yt_dlp/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     urljoin,
  21 )
  22 from .dailymotion import DailymotionIE
  23
  24
  25 class FranceTVBaseInfoExtractor(InfoExtractor):
  26     def _make_url_result(self, video_or_full_id, catalog=None):
  27         full_id = 'francetv:%s' % video_or_full_id
  28         if '@' not in video_or_full_id and catalog:
  29             full_id += '@%s' % catalog
  30         return self.url_result(
  31             full_id, ie=FranceTVIE.ie_key(),
  32             video_id=video_or_full_id.split('@')[0])
  33
  34
  35 class FranceTVIE(InfoExtractor):
  36     _VALID_URL = r'''(?x)
  37                     (?:
  38                         https?://
  39                             sivideo\.webservices\.francetelevisions\.fr/tools/getInfosOeuvre/v2/\?
  40                             .*?\bidDiffusion=[^&]+|
  41                         (?:
  42                             https?://videos\.francetv\.fr/video/|
  43                             francetv:
  44                         )
  45                         (?P<id>[^@]+)(?:@(?P<catalog>.+))?
  46                     )
  47                     '''
  48
  49     _TESTS = [{
  50         # without catalog
  51         'url': 'https://sivideo.webservices.francetelevisions.fr/tools/getInfosOeuvre/v2/?idDiffusion=162311093&callback=_jsonp_loader_callback_request_0',
  52         'md5': 'c2248a8de38c4e65ea8fae7b5df2d84f',
  53         'info_dict': {
  54             'id': '162311093',
  55             'ext': 'mp4',
  56             'title': '13h15, le dimanche... - Les mystères de Jésus',
  57             'description': 'md5:75efe8d4c0a8205e5904498ffe1e1a42',
  58             'timestamp': 1502623500,
  59             'upload_date': '20170813',
  60         },
  61     }, {
  62         # with catalog
  63         'url': 'https://sivideo.webservices.francetelevisions.fr/tools/getInfosOeuvre/v2/?idDiffusion=NI_1004933&catalogue=Zouzous&callback=_jsonp_loader_callback_request_4',
  64         'only_matching': True,
  65     }, {
  66         'url': 'http://videos.francetv.fr/video/NI_657393@Regions',
  67         'only_matching': True,
  68     }, {
  69         'url': 'francetv:162311093',
  70         'only_matching': True,
  71     }, {
  72         'url': 'francetv:NI_1004933@Zouzous',
  73         'only_matching': True,
  74     }, {
  75         'url': 'francetv:NI_983319@Info-web',
  76         'only_matching': True,
  77     }, {
  78         'url': 'francetv:NI_983319',
  79         'only_matching': True,
  80     }, {
  81         'url': 'francetv:NI_657393@Regions',
  82         'only_matching': True,
  83     }, {
  84         # france-3 live
  85         'url': 'francetv:SIM_France3',
  86         'only_matching': True,
  87     }]
  88
  89     def _extract_video(self, video_id, catalogue=None):
  90         # Videos are identified by idDiffusion so catalogue part is optional.
  91         # However when provided, some extra formats may be returned so we pass
  92         # it if available.
