]> Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/francetv.py
Merge pull request #3 from unit193/master
[youtubedl] / youtube_dl / extractor / francetv.py
1 # coding: utf-8
2
3 from __future__ import unicode_literals
4
5 import re
6
7 from .common import InfoExtractor
8 from ..compat import (
9 compat_str,
10 compat_urlparse,
11 )
12 from ..utils import (
13 clean_html,
14 determine_ext,
15 ExtractorError,
16 int_or_none,
17 parse_duration,
18 try_get,
19 url_or_none,
20 )
21 from .dailymotion import DailymotionIE
22
23
24 class FranceTVBaseInfoExtractor(InfoExtractor):
25 def _make_url_result(self, video_or_full_id, catalog=None):
26 full_id = 'francetv:%s' % video_or_full_id
27 if '@' not in video_or_full_id and catalog:
28 full_id += '@%s' % catalog
29 return self.url_result(
30 full_id, ie=FranceTVIE.ie_key(),
31 video_id=video_or_full_id.split('@')[0])
32
33
34 class FranceTVIE(InfoExtractor):
35 _VALID_URL = r'''(?x)
36 (?:
37 https?://
38 sivideo\.webservices\.francetelevisions\.fr/tools/getInfosOeuvre/v2/\?
39 .*?\bidDiffusion=[^&]+|
40 (?:
41 https?://videos\.francetv\.fr/video/|
42 francetv:
43 )
44 (?P<id>[^@]+)(?:@(?P<catalog>.+))?
45 )
46 '''
47
48 _TESTS = [{
49 # without catalog
50 'url': 'https://sivideo.webservices.francetelevisions.fr/tools/getInfosOeuvre/v2/?idDiffusion=162311093&callback=_jsonp_loader_callback_request_0',
51 'md5': 'c2248a8de38c4e65ea8fae7b5df2d84f',
52 'info_dict': {
53 'id': '162311093',
54 'ext': 'mp4',
55 'title': '13h15, le dimanche... - Les mystères de Jésus',
56 'description': 'md5:75efe8d4c0a8205e5904498ffe1e1a42',
57 'timestamp': 1502623500,
58 'upload_date': '20170813',
59 },
60 }, {
61 # with catalog
62 'url': 'https://sivideo.webservices.francetelevisions.fr/tools/getInfosOeuvre/v2/?idDiffusion=NI_1004933&catalogue=Zouzous&callback=_jsonp_loader_callback_request_4',
63 'only_matching': True,
64 }, {
65 'url': 'http://videos.francetv.fr/video/NI_657393@Regions',
66 'only_matching': True,
67 }, {
68 'url': 'francetv:162311093',
69 'only_matching': True,
70 }, {
71 'url': 'francetv:NI_1004933@Zouzous',
72 'only_matching': True,
73 }, {
74 'url': 'francetv:NI_983319@Info-web',
75 'only_matching': True,
76 }, {
77 'url': 'francetv:NI_983319',
78 'only_matching': True,
79 }, {
80 'url': 'francetv:NI_657393@Regions',
81 'only_matching': True,
82 }, {
83 # france-3 live
84 'url': 'francetv:SIM_France3',
85 'only_matching': True,
86 }]
87
88 def _extract_video(self, video_id, catalogue=None):
89 # Videos are identified by idDiffusion so catalogue part is optional.
90 # However when provided, some extra formats may be returned so we pass
91 # it if available.
