]> gitweb @ CieloNegro.org - youtube-dl.git/blob - youtube_dl/extractor/prosiebensat1.py
[ard] Improve _VALID_URL (closes #25134) (#25198)
[youtube-dl.git] / youtube_dl / extractor / prosiebensat1.py
1 # coding: 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     merge_dicts,
15     unified_strdate,
16 )
17
18
19 class ProSiebenSat1BaseIE(InfoExtractor):
20     _GEO_BYPASS = False
21     _ACCESS_ID = None
22     _SUPPORTED_PROTOCOLS = 'dash:clear,hls:clear,progressive:clear'
23     _V4_BASE_URL = 'https://vas-v4.p7s1video.net/4.0/get'
24
25     def _extract_video_info(self, url, clip_id):
26         client_location = url
27
28         video = self._download_json(
29             'http://vas.sim-technik.de/vas/live/v2/videos',
30             clip_id, 'Downloading videos JSON', query={
31                 'access_token': self._TOKEN,
32                 'client_location': client_location,
33                 'client_name': self._CLIENT_NAME,
34                 'ids': clip_id,
35             })[0]
36
37         if video.get('is_protected') is True:
38             raise ExtractorError('This video is DRM protected.', expected=True)
39
40         formats = []
41         if self._ACCESS_ID:
42             raw_ct = self._ENCRYPTION_KEY + clip_id + self._IV + self._ACCESS_ID
43             protocols = self._download_json(
44                 self._V4_BASE_URL + 'protocols', clip_id,
45                 'Downloading protocols JSON',
46                 headers=self.geo_verification_headers(), query={
47                     'access_id': self._ACCESS_ID,
48                     'client_token': sha1((raw_ct).encode()).hexdigest(),
49                     'video_id': clip_id,
50                 }, fatal=False, expected_status=(403,)) or {}
51             error = protocols.get('error') or {}
52             if error.get('title') == 'Geo check failed':
53                 self.raise_geo_restricted(countries=['AT', 'CH', 'DE'])
54             server_token = protocols.get('server_token')
55             if server_token:
56                 urls = (self._download_json(
57                     self._V4_BASE_URL + 'urls', clip_id, 'Downloading urls JSON', query={
58                         'access_id': self._ACCESS_ID,
59                         'client_token': sha1((raw_ct + server_token + self._SUPPORTED_PROTOCOLS).encode()).hexdigest(),
60                         'protocols': self._SUPPORTED_PROTOCOLS,
61                         'server_token': server_token,
62                         'video_id': clip_id,
63                     }, fatal=False) or {}).get('urls') or {}
64                 for protocol, variant in urls.items():
65                     source_url = variant.get('clear', {}).get('url')
66                     if not source_url:
67                         continue
68                     if protocol == 'dash':
69                         formats.extend(self._extract_mpd_formats(
70                             source_url, clip_id, mpd_id=protocol, fatal=False))
71                     elif protocol == 'hls':
72                         formats.extend(self._extract_m3u8_formats(
73                             source_url, clip_id, 'mp4', 'm3u8_native',
74                             m3u8_id=protocol, fatal=False))
75                     else:
76                         formats.append({
77                             'url': source_url,
78                             'format_id': protocol,
79                         })
80         if not formats:
81             source_ids = [compat_str(source['id']) for source in video['sources']]
82
83             client_id = self._SALT[:2] + sha1(''.join([clip_id, self._SALT, self._TOKEN, client_location, self._SALT, self._CLIENT_NAME]).encode('utf-8')).hexdigest()
84
85             sources = self._download_json(
86                 'http://vas.sim-technik.de/vas/live/v2/videos/%s/sources' % clip_id,
87                 clip_id, 'Downloading sources JSON', query={
88                     'access_token': self._TOKEN,
89                     'client_id': client_id,
90                     'client_location': client_location,
91                     'client_name': self._CLIENT_NAME,
92                 })
93             server_id = sources['server_id']
94
95             def fix_bitrate(bitrate):
96                 bitrate = int_or_none(bitrate)
97                 if not bitrate:
98                     return None
