yt_dlp/extractor/francetv.py

   1 from .common import InfoExtractor
   2 from .dailymotion import DailymotionIE
   3 from ..utils import (
   4     ExtractorError,
   5     determine_ext,
   6     format_field,
   7     int_or_none,
   8     join_nonempty,
   9     parse_iso8601,
  10     parse_qs,
  11 )
  12
  13
  14 class FranceTVBaseInfoExtractor(InfoExtractor):
  15     def _make_url_result(self, video_or_full_id, catalog=None):
  16         full_id = 'francetv:%s' % video_or_full_id
  17         if '@' not in video_or_full_id and catalog:
  18             full_id += '@%s' % catalog
  19         return self.url_result(
  20             full_id, ie=FranceTVIE.ie_key(),
  21             video_id=video_or_full_id.split('@')[0])
  22
  23
  24 class FranceTVIE(InfoExtractor):
  25     _VALID_URL = r'''(?x)
  26                     (?:
  27                         https?://
  28                             sivideo\.webservices\.francetelevisions\.fr/tools/getInfosOeuvre/v2/\?
  29                             .*?\bidDiffusion=[^&]+|
  30                         (?:
  31                             https?://videos\.francetv\.fr/video/|
  32                             francetv:
  33                         )
  34                         (?P<id>[^@]+)(?:@(?P<catalog>.+))?
  35                     )
  36                     '''
  37     _EMBED_REGEX = [r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?://)?embed\.francetv\.fr/\?ue=.+?)\1']
  38
  39     _TESTS = [{
  40         # without catalog
  41         'url': 'https://sivideo.webservices.francetelevisions.fr/tools/getInfosOeuvre/v2/?idDiffusion=162311093&callback=_jsonp_loader_callback_request_0',
  42         'md5': 'c2248a8de38c4e65ea8fae7b5df2d84f',
  43         'info_dict': {
  44             'id': '162311093',
  45             'ext': 'mp4',
  46             'title': '13h15, le dimanche... - Les mystères de Jésus',
  47             'description': 'md5:75efe8d4c0a8205e5904498ffe1e1a42',
  48             'timestamp': 1502623500,
  49             'upload_date': '20170813',
  50         },
  51     }, {
  52         # with catalog
  53         'url': 'https://sivideo.webservices.francetelevisions.fr/tools/getInfosOeuvre/v2/?idDiffusion=NI_1004933&catalogue=Zouzous&callback=_jsonp_loader_callback_request_4',
  54         'only_matching': True,
  55     }, {
  56         'url': 'http://videos.francetv.fr/video/NI_657393@Regions',
  57         'only_matching': True,
  58     }, {
  59         'url': 'francetv:162311093',
  60         'only_matching': True,
  61     }, {
  62         'url': 'francetv:NI_1004933@Zouzous',
  63         'only_matching': True,
  64     }, {
  65         'url': 'francetv:NI_983319@Info-web',
  66         'only_matching': True,
  67     }, {
  68         'url': 'francetv:NI_983319',
  69         'only_matching': True,
  70     }, {
  71         'url': 'francetv:NI_657393@Regions',
  72         'only_matching': True,
  73     }, {
  74         # france-3 live
  75         'url': 'francetv:SIM_France3',
  76         'only_matching': True,
  77     }]
  78
  79     def _extract_video(self, video_id, catalogue=None):
  80         # Videos are identified by idDiffusion so catalogue part is optional.
  81         # However when provided, some extra formats may be returned so we pass
  82         # it if available.