92 info = self._download_json(
93 'https://sivideo.webservices.francetelevisions.fr/tools/getInfosOeuvre/v2/',
94 video_id, 'Downloading video JSON', query={
95 'idDiffusion': video_id,
96 'catalogue': catalogue or '',
97 })
98
99 if info.get('status') == 'NOK':
100 raise ExtractorError(
101 '%s returned error: %s' % (self.IE_NAME, info['message']),
102 expected=True)
103 allowed_countries = info['videos'][0].get('geoblocage')
104 if allowed_countries:
105 georestricted = True
106 geo_info = self._download_json(
107 'http://geo.francetv.fr/ws/edgescape.json', video_id,
108 'Downloading geo restriction info')
109 country = geo_info['reponse']['geo_info']['country_code']
110 if country not in allowed_countries:
111 raise ExtractorError(
112 'The video is not available from your location',
113 expected=True)
114 else:
115 georestricted = False
116
117 def sign(manifest_url, manifest_id):
118 for host in ('hdfauthftv-a.akamaihd.net', 'hdfauth.francetv.fr'):
119 signed_url = url_or_none(self._download_webpage(
120 'https://%s/esi/TA' % host, video_id,
121 'Downloading signed %s manifest URL' % manifest_id,
122 fatal=False, query={
123 'url': manifest_url,
124 }))
125 if signed_url:
126 return signed_url
127 return manifest_url
128
129 is_live = None
130
131 formats = []
132 for video in info['videos']:
133 if video['statut'] != 'ONLINE':
134 continue
135 video_url = video['url']
136 if not video_url:
137 continue
138 if is_live is None:
139 is_live = (try_get(
140 video, lambda x: x['plages_ouverture'][0]['direct'],
141 bool) is True) or '/live.francetv.fr/' in video_url
142 format_id = video['format']
143 ext = determine_ext(video_url)
144 if ext == 'f4m':
145 if georestricted:
146 # See https://github.com/ytdl-org/youtube-dl/issues/3963
147 # m3u8 urls work fine
148 continue
149 formats.extend(self._extract_f4m_formats(
150 sign(video_url, format_id) + '&hdcore=3.7.0&plugin=aasp-3.7.0.39.44',
151 video_id, f4m_id=format_id, fatal=False))
152 elif ext == 'm3u8':
153 formats.extend(self._extract_m3u8_formats(
154 sign(video_url, format_id), video_id, 'mp4',
155 entry_protocol='m3u8_native', m3u8_id=format_id,
156 fatal=False))
157 elif video_url.startswith('rtmp'):
158 formats.append({
159 'url': video_url,
160 'format_id': 'rtmp-%s' % format_id,
161 'ext': 'flv',
162 })
163 else:
164 if self._is_valid_url(video_url, video_id, format_id):
165 formats.append({
166 'url': video_url,
167 'format_id': format_id,
168 })
169 self._sort_formats(formats)
170
171 title = info['titre']
172 subtitle = info.get('sous_titre')
173 if subtitle:
174 title += ' - %s' % subtitle
175 title = title.strip()
176
177 subtitles = {}
178 subtitles_list = [{
179 'url': subformat['url'],
180 'ext': subformat.get('format'),
181 } for subformat in info.get('subtitles', []) if subformat.get('url')]
182 if subtitles_list:
183 subtitles['fr'] = subtitles_list
184
185 return {
186 'id': video_id,
187 'title': self._live_title(title) if is_live else title,
188 'description': clean_html(info['synopsis']),
189 'thumbnail': compat_urlparse.urljoin('http://pluzz.francetv.fr', info['image']),
190 'duration': int_or_none(info.get('real_duration')) or parse_duration(info['duree']),
191 'timestamp': int_or_none(info['diffusion']['timestamp']),
192 'is_live': is_live,
193 'formats': formats,
194 'subtitles': subtitles,
195 }
196
197 def _real_extract(self, url):
198 mobj = re.match(self._VALID_URL, url)
199 video_id = mobj.group('id')
200 catalog = mobj.group('catalog')
201
202 if not video_id:
