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, Doctype
  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 class Video(dict):
  35     def __init__(self, *args, **kwargs):
  36         self.update(dict(*args, **kwargs))  # use the free update to set keys
  37
  38     def __setattr__(self, name, value):
  39         return self.__setitem__(name,value)
  40
  41     def __getattr__(self, name):
  42         return self.__getitem__(name)
  43
  44     def is_downloaded(self):
  45         raise("NotImplemented")
  46
  47 def scrape_player_page(video):
  48     """
  49     Try to scrape the site for video and download.
  50     """
  51     if not video['url'].startswith('http'):
  52         video['url'] = "http://www.svtplay.se" + video['url']
  53     soup = BeautifulSoup(requests.get(video['url']).text)
  54     video_player = soup.body('a',{'data-json-href':True})[0]
  55     if 'oppetarkiv.se' in video['url']:
  56         flashvars = requests.get("http://www.oppetarkiv.se/%s"%video_player.attrs['data-json-href']+"?output=json").json()
  57     else:
  58         if video_player.attrs['data-json-href'].startswith("/wd"):
  59             flashvars = requests.get("http://www.svt.se/%s"%video_player.attrs['data-json-href']).json()
  60         else:
  61             flashvars = requests.get("http://www.svtplay.se/%s"%video_player.attrs['data-json-href']+"?output=json").json()
  62     video['duration'] = video_player.attrs.get('data-length',0)
  63     if not video['title']:
  64         video['title'] = soup.find('meta',{'property':'og:title'}).attrs['content'].replace('|','_').replace('/','_')
  65     if not 'genre' in video:
  66         if soup.find(text='Kategori:'):
  67             video['genre'] = soup.find(text='Kategori:').parent.parent.a.text
  68         else:
  69             video['genre'] = 'Ingen Genre'
  70     if 'dynamicStreams' in flashvars:
  71         video['url'] = flashvars['dynamicStreams'][0].split('url:')[1].split('.mp4,')[0] +'.mp4'
  72         filename = video['title']+".mp4"
  73         print(Popen(["rtmpdump","-o"+filename,"-r", url], stdout=PIPE).communicate()[0])
  74     if 'pathflv' in flashvars:
  75         rtmp = flashvars['pathflv'][0]
  76         filename = video['title']+".flv"
  77         print(Popen(["mplayer","-dumpstream","-dumpfile",filename, rtmp], stdout=PIPE).communicate()[0])
  78     if 'video' in flashvars:
  79         for reference in flashvars['video']['videoReferences']:
  80             if 'm3u8' in reference['url']:
  81                 video['url']=reference['url']
  82                 video['filename'] = video['title']+'.ts'
  83                 if 'statistics' in flashvars:
  84                     video['category'] = flashvars['statistics']['category']
  85         download_from_playlist(video)
  86     if not 'url' in video:
  87         print("Could not find any streams")
  88         return False
  89     return video
  90
  91 def download_from_playlist(video):
  92     playlist = parse_playlist(requests.get(video['url']).text)
  93     if not playlist:
  94         return
  95     videourl = sorted(playlist, key=lambda k: int(k['BANDWIDTH']))[-1]['url']
  96     if not videourl.startswith('http'): #if relative path
  97         videourl = "{}/{}".format(os.path.dirname(video['url']), videourl)
  98     segments, metadata = parse_segment_playlist(videourl)
  99     if "EXT-X-KEY" in metadata:
 100         key = requests.get(metadata["EXT-X-KEY"]['URI'].strip('"')).text
 101         decrypt=True
 102     else:
 103         decrypt=False
 104     with open("%s"%video['filename'],"wb") as ofile:
 105         segment=0
 106         size = 0
 107         for url in segments:
 108             ufile = requests.get(url, stream=True).raw
 109             print("\r{0:.2f} MB".format(size/1024/1024))
 110             sys.stdout.flush()
 111             if decrypt:
 112                 iv=struct.pack("IIII",segment,0,0,0)
 113                 decryptor = AES.new(key, AES.MODE_CBC, iv)
 114             while(True):
 115                 buf = ufile.read(4096)
 116                 if not buf:
 117                     break
 118                 if decrypt:
 119                     buf = decryptor.decrypt(buf)
 120                 ofile.write(buf)
 121                 size += len(buf)
 122             segment += 1
 123
 124     if 'thumb-url' in video:
 125         video['thumb'] = requests.get(video['thumb-url'],stream=True).raw
 126
 127 def parse_playlist(playlist):
 128     if not playlist.startswith("#EXTM3U"):
 129         print(playlist)
 130         return False
 131     playlist = playlist.splitlines()
 132     while not 'EXT-X-STREAM-INF' in playlist[0]:
 133         playlist = playlist[1:]
 134     items=[]
 135     for (metadata_string,url) in zip(playlist[0::2], playlist[1::2]):
 136         md = Video()
 137         if not 'EXT-X-STREAM-INF' in metadata_string.split(':')[0]:
 138             continue
 139         for item in metadata_string.split(':')[1].split(','):
 140             if '=' in item:
 141                 md.update([item.split('='),])
 142         md['url']=url
 143         items.append(md)
 144     return items
 145
 146 def parse_segment_playlist(playlisturl):
 147     playlist = requests.get(playlisturl).text
 148     assert playlist.startswith("#EXTM3U")
 149     PATTERN = re.compile(r'''((?:[^,"']|"[^"]*"|'[^']*')+)''')
 150     segments = []
 151     next_is_url=False
 152     metadata = {}
 153     for row in playlist.splitlines():
 154         if next_is_url:
 155             if not row.startswith('http'): #if relative path
 156                 row = "{}/{}".format(os.path.dirname(playlisturl), row)
 157             segments.append(row)
