]> git.bitcoin.ninja Git - youtube-dl/blob - youtube_dl/extractor/prosiebensat1.py
[README.md] Make download URLs consistent
[youtube-dl] / youtube_dl / extractor / prosiebensat1.py
1 # encoding: utf-8
2 from __future__ import unicode_literals
3
4 import re
5
6 from hashlib import sha1
7 from .common import InfoExtractor
8 from ..compat import compat_str
9 from ..utils import (
10     ExtractorError,
11     determine_ext,
12     float_or_none,
13     int_or_none,
14     unified_strdate,
15 )
16
17
18 class ProSiebenSat1IE(InfoExtractor):
19     IE_NAME = 'prosiebensat1'
20     IE_DESC = 'ProSiebenSat.1 Digital'
21     _VALID_URL = r'https?://(?:www\.)?(?:(?:prosieben|prosiebenmaxx|sixx|sat1|kabeleins|the-voice-of-germany|7tv)\.(?:de|at|ch)|ran\.de|fem\.com)/(?P<id>.+)'
22
23     _TESTS = [
24         {
25             # Tests changes introduced in https://github.com/rg3/youtube-dl/pull/6242
26             # in response to fixing https://github.com/rg3/youtube-dl/issues/6215:
27             # - malformed f4m manifest support
28             # - proper handling of URLs starting with `https?://` in 2.0 manifests
29             # - recursive child f4m manifests extraction
30             'url': 'http://www.prosieben.de/tv/circus-halligalli/videos/218-staffel-2-episode-18-jahresrueckblick-ganze-folge',
31             'info_dict': {
32                 'id': '2104602',
33                 'ext': 'flv',
34                 'title': 'Episode 18 - Staffel 2',
35                 'description': 'md5:8733c81b702ea472e069bc48bb658fc1',
36                 'upload_date': '20131231',
37                 'duration': 5845.04,
38             },
39             'params': {
40                 # rtmp download
41                 'skip_download': True,
42             },
43         },
44         {
45             'url': 'http://www.prosieben.de/videokatalog/Gesellschaft/Leben/Trends/video-Lady-Umstyling-f%C3%BCr-Audrina-Rebekka-Audrina-Fergen-billig-aussehen-Battal-Modica-700544.html',
46             'info_dict': {
47                 'id': '2570327',
48                 'ext': 'mp4',
49                 'title': 'Lady-Umstyling für Audrina',
50                 'description': 'md5:4c16d0c17a3461a0d43ea4084e96319d',
51                 'upload_date': '20131014',
52                 'duration': 606.76,
53             },
54             'params': {
55                 # rtmp download
56                 'skip_download': True,
57             },
58             'skip': 'Seems to be broken',
59         },
60         {
61             'url': 'http://www.prosiebenmaxx.de/tv/experience/video/144-countdown-fuer-die-autowerkstatt-ganze-folge',
62             'info_dict': {
63                 'id': '2429369',
64                 'ext': 'mp4',
65                 'title': 'Countdown für die Autowerkstatt',
66                 'description': 'md5:809fc051a457b5d8666013bc40698817',
67                 'upload_date': '20140223',
68                 'duration': 2595.04,
69             },
70             'params': {
71                 # rtmp download
72                 'skip_download': True,
73             },
74             'skip': 'This video is unavailable',
75         },
76         {
77             'url': 'http://www.sixx.de/stars-style/video/sexy-laufen-in-ugg-boots-clip',
78             'info_dict': {
79                 'id': '2904997',
80                 'ext': 'mp4',
81                 'title': 'Sexy laufen in Ugg Boots',
82                 'description': 'md5:edf42b8bd5bc4e5da4db4222c5acb7d6',
83                 'upload_date': '20140122',
84                 'duration': 245.32,
85             },
86             'params': {
87                 # rtmp download
88                 'skip_download': True,
89             },
90             'skip': 'This video is unavailable',
91         },
92         {
93             'url': 'http://www.sat1.de/film/der-ruecktritt/video/im-interview-kai-wiesinger-clip',
94             'info_dict': {
95                 'id': '2906572',
96                 'ext': 'mp4',
97                 'title': 'Im Interview: Kai Wiesinger',
98                 'description': 'md5:e4e5370652ec63b95023e914190b4eb9',
99                 'upload_date': '20140203',
100                 'duration': 522.56,
101             },
102             'params': {
103                 # rtmp download
104                 'skip_download': True,
