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         subtitles = {}
 155         for video in videos:
 156             video_url = video.get('url')
 157             if not video_url:
 158                 continue
 159             if is_live is None:
 160                 is_live = (try_get(
 161                     video, lambda x: x['plages_ouverture'][0]['direct'], bool) is True
 162                     or video.get('is_live') is True
 163                     or '/live.francetv.fr/' in video_url)
 164             format_id = video.get('format')
 165             ext = determine_ext(video_url)
 166             if ext == 'f4m':
 167                 if georestricted:
 168                     # See https://github.com/ytdl-org/youtube-dl/issues/3963
 169                     # m3u8 urls work fine
 170                     continue
 171                 formats.extend(self._extract_f4m_formats(
 172                     sign(video_url, format_id) + '&hdcore=3.7.0&plugin=aasp-3.7.0.39.44',
 173                     video_id, f4m_id=format_id, fatal=False))
 174             elif ext == 'm3u8':
 175                 m3u8_fmts, m3u8_subs = self._extract_m3u8_formats_and_subtitles(
 176                     sign(video_url, format_id), video_id, 'mp4',
 177                     entry_protocol='m3u8_native', m3u8_id=format_id,
 178                     fatal=False)
 179                 formats.extend(m3u8_fmts)
 180                 subtitles = self._merge_subtitles(subtitles, m3u8_subs)
 181             elif ext == 'mpd':
 182                 formats.extend(self._extract_mpd_formats(
 183                     sign(video_url, format_id), video_id, mpd_id=format_id, fatal=False))
 184             elif video_url.startswith('rtmp'):
 185                 formats.append({
 186                     'url': video_url,
 187                     'format_id': 'rtmp-%s' % format_id,
 188                     'ext': 'flv',
 189                 })
 190             else:
 191                 if self._is_valid_url(video_url, video_id, format_id):
 192                     formats.append({
 193                         'url': video_url,
 194                         'format_id': format_id,
 195                     })
 196
 197         self._sort_formats(formats)
 198
 199         title = info['titre']
 200         subtitle = info.get('sous_titre')
 201         if subtitle:
 202             title += ' - %s' % subtitle
 203         title = title.strip()
 204
 205         subtitles.setdefault('fr', []).extend(
 206             [{
 207                 'url': subformat['url'],
 208                 'ext': subformat.get('format'),
 209             } for subformat in info.get('subtitles', []) if subformat.get('url')]
 210         )
 211
 212         return {
 213             'id': video_id,
 214             'title': self._live_title(title) if is_live else title,
 215             'description': clean_html(info.get('synopsis')),
 216             'thumbnail': urljoin('https://sivideo.webservices.francetelevisions.fr', info.get('image')),
 217             'duration': int_or_none(info.get('real_duration')) or parse_duration(info.get('duree')),
 218             'timestamp': int_or_none(try_get(info, lambda x: x['diffusion']['timestamp'])),
 219             'is_live': is_live,
 220             'formats': formats,
 221             'subtitles': subtitles,
 222         }
 223
 224     def _real_extract(self, url):
 225         mobj = re.match(self._VALID_URL, url)
 226         video_id = mobj.group('id')
 227         catalog = mobj.group('catalog')
 228
 229         if not video_id:
 230             qs = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
 231             video_id = qs.get('idDiffusion', [None])[0]
 232             catalog = qs.get('catalogue', [None])[0]
 233             if not video_id:
 234                 raise ExtractorError('Invalid URL', expected=True)
 235
 236         return self._extract_video(video_id, catalog)
 237
 238
 239 class FranceTVSiteIE(FranceTVBaseInfoExtractor):
 240     _VALID_URL = r'https?://(?:(?:www\.)?france\.tv|mobile\.france\.tv)/(?:[^/]+/)*(?P<id>[^/]+)\.html'
 241
 242     _TESTS = [{
 243         'url': 'https://www.france.tv/france-2/13h15-le-dimanche/140921-les-mysteres-de-jesus.html',
 244         'info_dict': {
 245             'id': 'ec217ecc-0733-48cf-ac06-af1347b849d1',
