YouPorn Python API

I programmed one, who's interested?

Take a look at it yourself. It isn't much, just the basics you can do via the homepage. Probably not even enough to call it API or something similiar. Anyways, have fun with it:
#!/usr/bin/env python

import html5lib
import mechanize

URL = "http://youporn.com/"
ENTER_URL = "%s?user_choice=Enter" % URL
BROWSE_URL = "%sbrowse/%s?page=%s" % (URL, "%s", "%d")
TOP_RATED_URL = "%stop_rated/%s?page=%s" % (URL, "%s", "%d")
MOST_VIEWED_URL = "%smost_viewed/%s?page=%s" % (URL, "%s", "%d")
SEARCH_URL = "%s%s?query=%s&type=%s&page=%s" % (URL, "%s", "%s", "%s", "%d")

def _join_url(a, *p):
    path = a
    for b in p:
        if b.startswith('/'):
            path = b
        elif path == '' or path.endswith('/'):
            path +=  b
        else:
            path += '/' + b
    return path


class YouPorn(object):
    def __init__(self):
        self._browser = mechanize.Browser()
        self._browser.addheaders = []
        self._enter()

    def _enter(self):
        self._browser.open(ENTER_URL)

    @staticmethod
    def _filter_videos(soup):
        watch = lambda href: href and "/watch/" in href
        videos = []
        for a in soup.findAll("a", {"href":watch}):
            videos.append(_join_url(URL, a["href"]))
        return videos

    def get_newest_videos(self, page=1, sort_by="rating"):
        return self._filter_videos(html5lib.parse(self._browser.open(
            BROWSE_URL % (sort_by, page)), "beautifulsoup"))

    def get_top_rated(self, page=1, sort_by="week"):
        return self._filter_videos(html5lib.parse(self._browser.open(
            TOP_RATED_URL % (sort_by, page)), "beautifulsoup"))

    def get_most_viewed(self, page=1, sort_by="week"):
        return self._filter_videos(html5lib.parse(self._browser.open(
            MOST_VIEWED_URL % (sort_by, page)),"beautifulsoup"))

    def search(self, query, page=1, sort_by="relevance", type="straight"):
        return self._filter_videos(html5lib.parse(self._browser.open(
            SEARCH_URL % (sort_by, query, type, page)), "beautifulsoup"))

    def download_video(self, url):
        soup = html5lib.parse(self._browser.open(url), "beautifulsoup")
        download = lambda href: "/download/" in href
        download_url = soup.find("a", {"href":download})["href"]
        self._browser.retrieve(download_url,
            self._browser.geturl().split("/")[-2] + ".flv")


def main():
    youporn = YouPorn()
    for video in youporn.get_most_viewed(sort_by="all")[1:]:
        print "Downloading %s..." % video
        youporn.download_video(video)

if __name__ == "__main__":
    main()

If I were you, had too less to do and were bored, I would probably implement a Video class.