105             },
106             'skip': 'This video is unavailable',
107         },
108         {
109             'url': 'http://www.kabeleins.de/tv/rosins-restaurants/videos/jagd-auf-fertigkost-im-elsthal-teil-2-ganze-folge',
110             'info_dict': {
111                 'id': '2992323',
112                 'ext': 'mp4',
113                 'title': 'Jagd auf Fertigkost im Elsthal - Teil 2',
114                 'description': 'md5:2669cde3febe9bce13904f701e774eb6',
115                 'upload_date': '20141014',
116                 'duration': 2410.44,
117             },
118             'params': {
119                 # rtmp download
120                 'skip_download': True,
121             },
122             'skip': 'This video is unavailable',
123         },
124         {
125             'url': 'http://www.ran.de/fussball/bundesliga/video/schalke-toennies-moechte-raul-zurueck-ganze-folge',
126             'info_dict': {
127                 'id': '3004256',
128                 'ext': 'mp4',
129                 'title': 'Schalke: Tönnies möchte Raul zurück',
130                 'description': 'md5:4b5b271d9bcde223b54390754c8ece3f',
131                 'upload_date': '20140226',
132                 'duration': 228.96,
133             },
134             'params': {
135                 # rtmp download
136                 'skip_download': True,
137             },
138             'skip': 'This video is unavailable',
139         },
140         {
141             'url': 'http://www.the-voice-of-germany.de/video/31-andreas-kuemmert-rocket-man-clip',
142             'info_dict': {
143                 'id': '2572814',
144                 'ext': 'flv',
145                 'title': 'Andreas Kümmert: Rocket Man',
146                 'description': 'md5:6ddb02b0781c6adf778afea606652e38',
147                 'upload_date': '20131017',
148                 'duration': 469.88,
149             },
150             'params': {
151                 'skip_download': True,
152             },
153         },
154         {
155             'url': 'http://www.fem.com/wellness/videos/wellness-video-clip-kurztripps-zum-valentinstag.html',
156             'info_dict': {
157                 'id': '2156342',
158                 'ext': 'flv',
159                 'title': 'Kurztrips zum Valentinstag',
160                 'description': 'Romantischer Kurztrip zum Valentinstag? Nina Heinemann verrät, was sich hier wirklich lohnt.',
161                 'duration': 307.24,
162             },
163             'params': {
164                 'skip_download': True,
165             },
166         },
167         {
168             'url': 'http://www.prosieben.de/tv/joko-gegen-klaas/videos/playlists/episode-8-ganze-folge-playlist',
169             'info_dict': {
170                 'id': '439664',
171                 'title': 'Episode 8 - Ganze Folge - Playlist',
172                 'description': 'md5:63b8963e71f481782aeea877658dec84',
173             },
174             'playlist_count': 2,
175         },
176         {
177             'url': 'http://www.7tv.de/circus-halligalli/615-best-of-circus-halligalli-ganze-folge',
178             'info_dict': {
179                 'id': '4187506',
180                 'ext': 'flv',
181                 'title': 'Best of Circus HalliGalli',
182                 'description': 'md5:8849752efd90b9772c9db6fdf87fb9e9',
183                 'upload_date': '20151229',
184             },
185             'params': {
186                 'skip_download': True,
187             },
188         },
189     ]
190
191     _CLIPID_REGEXES = [
192         r'"clip_id"\s*:\s+"(\d+)"',
193         r'clipid: "(\d+)"',
194         r'clip[iI]d=(\d+)',
195         r'clip[iI]d\s*=\s*["\'](\d+)',
196         r"'itemImageUrl'\s*:\s*'/dynamic/thumbnails/full/\d+/(\d+)",
197     ]
198     _TITLE_REGEXES = [
199         r'<h2 class="subtitle" itemprop="name">\s*(.+?)</h2>',
200         r'<header class="clearfix">\s*<h3>(.+?)</h3>',
201         r'<!-- start video -->\s*<h1>(.+?)</h1>',
202         r'<h1 class="att-name">\s*(.+?)</h1>',
203         r'<header class="module_header">\s*<h2>([^<]+)</h2>\s*</header>',
204         r'<h2 class="video-title" itemprop="name">\s*(.+?)</h2>',
205         r'<div[^>]+id="veeseoTitle"[^>]*>(.+?)</div>',
206     ]
207     _DESCRIPTION_REGEXES = [