  93         info = self._download_json(
  94             'https://sivideo.webservices.francetelevisions.fr/tools/getInfosOeuvre/v2/',
  95             video_id, 'Downloading video JSON', query={
  96                 'idDiffusion': video_id,
  97                 'catalogue': catalogue or '',
  98             })
  99
 100         if info.get('status') == 'NOK':
 101             raise ExtractorError(
 102                 '%s returned error: %s' % (self.IE_NAME, info['message']),
 103                 expected=True)
 104         allowed_countries = info['videos'][0].get('geoblocage')
 105         if allowed_countries:
 106             georestricted = True
 107             geo_info = self._download_json(
 108                 'http://geo.francetv.fr/ws/edgescape.json', video_id,
 109                 'Downloading geo restriction info')
 110             country = geo_info['reponse']['geo_info']['country_code']
 111             if country not in allowed_countries:
 112                 raise ExtractorError(
 113                     'The video is not available from your location',
 114                     expected=True)
 115         else:
 116             georestricted = False
 117
 118         def sign(manifest_url, manifest_id):
 119             for host in ('hdfauthftv-a.akamaihd.net', 'hdfauth.francetv.fr'):
 120                 signed_url = url_or_none(self._download_webpage(
 121                     'https://%s/esi/TA' % host, video_id,
 122                     'Downloading signed %s manifest URL' % manifest_id,
 123                     fatal=False, query={
 124                         'url': manifest_url,
 125                     }))
 126                 if signed_url:
 127                     return signed_url
 128             return manifest_url
 129
 130         is_live = None
 131
 132         videos = []
 133
 134         for video in (info.get('videos') or []):
 135             if video.get('statut') != 'ONLINE':
 136                 continue
 137             if not video.get('url'):
 138                 continue
 139             videos.append(video)
 140
 141         if not videos:
 142             for device_type in ['desktop', 'mobile']:
 143                 fallback_info = self._download_json(
 144                     'https://player.webservices.francetelevisions.fr/v1/videos/%s' % video_id,
 145                     video_id, 'Downloading fallback %s video JSON' % device_type, query={
 146                         'device_type': device_type,
 147                         'browser': 'chrome',
 148                     }, fatal=False)
 149
 150                 if fallback_info and fallback_info.get('video'):
 151                     videos.append(fallback_info['video'])
 152
 153         formats = []
 154         for video in videos:
 155             video_url = video.get('url')
 156             if not video_url:
 157                 continue
 158             if is_live is None:
 159                 is_live = (try_get(
 160                     video, lambda x: x['plages_ouverture'][0]['direct'], bool) is True
 161                     or video.get('is_live') is True
 162                     or '/live.francetv.fr/' in video_url)
 163             format_id = video.get('format')
 164             ext = determine_ext(video_url)
 165             if ext == 'f4m':
 166                 if georestricted:
 167                     # See https://github.com/ytdl-org/youtube-dl/issues/3963
 168                     # m3u8 urls work fine
 169                     continue
 170                 formats.extend(self._extract_f4m_formats(
 171                     sign(video_url, format_id) + '&hdcore=3.7.0&plugin=aasp-3.7.0.39.44',
 172                     video_id, f4m_id=format_id, fatal=False))
 173             elif ext == 'm3u8':
 174                 formats.extend(self._extract_m3u8_formats(
 175                     sign(video_url, format_id), video_id, 'mp4',
 176                     entry_protocol='m3u8_native', m3u8_id=format_id,