99                 return (bitrate // 1000) if bitrate % 1000 == 0 else bitrate
100
101             for source_id in source_ids:
102                 client_id = self._SALT[:2] + sha1(''.join([self._SALT, clip_id, self._TOKEN, server_id, client_location, source_id, self._SALT, self._CLIENT_NAME]).encode('utf-8')).hexdigest()
103                 urls = self._download_json(
104                     'http://vas.sim-technik.de/vas/live/v2/videos/%s/sources/url' % clip_id,
105                     clip_id, 'Downloading urls JSON', fatal=False, query={
106                         'access_token': self._TOKEN,
107                         'client_id': client_id,
108                         'client_location': client_location,
109                         'client_name': self._CLIENT_NAME,
110                         'server_id': server_id,
111                         'source_ids': source_id,
112                     })
113                 if not urls:
114                     continue
115                 if urls.get('status_code') != 0:
116                     raise ExtractorError('This video is unavailable', expected=True)
117                 urls_sources = urls['sources']
118                 if isinstance(urls_sources, dict):
119                     urls_sources = urls_sources.values()
120                 for source in urls_sources:
121                     source_url = source.get('url')
122                     if not source_url:
123                         continue
124                     protocol = source.get('protocol')
125                     mimetype = source.get('mimetype')
126                     if mimetype == 'application/f4m+xml' or 'f4mgenerator' in source_url or determine_ext(source_url) == 'f4m':
127                         formats.extend(self._extract_f4m_formats(
128                             source_url, clip_id, f4m_id='hds', fatal=False))
129                     elif mimetype == 'application/x-mpegURL':
130                         formats.extend(self._extract_m3u8_formats(
131                             source_url, clip_id, 'mp4', 'm3u8_native',
132                             m3u8_id='hls', fatal=False))
133                     elif mimetype == 'application/dash+xml':
134                         formats.extend(self._extract_mpd_formats(
135                             source_url, clip_id, mpd_id='dash', fatal=False))
136                     else:
137                         tbr = fix_bitrate(source['bitrate'])
138                         if protocol in ('rtmp', 'rtmpe'):
139                             mobj = re.search(r'^(?P<url>rtmpe?://[^/]+)/(?P<path>.+)$', source_url)
140                             if not mobj:
141                                 continue
142                             path = mobj.group('path')
143                             mp4colon_index = path.rfind('mp4:')
144                             app = path[:mp4colon_index]
145                             play_path = path[mp4colon_index:]
146                             formats.append({
147                                 'url': '%s/%s' % (mobj.group('url'), app),
148                                 'app': app,
149                                 'play_path': play_path,
150                                 'player_url': 'http://livepassdl.conviva.com/hf/ver/2.79.0.17083/LivePassModuleMain.swf',
151                                 'page_url': 'http://www.prosieben.de',
152                                 'tbr': tbr,
153                                 'ext': 'flv',
154                                 'format_id': 'rtmp%s' % ('-%d' % tbr if tbr else ''),
155                             })
156                         else:
157                             formats.append({
158                                 'url': source_url,
159                                 'tbr': tbr,
160                                 'format_id': 'http%s' % ('-%d' % tbr if tbr else ''),
161                             })
162         self._sort_formats(formats)
163
164         return {
165             'duration': float_or_none(video.get('duration')),
166             'formats': formats,
167         }
168
169
170 class ProSiebenSat1IE(ProSiebenSat1BaseIE):
171     IE_NAME = 'prosiebensat1'
172     IE_DESC = 'ProSiebenSat.1 Digital'
173     _VALID_URL = r'''(?x)
174                     https?://
175                         (?:www\.)?
176                         (?:
177                             (?:beta\.)?
