svtplaydump.py

   1 #!/usr/bin/env python
   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 BeautifulSoup import BeautifulSoup
  26 from subprocess import *
  27 import re
  28 import json
  29 from Crypto.Cipher import AES
  30 import struct
  31 import argparse
  32 import feedparser
  33 try:
  34     import urlparse
  35 except ImportError:
  36     pass
  37 import urllib2
  38 try:
  39     import urllib2.urlparse as urlparse
  40 except ImportError:
  41     pass
  42 import sys, os
  43
  44 def scrape_player_page(url, title):
  45     """
  46     Try to scrape the site for video and download.
  47     """
  48     if not url.startswith('http'):
  49         url = "http://www.svtplay.se" + url
  50     video = {}
  51     page = urllib2.urlopen(url).read()
  52     soup = BeautifulSoup(page,convertEntities=BeautifulSoup.HTML_ENTITIES)
  53     video_player = soup.body('a',{'data-json-href':True})[0]
  54     if video_player.attrMap['data-json-href'].startswith("/wd"):
  55         flashvars = json.loads(urllib2.urlopen("http://www.svt.se/%s"%video_player.attrMap['data-json-href']).read())
  56     else:
  57         flashvars = json.loads(urllib2.urlopen("http://www.svtplay.se/%s"%video_player.attrMap['data-json-href']+"?output=json").read())
  58     video['duration'] = video_player.attrMap.get('data-length',0)
  59     video['title'] = title
  60     if not title:
  61         video['title'] = soup.find('meta',{'property':'og:title'}).attrMap['content'].replace('|','_').replace('/','_')
  62     if 'dynamicStreams' in flashvars:
  63         video['url'] = flashvars['dynamicStreams'][0].split('url:')[1].split('.mp4,')[0] +'.mp4'
  64         filename = video['title']+".mp4"
  65         print Popen(["rtmpdump",u"-o"+filename,"-r", url], stdout=PIPE).communicate()[0]
  66     if 'pathflv' in flashvars:
  67         rtmp = flashvars['pathflv'][0]
  68         filename = video['title']+".flv"
  69         print Popen(["mplayer","-dumpstream","-dumpfile",filename, rtmp], stdout=PIPE).communicate()[0]
  70     if 'video' in flashvars:
  71         for reference in flashvars['video']['videoReferences']:
  72             if reference['url'].endswith("m3u8"):
  73                 video['url']=reference['url']
  74                 video['filename'] = video['title']+'.ts'
  75                 if 'statistics' in flashvars:
  76                     video['category'] = flashvars['statistics']['category']
  77         download_from_playlist(video)
  78     else:
  79         print "Could not find any streams"
  80         return
  81     return video
  82
  83 def download_from_playlist(video):
  84     playlist = parse_playlist(urllib2.urlopen(video['url']).read())
  85     videourl = sorted(playlist, key=lambda k: int(k['BANDWIDTH']))[-1]['url']
  86     segments, metadata = parse_segment_playlist(urllib2.urlopen(videourl).read())
  87     if "EXT-X-KEY" in metadata:
  88         key = urllib2.urlopen(metadata["EXT-X-KEY"]['URI'].strip('"')).read()
  89         decrypt=True
  90     else:
  91         decrypt=False
  92     with open("%s"%video['filename'],"w") as ofile:
  93         segment=0
  94         size = 0
  95         for url in segments:
  96             ufile = urllib2.urlopen(url)
  97             print "\r{} MB".format(size/1024/1024),
  98             sys.stdout.flush()
  99             if decrypt:
 100                 iv=struct.pack("IIII",segment,0,0,0)
 101                 decryptor = AES.new(key, AES.MODE_CBC, iv)
 102             while(True):
 103                 buf = ufile.read(1024)
 104                 if buf:
 105                     if decrypt:
 106                         buf = decryptor.decrypt(buf)
 107                     ofile.write(buf)
 108                     size += len(buf)
 109                 else:
 110                     ufile.close()
 111                     break
 112             segment += 1
 113
 114 def parse_playlist(playlist):
 115     if not playlist.startswith("#EXTM3U"):
 116         print playlist
 117         return False
 118     playlist = playlist.splitlines()[1:]
 119     items=[]
 120     for (metadata_string,url) in zip(playlist[0::2], playlist[1::2]):
 121         md = dict()
 122         assert 'EXT-X-STREAM-INF' in metadata_string.split(':')[0]