  83         is_live = None
  84         videos = []
  85         title = None
  86         subtitle = None
  87         episode_number = None
  88         season_number = None
  89         image = None
  90         duration = None
  91         timestamp = None
  92         spritesheets = None
  93
  94         for device_type in ('desktop', 'mobile'):
  95             dinfo = self._download_json(
  96                 'https://player.webservices.francetelevisions.fr/v1/videos/%s' % video_id,
  97                 video_id, 'Downloading %s video JSON' % device_type, query={
  98                     'device_type': device_type,
  99                     'browser': 'chrome',
 100                 }, fatal=False)
 101
 102             if not dinfo:
 103                 continue
 104
 105             video = dinfo.get('video')
 106             if video:
 107                 videos.append(video)
 108                 if duration is None:
 109                     duration = video.get('duration')
 110                 if is_live is None:
 111                     is_live = video.get('is_live')
 112                 if spritesheets is None:
 113                     spritesheets = video.get('spritesheets')
 114
 115             meta = dinfo.get('meta')
 116             if meta:
 117                 if title is None:
 118                     title = meta.get('title')
 119                 # meta['pre_title'] contains season and episode number for series in format "S<ID> E<ID>"
 120                 season_number, episode_number = self._search_regex(
 121                     r'S(\d+)\s*E(\d+)', meta.get('pre_title'), 'episode info', group=(1, 2), default=(None, None))
 122                 if subtitle is None:
 123                     subtitle = meta.get('additional_title')
 124                 if image is None:
 125                     image = meta.get('image_url')
 126                 if timestamp is None:
 127                     timestamp = parse_iso8601(meta.get('broadcasted_at'))
 128
 129         formats = []
 130         subtitles = {}
 131         for video in videos:
 132             format_id = video.get('format')
 133
 134             video_url = None
 135             if video.get('workflow') == 'token-akamai':
 136                 token_url = video.get('token')
 137                 if token_url:
 138                     token_json = self._download_json(
 139                         token_url, video_id,
 140                         'Downloading signed %s manifest URL' % format_id)
 141                     if token_json:
 142                         video_url = token_json.get('url')
 143             if not video_url:
 144                 video_url = video.get('url')
 145
 146             ext = determine_ext(video_url)
 147             if ext == 'f4m':
 148                 formats.extend(self._extract_f4m_formats(
 149                     video_url, video_id, f4m_id=format_id, fatal=False))
 150             elif ext == 'm3u8':
 151                 fmts, subs = self._extract_m3u8_formats_and_subtitles(
 152                     video_url, video_id, 'mp4',
 153                     entry_protocol='m3u8_native', m3u8_id=format_id,
 154                     fatal=False)
 155                 formats.extend(fmts)
 156                 self._merge_subtitles(subs, target=subtitles)
 157             elif ext == 'mpd':
 158                 fmts, subs = self._extract_mpd_formats_and_subtitles(
 159                     video_url, video_id, mpd_id=format_id, fatal=False)
 160                 formats.extend(fmts)
 161                 self._merge_subtitles(subs, target=subtitles)
 162             elif video_url.startswith('rtmp'):
 163                 formats.append({
 164                     'url': video_url,
 165                     'format_id': 'rtmp-%s' % format_id,
 166                     'ext': 'flv',
 167                 })
 168             else:
 169                 if self._is_valid_url(video_url, video_id, format_id):
 170                     formats.append({
 171                         'url': video_url,
 172                         'format_id': format_id,
 173                     })
 174
 175             # XXX: what is video['captions']?
