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