178                             (?:
179                                 prosieben(?:maxx)?|sixx|sat1(?:gold)?|kabeleins(?:doku)?|the-voice-of-germany|advopedia
180                             )\.(?:de|at|ch)|
181                             ran\.de|fem\.com|advopedia\.de|galileo\.tv/video
182                         )
183                         /(?P<id>.+)
184                     '''
185
186     _TESTS = [
187         {
188             # Tests changes introduced in https://github.com/ytdl-org/youtube-dl/pull/6242
189             # in response to fixing https://github.com/ytdl-org/youtube-dl/issues/6215:
190             # - malformed f4m manifest support
191             # - proper handling of URLs starting with `https?://` in 2.0 manifests
192             # - recursive child f4m manifests extraction
193             'url': 'http://www.prosieben.de/tv/circus-halligalli/videos/218-staffel-2-episode-18-jahresrueckblick-ganze-folge',
194             'info_dict': {
195                 'id': '2104602',
196                 'ext': 'mp4',
197                 'title': 'CIRCUS HALLIGALLI - Episode 18 - Staffel 2',
198                 'description': 'md5:8733c81b702ea472e069bc48bb658fc1',
199                 'upload_date': '20131231',
200                 'duration': 5845.04,
201                 'series': 'CIRCUS HALLIGALLI',
202                 'season_number': 2,
203                 'episode': 'Episode 18 - Staffel 2',
204                 'episode_number': 18,
205             },
206         },
207         {
208             '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',
209             'info_dict': {
210                 'id': '2570327',
211                 'ext': 'mp4',
212                 'title': 'Lady-Umstyling für Audrina',
213                 'description': 'md5:4c16d0c17a3461a0d43ea4084e96319d',
214                 'upload_date': '20131014',
215                 'duration': 606.76,
216             },
217             'params': {
218                 # rtmp download
219                 'skip_download': True,
220             },
221             'skip': 'Seems to be broken',
222         },
223         {
224             'url': 'http://www.prosiebenmaxx.de/tv/experience/video/144-countdown-fuer-die-autowerkstatt-ganze-folge',
225             'info_dict': {
226                 'id': '2429369',
227                 'ext': 'mp4',
228                 'title': 'Countdown für die Autowerkstatt',
229                 'description': 'md5:809fc051a457b5d8666013bc40698817',
230                 'upload_date': '20140223',
231                 'duration': 2595.04,
232             },
233             'params': {
234                 # rtmp download
235                 'skip_download': True,
236             },
237             'skip': 'This video is unavailable',
238         },
239         {
240             'url': 'http://www.sixx.de/stars-style/video/sexy-laufen-in-ugg-boots-clip',
241             'info_dict': {
242                 'id': '2904997',
243                 'ext': 'mp4',
244                 'title': 'Sexy laufen in Ugg Boots',
245                 'description': 'md5:edf42b8bd5bc4e5da4db4222c5acb7d6',
246                 'upload_date': '20140122',
247                 'duration': 245.32,
248             },
249             'params': {
250                 # rtmp download
251                 'skip_download': True,
252             },
253             'skip': 'This video is unavailable',
254         },
255         {
256             'url': 'http://www.sat1.de/film/der-ruecktritt/video/im-interview-kai-wiesinger-clip',
257             'info_dict': {
258                 'id': '2906572',
259                 'ext': 'mp4',
260                 'title': 'Im Interview: Kai Wiesinger',
261                 'description': 'md5:e4e5370652ec63b95023e914190b4eb9',
262                 'upload_date': '20140203',
263                 'duration': 522.56,
264             },
265             'params': {
266                 # rtmp download
267                 'skip_download': True,
268             },
269             'skip': 'This video is unavailable',
270         },
271         {
272             'url': 'http://www.kabeleins.de/tv/rosins-restaurants/videos/jagd-auf-fertigkost-im-elsthal-teil-2-ganze-folge',
273             'info_dict': {
274                 'id': '2992323',
275                 'ext': 'mp4',
276                 'title': 'Jagd auf Fertigkost im Elsthal - Teil 2',
277                 'description': 'md5:2669cde3febe9bce13904f701e774eb6',
278                 'upload_date': '20141014',
279                 'duration': 2410.44,
280             },
281             'params': {
282                 # rtmp download
283                 'skip_download': True,
284             },
285             'skip': 'This video is unavailable',
286         },
287         {
