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 import socket
  34
  35 class Video(dict):
  36     def __init__(self, *args, **kwargs):
  37         self.update(dict(*args, **kwargs))  # use the free update to set keys
  38
  39     def __setattr__(self, name, value):
  40         return self.__setitem__(name,value)
  41
  42     def __getattr__(self, name):
  43         return self.__getitem__(name)
  44
  45     def is_downloaded(self):
  46         raise("NotImplemented")
  47
  48 def scrape_player_page(video):
  49     """
  50     Try to scrape the site for video and download.
  51     """
  52     if not video['url'].startswith('http'):
  53         video['url'] = "http://www.svtplay.se" + video['url']
  54     soup = BeautifulSoup(requests.get(video['url']).text)
  55     video_player = soup.body('a',{'data-json-href':True})[0]
  56     if 'oppetarkiv.se' in video['url']:
  57         flashvars = requests.get("http://www.oppetarkiv.se/%s"%video_player.attrs['data-json-href']+"?output=json").json()
  58     else:
  59         if video_player.attrs['data-json-href'].startswith("/wd"):
  60             flashvars = requests.get("http://www.svt.se/%s"%video_player.attrs['data-json-href']).json()
  61         else:
  62             flashvars = requests.get("http://www.svtplay.se/%s"%video_player.attrs['data-json-href']+"?output=json").json()
  63     video['duration'] = video_player.attrs.get('data-length',0)
  64     if not video['title']:
  65         video['title'] = soup.find('meta',{'property':'og:title'}).attrs['content'].replace('|','_').replace('/','_')
  66     if not 'genre' in video:
  67         if soup.find(text='Kategori:'):
  68             video['genre'] = soup.find(text='Kategori:').parent.parent.a.text
  69         else:
  70             video['genre'] = 'Ingen Genre'
  71     if 'dynamicStreams' in flashvars:
  72         video['url'] = flashvars['dynamicStreams'][0].split('url:')[1].split('.mp4,')[0] +'.mp4'
  73         filename = video['title']+".mp4"
  74         print(Popen(["rtmpdump","-o"+filename,"-r", url], stdout=PIPE).communicate()[0])
  75     if 'pathflv' in flashvars:
  76         rtmp = flashvars['pathflv'][0]
  77         filename = video['title']+".flv"
  78         print(Popen(["mplayer","-dumpstream","-dumpfile",filename, rtmp], stdout=PIPE).communicate()[0])
  79     if 'video' in flashvars:
  80         for reference in flashvars['video']['videoReferences']:
  81             if 'm3u8' in reference['url']:
  82                 video['url']=reference['url']
  83                 video['filename'] = video['title']+'.ts'
  84                 if 'statistics' in flashvars:
  85                     video['category'] = flashvars['statistics']['category']
  86         download_from_playlist(video)
  87     if not 'url' in video:
  88         print("Could not find any streams")
  89         return False
  90     return video
  91
  92 def download_from_playlist(video):
  93     playlist = parse_playlist(requests.get(video['url']).text)
  94     if not playlist:
  95         return
  96     videourl = sorted(playlist, key=lambda k: int(k['BANDWIDTH']))[-1]['url']
  97     if not videourl.startswith('http'): #if relative path
  98         videourl = "{}/{}".format(os.path.dirname(video['url']), videourl)
  99     segments, metadata = parse_segment_playlist(videourl)
 100     if "EXT-X-KEY" in metadata:
 101         key = requests.get(metadata["EXT-X-KEY"]['URI'].strip('"')).text
 102         decrypt=True
 103     else:
 104         decrypt=False
 105     with open("%s"%video['filename'],"wb") as ofile:
 106         segment=0
 107         size = 0
 108         for url in segments:
 109             ufile = requests.get(url, stream=True).raw
 110             print("\r{0:.2f} MB".format(size/1024/1024),end="")
 111             sys.stdout.flush()
 112             if decrypt:
 113                 iv=struct.pack("IIII",segment,0,0,0)
 114                 decryptor = AES.new(key, AES.MODE_CBC, iv)
 115             while(True):
 116                 try:
 117                     buf = ufile.read(4096)
 118                 except socket.error as e:
 119                     print("Error reading, skipping file")
 120                     print(e)
 121                     return
 122                 if not buf:
 123                     break
 124                 if decrypt:
 125                     buf = decryptor.decrypt(buf)
 126                 ofile.write(buf)
 127                 size += len(buf)
 128             segment += 1
 129
 130     if 'thumb-url' in video:
 131         video['thumb'] = requests.get(video['thumb-url'],stream=True).raw
 132
 133 def parse_playlist(playlist):
 134     if not playlist.startswith("#EXTM3U"):
 135         print(playlist)
 136         return False
 137     playlist = playlist.splitlines()
 138     while not 'EXT-X-STREAM-INF' in playlist[0]:
 139         playlist = playlist[1:]
 140     items=[]
 141     for (metadata_string,url) in zip(playlist[0::2], playlist[1::2]):
 142         md = Video()
 143         if not 'EXT-X-STREAM-INF' in metadata_string.split(':')[0]:
 144             continue
 145         for item in metadata_string.split(':')[1].split(','):
 146             if '=' in item:
 147                 md.update([item.split('='),])
 148         md['url']=url
 149         items.append(md)
 150     return items
 151
 152 def parse_segment_playlist(playlisturl):
 153     playlist = requests.get(playlisturl).text
 154     assert playlist.startswith("#EXTM3U")
 155     PATTERN = re.compile(r'''((?:[^,"']|"[^"]*"|'[^']*')+)''')
 156     segments = []
 157     next_is_url=False
 158     metadata = {}
 159     for row in playlist.splitlines():
 160         if next_is_url:
 161             if not row.startswith('http'): #if relative path
 162                 row = "{}/{}".format(os.path.dirname(playlisturl), row)