 246             'ext': 'mp4',
 247             'title': '13h15, le dimanche... - Les mystères de Jésus',
 248             'description': 'md5:75efe8d4c0a8205e5904498ffe1e1a42',
 249             'timestamp': 1502623500,
 250             'upload_date': '20170813',
 251         },
 252         'params': {
 253             'skip_download': True,
 254         },
 255         'add_ie': [FranceTVIE.ie_key()],
 256     }, {
 257         # france3
 258         'url': 'https://www.france.tv/france-3/des-chiffres-et-des-lettres/139063-emission-du-mardi-9-mai-2017.html',
 259         'only_matching': True,
 260     }, {
 261         # france4
 262         'url': 'https://www.france.tv/france-4/hero-corp/saison-1/134151-apres-le-calme.html',
 263         'only_matching': True,
 264     }, {
 265         # france5
 266         'url': 'https://www.france.tv/france-5/c-a-dire/saison-10/137013-c-a-dire.html',
 267         'only_matching': True,
 268     }, {
 269         # franceo
 270         'url': 'https://www.france.tv/france-o/archipels/132249-mon-ancetre-l-esclave.html',
 271         'only_matching': True,
 272     }, {
 273         # france2 live
 274         'url': 'https://www.france.tv/france-2/direct.html',
 275         'only_matching': True,
 276     }, {
 277         'url': 'https://www.france.tv/documentaires/histoire/136517-argentine-les-500-bebes-voles-de-la-dictature.html',
 278         'only_matching': True,
 279     }, {
 280         'url': 'https://www.france.tv/jeux-et-divertissements/divertissements/133965-le-web-contre-attaque.html',
 281         'only_matching': True,
 282     }, {
 283         'url': 'https://mobile.france.tv/france-5/c-dans-l-air/137347-emission-du-vendredi-12-mai-2017.html',
 284         'only_matching': True,
 285     }, {
 286         'url': 'https://www.france.tv/142749-rouge-sang.html',
 287         'only_matching': True,
 288     }, {
 289         # france-3 live
 290         'url': 'https://www.france.tv/france-3/direct.html',
 291         'only_matching': True,
 292     }]
 293
 294     def _real_extract(self, url):
 295         display_id = self._match_id(url)
 296
 297         webpage = self._download_webpage(url, display_id)
 298
 299         catalogue = None
 300         video_id = self._search_regex(
 301             r'(?:data-main-video\s*=|videoId["\']?\s*[:=])\s*(["\'])(?P<id>(?:(?!\1).)+)\1',
 302             webpage, 'video id', default=None, group='id')
 303
 304         if not video_id:
 305             video_id, catalogue = self._html_search_regex(
 306                 r'(?:href=|player\.setVideo\(\s*)"http://videos?\.francetv\.fr/video/([^@]+@[^"]+)"',
 307                 webpage, 'video ID').split('@')
 308
 309         return self._make_url_result(video_id, catalogue)
 310
 311
 312 class FranceTVEmbedIE(FranceTVBaseInfoExtractor):
 313     _VALID_URL = r'https?://embed\.francetv\.fr/*\?.*?\bue=(?P<id>[^&]+)'
 314
 315     _TESTS = [{
 316         'url': 'http://embed.francetv.fr/?ue=7fd581a2ccf59d2fc5719c5c13cf6961',
 317         'info_dict': {
 318             'id': 'NI_983319',
 319             'ext': 'mp4',
 320             'title': 'Le Pen Reims',
 321             'upload_date': '20170505',
 322             'timestamp': 1493981780,
 323             'duration': 16,
 324         },
 325         'params': {
 326             'skip_download': True,
 327         },
 328         'add_ie': [FranceTVIE.ie_key()],
 329     }]
 330
 331     def _real_extract(self, url):
 332         video_id = self._match_id(url)
 333
 334         video = self._download_json(
 335             'http://api-embed.webservices.francetelevisions.fr/key/%s' % video_id,
 336             video_id)
 337
 338         return self._make_url_result(video['video_id'], video.get('catalog'))
 339
 340
 341 class FranceTVInfoIE(FranceTVBaseInfoExtractor):
 342     IE_NAME = 'francetvinfo.fr'
 343     _VALID_URL = r'https?://(?:www|mobile|france3-regions)\.francetvinfo\.fr/(?:[^/]+/)*(?P<id>[^/?#&.]+)'
 344
 345     _TESTS = [{
 346         'url': 'https://www.francetvinfo.fr/replay-jt/france-3/soir-3/jt-grand-soir-3-jeudi-22-aout-2019_3561461.html',
 347         'info_dict': {
 348             'id': 'd12458ee-5062-48fe-bfdd-a30d6a01b793',
 349             'ext': 'mp4',
 350             'title': 'Soir 3',
 351             'upload_date': '20190822',
 352             'timestamp': 1566510900,