288             'url': 'http://www.ran.de/fussball/bundesliga/video/schalke-toennies-moechte-raul-zurueck-ganze-folge',
289             'info_dict': {
290                 'id': '3004256',
291                 'ext': 'mp4',
292                 'title': 'Schalke: Tönnies möchte Raul zurück',
293                 'description': 'md5:4b5b271d9bcde223b54390754c8ece3f',
294                 'upload_date': '20140226',
295                 'duration': 228.96,
296             },
297             'params': {
298                 # rtmp download
299                 'skip_download': True,
300             },
301             'skip': 'This video is unavailable',
302         },
303         {
304             'url': 'http://www.the-voice-of-germany.de/video/31-andreas-kuemmert-rocket-man-clip',
305             'info_dict': {
306                 'id': '2572814',
307                 'ext': 'mp4',
308                 'title': 'The Voice of Germany - Andreas Kümmert: Rocket Man',
309                 'description': 'md5:6ddb02b0781c6adf778afea606652e38',
310                 'timestamp': 1382041620,
311                 'upload_date': '20131017',
312                 'duration': 469.88,
313             },
314             'params': {
315                 'skip_download': True,
316             },
317         },
318         {
319             'url': 'http://www.fem.com/videos/beauty-lifestyle/kurztrips-zum-valentinstag',
320             'info_dict': {
321                 'id': '2156342',
322                 'ext': 'mp4',
323                 'title': 'Kurztrips zum Valentinstag',
324                 'description': 'Romantischer Kurztrip zum Valentinstag? Nina Heinemann verrät, was sich hier wirklich lohnt.',
325                 'duration': 307.24,
326             },
327             'params': {
328                 'skip_download': True,
329             },
330         },
331         {
332             'url': 'http://www.prosieben.de/tv/joko-gegen-klaas/videos/playlists/episode-8-ganze-folge-playlist',
333             'info_dict': {
334                 'id': '439664',
335                 'title': 'Episode 8 - Ganze Folge - Playlist',
336                 'description': 'md5:63b8963e71f481782aeea877658dec84',
337             },
338             'playlist_count': 2,
339             'skip': 'This video is unavailable',
340         },
341         {
342             # title in <h2 class="subtitle">
343             'url': 'http://www.prosieben.de/stars/oscar-award/videos/jetzt-erst-enthuellt-das-geheimnis-von-emma-stones-oscar-robe-clip',
344             'info_dict': {
345                 'id': '4895826',
346                 'ext': 'mp4',
347                 'title': 'Jetzt erst enthüllt: Das Geheimnis von Emma Stones Oscar-Robe',
348                 'description': 'md5:e5ace2bc43fadf7b63adc6187e9450b9',
349                 'upload_date': '20170302',
350             },
351             'params': {
352                 'skip_download': True,
353             },
354             'skip': 'geo restricted to Germany',
355         },
356         {
357             # geo restricted to Germany
358             'url': 'http://www.kabeleinsdoku.de/tv/mayday-alarm-im-cockpit/video/102-notlandung-im-hudson-river-ganze-folge',
359             'only_matching': True,
360         },
361         {
362             # geo restricted to Germany
363             'url': 'http://www.sat1gold.de/tv/edel-starck/video/11-staffel-1-episode-1-partner-wider-willen-ganze-folge',
364             'only_matching': True,
365         },
366         {
367             # geo restricted to Germany
368             'url': 'https://www.galileo.tv/video/diese-emojis-werden-oft-missverstanden',
369             'only_matching': True,
370         },
371         {
372             'url': 'http://www.sat1gold.de/tv/edel-starck/playlist/die-gesamte-1-staffel',
373             'only_matching': True,
374         },
375         {
376             'url': 'http://www.advopedia.de/videos/lenssen-klaert-auf/lenssen-klaert-auf-folge-8-staffel-3-feiertage-und-freie-tage',
377             'only_matching': True,
378         },
379     ]
380
381     _TOKEN = 'prosieben'
382     _SALT = '01!8d8F_)r9]4s[qeuXfP%'
383     _CLIENT_NAME = 'kolibri-2.0.19-splec4'
384
385     _ACCESS_ID = 'x_prosiebenmaxx-de'
386     _ENCRYPTION_KEY = 'Eeyeey9oquahthainoofashoyoikosag'
387     _IV = 'Aeluchoc6aevechuipiexeeboowedaok'
388
389     _CLIPID_REGEXES = [
390         r'"clip_id"\s*:\s+"(\d+)"',
391         r'clipid: "(\d+)"',
392         r'clip[iI]d=(\d+)',
393         r'clip[iI][dD]\s*=\s*["\'](\d+)',
394         r"'itemImageUrl'\s*:\s*'/dynamic/thumbnails/full/\d+/(\d+)",
395         r'proMamsId&quot;\s*:\s*&quot;(\d+)',
396         r'proMamsId"\s*:\s*"(\d+)',
397     ]
398     _TITLE_REGEXES = [
399         r'<h2 class="subtitle" itemprop="name">\s*(.+?)</h2>',