203 qs = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
204 video_id = qs.get('idDiffusion', [None])[0]
205 catalog = qs.get('catalogue', [None])[0]
206 if not video_id:
207 raise ExtractorError('Invalid URL', expected=True)
208
209 return self._extract_video(video_id, catalog)
210
211
212 class FranceTVSiteIE(FranceTVBaseInfoExtractor):
213 _VALID_URL = r'https?://(?:(?:www\.)?france\.tv|mobile\.france\.tv)/(?:[^/]+/)*(?P<id>[^/]+)\.html'
214
215 _TESTS = [{
216 'url': 'https://www.france.tv/france-2/13h15-le-dimanche/140921-les-mysteres-de-jesus.html',
217 'info_dict': {
218 'id': 'ec217ecc-0733-48cf-ac06-af1347b849d1',
219 'ext': 'mp4',
220 'title': '13h15, le dimanche... - Les mystères de Jésus',
221 'description': 'md5:75efe8d4c0a8205e5904498ffe1e1a42',
222 'timestamp': 1502623500,
223 'upload_date': '20170813',
224 },
225 'params': {
226 'skip_download': True,
227 },
228 'add_ie': [FranceTVIE.ie_key()],
229 }, {
230 # france3
231 'url': 'https://www.france.tv/france-3/des-chiffres-et-des-lettres/139063-emission-du-mardi-9-mai-2017.html',
232 'only_matching': True,
233 }, {
234 # france4
235 'url': 'https://www.france.tv/france-4/hero-corp/saison-1/134151-apres-le-calme.html',
236 'only_matching': True,
237 }, {
238 # france5
239 'url': 'https://www.france.tv/france-5/c-a-dire/saison-10/137013-c-a-dire.html',
240 'only_matching': True,
241 }, {
242 # franceo
243 'url': 'https://www.france.tv/france-o/archipels/132249-mon-ancetre-l-esclave.html',
244 'only_matching': True,
245 }, {
246 # france2 live
247 'url': 'https://www.france.tv/france-2/direct.html',
248 'only_matching': True,
249 }, {
250 'url': 'https://www.france.tv/documentaires/histoire/136517-argentine-les-500-bebes-voles-de-la-dictature.html',
251 'only_matching': True,
252 }, {
253 'url': 'https://www.france.tv/jeux-et-divertissements/divertissements/133965-le-web-contre-attaque.html',
254 'only_matching': True,
255 }, {
256 'url': 'https://mobile.france.tv/france-5/c-dans-l-air/137347-emission-du-vendredi-12-mai-2017.html',
257 'only_matching': True,
258 }, {
259 'url': 'https://www.france.tv/142749-rouge-sang.html',
260 'only_matching': True,
261 }, {
262 # france-3 live
263 'url': 'https://www.france.tv/france-3/direct.html',
264 'only_matching': True,
265 }]
266
267 def _real_extract(self, url):
268 display_id = self._match_id(url)
269
270 webpage = self._download_webpage(url, display_id)
271
272 catalogue = None
273 video_id = self._search_regex(
274 r'(?:data-main-video\s*=|videoId["\']?\s*[:=])\s*(["\'])(?P<id>(?:(?!\1).)+)\1',
275 webpage, 'video id', default=None, group='id')
276
277 if not video_id:
278 video_id, catalogue = self._html_search_regex(
279 r'(?:href=|player\.setVideo\(\s*)"http://videos?\.francetv\.fr/video/([^@]+@[^"]+)"',
280 webpage, 'video ID').split('@')
281
282 return self._make_url_result(video_id, catalogue)
283
284
285 class FranceTVEmbedIE(FranceTVBaseInfoExtractor):
286 _VALID_URL = r'https?://embed\.francetv\.fr/*\?.*?\bue=(?P<id>[^&]+)'
287
288 _TESTS = [{
289 'url': 'http://embed.francetv.fr/?ue=7fd581a2ccf59d2fc5719c5c13cf6961',
290 'info_dict': {
291 'id': 'NI_983319',
292 'ext': 'mp4',
293 'title': 'Le Pen Reims',
294 'upload_date': '20170505',
295 'timestamp': 1493981780,
296 'duration': 16,
297 },
298 'params': {
299 'skip_download': True,
300 },
301 'add_ie': [FranceTVIE.ie_key()],
302 }]
303
304 def _real_extract(self, url):
305 video_id = self._match_id(url)
306
307 video = self._download_json(
308 'http://api-embed.webservices.francetelevisions.fr/key/%s' % video_id,
309 video_id)