 123         for item in metadata_string.split(':')[1].split(','):
 124             if '=' in item:
 125                 md.update([item.split('='),])
 126         md['url']=url
 127         items.append(md)
 128     return items
 129
 130 def parse_segment_playlist(playlist):
 131     assert playlist.startswith("#EXTM3U")
 132     PATTERN = re.compile(r'''((?:[^,"']|"[^"]*"|'[^']*')+)''')
 133     segments = []
 134     next_is_url=False
 135     metadata = {}
 136     for row in playlist.splitlines():
 137         if next_is_url:
 138             segments.append(row)
 139             next_is_url=False
 140             continue
 141         if 'EXTINF' in row:
 142             next_is_url=True
 143         if "EXT-X-KEY" in row:
 144              row = row.split(':',1)[1] #skip first part
 145              parts = PATTERN.split(row)[1:-1] #do magic re split and keep quotes
 146              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
 147     return(segments, metadata)
 148 def parse_videolist():
 149     page_num = 1
 150     page = urllib2.urlopen("http://www.svtplay.se/ajax/videospager").read() #this call does not work for getting the pages, we use it for the page totals only
 151     soup = BeautifulSoup(page,convertEntities=BeautifulSoup.HTML_ENTITIES)
 152     page_tot = int(soup.find('a',{'data-currentpage':True}).attrMap['data-lastpage'])
 153     videos_per_page = 8
 154     video_num = 0
 155     while(page_num <= page_tot):
 156         base_url = "http://www.svtplay.se/ajax/videos?sida={}".format(page_num)
 157         page = urllib2.urlopen(base_url).read()
 158         soup = BeautifulSoup(page,convertEntities=BeautifulSoup.HTML_ENTITIES)
 159         for article in soup.findAll('article'):
 160             meta = dict(article.attrs)
 161             video = {}
 162             video['title'] = meta['data-title']
 163             video['description'] = meta['data-description']
 164             video['url'] = dict(article.find('a').attrs)['href']
 165             video['thumb-url'] = dict(article.find('img',{}).attrs)['src']
 166             video['num'] = video_num
 167             video['total'] = page_tot * videos_per_page
 168             video_num += 1
 169             yield video
 170         page_num += 1
 171
 172
 173 if __name__ == "__main__":
 174     parser = argparse.ArgumentParser()
 175     group = parser.add_mutually_exclusive_group(required=True)
 176     group.add_argument("-r", "--rss", help="Download all files in rss")
 177     group.add_argument("-u", "--url", help="Download video in url")
 178     group.add_argument("-m", "--mirror", help="Mirror all files", action="store_true")
 179     parser.add_argument("-n", "--no_act", help="Just print what would be done, don't do any downloading.", action="store_true")
 180     args = parser.parse_args()
 181     if args.rss:
 182         d = feedparser.parse(args.rss)
 183         for e in d.entries:
 184             print("Downloading: %s"%e.title)
 185             if args.no_act:
 186                 continue
 187             filename = scrape_player_page(e.link, e.title)
 188             print Popen(["avconv","-i",filename,"-vcodec","copy","-acodec","copy", filename+'.mkv'], stdout=PIPE).communicate()[0]
 189         #print(e.description)
 190     if args.mirror:
 191         for video in parse_videolist():
 192             video['title'] = video['title'].replace('/','_')
 193             print video['title']+'.mkv',
 194             print u"{} of {}".format(video['num'], video['total'])
 195             if os.path.exists(video['title']+'.mkv'):
 196                 print "Skipping"
 197                 continue
 198             print("Downloading...")
 199             if args.no_act:
 200                 continue
 201             ret = scrape_player_page(video['url'], video['title'])
 202             print ret
 203             print Popen(["avconv","-i",video['title']+'.ts',"-vcodec","copy","-acodec","copy", video['title']+'.mkv'], stdout=PIPE).communicate()[0]
 204             try:
 205                 os.unlink(video['title']+'.ts')
 206             except:
 207                 import pdb;pdb.set_trace()
 208     else:
 209         if not args.no_act:
 210             video = scrape_player_page(args.url, None)
 211         print(u"Downloaded {}".format(args.url))