]> Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/prosiebensat1.py
Set upstream metadata fields: Bug-Database, Bug-Submit, Repository, Repository-Browse.
[youtubedl] / 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 unified_strdate,
15 )
16
17
18 class ProSiebenSat1BaseIE(InfoExtractor):
19 _GEO_BYPASS = False
20 _ACCESS_ID = None
21 _SUPPORTED_PROTOCOLS = 'dash:clear,hls:clear,progressive:clear'
22 _V4_BASE_URL = 'https://vas-v4.p7s1video.net/4.0/get'
23
24 def _extract_video_info(self, url, clip_id):
25 client_location = url
26
27 video = self._download_json(
28 'http://vas.sim-technik.de/vas/live/v2/videos',
29 clip_id, 'Downloading videos JSON', query={
30 'access_token': self._TOKEN,
31 'client_location': client_location,
32 'client_name': self._CLIENT_NAME,
33 'ids': clip_id,
34 })[0]
35
36 if video.get('is_protected') is True:
37 raise ExtractorError('This video is DRM protected.', expected=True)
38
39 formats = []
40 if self._ACCESS_ID:
41 raw_ct = self._ENCRYPTION_KEY + clip_id + self._IV + self._ACCESS_ID
42 protocols = self._download_json(
43 self._V4_BASE_URL + 'protocols', clip_id,
44 'Downloading protocols JSON',
45 headers=self.geo_verification_headers(), query={
46 'access_id': self._ACCESS_ID,
47 'client_token': sha1((raw_ct).encode()).hexdigest(),
48 'video_id': clip_id,
49 }, fatal=False, expected_status=(403,)) or {}
50 error = protocols.get('error') or {}
51 if error.get('title') == 'Geo check failed':
52 self.raise_geo_restricted(countries=['AT', 'CH', 'DE'])
53 server_token = protocols.get('server_token')
54 if server_token:
55 urls = (self._download_json(
56 self._V4_BASE_URL + 'urls', clip_id, 'Downloading urls JSON', query={
57 'access_id': self._ACCESS_ID,
58 'client_token': sha1((raw_ct + server_token + self._SUPPORTED_PROTOCOLS).encode()).hexdigest(),
59 'protocols': self._SUPPORTED_PROTOCOLS,
60 'server_token': server_token,
61 'video_id': clip_id,
62 }, fatal=False) or {}).get('urls') or {}
63 for protocol, variant in urls.items():
64 source_url = variant.get('clear', {}).get('url')
65 if not source_url:
66 continue
67 if protocol == 'dash':
68 formats.extend(self._extract_mpd_formats(
69 source_url, clip_id, mpd_id=protocol, fatal=False))
70 elif protocol == 'hls':
71 formats.extend(self._extract_m3u8_formats(
72 source_url, clip_id, 'mp4', 'm3u8_native',
73 m3u8_id=protocol, fatal=False))
74 else:
75 formats.append({
76 'url': source_url,
77 'format_id': protocol,
78 })
79 if not formats:
80 source_ids = [compat_str(source['id']) for source in video['sources']]
81
82 client_id = self._SALT[:2] + sha1(''.join([clip_id, self._SALT, self._TOKEN, client_location, self._SALT, self._CLIENT_NAME]).encode('utf-8')).hexdigest()
83
84 sources = self._download_json(
85 'http://vas.sim-technik.de/vas/live/v2/videos/%s/sources' % clip_id,
86 clip_id, 'Downloading sources JSON', query={
87 'access_token': self._TOKEN,
88 'client_id': client_id,
89 'client_location': client_location,
90 'client_name': self._CLIENT_NAME,
91 })
92 server_id = sources['server_id']
93
94 def fix_bitrate(bitrate):
95 bitrate = int_or_none(bitrate)
96 if not bitrate:
97 return None
98 return (bitrate // 1000) if bitrate % 1000 == 0 else bitrate
99
100 for source_id in source_ids:
101 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()