310
311 return self._make_url_result(video['video_id'], video.get('catalog'))
312
313
314 class FranceTVInfoIE(FranceTVBaseInfoExtractor):
315 IE_NAME = 'francetvinfo.fr'
316 _VALID_URL = r'https?://(?:www|mobile|france3-regions)\.francetvinfo\.fr/(?:[^/]+/)*(?P<id>[^/?#&.]+)'
317
318 _TESTS = [{
319 'url': 'http://www.francetvinfo.fr/replay-jt/france-3/soir-3/jt-grand-soir-3-lundi-26-aout-2013_393427.html',
320 'info_dict': {
321 'id': '84981923',
322 'ext': 'mp4',
323 'title': 'Soir 3',
324 'upload_date': '20130826',
325 'timestamp': 1377548400,
326 'subtitles': {
327 'fr': 'mincount:2',
328 },
329 },
330 'params': {
331 'skip_download': True,
332 },
333 'add_ie': [FranceTVIE.ie_key()],
334 }, {
335 'url': 'http://www.francetvinfo.fr/elections/europeennes/direct-europeennes-regardez-le-debat-entre-les-candidats-a-la-presidence-de-la-commission_600639.html',
336 'only_matching': True,
337 }, {
338 'url': 'http://www.francetvinfo.fr/economie/entreprises/les-entreprises-familiales-le-secret-de-la-reussite_933271.html',
339 'only_matching': True,
340 }, {
341 'url': 'http://france3-regions.francetvinfo.fr/bretagne/cotes-d-armor/thalassa-echappee-breizh-ce-venredi-dans-les-cotes-d-armor-954961.html',
342 'only_matching': True,
343 }, {
344 # Dailymotion embed
345 'url': 'http://www.francetvinfo.fr/politique/notre-dame-des-landes/video-sur-france-inter-cecile-duflot-denonce-le-regard-meprisant-de-patrick-cohen_1520091.html',
346 'md5': 'ee7f1828f25a648addc90cb2687b1f12',
347 'info_dict': {
348 'id': 'x4iiko0',
349 'ext': 'mp4',
350 'title': 'NDDL, référendum, Brexit : Cécile Duflot répond à Patrick Cohen',
351 'description': 'Au lendemain de la victoire du "oui" au référendum sur l\'aéroport de Notre-Dame-des-Landes, l\'ancienne ministre écologiste est l\'invitée de Patrick Cohen. Plus d\'info : https://www.franceinter.fr/emissions/le-7-9/le-7-9-27-juin-2016',
352 'timestamp': 1467011958,
353 'upload_date': '20160627',
354 'uploader': 'France Inter',
355 'uploader_id': 'x2q2ez',
356 },
357 'add_ie': ['Dailymotion'],
358 }, {
359 'url': 'http://france3-regions.francetvinfo.fr/limousin/emissions/jt-1213-limousin',
360 'only_matching': True,
361 }]
362
363 def _real_extract(self, url):
364 display_id = self._match_id(url)
365
366 webpage = self._download_webpage(url, display_id)
367
368 dailymotion_urls = DailymotionIE._extract_urls(webpage)
369 if dailymotion_urls:
370 return self.playlist_result([
371 self.url_result(dailymotion_url, DailymotionIE.ie_key())
372 for dailymotion_url in dailymotion_urls])
373
374 video_id = self._search_regex(
375 (r'player\.load[^;]+src:\s*["\']([^"\']+)',
376 r'id-video=([^@]+@[^"]+)',
377 r'<a[^>]+href="(?:https?:)?//videos\.francetv\.fr/video/([^@]+@[^"]+)"'),
378 webpage, 'video id')
379
380 return self._make_url_result(video_id)
381
382
383 class FranceTVInfoSportIE(FranceTVBaseInfoExtractor):
384 IE_NAME = 'sport.francetvinfo.fr'
385 _VALID_URL = r'https?://sport\.francetvinfo\.fr/(?:[^/]+/)*(?P<id>[^/?#&]+)'
386 _TESTS = [{
387 'url': 'https://sport.francetvinfo.fr/les-jeux-olympiques/retour-sur-les-meilleurs-moments-de-pyeongchang-2018',
388 'info_dict': {
389 'id': '6e49080e-3f45-11e8-b459-000d3a2439ea',
390 'ext': 'mp4',
391 'title': 'Retour sur les meilleurs moments de Pyeongchang 2018',
392 'timestamp': 1523639962,
393 'upload_date': '20180413',
394 },
395 'params': {
396 'skip_download': True,
397 },
398 'add_ie': [FranceTVIE.ie_key()],
399 }]
400
401 def _real_extract(self, url):
402 display_id = self._match_id(url)
