title: //article/header/h1 author: //article/header/section[@class='byline']/span[contains(@class, 'author')]/a date: //article/header/section[@class='byline']/span[@class='published']/span body: //article/section[@class='body'] convert_double_br_tags: yes # This is required, because Tidy chokes on the HTML5 tags... tidy: no test_url: http://www.version2.dk/artikel/17069-amerikansk-hit-investor-er-vild-med-danske-net-ivaerksaettere