102 urls = self._download_json(
103 'http://vas.sim-technik.de/vas/live/v2/videos/%s/sources/url' % clip_id,
104 clip_id, 'Downloading urls JSON', fatal=False, query={
105 'access_token': self._TOKEN,
106 'client_id': client_id,
107 'client_location': client_location,
108 'client_name': self._CLIENT_NAME,
109 'server_id': server_id,
110 'source_ids': source_id,
111 })
112 if not urls:
113 continue
114 if urls.get('status_code') != 0:
115 raise ExtractorError('This video is unavailable', expected=True)
116 urls_sources = urls['sources']
117 if isinstance(urls_sources, dict):
118 urls_sources = urls_sources.values()
119 for source in urls_sources:
120 source_url = source.get('url')
121 if not source_url:
122 continue
123 protocol = source.get('protocol')
124 mimetype = source.get('mimetype')
125 if mimetype == 'application/f4m+xml' or 'f4mgenerator' in source_url or determine_ext(source_url) == 'f4m':
126 formats.extend(self._extract_f4m_formats(
127 source_url, clip_id, f4m_id='hds', fatal=False))
128 elif mimetype == 'application/x-mpegURL':
129 formats.extend(self._extract_m3u8_formats(
130 source_url, clip_id, 'mp4', 'm3u8_native',
131 m3u8_id='hls', fatal=False))
132 elif mimetype == 'application/dash+xml':
133 formats.extend(self._extract_mpd_formats(
134 source_url, clip_id, mpd_id='dash', fatal=False))
135 else:
136 tbr = fix_bitrate(source['bitrate'])
137 if protocol in ('rtmp', 'rtmpe'):
138 mobj = re.search(r'^(?P<url>rtmpe?://[^/]+)/(?P<path>.+)$', source_url)
139 if not mobj:
140 continue
141 path = mobj.group('path')
142 mp4colon_index = path.rfind('mp4:')
143 app = path[:mp4colon_index]
144 play_path = path[mp4colon_index:]
145 formats.append({
146 'url': '%s/%s' % (mobj.group('url'), app),
147 'app': app,
148 'play_path': play_path,
149 'player_url': 'http://livepassdl.conviva.com/hf/ver/2.79.0.17083/LivePassModuleMain.swf',
150 'page_url': 'http://www.prosieben.de',
151 'tbr': tbr,
152 'ext': 'flv',
153 'format_id': 'rtmp%s' % ('-%d' % tbr if tbr else ''),
154 })
155 else:
156 formats.append({
157 'url': source_url,
158 'tbr': tbr,
159 'format_id': 'http%s' % ('-%d' % tbr if tbr else ''),
160 })
161 self._sort_formats(formats)
162
163 return {
164 'duration': float_or_none(video.get('duration')),
165 'formats': formats,
166 }
167
168
169 class ProSiebenSat1IE(ProSiebenSat1BaseIE):
170 IE_NAME = 'prosiebensat1'
171 IE_DESC = 'ProSiebenSat.1 Digital'
172 _VALID_URL = r'''(?x)
173 https?://
174 (?:www\.)?
175 (?:
176 (?:beta\.)?
177 (?:
178 prosieben(?:maxx)?|sixx|sat1(?:gold)?|kabeleins(?:doku)?|the-voice-of-germany|7tv|advopedia
179 )\.(?:de|at|ch)|
180 ran\.de|fem\.com|advopedia\.de|galileo\.tv/video
181 )
182 /(?P<id>.+)
183 '''
184
185 _TESTS = [
186 {
187 # Tests changes introduced in https://github.com/ytdl-org/youtube-dl/pull/6242
188 # in response to fixing https://github.com/ytdl-org/youtube-dl/issues/6215:
189 # - malformed f4m manifest support
190 # - proper handling of URLs starting with `https?://` in 2.0 manifests
191 # - recursive child f4m manifests extraction
192 'url': 'http://www.prosieben.de/tv/circus-halligalli/videos/218-staffel-2-episode-18-jahresrueckblick-ganze-folge',
193 'info_dict': {
194 'id': '2104602',
195 'ext': 'mp4',
196 'title': 'Episode 18 - Staffel 2',
197 'description': 'md5:8733c81b702ea472e069bc48bb658fc1',
