Get the video url

get_videos.py

# -*- coding = utf-8 -*-

import urllib, urllib2, sys
from BeautifulSoup import BeautifulSoup # for html

def fetch_url(link):
	''' '''
	f = urllib.urlopen(link)
	contents = f.read()	
	f.close()
	
	'''
	f.open(filename, 'w')
	f.write(contents)
	f.close()
	'''
	
	return BeautifulSoup(''.join(contents))
	
def listLinks(soup):
	''' '''
	links = []
	for link in soup.findAll('a', rel = 'lecture-link'):
		links.append(link.get('href'))	
	return links
	
def findMP4link(link):
	''' '''	
	opener = urllib2.build_opener()
	f = opener.open(link)
	contents = f.read()
	f.close()
	opener.close()
		
	soup = BeautifulSoup(''.join(contents))
	links = soup.findAll('source', type = 'video/mp4')
	assert len(links) == 1
	file = links[0]	
	return file.get('src')
	
	'''
	files = []
	for file in soup.findAll('source', type = 'video/mp4'):
		files.append(file.get('src'))
	
	return files
	'''
	
if __name__ == "__main__":
	''' '''
	# for gpm
	# link = "https://class.coursera.org/pgm/lecture/preview"
	# soup = BeautifulSoup(''.join(contents))
	
	# for ml
	# link = "https://class.coursera.org/ml/lecture/preview"
	
	# for nlp
	link = "https://class.coursera.org/nlp/lecture/preview"
	
	soup = fetch_url(link)
	# f = open('contents.html', 'r')
	# contents = f.read()
	# f.close()
		
	# print soup.prettify()	
	links = listLinks(soup)
	# print links	
	f = open('nlp', 'w')
	for link in links:		
		mp4_link = findMP4link(link)		
		f.write(mp4_link)
		f.write("\n")
	f.close()