svtplaydump.py

   1 #!/usr/bin/env python3
   2 # -*- coding: utf-8 -*-
   3 #
   4 #   (C) Copyright 2010 Mikael Frykholm <mikael@frykholm.com>
   5 #
   6 #   This program is free software: you can redistribute it and/or modify
   7 #   it under the terms of the GNU General Public License as published by
   8 #   the Free Software Foundation, either version 3 of the License, or
   9 #   (at your option) any later version.
  10 #
  11 #   This program is distributed in the hope that it will be useful,
  12 #   but WITHOUT ANY WARRANTY; without even the implied warranty of
  13 #   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  14 #   GNU General Public License for more details.
  15 #
  16 #   You should have received a copy of the GNU General Public License
  17 #   along with this program.  If not, see <http://www.gnu.org/licenses/>
  18 #
  19 # Changelog:
  20 # 0.4 added mirror mode.
  21 # 0.3 added apple streaming playlist parsing and decryption
  22 # 0.2 added python 2.4 urlparse compatibility
  23 # 0.1 initial release
  24
  25 from bs4 import BeautifulSoup
  26 from subprocess import *
  27 import re
  28 from Crypto.Cipher import AES
  29 import struct
  30 import argparse
  31 import requests
  32 import sys, os
  33
  34 def scrape_player_page(url, title):
  35     """
  36     Try to scrape the site for video and download.
  37     """
  38     if not url.startswith('http'):
  39         url = "http://www.svtplay.se" + url
  40     video = {}
  41     soup = BeautifulSoup(requests.get(url).text)
  42     video_player = soup.body('a',{'data-json-href':True})[0]
  43     if 'oppetarkiv.se' in url:
  44         flashvars = requests.get("http://www.oppetarkiv.se/%s"%video_player.attrs['data-json-href']+"?output=json").json()
  45     else:
  46         if video_player.attrs['data-json-href'].startswith("/wd"):
  47             flashvars = requests.get("http://www.svt.se/%s"%video_player.attrs['data-json-href']).json()
  48         else:
  49             flashvars = requests.get("http://www.svtplay.se/%s"%video_player.attrs['data-json-href']+"?output=json").json()
  50     video['duration'] = video_player.attrs.get('data-length',0)
  51     video['title'] = title
  52     if not title:
  53         video['title'] = soup.find('meta',{'property':'og:title'}).attrs['content'].replace('|','_').replace('/','_')
  54     if 'dynamicStreams' in flashvars:
  55         video['url'] = flashvars['dynamicStreams'][0].split('url:')[1].split('.mp4,')[0] +'.mp4'
  56         filename = video['title']+".mp4"
  57         print(Popen(["rtmpdump","-o"+filename,"-r", url], stdout=PIPE).communicate()[0])
  58     if 'pathflv' in flashvars:
  59         rtmp = flashvars['pathflv'][0]
  60         filename = video['title']+".flv"
  61         print(Popen(["mplayer","-dumpstream","-dumpfile",filename, rtmp], stdout=PIPE).communicate()[0])
  62     if 'video' in flashvars:
  63         for reference in flashvars['video']['videoReferences']:
  64             if reference['url'].endswith("m3u8"):
  65                 video['url']=reference['url']
  66                 video['filename'] = video['title']+'.ts'
  67                 if 'statistics' in flashvars:
  68                     video['category'] = flashvars['statistics']['category']
  69         download_from_playlist(video)
  70     else:
  71         print("Could not find any streams")
  72         return
  73     return video
  74
  75 def download_from_playlist(video):
  76     playlist = parse_playlist(requests.get(video['url']).text)
  77     if not playlist:
  78         return
  79     videourl = sorted(playlist, key=lambda k: int(k['BANDWIDTH']))[-1]['url']
  80     segments, metadata = parse_segment_playlist(requests.get(videourl).text)
  81     if "EXT-X-KEY" in metadata:
  82         key = requests.get(metadata["EXT-X-KEY"]['URI'].strip('"')).text
  83         decrypt=True
  84     else:
  85         decrypt=False
  86     with open("%s"%video['filename'],"wb") as ofile:
  87         segment=0
  88         size = 0
  89         for url in segments:
  90             ufile = requests.get(url, stream=True).raw
  91             print("\r{} MB".format(size/1024/1024))
  92             sys.stdout.flush()
  93             if decrypt:
  94                 iv=struct.pack("IIII",segment,0,0,0)
  95                 decryptor = AES.new(key, AES.MODE_CBC, iv)
  96             while(True):
  97                 buf = ufile.read(4096)
  98                 if not buf:
  99                     break
 100                 if decrypt:
 101                     buf = decryptor.decrypt(buf)
 102                 ofile.write(buf)
 103                 size += len(buf)
