PHP || XML || Convert HTML Docs to CSV format?

Hey all. Looking for a way to do the following: I'm working with a group of high school athletic departments who want to post their team rosters on our website. The problem is coaches don't have time to fill out our online roster submission form. The rosters do exist, however, in varying formats (HTML, Word, PDF, etc.) on each school's website -- sounds like a job for XML, no? Is there a way to transform exisiting team roster data on each school's website into a CSV file, or other db friendly format. The goal is to have a set of XML docs that will be able to transform the data set for each school and import it into our rosters db. Any ideas, leads, suggestions? Thanks in advance, --Noah