 158             next_is_url=False
 159             continue
 160         if 'EXTINF' in row:
 161             next_is_url=True
 162         if "EXT-X-KEY" in row:
 163              row = row.split(':',1)[1] #skip first part
 164              parts = PATTERN.split(row)[1:-1] #do magic re split and keep quotes
 165              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
 166     return(segments, metadata)
 167
 168 def parse_videolist():
 169     page_num = 1
 170     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
 171     page_tot = int(soup.find('a',{'data-currentpage':True}).attrs['data-lastpage'])
 172     videos_per_page = 8
 173     video_num = 0
 174     while(page_num <= page_tot):
 175         base_url = "http://www.svtplay.se/ajax/videos?sida={}".format(page_num)
 176         soup = BeautifulSoup(requests.get(base_url).text)
 177         for article in soup.findAll('article'):
 178             meta = dict(article.attrs)
 179             video = Video()
 180             video['title'] = meta['data-title']
 181             video['description'] = meta['data-description']
 182             video['url'] = dict(article.find('a').attrs)['href']
 183             video['thumb-url'] = dict(article.find('img',{}).attrs)['src']
 184             video['num'] = video_num
 185             video['total'] = page_tot * videos_per_page
 186             video_num += 1
 187             yield video
 188         page_num += 1
 189
 190 def remux(video, xml=None):
 191     basename = video['filename'].split('.ts')[0]
 192     if 'genre' in video:
 193         if not os.path.exists(video['genre']):
 194             os.mkdir(video['genre'])
 195     video['path'] = os.path.join(video['genre'],basename+'.mkv')
 196     command = ["mkvmerge","-o",video['path'], '--title',video['title']]
 197
 198     if xml:
 199         with open(basename+'.xml','w') as f:
 200             f.write(xml)
 201             command.extend(['--global-tags',basename+'.xml'])
 202     if 'thumb' in video:
 203         with open('thumbnail.jpg','wb') as f: #FIXME use title instead for many downloaders
 204             f.write(video['thumb'].read())
 205             command.extend(['--attachment-description', "Thumbnail",
 206                  '--attachment-mime-type', 'image/jpeg',
 207                  '--attach-file', 'thumbnail.jpg'])
 208     command.append(video['filename'])
 209     print(Popen(command, stdout=PIPE).communicate()[0])
 210     for fname in (video['filename'], basename+'.xml','thumbnail.jpg'):
 211         try:
 212             os.unlink(fname)
 213         except:
 214             pass
 215
 216 def mkv_metadata(video):
 217     root = BeautifulSoup(features='xml')
 218     root.append(Doctype('Tags SYSTEM "matroskatags.dtd"'))
 219     tags = root.new_tag("Tags")
 220     tag = root.new_tag("Tag")
 221     tags.append(tag)
 222     root.append(tags)
 223     keep = ('title','description', 'url','genre')
 224     targets = root.new_tag("Targets")
 225     ttv = root.new_tag("TargetTypeValue")
 226     ttv.string = str(50)
 227     targets.append(ttv)
 228     tag.append(targets)
 229     for key in video:
 230         if not key in keep:
 231             continue
 232         simple = root.new_tag('Simple')
 233         name = root.new_tag('Name')
 234         name.string=key.upper()
 235         simple.append(name)
 236         sstring = root.new_tag('String')
 237         sstring.string=video[key]
 238         simple.append(sstring)
 239         tag.append(simple)
 240     return str(root)
 241
 242 if __name__ == "__main__":
 243     parser = argparse.ArgumentParser()
 244     group = parser.add_mutually_exclusive_group(required=True)
 245     group.add_argument("-r", "--rss", help="Download all files in rss")
 246     group.add_argument("-u", "--url", help="Download video in url")
 247     group.add_argument("-m", "--mirror", help="Mirror all files", action="store_true")
 248     parser.add_argument("-n", "--no_act", help="Just print what would be done, don't do any downloading.", action="store_true")
 249     parser.add_argument("--no_remux", help="Don't remux into mkv", action="store_true")
 250
 251     args = parser.parse_args()
 252     if args.rss:
 253         import feedparser
 254         d = feedparser.parse(args.rss)
 255         for e in d.entries:
 256             print(("Downloading: %s"%e.title))
 257             if args.no_act:
 258                 continue
 259             video = scrape_player_page({'title':e.title,'url':e.link})
 260             if args.no_remux:
 261                 continue
 262             self.remux(video)
 263         #print(e.description)
 264     if args.mirror:
 265         if not os.path.exists('.seen'):
 266             os.mkdir('.seen')
 267         for video in parse_videolist():
 268             video['title'] = video['title'].replace('/','_')
 269             print(video['title']+'.mkv')
 270             print("{} of {}".format(video['num'], video['total']))
 271
 272             if os.path.exists(os.path.join('.seen',video['title'])):
 273                 print("Skipping")
 274                 continue
 275             print("Downloading...")
 276             if args.no_act:
 277                 continue
 278             open(os.path.join('.seen',video['title']),'w').close() #touch
 279             video = scrape_player_page(video)
 280             if args.no_remux:
 281                 continue
 282             xml = mkv_metadata(video)
 283             remux(video, xml)
 284
 285     else:
 286         if not args.no_act:
 287             video = scrape_player_page({'url':args.url})
 288         if not args.no_remux:
 289             remux({'title':e.title})
 290         print(("Downloaded {}".format(args.url)))