198 'upload_date': '20131231',
199 'duration': 5845.04,
200 },
201 },
202 {
203 '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',
204 'info_dict': {
205 'id': '2570327',
206 'ext': 'mp4',
207 'title': 'Lady-Umstyling für Audrina',
208 'description': 'md5:4c16d0c17a3461a0d43ea4084e96319d',
209 'upload_date': '20131014',
210 'duration': 606.76,
211 },
212 'params': {
213 # rtmp download
214 'skip_download': True,
215 },
216 'skip': 'Seems to be broken',
217 },
218 {
219 'url': 'http://www.prosiebenmaxx.de/tv/experience/video/144-countdown-fuer-die-autowerkstatt-ganze-folge',
220 'info_dict': {
221 'id': '2429369',
222 'ext': 'mp4',
223 'title': 'Countdown für die Autowerkstatt',
224 'description': 'md5:809fc051a457b5d8666013bc40698817',
225 'upload_date': '20140223',
226 'duration': 2595.04,
227 },
228 'params': {
229 # rtmp download
230 'skip_download': True,
231 },
232 'skip': 'This video is unavailable',
233 },
234 {
235 'url': 'http://www.sixx.de/stars-style/video/sexy-laufen-in-ugg-boots-clip',
236 'info_dict': {
237 'id': '2904997',
238 'ext': 'mp4',
239 'title': 'Sexy laufen in Ugg Boots',
240 'description': 'md5:edf42b8bd5bc4e5da4db4222c5acb7d6',
241 'upload_date': '20140122',
242 'duration': 245.32,
243 },
244 'params': {
245 # rtmp download
246 'skip_download': True,
247 },
248 'skip': 'This video is unavailable',
249 },
250 {
251 'url': 'http://www.sat1.de/film/der-ruecktritt/video/im-interview-kai-wiesinger-clip',
252 'info_dict': {
253 'id': '2906572',
254 'ext': 'mp4',
255 'title': 'Im Interview: Kai Wiesinger',
256 'description': 'md5:e4e5370652ec63b95023e914190b4eb9',
257 'upload_date': '20140203',
258 'duration': 522.56,
259 },
260 'params': {
261 # rtmp download
262 'skip_download': True,
263 },
264 'skip': 'This video is unavailable',
265 },
266 {
267 'url': 'http://www.kabeleins.de/tv/rosins-restaurants/videos/jagd-auf-fertigkost-im-elsthal-teil-2-ganze-folge',
268 'info_dict': {
269 'id': '2992323',
270 'ext': 'mp4',
271 'title': 'Jagd auf Fertigkost im Elsthal - Teil 2',
272 'description': 'md5:2669cde3febe9bce13904f701e774eb6',
273 'upload_date': '20141014',
274 'duration': 2410.44,
275 },
276 'params': {
277 # rtmp download
278 'skip_download': True,
279 },
280 'skip': 'This video is unavailable',
281 },
282 {
283 'url': 'http://www.ran.de/fussball/bundesliga/video/schalke-toennies-moechte-raul-zurueck-ganze-folge',
284 'info_dict': {
285 'id': '3004256',
286 'ext': 'mp4',
287 'title': 'Schalke: Tönnies möchte Raul zurück',
288 'description': 'md5:4b5b271d9bcde223b54390754c8ece3f',
289 'upload_date': '20140226',
290 'duration': 228.96,
291 },
292 'params': {
293 # rtmp download
294 'skip_download': True,
295 },
296 'skip': 'This video is unavailable',
297 },
298 {
299 'url': 'http://www.the-voice-of-germany.de/video/31-andreas-kuemmert-rocket-man-clip',
300 'info_dict': {
301 'id': '2572814',
302 'ext': 'mp4',
303 'title': 'Andreas Kümmert: Rocket Man',
304 'description': 'md5:6ddb02b0781c6adf778afea606652e38',
305 'upload_date': '20131017',
306 'duration': 469.88,
307 },
308 'params': {
309 'skip_download': True,
310 },
311 },
312 {
313 'url': 'http://www.fem.com/wellness/videos/wellness-video-clip-kurztripps-zum-valentinstag.html',
314 'info_dict': {
315 'id': '2156342',
316 'ext': 'mp4',
317 'title': 'Kurztrips zum Valentinstag',