 177                     fatal=False))
 178             elif ext == 'mpd':
 179                 formats.extend(self._extract_mpd_formats(
 180                     sign(video_url, format_id), video_id, mpd_id=format_id, fatal=False))
 181             elif video_url.startswith('rtmp'):
 182                 formats.append({
 183                     'url': video_url,
 184                     'format_id': 'rtmp-%s' % format_id,
 185                     'ext': 'flv',
 186                 })
 187             else:
 188                 if self._is_valid_url(video_url, video_id, format_id):
 189                     formats.append({
 190                         'url': video_url,
 191                         'format_id': format_id,
 192                     })
 193
 194         self._sort_formats(formats)
 195
 196         title = info['titre']
 197         subtitle = info.get('sous_titre')
 198         if subtitle:
 199             title += ' - %s' % subtitle
 200         title = title.strip()
 201
 202         subtitles = {}
 203         subtitles_list = [{
 204             'url': subformat['url'],
 205             'ext': subformat.get('format'),
 206         } for subformat in info.get('subtitles', []) if subformat.get('url')]
 207         if subtitles_list:
 208             subtitles['fr'] = subtitles_list
 209
 210         return {
 211             'id': video_id,
 212             'title': self._live_title(title) if is_live else title,
 213             'description': clean_html(info.get('synopsis')),
 214             'thumbnail': urljoin('https://sivideo.webservices.francetelevisions.fr', info.get('image')),
 215             'duration': int_or_none(info.get('real_duration')) or parse_duration(info.get('duree')),
 216             'timestamp': int_or_none(try_get(info, lambda x: x['diffusion']['timestamp'])),
 217             'is_live': is_live,
 218             'formats': formats,
 219             'subtitles': subtitles,
 220         }
 221
 222     def _real_extract(self, url):
 223         mobj = re.match(self._VALID_URL, url)
 224         video_id = mobj.group('id')
 225         catalog = mobj.group('catalog')
 226
 227         if not video_id:
 228             qs = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
 229             video_id = qs.get('idDiffusion', [None])[0]
 230             catalog = qs.get('catalogue', [None])[0]
 231             if not video_id:
 232                 raise ExtractorError('Invalid URL', expected=True)
 233
 234         return self._extract_video(video_id, catalog)
 235
 236
 237 class FranceTVSiteIE(FranceTVBaseInfoExtractor):
 238     _VALID_URL = r'https?://(?:(?:www\.)?france\.tv|mobile\.france\.tv)/(?:[^/]+/)*(?P<id>[^/]+)\.html'
 239
 240     _TESTS = [{
 241         'url': 'https://www.france.tv/france-2/13h15-le-dimanche/140921-les-mysteres-de-jesus.html',
 242         'info_dict': {
 243             'id': 'ec217ecc-0733-48cf-ac06-af1347b849d1',
 244             'ext': 'mp4',
 245             'title': '13h15, le dimanche... - Les mystères de Jésus',
 246             'description': 'md5:75efe8d4c0a8205e5904498ffe1e1a42',
 247             'timestamp': 1502623500,
 248             'upload_date': '20170813',
 249         },
 250         'params': {
 251             'skip_download': True,
 252         },
 253         'add_ie': [FranceTVIE.ie_key()],
 254     }, {
 255         # france3
 256         'url': 'https://www.france.tv/france-3/des-chiffres-et-des-lettres/139063-emission-du-mardi-9-mai-2017.html',
 257         'only_matching': True,
 258     }, {
 259         # france4
 260         'url': 'https://www.france.tv/france-4/hero-corp/saison-1/134151-apres-le-calme.html',
 261         'only_matching': True,
 262     }, {
 263         # france5
 264         'url': 'https://www.france.tv/france-5/c-a-dire/saison-10/137013-c-a-dire.html',
 265         'only_matching': True,
 266     }, {
 267         # franceo
 268         'url': 'https://www.france.tv/france-o/archipels/132249-mon-ancetre-l-esclave.html',
