repopular_projects.py

#!/usr/bin/env python
import urllib, simplejson

def is_github_project(href):
    "Quick and dirty way to find if a link is a github project"
    if href.startswith("http://github.com") and len(href.split("/")) == 5:
        return True
    return False

def get_popular_projects():
    yql_endpoint = "http://query.yahooapis.com/v1/public/yql?q=use%20'http%3A%2F%2Fyqlblog.net%2Fsamples%2Fdata.html.cssselect.xml'%20as%20data.html.cssselect%3B%20select%20*%20from%20data.html.cssselect%20where%20url%3D%22repopular.com%22%20and%20css%3D%22div.pad%20a%22&format=json&diagnostics=true&callback="
    data = urllib.urlopen(yql_endpoint).read()
    repopular_json = simplejson.loads(data)
    links = repopular_json["query"]["results"]["results"]["a"]
    projects = set([link["href"] for link in links if "href" in link and is_github_project(link["href"])])
    return projects



if __name__ == "__main__":
    popular_projects = get_popular_projects()
    print popular_projects