 104             segment += 1
 105
 106 def parse_playlist(playlist):
 107     if not playlist.startswith("#EXTM3U"):
 108         print(playlist)
 109         return False
 110     playlist = playlist.splitlines()[1:]
 111     items=[]
 112     for (metadata_string,url) in zip(playlist[0::2], playlist[1::2]):
 113         md = dict()
 114         assert 'EXT-X-STREAM-INF' in metadata_string.split(':')[0]
 115         for item in metadata_string.split(':')[1].split(','):
 116             if '=' in item:
 117                 md.update([item.split('='),])
 118         md['url']=url
 119         items.append(md)
 120     return items
 121
 122 def parse_segment_playlist(playlist):
 123     assert playlist.startswith("#EXTM3U")
 124     PATTERN = re.compile(r'''((?:[^,"']|"[^"]*"|'[^']*')+)''')
 125     segments = []
 126     next_is_url=False
 127     metadata = {}
 128     for row in playlist.splitlines():
 129         if next_is_url:
 130             segments.append(row)
 131             next_is_url=False
 132             continue
 133         if 'EXTINF' in row:
 134             next_is_url=True
 135         if "EXT-X-KEY" in row:
 136              row = row.split(':',1)[1] #skip first part
 137              parts = PATTERN.split(row)[1:-1] #do magic re split and keep quotes
 138              metadata["EXT-X-KEY"] = dict([part.split('=',1) for part in parts if '=' in part]) #throw away the commas and make dict of the pairs
 139     return(segments, metadata)
 140
 141 def parse_videolist():
 142     page_num = 1
 143     soup = BeautifulSoup(requests.get("http://www.svtplay.se/ajax/videospager").text)#this call does not work for getting the pages, we use it for the page totals only
 144     page_tot = int(soup.find('a',{'data-currentpage':True}).attrs['data-lastpage'])
 145     videos_per_page = 8
 146     video_num = 0
 147     while(page_num <= page_tot):
 148         base_url = "http://www.svtplay.se/ajax/videos?sida={}".format(page_num)
 149         soup = BeautifulSoup(requests.get(base_url).text)
 150         for article in soup.findAll('article'):
 151             meta = dict(article.attrs)
 152             video = {}
 153             video['title'] = meta['data-title']
 154             video['description'] = meta['data-description']
 155             video['url'] = dict(article.find('a').attrs)['href']
 156             video['thumb-url'] = dict(article.find('img',{}).attrs)['src']
 157             video['num'] = video_num
 158             video['total'] = page_tot * videos_per_page
 159             video_num += 1
 160             yield video
 161         page_num += 1
 162
 163
 164 if __name__ == "__main__":
 165     parser = argparse.ArgumentParser()
 166     group = parser.add_mutually_exclusive_group(required=True)
 167     group.add_argument("-r", "--rss", help="Download all files in rss")
 168     group.add_argument("-u", "--url", help="Download video in url")
 169     group.add_argument("-m", "--mirror", help="Mirror all files", action="store_true")
 170     parser.add_argument("-n", "--no_act", help="Just print what would be done, don't do any downloading.", action="store_true")
 171     args = parser.parse_args()
 172     if args.rss:
 173         import feedparser
 174         d = feedparser.parse(args.rss)
 175         for e in d.entries:
 176             print(("Downloading: %s"%e.title))
 177             if args.no_act:
 178                 continue
 179             filename = scrape_player_page(e.link, e.title)
 180             print(Popen(["avconv","-i",filename,"-vcodec","copy","-acodec","copy", filename+'.mkv'], stdout=PIPE).communicate()[0])
 181         #print(e.description)
 182     if args.mirror:
 183         for video in parse_videolist():
 184             video['title'] = video['title'].replace('/','_')
 185             print(video['title']+'.mkv')
 186             print("{} of {}".format(video['num'], video['total']))
 187             if os.path.exists(video['title']+'.mkv'):
 188                 print("Skipping")
 189                 continue
 190             print("Downloading...")
 191             if args.no_act:
 192                 continue
 193             ret = scrape_player_page(video['url'], video['title'])
 194             print(ret)
 195             print(Popen(["avconv","-i",video['title']+'.ts',"-vcodec","copy","-acodec","copy", video['title']+'.mkv'], stdout=PIPE).communicate()[0])
 196             try:
 197                 os.unlink(video['title']+'.ts')
 198             except:
 199                 import pdb;pdb.set_trace()
 200     else:
 201         if not args.no_act:
 202             video = scrape_player_page(args.url, None)
 203         print(("Downloaded {}".format(args.url)))