 269         'only_matching': True,
 270     }, {
 271         # france2 live
 272         'url': 'https://www.france.tv/france-2/direct.html',
 273         'only_matching': True,
 274     }, {
 275         'url': 'https://www.france.tv/documentaires/histoire/136517-argentine-les-500-bebes-voles-de-la-dictature.html',
 276         'only_matching': True,
 277     }, {
 278         'url': 'https://www.france.tv/jeux-et-divertissements/divertissements/133965-le-web-contre-attaque.html',
 279         'only_matching': True,
 280     }, {
 281         'url': 'https://mobile.france.tv/france-5/c-dans-l-air/137347-emission-du-vendredi-12-mai-2017.html',
 282         'only_matching': True,
 283     }, {
 284         'url': 'https://www.france.tv/142749-rouge-sang.html',
 285         'only_matching': True,
 286     }, {
 287         # france-3 live
 288         'url': 'https://www.france.tv/france-3/direct.html',
 289         'only_matching': True,
 290     }]
 291
 292     def _real_extract(self, url):
 293         display_id = self._match_id(url)
 294
 295         webpage = self._download_webpage(url, display_id)
 296
 297         catalogue = None
 298         video_id = self._search_regex(
 299             r'(?:data-main-video\s*=|videoId["\']?\s*[:=])\s*(["\'])(?P<id>(?:(?!\1).)+)\1',
 300             webpage, 'video id', default=None, group='id')
 301
 302         if not video_id:
 303             video_id, catalogue = self._html_search_regex(
 304                 r'(?:href=|player\.setVideo\(\s*)"http://videos?\.francetv\.fr/video/([^@]+@[^"]+)"',
 305                 webpage, 'video ID').split('@')
 306
 307         return self._make_url_result(video_id, catalogue)
 308
 309
 310 class FranceTVEmbedIE(FranceTVBaseInfoExtractor):
 311     _VALID_URL = r'https?://embed\.francetv\.fr/*\?.*?\bue=(?P<id>[^&]+)'
 312
 313     _TESTS = [{
 314         'url': 'http://embed.francetv.fr/?ue=7fd581a2ccf59d2fc5719c5c13cf6961',
 315         'info_dict': {
 316             'id': 'NI_983319',
 317             'ext': 'mp4',
 318             'title': 'Le Pen Reims',
 319             'upload_date': '20170505',
 320             'timestamp': 1493981780,
 321             'duration': 16,
 322         },
 323         'params': {
 324             'skip_download': True,
 325         },
 326         'add_ie': [FranceTVIE.ie_key()],
 327     }]
 328
 329     def _real_extract(self, url):
 330         video_id = self._match_id(url)
 331
 332         video = self._download_json(
 333             'http://api-embed.webservices.francetelevisions.fr/key/%s' % video_id,
 334             video_id)
 335
 336         return self._make_url_result(video['video_id'], video.get('catalog'))
 337
 338
 339 class FranceTVInfoIE(FranceTVBaseInfoExtractor):
 340     IE_NAME = 'francetvinfo.fr'
 341     _VALID_URL = r'https?://(?:www|mobile|france3-regions)\.francetvinfo\.fr/(?:[^/]+/)*(?P<id>[^/?#&.]+)'
 342
 343     _TESTS = [{
 344         'url': 'https://www.francetvinfo.fr/replay-jt/france-3/soir-3/jt-grand-soir-3-jeudi-22-aout-2019_3561461.html',
 345         'info_dict': {
 346             'id': 'd12458ee-5062-48fe-bfdd-a30d6a01b793',
 347             'ext': 'mp4',
 348             'title': 'Soir 3',
 349             'upload_date': '20190822',
 350             'timestamp': 1566510900,
 351             'description': 'md5:72d167097237701d6e8452ff03b83c00',
 352             'subtitles': {
 353                 'fr': 'mincount:2',
 354             },
 355         },
 356         'params': {
 357             'skip_download': True,
 358         },
 359         'add_ie': [FranceTVIE.ie_key()],
 360     }, {
 361         'note': 'Only an image exists in initial webpage instead of the video',
 362         'url': 'https://www.francetvinfo.fr/sante/maladie/coronavirus/covid-19-en-inde-une-situation-catastrophique-a-new-dehli_4381095.html',
 363         'info_dict': {
 364             'id': '7d204c9e-a2d3-11eb-9e4c-000d3a23d482',
 365             'ext': 'mp4',
