C+ broadcasts to RSS

cpdl.py

import re
import xml.etree.ElementTree as ET
import urllib
import m3u8
import shutil
import os
import pytz
import json
from datetime import datetime
from feedgen.feed import FeedGenerator

RSS_FILENAME = "rss.xml"
ROOT = "https://82.237.11.61/lpj/"
MAX_ITEMS = 10

fg = FeedGenerator()
fg.title('Youtube Subscriptions')
fg.id(ROOT)
fg.language('en')
fg.updated(datetime.now(tz=pytz.utc))
fg.link({'href':ROOT + RSS_FILENAME, 'rel':'self'})
fg.load_extension('podcast')
fg.description("Youtube subscriptions")

downloaded = json.load(open("downloads.json"))
downloaded_cpy = [i for i in downloaded]
maintree = ET.parse(urllib.urlopen("http://service.canal-plus.com/video/rest/getMEAs/cplus/249"))

counter = 0
for i in maintree.findall('.//MEA/ID'):
	counter += 1

	tree = ET.parse(urllib.urlopen("http://service.canal-plus.com/video/rest/getVideos/cplus/" + i.text))
	playlist = m3u8.load(ET.tostring(tree.findall('.//VIDEOS/HLS')[0], method='text').decode("ascii"))
	name = tree.findall('.//TITRAGE/TITRE')[0].text + " - " + tree.findall('.//TITRAGE/SOUS_TITRE')[0].text

	fe = fg.add_entry()
	fe.id("http://service.canal-plus.com/video/rest/getVideos/cplus/" + i.text)
	fe.title(name)
	fe.description(tree.findall('.//INFOS/DESCRIPTION')[0].text)
	#fe.published(downloaded[download]["published"])
	fe.link({'href':"http://service.canal-plus.com/video/rest/getVideos/cplus/" + i.text, 'rel':'alternate', 'type':'text/html'})
	fe.author({'name':ET.tostring(tree.findall('.//INFOS/AUTEUR')[0], method='text').decode("utf-8")})
	fe.category({'term': ET.tostring(tree.findall('.//RUBRIQUAGE/UNIVERS')[0], method='text').decode("utf-8")})
	fe.podcast.itunes_image(ET.tostring(tree.findall('.//IMAGES/GRAND')[0], method='text').decode("utf-8"))

	if counter > MAX_ITEMS : break
	if i.text in downloaded_cpy :
		del downloaded_cpy[downloaded_cpy.index(i.text)]
		fe.enclosure(ROOT + i.text + ".mp4", type = "video/mp4", length = str(os.path.getsize(i.text + ".mp4")))
		continue
	else : downloaded.append(i.text)

	print ("Processing " + name)
	for chunk in playlist.playlists :
		maximum = chunk.uri

	filelist = m3u8.load(maximum)

	urllist = []
	for segment in filelist.segments:
		urllist.append(segment.uri)

	urllist.sort(key = lambda a : int(re.search(r"\/segment(\d+)_.+\.ts+", a).group(1)))


	for uri in urllist : 
		print("Downloading part "  + re.search(r"\/segment(\d+)_.+\.ts+", uri).group(1))
		dlobject = urllib.urlopen(uri)
		shutil.copyfileobj(dlobject, open(i.text + ".mp4", "ab"))

	fe.enclosure(ROOT + i.text + ".mp4", type = "video/mp4", length = str(os.path.getsize(i.text + ".mp4")))
	json.dump(downloaded, open("downloads.json", "w"))

for i in downloaded_cpy :
	del downloaded[downloaded.index(i)]
	os.remove(i + ".mp4")

json.dump(downloaded, open("downloads.json", "w"))

fg.rss_file(RSS_FILENAME)