400         r'<header class="clearfix">\s*<h3>(.+?)</h3>',
401         r'<!-- start video -->\s*<h1>(.+?)</h1>',
402         r'<h1 class="att-name">\s*(.+?)</h1>',
403         r'<header class="module_header">\s*<h2>([^<]+)</h2>\s*</header>',
404         r'<h2 class="video-title" itemprop="name">\s*(.+?)</h2>',
405         r'<div[^>]+id="veeseoTitle"[^>]*>(.+?)</div>',
406         r'<h2[^>]+class="subtitle"[^>]*>([^<]+)</h2>',
407     ]
408     _DESCRIPTION_REGEXES = [
409         r'<p itemprop="description">\s*(.+?)</p>',
410         r'<div class="videoDecription">\s*<p><strong>Beschreibung</strong>: (.+?)</p>',
411         r'<div class="g-plusone" data-size="medium"></div>\s*</div>\s*</header>\s*(.+?)\s*<footer>',
412         r'<p class="att-description">\s*(.+?)\s*</p>',
413         r'<p class="video-description" itemprop="description">\s*(.+?)</p>',
414         r'<div[^>]+id="veeseoDescription"[^>]*>(.+?)</div>',
415     ]
416     _UPLOAD_DATE_REGEXES = [
417         r'<span>\s*(\d{2}\.\d{2}\.\d{4} \d{2}:\d{2}) \|\s*<span itemprop="duration"',
418         r'<footer>\s*(\d{2}\.\d{2}\.\d{4}) \d{2}:\d{2} Uhr',
419         r'<span style="padding-left: 4px;line-height:20px; color:#404040">(\d{2}\.\d{2}\.\d{4})</span>',
420         r'(\d{2}\.\d{2}\.\d{4}) \| \d{2}:\d{2} Min<br/>',
421     ]
422     _PAGE_TYPE_REGEXES = [
423         r'<meta name="page_type" content="([^"]+)">',
424         r"'itemType'\s*:\s*'([^']*)'",
425     ]
426     _PLAYLIST_ID_REGEXES = [
427         r'content[iI]d=(\d+)',
428         r"'itemId'\s*:\s*'([^']*)'",
429     ]
430     _PLAYLIST_CLIP_REGEXES = [
431         r'(?s)data-qvt=.+?<a href="([^"]+)"',
432     ]
433
434     def _extract_clip(self, url, webpage):
435         clip_id = self._html_search_regex(
436             self._CLIPID_REGEXES, webpage, 'clip id')
437         title = self._html_search_regex(
438             self._TITLE_REGEXES, webpage, 'title',
439             default=None) or self._og_search_title(webpage)
440         info = self._extract_video_info(url, clip_id)
441         description = self._html_search_regex(
442             self._DESCRIPTION_REGEXES, webpage, 'description', default=None)
443         if description is None:
444             description = self._og_search_description(webpage)
445         thumbnail = self._og_search_thumbnail(webpage)
446         upload_date = unified_strdate(
447             self._html_search_meta('og:published_time', webpage,
448                                    'upload date', default=None)
449             or self._html_search_regex(self._UPLOAD_DATE_REGEXES,
450                                        webpage, 'upload date', default=None))
451
452         json_ld = self._search_json_ld(webpage, clip_id, default={})
453
454         return merge_dicts(info, {
455             'id': clip_id,
456             'title': title,
457             'description': description,
458             'thumbnail': thumbnail,
459             'upload_date': upload_date,
460         }, json_ld)
461
462     def _extract_playlist(self, url, webpage):
463         playlist_id = self._html_search_regex(
464             self._PLAYLIST_ID_REGEXES, webpage, 'playlist id')
465         playlist = self._parse_json(
466             self._search_regex(
467                 r'var\s+contentResources\s*=\s*(\[.+?\]);\s*</script',
468                 webpage, 'playlist'),
469             playlist_id)
470         entries = []
471         for item in playlist:
472             clip_id = item.get('id') or item.get('upc')
473             if not clip_id:
474                 continue
475             info = self._extract_video_info(url, clip_id)
476             info.update({
477                 'id': clip_id,
478                 'title': item.get('title') or item.get('teaser', {}).get('headline'),
479                 'description': item.get('teaser', {}).get('description'),
480                 'thumbnail': item.get('poster'),
481                 'duration': float_or_none(item.get('duration')),
482                 'series': item.get('tvShowTitle'),
483                 'uploader': item.get('broadcastPublisher'),
484             })
485             entries.append(info)
486         return self.playlist_result(entries, playlist_id)
487
488     def _real_extract(self, url):
489         video_id = self._match_id(url)
490         webpage = self._download_webpage(url, video_id)
491         page_type = self._search_regex(
492             self._PAGE_TYPE_REGEXES, webpage,
493             'page type', default='clip').lower()
494         if page_type == 'clip':
495             return self._extract_clip(url, webpage)
496         elif page_type == 'playlist':
497             return self._extract_playlist(url, webpage)
498         else:
499             raise ExtractorError(
500                 'Unsupported page type %s' % page_type, expected=True)