 353             'description': 'md5:72d167097237701d6e8452ff03b83c00',
 354             'subtitles': {
 355                 'fr': 'mincount:2',
 356             },
 357         },
 358         'params': {
 359             'skip_download': True,
 360         },
 361         'add_ie': [FranceTVIE.ie_key()],
 362     }, {
 363         'note': 'Only an image exists in initial webpage instead of the video',
 364         'url': 'https://www.francetvinfo.fr/sante/maladie/coronavirus/covid-19-en-inde-une-situation-catastrophique-a-new-dehli_4381095.html',
 365         'info_dict': {
 366             'id': '7d204c9e-a2d3-11eb-9e4c-000d3a23d482',
 367             'ext': 'mp4',
 368             'title': 'Covid-19 : une situation catastrophique à New Dehli',
 369             'thumbnail': str,
 370             'duration': 76,
 371             'timestamp': 1619028518,
 372             'upload_date': '20210421',
 373         },
 374         'params': {
 375             'skip_download': True,
 376         },
 377         'add_ie': [FranceTVIE.ie_key()],
 378     }, {
 379         'url': 'http://www.francetvinfo.fr/elections/europeennes/direct-europeennes-regardez-le-debat-entre-les-candidats-a-la-presidence-de-la-commission_600639.html',
 380         'only_matching': True,
 381     }, {
 382         'url': 'http://www.francetvinfo.fr/economie/entreprises/les-entreprises-familiales-le-secret-de-la-reussite_933271.html',
 383         'only_matching': True,
 384     }, {
 385         'url': 'http://france3-regions.francetvinfo.fr/bretagne/cotes-d-armor/thalassa-echappee-breizh-ce-venredi-dans-les-cotes-d-armor-954961.html',
 386         'only_matching': True,
 387     }, {
 388         # Dailymotion embed
 389         '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',
 390         'md5': 'ee7f1828f25a648addc90cb2687b1f12',
 391         'info_dict': {
 392             'id': 'x4iiko0',
 393             'ext': 'mp4',
 394             'title': 'NDDL, référendum, Brexit : Cécile Duflot répond à Patrick Cohen',
 395             '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',
 396             'timestamp': 1467011958,
 397             'upload_date': '20160627',
 398             'uploader': 'France Inter',
 399             'uploader_id': 'x2q2ez',
 400         },
 401         'add_ie': ['Dailymotion'],
 402     }, {
 403         'url': 'http://france3-regions.francetvinfo.fr/limousin/emissions/jt-1213-limousin',
 404         'only_matching': True,
 405     }]
 406
 407     def _real_extract(self, url):
 408         display_id = self._match_id(url)
 409
 410         webpage = self._download_webpage(url, display_id)
 411
 412         dailymotion_urls = DailymotionIE._extract_urls(webpage)
 413         if dailymotion_urls:
 414             return self.playlist_result([
 415                 self.url_result(dailymotion_url, DailymotionIE.ie_key())
 416                 for dailymotion_url in dailymotion_urls])
 417
 418         video_id = self._search_regex(
 419             (r'player\.load[^;]+src:\s*["\']([^"\']+)',
 420              r'id-video=([^@]+@[^"]+)',
 421              r'<a[^>]+href="(?:https?:)?//videos\.francetv\.fr/video/([^@]+@[^"]+)"',
 422              r'data-id=["\']([\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})',
 423              r'<figure[^>]+id=["\']([\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})'),
 424             webpage, 'video id')
 425
 426         return self._make_url_result(video_id)
 427
 428
 429 class FranceTVInfoSportIE(FranceTVBaseInfoExtractor):
 430     IE_NAME = 'sport.francetvinfo.fr'
 431     _VALID_URL = r'https?://sport\.francetvinfo\.fr/(?:[^/]+/)*(?P<id>[^/?#&]+)'
 432     _TESTS = [{
 433         'url': 'https://sport.francetvinfo.fr/les-jeux-olympiques/retour-sur-les-meilleurs-moments-de-pyeongchang-2018',
 434         'info_dict': {
 435             'id': '6e49080e-3f45-11e8-b459-000d3a2439ea',
 436             'ext': 'mp4',
 437             'title': 'Retour sur les meilleurs moments de Pyeongchang 2018',
 438             'timestamp': 1523639962,
 439             'upload_date': '20180413',
 440         },
 441         'params': {
 442             'skip_download': True,
 443         },
 444         'add_ie': [FranceTVIE.ie_key()],
 445     }]
 446
 447     def _real_extract(self, url):