318 'description': 'Romantischer Kurztrip zum Valentinstag? Nina Heinemann verrät, was sich hier wirklich lohnt.',
319 'duration': 307.24,
320 },
321 'params': {
322 'skip_download': True,
323 },
324 },
325 {
326 'url': 'http://www.prosieben.de/tv/joko-gegen-klaas/videos/playlists/episode-8-ganze-folge-playlist',
327 'info_dict': {
328 'id': '439664',
329 'title': 'Episode 8 - Ganze Folge - Playlist',
330 'description': 'md5:63b8963e71f481782aeea877658dec84',
331 },
332 'playlist_count': 2,
333 'skip': 'This video is unavailable',
334 },
335 {
336 'url': 'http://www.7tv.de/circus-halligalli/615-best-of-circus-halligalli-ganze-folge',
337 'info_dict': {
338 'id': '4187506',
339 'ext': 'mp4',
340 'title': 'Best of Circus HalliGalli',
341 'description': 'md5:8849752efd90b9772c9db6fdf87fb9e9',
342 'upload_date': '20151229',
343 },
344 'params': {
345 'skip_download': True,
346 },
347 },
348 {
349 # title in <h2 class="subtitle">
350 'url': 'http://www.prosieben.de/stars/oscar-award/videos/jetzt-erst-enthuellt-das-geheimnis-von-emma-stones-oscar-robe-clip',
351 'info_dict': {
352 'id': '4895826',
353 'ext': 'mp4',
354 'title': 'Jetzt erst enthüllt: Das Geheimnis von Emma Stones Oscar-Robe',
355 'description': 'md5:e5ace2bc43fadf7b63adc6187e9450b9',
356 'upload_date': '20170302',
357 },
358 'params': {
359 'skip_download': True,
360 },
361 'skip': 'geo restricted to Germany',
362 },
363 {
364 # geo restricted to Germany
365 'url': 'http://www.kabeleinsdoku.de/tv/mayday-alarm-im-cockpit/video/102-notlandung-im-hudson-river-ganze-folge',
366 'only_matching': True,
367 },
368 {
369 # geo restricted to Germany
370 'url': 'http://www.sat1gold.de/tv/edel-starck/video/11-staffel-1-episode-1-partner-wider-willen-ganze-folge',
371 'only_matching': True,
372 },
373 {
374 # geo restricted to Germany
375 'url': 'https://www.galileo.tv/video/diese-emojis-werden-oft-missverstanden',
376 'only_matching': True,
377 },
378 {
379 'url': 'http://www.sat1gold.de/tv/edel-starck/playlist/die-gesamte-1-staffel',
380 'only_matching': True,
381 },
382 {
383 'url': 'http://www.advopedia.de/videos/lenssen-klaert-auf/lenssen-klaert-auf-folge-8-staffel-3-feiertage-und-freie-tage',
384 'only_matching': True,
385 },
386 ]
387
388 _TOKEN = 'prosieben'
389 _SALT = '01!8d8F_)r9]4s[qeuXfP%'
390 _CLIENT_NAME = 'kolibri-2.0.19-splec4'
391
392 _ACCESS_ID = 'x_prosiebenmaxx-de'
393 _ENCRYPTION_KEY = 'Eeyeey9oquahthainoofashoyoikosag'
394 _IV = 'Aeluchoc6aevechuipiexeeboowedaok'
395
396 _CLIPID_REGEXES = [
397 r'"clip_id"\s*:\s+"(\d+)"',
398 r'clipid: "(\d+)"',
399 r'clip[iI]d=(\d+)',
400 r'clip[iI][dD]\s*=\s*["\'](\d+)',
401 r"'itemImageUrl'\s*:\s*'/dynamic/thumbnails/full/\d+/(\d+)",
402 r'proMamsId&quot;\s*:\s*&quot;(\d+)',
403 r'proMamsId"\s*:\s*"(\d+)',
404 ]
405 _TITLE_REGEXES = [
406 r'<h2 class="subtitle" itemprop="name">\s*(.+?)</h2>',
407 r'<header class="clearfix">\s*<h3>(.+?)</h3>',
408 r'<!-- start video -->\s*<h1>(.+?)</h1>',
409 r'<h1 class="att-name">\s*(.+?)</h1>',
410 r'<header class="module_header">\s*<h2>([^<]+)</h2>\s*</header>',
411 r'<h2 class="video-title" itemprop="name">\s*(.+?)</h2>',
412 r'<div[^>]+id="veeseoTitle"[^>]*>(.+?)</div>',
413 r'<h2[^>]+class="subtitle"[^>]*>([^<]+)</h2>',
414 ]
415 _DESCRIPTION_REGEXES = [
416 r'<p itemprop="description">\s*(.+?)</p>',