403 webpage = self._download_webpage(url, display_id)
404 video_id = self._search_regex(r'data-video="([^"]+)"', webpage, 'video_id')
405 return self._make_url_result(video_id, 'Sport-web')
406
407
408 class GenerationWhatIE(InfoExtractor):
409 IE_NAME = 'france2.fr:generation-what'
410 _VALID_URL = r'https?://generation-what\.francetv\.fr/[^/]+/video/(?P<id>[^/?#&]+)'
411
412 _TESTS = [{
413 'url': 'http://generation-what.francetv.fr/portrait/video/present-arms',
414 'info_dict': {
415 'id': 'wtvKYUG45iw',
416 'ext': 'mp4',
417 'title': 'Generation What - Garde à vous - FRA',
418 'uploader': 'Generation What',
419 'uploader_id': 'UCHH9p1eetWCgt4kXBYCb3_w',
420 'upload_date': '20160411',
421 },
422 'params': {
423 'skip_download': True,
424 },
425 'add_ie': ['Youtube'],
426 }, {
427 'url': 'http://generation-what.francetv.fr/europe/video/present-arms',
428 'only_matching': True,
429 }]
430
431 def _real_extract(self, url):
432 display_id = self._match_id(url)
433
434 webpage = self._download_webpage(url, display_id)
435
436 youtube_id = self._search_regex(
437 r"window\.videoURL\s*=\s*'([0-9A-Za-z_-]{11})';",
438 webpage, 'youtube id')
439
440 return self.url_result(youtube_id, ie='Youtube', video_id=youtube_id)
441
442
443 class CultureboxIE(FranceTVBaseInfoExtractor):
444 _VALID_URL = r'https?://(?:m\.)?culturebox\.francetvinfo\.fr/(?:[^/]+/)*(?P<id>[^/?#&]+)'
445
446 _TESTS = [{
447 'url': 'https://culturebox.francetvinfo.fr/opera-classique/musique-classique/c-est-baroque/concerts/cantates-bwv-4-106-et-131-de-bach-par-raphael-pichon-57-268689',
448 'info_dict': {
449 'id': 'EV_134885',
450 'ext': 'mp4',
451 'title': 'Cantates BWV 4, 106 et 131 de Bach par Raphaël Pichon 5/7',
452 'description': 'md5:19c44af004b88219f4daa50fa9a351d4',
453 'upload_date': '20180206',
454 'timestamp': 1517945220,
455 'duration': 5981,
456 },
457 'params': {
458 'skip_download': True,
459 },
460 'add_ie': [FranceTVIE.ie_key()],
461 }]
462
463 def _real_extract(self, url):
464 display_id = self._match_id(url)
465
466 webpage = self._download_webpage(url, display_id)
467
468 if ">Ce live n'est plus disponible en replay<" in webpage:
469 raise ExtractorError(
470 'Video %s is not available' % display_id, expected=True)
471
472 video_id, catalogue = self._search_regex(
473 r'["\'>]https?://videos\.francetv\.fr/video/([^@]+@.+?)["\'<]',
474 webpage, 'video id').split('@')
475
476 return self._make_url_result(video_id, catalogue)
477
478
479 class FranceTVJeunesseIE(FranceTVBaseInfoExtractor):
480 _VALID_URL = r'(?P<url>https?://(?:www\.)?(?:zouzous|ludo)\.fr/heros/(?P<id>[^/?#&]+))'
481
482 _TESTS = [{
483 'url': 'https://www.zouzous.fr/heros/simon',
484 'info_dict': {
485 'id': 'simon',
486 },
487 'playlist_count': 9,
488 }, {
489 'url': 'https://www.ludo.fr/heros/ninjago',
490 'info_dict': {
491 'id': 'ninjago',
492 },
493 'playlist_count': 10,
494 }, {
495 'url': 'https://www.zouzous.fr/heros/simon?abc',
496 'only_matching': True,
497 }]
498
499 def _real_extract(self, url):
500 mobj = re.match(self._VALID_URL, url)
501 playlist_id = mobj.group('id')
502
503 playlist = self._download_json(
504 '%s/%s' % (mobj.group('url'), 'playlist'), playlist_id)
505
506 if not playlist.get('count'):
507 raise ExtractorError(
508 '%s is not available' % playlist_id, expected=True)
509
510 entries = []
511 for item in playlist['items']:
512 identity = item.get('identity')
513 if identity and isinstance(identity, compat_str):
514 entries.append(self._make_url_result(identity))
515
516 return self.playlist_result(entries, playlist_id)