208         r'<p itemprop="description">\s*(.+?)</p>',
209         r'<div class="videoDecription">\s*<p><strong>Beschreibung</strong>: (.+?)</p>',
210         r'<div class="g-plusone" data-size="medium"></div>\s*</div>\s*</header>\s*(.+?)\s*<footer>',
211         r'<p class="att-description">\s*(.+?)\s*</p>',
212         r'<p class="video-description" itemprop="description">\s*(.+?)</p>',
213         r'<div[^>]+id="veeseoDescription"[^>]*>(.+?)</div>',
214     ]
215     _UPLOAD_DATE_REGEXES = [
216         r'<meta property="og:published_time" content="(.+?)">',
217         r'<span>\s*(\d{2}\.\d{2}\.\d{4} \d{2}:\d{2}) \|\s*<span itemprop="duration"',
218         r'<footer>\s*(\d{2}\.\d{2}\.\d{4}) \d{2}:\d{2} Uhr',
219         r'<span style="padding-left: 4px;line-height:20px; color:#404040">(\d{2}\.\d{2}\.\d{4})</span>',
220         r'(\d{2}\.\d{2}\.\d{4}) \| \d{2}:\d{2} Min<br/>',
221     ]
222     _PAGE_TYPE_REGEXES = [
223         r'<meta name="page_type" content="([^"]+)">',
224         r"'itemType'\s*:\s*'([^']*)'",
225     ]
226     _PLAYLIST_ID_REGEXES = [
227         r'content[iI]d=(\d+)',
228         r"'itemId'\s*:\s*'([^']*)'",
229     ]
230     _PLAYLIST_CLIP_REGEXES = [
231         r'(?s)data-qvt=.+?<a href="([^"]+)"',
232     ]
233
234     def _extract_clip(self, url, webpage):
235         clip_id = self._html_search_regex(
236             self._CLIPID_REGEXES, webpage, 'clip id')
237
238         access_token = 'prosieben'
239         client_name = 'kolibri-2.0.19-splec4'
240         client_location = url
241
242         video = self._download_json(
243             'http://vas.sim-technik.de/vas/live/v2/videos',
244             clip_id, 'Downloading videos JSON', query={
245                 'access_token': access_token,
246                 'client_location': client_location,
247                 'client_name': client_name,
248                 'ids': clip_id,
249             })[0]
250
251         if video.get('is_protected') is True:
252             raise ExtractorError('This video is DRM protected.', expected=True)
253
254         duration = float_or_none(video.get('duration'))
255         source_ids = [compat_str(source['id']) for source in video['sources']]
256
257         g = '01!8d8F_)r9]4s[qeuXfP%'
258         client_id = g[:2] + sha1(''.join([clip_id, g, access_token, client_location, g, client_name]).encode('utf-8')).hexdigest()
259
260         sources = self._download_json(
261             'http://vas.sim-technik.de/vas/live/v2/videos/%s/sources' % clip_id,
262             clip_id, 'Downloading sources JSON', query={
263                 'access_token': access_token,
264                 'client_id': client_id,
265                 'client_location': client_location,
266                 'client_name': client_name,
267             })
268         server_id = sources['server_id']
269
270         title = self._html_search_regex(self._TITLE_REGEXES, webpage, 'title')
271
272         def fix_bitrate(bitrate):
273             bitrate = int_or_none(bitrate)
274             if not bitrate:
275                 return None
276             return (bitrate // 1000) if bitrate % 1000 == 0 else bitrate
277
278         formats = []
279         for source_id in source_ids:
280             client_id = g[:2] + sha1(''.join([g, clip_id, access_token, server_id, client_location, source_id, g, client_name]).encode('utf-8')).hexdigest()
281             urls = self._download_json(
282                 'http://vas.sim-technik.de/vas/live/v2/videos/%s/sources/url' % clip_id,
283                 clip_id, 'Downloading urls JSON', fatal=False, query={
284                     'access_token': access_token,
285                     'client_id': client_id,
286                     'client_location': client_location,
287                     'client_name': client_name,
288                     'server_id': server_id,
289                     'source_ids': source_id,
290                 })
291             if not urls:
292                 continue
293             if urls.get('status_code') != 0:
294                 raise ExtractorError('This video is unavailable', expected=True)
295             urls_sources = urls['sources']
296             if isinstance(urls_sources, dict):
297                 urls_sources = urls_sources.values()
298             for source in urls_sources:
