#!/usr/bin/python
import urllib
import re

course = "nlp"
maindoc = urllib.urlopen("https://class.coursera.org/%s/lecture/preview" % course).read()

files = []
for line in maindoc.split("<li"):
    pattern = r'a href="(https://class\.coursera\.org/%s/lecture/preview_view\?lecture_id=\d+)[^>]+>([^<]+)<' % course
    m = re.search(pattern, line)
    if m:
        (url, title) =  m.group(1), m.group(2).strip()
        files.append((url, title))

cnt = 1
for url, title in files:
    doc = urllib.urlopen(url).read()
    pattern = r'<source type="video/mp4" src="([^\"]+)"'
    m = re.search(pattern, doc)
    print "downloading %.2d - %s.mp4" % (cnt, re.sub("/","-",title))
    urllib.urlretrieve (m.group(1), "%.2d - %s.mp4" % (cnt, re.sub("/","-",title)))
    cnt += 1