Bugfixes to make download work again with separate subtitles.(the subs are ignored...
[svtplaydump.git] / 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 'm3u8' in reference['url']:
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 if not 'url' in video:
71 print("Could not find any streams")
72 return False
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 if not videourl.startswith('http'): #if relative path
81 videourl = "{}/{}".format(os.path.dirname(video['url']), videourl)
82 segments, metadata = parse_segment_playlist(videourl)
83 if "EXT-X-KEY" in metadata:
84 key = requests.get(metadata["EXT-X-KEY"]['URI'].strip('"')).text
85 decrypt=True
86 else:
87 decrypt=False
88 with open("%s"%video['filename'],"wb") as ofile:
89 segment=0
90 size = 0
91 for url in segments:
92 ufile = requests.get(url, stream=True).raw
93 print("\r{} MB".format(size/1024/1024))
94 sys.stdout.flush()
95 if decrypt:
96 iv=struct.pack("IIII",segment,0,0,0)
97 decryptor = AES.new(key, AES.MODE_CBC, iv)
98 while(True):
99 buf = ufile.read(4096)
100 if not buf:
101 break
102 if decrypt:
103 buf = decryptor.decrypt(buf)
104 ofile.write(buf)
105 size += len(buf)
106 segment += 1
107
108 def parse_playlist(playlist):
109 if not playlist.startswith("#EXTM3U"):
110 print(playlist)
111 return False
112 playlist = playlist.splitlines()
113 while not 'EXT-X-STREAM-INF' in playlist[0]:
114 playlist = playlist[1:]
115 items=[]
116 for (metadata_string,url) in zip(playlist[0::2], playlist[1::2]):
117 md = dict()
118 if not 'EXT-X-STREAM-INF' in metadata_string.split(':')[0]:
119 continue
120 for item in metadata_string.split(':')[1].split(','):
121 if '=' in item:
122 md.update([item.split('='),])
123 md['url']=url
124 items.append(md)
125 return items
126
127 def parse_segment_playlist(playlisturl):
128 playlist = requests.get(playlisturl).text
129 assert playlist.startswith("#EXTM3U")
130 PATTERN = re.compile(r'''((?:[^,"']|"[^"]*"|'[^']*')+)''')
131 segments = []
132 next_is_url=False
133 metadata = {}
134 for row in playlist.splitlines():
135 if next_is_url:
136 if not row.startswith('http'): #if relative path
137 row = "{}/{}".format(os.path.dirname(playlisturl), row)
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
149 def parse_videolist():
150 page_num = 1
151 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
152 page_tot = int(soup.find('a',{'data-currentpage':True}).attrs['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 soup = BeautifulSoup(requests.get(base_url).text)
158 for article in soup.findAll('article'):
159 meta = dict(article.attrs)
160 video = {}
161 video['title'] = meta['data-title']
162 video['description'] = meta['data-description']
163 video['url'] = dict(article.find('a').attrs)['href']
164 video['thumb-url'] = dict(article.find('img',{}).attrs)['src']
165 video['num'] = video_num
166 video['total'] = page_tot * videos_per_page
167 video_num += 1
168 yield video
169 page_num += 1
170
171 def remux(video):
172 basename = video['filename'].split('.ts')[0]
173 print(Popen(["avconv","-i",video['filename'],"-vcodec","copy","-acodec","copy", basename+'.mkv'], stdout=PIPE).communicate()[0])
174 try:
175 os.unlink(video['filename'])
176 except:
177 pass
178
179 if __name__ == "__main__":
180 parser = argparse.ArgumentParser()
181 group = parser.add_mutually_exclusive_group(required=True)
182 group.add_argument("-r", "--rss", help="Download all files in rss")
183 group.add_argument("-u", "--url", help="Download video in url")
184 group.add_argument("-m", "--mirror", help="Mirror all files", action="store_true")
185 parser.add_argument("-n", "--no_act", help="Just print what would be done, don't do any downloading.", action="store_true")
186 parser.add_argument("--no_remux", help="Don't remux into mkv", action="store_true")
187
188 args = parser.parse_args()
189 if args.rss:
190 import feedparser
191 d = feedparser.parse(args.rss)
192 for e in d.entries:
193 print(("Downloading: %s"%e.title))
194 if args.no_act:
195 continue
196 filename = scrape_player_page(e.link, e.title)
197 if args.no_remux:
198 continue
199 self.remux({'title':e.title})
200 #print(e.description)
201 if args.mirror:
202 for video in parse_videolist():
203 video['title'] = video['title'].replace('/','_')
204 print(video['title']+'.mkv')
205 print("{} of {}".format(video['num'], video['total']))
206 if os.path.exists(video['title']+'.mkv'):
207 print("Skipping")
208 continue
209 print("Downloading...")
210 if args.no_act:
211 continue
212 video = scrape_player_page(video['url'], video['title'])
213 if args.no_remux:
214 continue
215 remux(video)
216 else:
217 if not args.no_act:
218 video = scrape_player_page(args.url, None)
219 if not args.no_remux:
220 remux({'title':e.title})
221 print(("Downloaded {}".format(args.url)))