 176
 177         for f in formats:
 178             if f.get('acodec') != 'none' and f.get('language') in ('qtz', 'qad'):
 179                 f['language_preference'] = -10
 180                 f['format_note'] = 'audio description%s' % format_field(f, 'format_note', ', %s')
 181
 182         if spritesheets:
 183             formats.append({
 184                 'format_id': 'spritesheets',
 185                 'format_note': 'storyboard',
 186                 'acodec': 'none',
 187                 'vcodec': 'none',
 188                 'ext': 'mhtml',
 189                 'protocol': 'mhtml',
 190                 'url': 'about:invalid',
 191                 'fragments': [{
 192                     'url': sheet,
 193                     # XXX: not entirely accurate; each spritesheet seems to be
 194                     # a 10×10 grid of thumbnails corresponding to approximately
 195                     # 2 seconds of the video; the last spritesheet may be shorter
 196                     'duration': 200,
 197                 } for sheet in spritesheets]
 198             })
 199
 200         return {
 201             'id': video_id,
 202             'title': join_nonempty(title, subtitle, delim=' - ').strip(),
 203             'thumbnail': image,
 204             'duration': duration,
 205             'timestamp': timestamp,
 206             'is_live': is_live,
 207             'formats': formats,
 208             'subtitles': subtitles,
 209             'episode': subtitle if episode_number else None,
 210             'series': title if episode_number else None,
 211             'episode_number': int_or_none(episode_number),
 212             'season_number': int_or_none(season_number),
 213         }
 214
 215     def _real_extract(self, url):
 216         mobj = self._match_valid_url(url)
 217         video_id = mobj.group('id')
 218         catalog = mobj.group('catalog')
 219
 220         if not video_id:
 221             qs = parse_qs(url)
 222             video_id = qs.get('idDiffusion', [None])[0]
 223             catalog = qs.get('catalogue', [None])[0]
 224             if not video_id:
 225                 raise ExtractorError('Invalid URL', expected=True)
 226
 227         return self._extract_video(video_id, catalog)
 228
 229
 230 class FranceTVSiteIE(FranceTVBaseInfoExtractor):
 231     _VALID_URL = r'https?://(?:(?:www\.)?france\.tv|mobile\.france\.tv)/(?:[^/]+/)*(?P<id>[^/]+)\.html'
 232
 233     _TESTS = [{
 234         'url': 'https://www.france.tv/france-2/13h15-le-dimanche/140921-les-mysteres-de-jesus.html',
 235         'info_dict': {
 236             'id': 'ec217ecc-0733-48cf-ac06-af1347b849d1',
 237             'ext': 'mp4',
 238             'title': '13h15, le dimanche... - Les mystères de Jésus',
 239             'timestamp': 1502623500,
 240             'duration': 2580,
 241             'thumbnail': r're:^https?://.*\.jpg$',
 242             'upload_date': '20170813',
 243         },
 244         'params': {
 245             'skip_download': True,
 246         },
 247         'add_ie': [FranceTVIE.ie_key()],
 248     }, {
 249         'url': 'https://www.france.tv/enfants/six-huit-ans/foot2rue/saison-1/3066387-duel-au-vieux-port.html',
 250         'info_dict': {
 251             'id': 'a9050959-eedd-4b4a-9b0d-de6eeaa73e44',
 252             'ext': 'mp4',
 253             'title': 'Foot2Rue - Duel au vieux port',
 254             'episode': 'Duel au vieux port',
 255             'series': 'Foot2Rue',
 256             'episode_number': 1,
 257             'season_number': 1,
 258             'timestamp': 1642761360,
 259             'upload_date': '20220121',
 260             'season': 'Season 1',
 261             'thumbnail': r're:^https?://.*\.jpg$',
 262             'duration': 1441,
 263         },
 264     }, {
 265         # france3
 266         'url': 'https://www.france.tv/france-3/des-chiffres-et-des-lettres/139063-emission-du-mardi-9-mai-2017.html',
 267         'only_matching': True,
 268     }, {
 269         # france4
 270         'url': 'https://www.france.tv/france-4/hero-corp/saison-1/134151-apres-le-calme.html',
 271         'only_matching': True,
 272     }, {