 366             'title': 'Covid-19 : une situation catastrophique à New Dehli',
 367             'thumbnail': str,
 368             'duration': 76,
 369             'timestamp': 1619028518,
 370             'upload_date': '20210421',
 371         },
 372         'params': {
 373             'skip_download': True,
 374         },
 375         'add_ie': [FranceTVIE.ie_key()],
 376     }, {
 377         'url': 'http://www.francetvinfo.fr/elections/europeennes/direct-europeennes-regardez-le-debat-entre-les-candidats-a-la-presidence-de-la-commission_600639.html',
 378         'only_matching': True,
 379     }, {
 380         'url': 'http://www.francetvinfo.fr/economie/entreprises/les-entreprises-familiales-le-secret-de-la-reussite_933271.html',
 381         'only_matching': True,
 382     }, {
 383         'url': 'http://france3-regions.francetvinfo.fr/bretagne/cotes-d-armor/thalassa-echappee-breizh-ce-venredi-dans-les-cotes-d-armor-954961.html',
 384         'only_matching': True,
 385     }, {
 386         # Dailymotion embed
 387         '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',
 388         'md5': 'ee7f1828f25a648addc90cb2687b1f12',
 389         'info_dict': {
 390             'id': 'x4iiko0',
 391             'ext': 'mp4',
 392             'title': 'NDDL, référendum, Brexit : Cécile Duflot répond à Patrick Cohen',
 393             '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',
 394             'timestamp': 1467011958,
 395             'upload_date': '20160627',
 396             'uploader': 'France Inter',
 397             'uploader_id': 'x2q2ez',
 398         },
 399         'add_ie': ['Dailymotion'],
 400     }, {
 401         'url': 'http://france3-regions.francetvinfo.fr/limousin/emissions/jt-1213-limousin',
 402         'only_matching': True,
 403     }]
 404
 405     def _real_extract(self, url):
 406         display_id = self._match_id(url)
 407
 408         webpage = self._download_webpage(url, display_id)
 409
 410         dailymotion_urls = DailymotionIE._extract_urls(webpage)
 411         if dailymotion_urls:
 412             return self.playlist_result([
 413                 self.url_result(dailymotion_url, DailymotionIE.ie_key())
 414                 for dailymotion_url in dailymotion_urls])
 415
 416         video_id = self._search_regex(
 417             (r'player\.load[^;]+src:\s*["\']([^"\']+)',
 418              r'id-video=([^@]+@[^"]+)',
 419              r'<a[^>]+href="(?:https?:)?//videos\.francetv\.fr/video/([^@]+@[^"]+)"',
 420              r'data-id=["\']([\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})',
 421              r'<figure[^>]+id=["\']([\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})'),
 422             webpage, 'video id')
 423
 424         return self._make_url_result(video_id)
 425
 426
 427 class FranceTVInfoSportIE(FranceTVBaseInfoExtractor):
 428     IE_NAME = 'sport.francetvinfo.fr'
 429     _VALID_URL = r'https?://sport\.francetvinfo\.fr/(?:[^/]+/)*(?P<id>[^/?#&]+)'
 430     _TESTS = [{
 431         'url': 'https://sport.francetvinfo.fr/les-jeux-olympiques/retour-sur-les-meilleurs-moments-de-pyeongchang-2018',
 432         'info_dict': {
 433             'id': '6e49080e-3f45-11e8-b459-000d3a2439ea',
 434             'ext': 'mp4',
 435             'title': 'Retour sur les meilleurs moments de Pyeongchang 2018',
 436             'timestamp': 1523639962,
 437             'upload_date': '20180413',
 438         },
 439         'params': {
 440             'skip_download': True,
 441         },
 442         'add_ie': [FranceTVIE.ie_key()],
 443     }]
 444
 445     def _real_extract(self, url):
 446         display_id = self._match_id(url)
 447         webpage = self._download_webpage(url, display_id)
 448         video_id = self._search_regex(r'data-video="([^"]+)"', webpage, 'video_id')
 449         return self._make_url_result(video_id, 'Sport-web')
 450
 451
 452 class GenerationWhatIE(InfoExtractor):
 453     IE_NAME = 'france2.fr:generation-what'