 163             segments.append(row)
 164             next_is_url=False
 165             continue
 166         if 'EXTINF' in row:
 167             next_is_url=True
 168         if "EXT-X-KEY" in row:
 169              row = row.split(':',1)[1] #skip first part
 170              parts = PATTERN.split(row)[1:-1] #do magic re split and keep quotes
 171              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
 172     return(segments, metadata)
 173
 174 def parse_videolist():
 175     page_num = 1
 176     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
 177     page_tot = int(soup.find('a',{'data-currentpage':True}).attrs['data-lastpage'])
 178     videos_per_page = 8
 179     video_num = 0
 180     while(page_num <= page_tot):
 181         base_url = "http://www.svtplay.se/ajax/videos?sida={}".format(page_num)
 182         soup = BeautifulSoup(requests.get(base_url).text)
 183         for article in soup.findAll('article'):
 184             meta = dict(article.attrs)
 185             video = Video()
 186             video['title'] = meta['data-title']
 187             video['description'] = meta['data-description']
 188             video['url'] = dict(article.find('a').attrs)['href']
 189             video['thumb-url'] = dict(article.find('img',{}).attrs)['src']
 190             video['num'] = video_num
 191             video['total'] = page_tot * videos_per_page
 192             video_num += 1
 193             yield video
 194         page_num += 1
 195
 196 def remux(video, xml=None):
 197     basename = video['filename'].split('.ts')[0]
 198     if 'genre' in video:
 199         if not os.path.exists(video['genre']):
 200             os.mkdir(video['genre'])
 201     video['path'] = os.path.join(video['genre'],basename+'.mkv')
 202     command = ["mkvmerge","-o",video['path'], '--title',video['title']]
 203
 204     if xml:
 205         with open(basename+'.xml','w') as f:
 206             f.write(xml)
 207             command.extend(['--global-tags',basename+'.xml'])
 208     if 'thumb' in video:
 209         with open('thumbnail.jpg','wb') as f: #FIXME use title instead for many downloaders
 210             f.write(video['thumb'].read())
 211             command.extend(['--attachment-description', "Thumbnail",
 212                  '--attachment-mime-type', 'image/jpeg',
 213                  '--attach-file', 'thumbnail.jpg'])
 214     command.append(video['filename'])
 215     print(Popen(command, stdout=PIPE).communicate()[0])
 216     for fname in (video['filename'], basename+'.xml','thumbnail.jpg'):
 217         try:
 218             os.unlink(fname)
 219         except:
 220             pass
 221
 222 def mkv_metadata(video):
 223     root = BeautifulSoup(features='xml')
 224     root.append(Doctype('Tags SYSTEM "matroskatags.dtd"'))
 225     tags = root.new_tag("Tags")
 226     tag = root.new_tag("Tag")
 227     tags.append(tag)
 228     root.append(tags)
 229     keep = ('title','description', 'url','genre')
 230     targets = root.new_tag("Targets")
 231     ttv = root.new_tag("TargetTypeValue")
 232     ttv.string = str(50)
 233     targets.append(ttv)
 234     tag.append(targets)
 235     for key in video:
 236         if not key in keep:
 237             continue
 238         simple = root.new_tag('Simple')
 239         name = root.new_tag('Name')
 240         name.string=key.upper()
 241         simple.append(name)
 242         sstring = root.new_tag('String')
 243         sstring.string=video[key]
 244         simple.append(sstring)
 245         tag.append(simple)
 246     return str(root)
 247
 248 if __name__ == "__main__":
 249     parser = argparse.ArgumentParser()
 250     group = parser.add_mutually_exclusive_group(required=True)
 251     group.add_argument("-r", "--rss", help="Download all files in rss")
 252     group.add_argument("-u", "--url", help="Download video in url")
 253     group.add_argument("-m", "--mirror", help="Mirror all files", action="store_true")
 254     parser.add_argument("-n", "--no_act", help="Just print what would be done, don't do any downloading.", action="store_true")
 255     parser.add_argument("--no_remux", help="Don't remux into mkv", action="store_true")
 256
 257     args = parser.parse_args()
 258     if args.rss:
 259         import feedparser
 260         d = feedparser.parse(args.rss)
 261         for e in d.entries:
 262             print(("Downloading: %s"%e.title))
 263             if args.no_act:
 264                 continue
 265             video = scrape_player_page({'title':e.title,'url':e.link})
 266             if args.no_remux:
 267                 continue
 268             self.remux(video)
 269         #print(e.description)
 270     if args.mirror:
 271         if not os.path.exists('.seen'):
 272             os.mkdir('.seen')
 273         for video in parse_videolist():
 274             video['title'] = video['title'].replace('/','_')
 275             print(video['title']+'.mkv')
 276             print("{} of {}".format(video['num'], video['total']))
 277
 278             if os.path.exists(os.path.join('.seen',video['title'])):
 279                 print("Skipping")
 280                 continue
 281             print("Downloading...")
 282             if args.no_act:
 283                 continue
 284             open(os.path.join('.seen',video['title']),'w').close() #touch
 285             video = scrape_player_page(video)
 286             if args.no_remux:
 287                 continue
 288             xml = mkv_metadata(video)
 289             remux(video, xml)
 290
 291     else:
 292         if not args.no_act:
 293             video = scrape_player_page({'url':args.url})
 294         if not args.no_remux:
 295             remux({'title':e.title})
 296         print(("Downloaded {}".format(args.url)))