 448         display_id = self._match_id(url)
 449         webpage = self._download_webpage(url, display_id)
 450         video_id = self._search_regex(r'data-video="([^"]+)"', webpage, 'video_id')
 451         return self._make_url_result(video_id, 'Sport-web')
 452
 453
 454 class GenerationWhatIE(InfoExtractor):
 455     IE_NAME = 'france2.fr:generation-what'
 456     _VALID_URL = r'https?://generation-what\.francetv\.fr/[^/]+/video/(?P<id>[^/?#&]+)'
 457
 458     _TESTS = [{
 459         'url': 'http://generation-what.francetv.fr/portrait/video/present-arms',
 460         'info_dict': {
 461             'id': 'wtvKYUG45iw',
 462             'ext': 'mp4',
 463             'title': 'Generation What - Garde à vous - FRA',
 464             'uploader': 'Generation What',
 465             'uploader_id': 'UCHH9p1eetWCgt4kXBYCb3_w',
 466             'upload_date': '20160411',
 467         },
 468         'params': {
 469             'skip_download': True,
 470         },
 471         'add_ie': ['Youtube'],
 472     }, {
 473         'url': 'http://generation-what.francetv.fr/europe/video/present-arms',
 474         'only_matching': True,
 475     }]
 476
 477     def _real_extract(self, url):
 478         display_id = self._match_id(url)
 479
 480         webpage = self._download_webpage(url, display_id)
 481
 482         youtube_id = self._search_regex(
 483             r"window\.videoURL\s*=\s*'([0-9A-Za-z_-]{11})';",
 484             webpage, 'youtube id')
 485
 486         return self.url_result(youtube_id, ie='Youtube', video_id=youtube_id)
 487
 488
 489 class CultureboxIE(FranceTVBaseInfoExtractor):
 490     _VALID_URL = r'https?://(?:m\.)?culturebox\.francetvinfo\.fr/(?:[^/]+/)*(?P<id>[^/?#&]+)'
 491
 492     _TESTS = [{
 493         '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',
 494         'info_dict': {
 495             'id': 'EV_134885',
 496             'ext': 'mp4',
 497             'title': 'Cantates BWV 4, 106 et 131 de Bach par Raphaël Pichon 5/7',
 498             'description': 'md5:19c44af004b88219f4daa50fa9a351d4',
 499             'upload_date': '20180206',
 500             'timestamp': 1517945220,
 501             'duration': 5981,
 502         },
 503         'params': {
 504             'skip_download': True,
 505         },
 506         'add_ie': [FranceTVIE.ie_key()],
 507     }]
 508
 509     def _real_extract(self, url):
 510         display_id = self._match_id(url)
 511
 512         webpage = self._download_webpage(url, display_id)
 513
 514         if ">Ce live n'est plus disponible en replay<" in webpage:
 515             raise ExtractorError(
 516                 'Video %s is not available' % display_id, expected=True)
 517
 518         video_id, catalogue = self._search_regex(
 519             r'["\'>]https?://videos\.francetv\.fr/video/([^@]+@.+?)["\'<]',
 520             webpage, 'video id').split('@')
 521
 522         return self._make_url_result(video_id, catalogue)
 523
 524
 525 class FranceTVJeunesseIE(FranceTVBaseInfoExtractor):
 526     _VALID_URL = r'(?P<url>https?://(?:www\.)?(?:zouzous|ludo)\.fr/heros/(?P<id>[^/?#&]+))'
 527
 528     _TESTS = [{
 529         'url': 'https://www.zouzous.fr/heros/simon',
 530         'info_dict': {
 531             'id': 'simon',
 532         },
 533         'playlist_count': 9,
 534     }, {
 535         'url': 'https://www.ludo.fr/heros/ninjago',
 536         'info_dict': {
 537             'id': 'ninjago',
 538         },
 539         'playlist_count': 10,
 540     }, {
 541         'url': 'https://www.zouzous.fr/heros/simon?abc',
 542         'only_matching': True,
 543     }]
 544
 545     def _real_extract(self, url):
 546         mobj = re.match(self._VALID_URL, url)
 547         playlist_id = mobj.group('id')
 548
 549         playlist = self._download_json(
 550             '%s/%s' % (mobj.group('url'), 'playlist'), playlist_id)
 551
 552         if not playlist.get('count'):
 553             raise ExtractorError(
 554                 '%s is not available' % playlist_id, expected=True)
 555
 556         entries = []
 557         for item in playlist['items']:
 558             identity = item.get('identity')
 559             if identity and isinstance(identity, compat_str):
 560                 entries.append(self._make_url_result(identity))
 561
 562         return self.playlist_result(entries, playlist_id)