 454     _VALID_URL = r'https?://generation-what\.francetv\.fr/[^/]+/video/(?P<id>[^/?#&]+)'
 455
 456     _TESTS = [{
 457         'url': 'http://generation-what.francetv.fr/portrait/video/present-arms',
 458         'info_dict': {
 459             'id': 'wtvKYUG45iw',
 460             'ext': 'mp4',
 461             'title': 'Generation What - Garde à vous - FRA',
 462             'uploader': 'Generation What',
 463             'uploader_id': 'UCHH9p1eetWCgt4kXBYCb3_w',
 464             'upload_date': '20160411',
 465         },
 466         'params': {
 467             'skip_download': True,
 468         },
 469         'add_ie': ['Youtube'],
 470     }, {
 471         'url': 'http://generation-what.francetv.fr/europe/video/present-arms',
 472         'only_matching': True,
 473     }]
 474
 475     def _real_extract(self, url):
 476         display_id = self._match_id(url)
 477
 478         webpage = self._download_webpage(url, display_id)
 479
 480         youtube_id = self._search_regex(
 481             r"window\.videoURL\s*=\s*'([0-9A-Za-z_-]{11})';",
 482             webpage, 'youtube id')
 483
 484         return self.url_result(youtube_id, ie='Youtube', video_id=youtube_id)
 485
 486
 487 class CultureboxIE(FranceTVBaseInfoExtractor):
 488     _VALID_URL = r'https?://(?:m\.)?culturebox\.francetvinfo\.fr/(?:[^/]+/)*(?P<id>[^/?#&]+)'
 489
 490     _TESTS = [{
 491         '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',
 492         'info_dict': {
 493             'id': 'EV_134885',
 494             'ext': 'mp4',
 495             'title': 'Cantates BWV 4, 106 et 131 de Bach par Raphaël Pichon 5/7',
 496             'description': 'md5:19c44af004b88219f4daa50fa9a351d4',
 497             'upload_date': '20180206',
 498             'timestamp': 1517945220,
 499             'duration': 5981,
 500         },
 501         'params': {
 502             'skip_download': True,
 503         },
 504         'add_ie': [FranceTVIE.ie_key()],
 505     }]
 506
 507     def _real_extract(self, url):
 508         display_id = self._match_id(url)
 509
 510         webpage = self._download_webpage(url, display_id)
 511
 512         if ">Ce live n'est plus disponible en replay<" in webpage:
 513             raise ExtractorError(
 514                 'Video %s is not available' % display_id, expected=True)
 515
 516         video_id, catalogue = self._search_regex(
 517             r'["\'>]https?://videos\.francetv\.fr/video/([^@]+@.+?)["\'<]',
 518             webpage, 'video id').split('@')
 519
 520         return self._make_url_result(video_id, catalogue)
 521
 522
 523 class FranceTVJeunesseIE(FranceTVBaseInfoExtractor):
 524     _VALID_URL = r'(?P<url>https?://(?:www\.)?(?:zouzous|ludo)\.fr/heros/(?P<id>[^/?#&]+))'
 525
 526     _TESTS = [{
 527         'url': 'https://www.zouzous.fr/heros/simon',
 528         'info_dict': {
 529             'id': 'simon',
 530         },
 531         'playlist_count': 9,
 532     }, {
 533         'url': 'https://www.ludo.fr/heros/ninjago',
 534         'info_dict': {
 535             'id': 'ninjago',
 536         },
 537         'playlist_count': 10,
 538     }, {
 539         'url': 'https://www.zouzous.fr/heros/simon?abc',
 540         'only_matching': True,
 541     }]
 542
 543     def _real_extract(self, url):
 544         mobj = re.match(self._VALID_URL, url)
 545         playlist_id = mobj.group('id')
 546
 547         playlist = self._download_json(
 548             '%s/%s' % (mobj.group('url'), 'playlist'), playlist_id)
 549
 550         if not playlist.get('count'):
 551             raise ExtractorError(
 552                 '%s is not available' % playlist_id, expected=True)
 553
 554         entries = []
 555         for item in playlist['items']:
 556             identity = item.get('identity')
 557             if identity and isinstance(identity, compat_str):
 558                 entries.append(self._make_url_result(identity))
 559
 560         return self.playlist_result(entries, playlist_id)