challenging task: Most scraping programs are still developed as ad-hoc solutions using a complex stack of languages and tools. Where comprehensive extraction solutions exist, they are expensive, heavyweight, and proprietary.