Wikipedia Crawl or DB project

I need to gather information about all of the schools in the USA. I noticed wikipedia has a lot of the data we need. I know with wikipedia there are 2 approaches, 1 use their SQL releases to find and pull the data out or devise a way to crawl the site (scrapy pref if we crawl). here is an example of what i found schools by USA [login to view URL] discricts [login to view URL] then to a specific school [login to view URL] more school examples [login to view URL](South_Dakota) [login to view URL] I need the data you will mainly find in the box on the right, the most important would be the school website url, but ideally we can figure out how to store all of the data in there in some normalized name value pairs. I like the crawl idea but lets not be overly passive of the SQL dumps [login to view URL]:Database_download you dont need to scope the process of running the crawler, after you develop and test and confirm it works, i can load it up on a server and run it for the entire data set. if we use the SQL dumps then i would like the target db to be mongoDB and or some json structure per school and per district