299                 source_url = source.get('url')
300                 if not source_url:
301                     continue
302                 protocol = source.get('protocol')
303                 mimetype = source.get('mimetype')
304                 if mimetype == 'application/f4m+xml' or 'f4mgenerator' in source_url or determine_ext(source_url) == 'f4m':
305                     formats.extend(self._extract_f4m_formats(
306                         source_url, clip_id, f4m_id='hds', fatal=False))
307                 elif mimetype == 'application/x-mpegURL':
308                     formats.extend(self._extract_m3u8_formats(
309                         source_url, clip_id, 'mp4', 'm3u8_native',
310                         m3u8_id='hls', fatal=False))
311                 else:
312                     tbr = fix_bitrate(source['bitrate'])
313                     if protocol in ('rtmp', 'rtmpe'):
314                         mobj = re.search(r'^(?P<url>rtmpe?://[^/]+)/(?P<path>.+)$', source_url)
315                         if not mobj:
316                             continue
317                         path = mobj.group('path')
318                         mp4colon_index = path.rfind('mp4:')
319                         app = path[:mp4colon_index]
320                         play_path = path[mp4colon_index:]
321                         formats.append({
322                             'url': '%s/%s' % (mobj.group('url'), app),
323                             'app': app,
324                             'play_path': play_path,
325                             'player_url': 'http://livepassdl.conviva.com/hf/ver/2.79.0.17083/LivePassModuleMain.swf',
326                             'page_url': 'http://www.prosieben.de',
327                             'tbr': tbr,
328                             'ext': 'flv',
329                             'format_id': 'rtmp%s' % ('-%d' % tbr if tbr else ''),
330                         })
331                     else:
332                         formats.append({
333                             'url': source_url,
334                             'tbr': tbr,
335                             'format_id': 'http%s' % ('-%d' % tbr if tbr else ''),
336                         })
337         self._sort_formats(formats)
338
339         description = self._html_search_regex(
340             self._DESCRIPTION_REGEXES, webpage, 'description', fatal=False)
341         thumbnail = self._og_search_thumbnail(webpage)
342         upload_date = unified_strdate(self._html_search_regex(
343             self._UPLOAD_DATE_REGEXES, webpage, 'upload date', default=None))
344
345         return {
346             'id': clip_id,
347             'title': title,
348             'description': description,
349             'thumbnail': thumbnail,
350             'upload_date': upload_date,
351             'duration': duration,
352             'formats': formats,
353         }
354
355     def _extract_playlist(self, url, webpage):
356         playlist_id = self._html_search_regex(
357             self._PLAYLIST_ID_REGEXES, webpage, 'playlist id')
358         for regex in self._PLAYLIST_CLIP_REGEXES:
359             playlist_clips = re.findall(regex, webpage)
360             if playlist_clips:
361                 title = self._html_search_regex(
362                     self._TITLE_REGEXES, webpage, 'title')
363                 description = self._html_search_regex(
364                     self._DESCRIPTION_REGEXES, webpage, 'description', fatal=False)
365                 entries = [
366                     self.url_result(
367                         re.match('(.+?//.+?)/', url).group(1) + clip_path,
368                         'ProSiebenSat1')
369                     for clip_path in playlist_clips]
370                 return self.playlist_result(entries, playlist_id, title, description)
371
372     def _real_extract(self, url):
373         video_id = self._match_id(url)
374         webpage = self._download_webpage(url, video_id)
375         page_type = self._search_regex(
376             self._PAGE_TYPE_REGEXES, webpage,
377             'page type', default='clip').lower()
378         if page_type == 'clip':
379             return self._extract_clip(url, webpage)
380         elif page_type == 'playlist':
381             return self._extract_playlist(url, webpage)