417 r'<div class="videoDecription">\s*<p><strong>Beschreibung</strong>: (.+?)</p>',
418 r'<div class="g-plusone" data-size="medium"></div>\s*</div>\s*</header>\s*(.+?)\s*<footer>',
419 r'<p class="att-description">\s*(.+?)\s*</p>',
420 r'<p class="video-description" itemprop="description">\s*(.+?)</p>',
421 r'<div[^>]+id="veeseoDescription"[^>]*>(.+?)</div>',
422 ]
423 _UPLOAD_DATE_REGEXES = [
424 r'<meta property="og:published_time" content="(.+?)">',
425 r'<span>\s*(\d{2}\.\d{2}\.\d{4} \d{2}:\d{2}) \|\s*<span itemprop="duration"',
426 r'<footer>\s*(\d{2}\.\d{2}\.\d{4}) \d{2}:\d{2} Uhr',
427 r'<span style="padding-left: 4px;line-height:20px; color:#404040">(\d{2}\.\d{2}\.\d{4})</span>',
428 r'(\d{2}\.\d{2}\.\d{4}) \| \d{2}:\d{2} Min<br/>',
429 ]
430 _PAGE_TYPE_REGEXES = [
431 r'<meta name="page_type" content="([^"]+)">',
432 r"'itemType'\s*:\s*'([^']*)'",
433 ]
434 _PLAYLIST_ID_REGEXES = [
435 r'content[iI]d=(\d+)',
436 r"'itemId'\s*:\s*'([^']*)'",
437 ]
438 _PLAYLIST_CLIP_REGEXES = [
439 r'(?s)data-qvt=.+?<a href="([^"]+)"',
440 ]
441
442 def _extract_clip(self, url, webpage):
443 clip_id = self._html_search_regex(
444 self._CLIPID_REGEXES, webpage, 'clip id')
445 title = self._html_search_regex(
446 self._TITLE_REGEXES, webpage, 'title',
447 default=None) or self._og_search_title(webpage)
448 info = self._extract_video_info(url, clip_id)
449 description = self._html_search_regex(
450 self._DESCRIPTION_REGEXES, webpage, 'description', default=None)
451 if description is None:
452 description = self._og_search_description(webpage)
453 thumbnail = self._og_search_thumbnail(webpage)
454 upload_date = unified_strdate(self._html_search_regex(
455 self._UPLOAD_DATE_REGEXES, webpage, 'upload date', default=None))
456
457 info.update({
458 'id': clip_id,
459 'title': title,
460 'description': description,
461 'thumbnail': thumbnail,
462 'upload_date': upload_date,
463 })
464 return info
465
466 def _extract_playlist(self, url, webpage):
467 playlist_id = self._html_search_regex(
468 self._PLAYLIST_ID_REGEXES, webpage, 'playlist id')
469 playlist = self._parse_json(
470 self._search_regex(
471 r'var\s+contentResources\s*=\s*(\[.+?\]);\s*</script',
472 webpage, 'playlist'),
473 playlist_id)
474 entries = []
475 for item in playlist:
476 clip_id = item.get('id') or item.get('upc')
477 if not clip_id:
478 continue
479 info = self._extract_video_info(url, clip_id)
480 info.update({
481 'id': clip_id,
482 'title': item.get('title') or item.get('teaser', {}).get('headline'),
483 'description': item.get('teaser', {}).get('description'),
484 'thumbnail': item.get('poster'),
485 'duration': float_or_none(item.get('duration')),
486 'series': item.get('tvShowTitle'),
487 'uploader': item.get('broadcastPublisher'),
488 })
489 entries.append(info)
490 return self.playlist_result(entries, playlist_id)
491
492 def _real_extract(self, url):
493 video_id = self._match_id(url)
494 webpage = self._download_webpage(url, video_id)
495 page_type = self._search_regex(
496 self._PAGE_TYPE_REGEXES, webpage,
497 'page type', default='clip').lower()
498 if page_type == 'clip':
499 return self._extract_clip(url, webpage)
500 elif page_type == 'playlist':
501 return self._extract_playlist(url, webpage)
502 else:
503 raise ExtractorError(
504 'Unsupported page type %s' % page_type, expected=True)