Further to Sam's answer, you could also extract the text pieces from the page return soup.get_text(separator=',') Then find the headers to validate format and proceed to parse the coordinates.