 273         # france5
 274         'url': 'https://www.france.tv/france-5/c-a-dire/saison-10/137013-c-a-dire.html',
 275         'only_matching': True,
 276     }, {
 277         # franceo
 278         'url': 'https://www.france.tv/france-o/archipels/132249-mon-ancetre-l-esclave.html',
 279         'only_matching': True,
 280     }, {
 281         # france2 live
 282         'url': 'https://www.france.tv/france-2/direct.html',
 283         'only_matching': True,
 284     }, {
 285         'url': 'https://www.france.tv/documentaires/histoire/136517-argentine-les-500-bebes-voles-de-la-dictature.html',
 286         'only_matching': True,
 287     }, {
 288         'url': 'https://www.france.tv/jeux-et-divertissements/divertissements/133965-le-web-contre-attaque.html',
 289         'only_matching': True,
 290     }, {
 291         'url': 'https://mobile.france.tv/france-5/c-dans-l-air/137347-emission-du-vendredi-12-mai-2017.html',
 292         'only_matching': True,
 293     }, {
 294         'url': 'https://www.france.tv/142749-rouge-sang.html',
 295         'only_matching': True,
 296     }, {
 297         # france-3 live
 298         'url': 'https://www.france.tv/france-3/direct.html',
 299         'only_matching': True,
 300     }]
 301
 302     def _real_extract(self, url):
 303         display_id = self._match_id(url)
 304
 305         webpage = self._download_webpage(url, display_id)
 306
 307         catalogue = None
 308         video_id = self._search_regex(
 309             r'(?:data-main-video\s*=|videoId["\']?\s*[:=])\s*(["\'])(?P<id>(?:(?!\1).)+)\1',
 310             webpage, 'video id', default=None, group='id')
 311
 312         if not video_id:
 313             video_id, catalogue = self._html_search_regex(
 314                 r'(?:href=|player\.setVideo\(\s*)"http://videos?\.francetv\.fr/video/([^@]+@[^"]+)"',
 315                 webpage, 'video ID').split('@')
 316
 317         return self._make_url_result(video_id, catalogue)
 318
 319
 320 class FranceTVInfoIE(FranceTVBaseInfoExtractor):
 321     IE_NAME = 'francetvinfo.fr'
 322     _VALID_URL = r'https?://(?:www|mobile|france3-regions)\.francetvinfo\.fr/(?:[^/]+/)*(?P<id>[^/?#&.]+)'
 323
 324     _TESTS = [{
 325         'url': 'https://www.francetvinfo.fr/replay-jt/france-3/soir-3/jt-grand-soir-3-jeudi-22-aout-2019_3561461.html',
 326         'info_dict': {
 327             'id': 'd12458ee-5062-48fe-bfdd-a30d6a01b793',
 328             'ext': 'mp4',
 329             'title': 'Soir 3',
 330             'upload_date': '20190822',
 331             'timestamp': 1566510900,
 332             'description': 'md5:72d167097237701d6e8452ff03b83c00',
 333             'subtitles': {
 334                 'fr': 'mincount:2',
 335             },
 336         },
 337         'params': {
 338             'skip_download': True,
 339         },
 340         'add_ie': [FranceTVIE.ie_key()],
 341     }, {
 342         'note': 'Only an image exists in initial webpage instead of the video',
 343         'url': 'https://www.francetvinfo.fr/sante/maladie/coronavirus/covid-19-en-inde-une-situation-catastrophique-a-new-dehli_4381095.html',
 344         'info_dict': {
 345             'id': '7d204c9e-a2d3-11eb-9e4c-000d3a23d482',
 346             'ext': 'mp4',
 347             'title': 'Covid-19 : une situation catastrophique à New Dehli',
 348             'thumbnail': str,
 349             'duration': 76,
 350             'timestamp': 1619028518,
 351             'upload_date': '20210421',
 352         },
 353         'params': {
 354             'skip_download': True,
 355         },
 356         'add_ie': [FranceTVIE.ie_key()],
 357     }, {
 358         'url': 'http://www.francetvinfo.fr/elections/europeennes/direct-europeennes-regardez-le-debat-entre-les-candidats-a-la-presidence-de-la-commission_600639.html',
 359         'only_matching': True,
 360     }, {
 361         'url': 'http://www.francetvinfo.fr/economie/entreprises/les-entreprises-familiales-le-secret-de-la-reussite_933271.html',
 362         'only_matching': True,
 363     }, {
 364         'url': 'http://france3-regions.francetvinfo.fr/bretagne/cotes-d-armor/thalassa-echappee-breizh-ce-venredi-dans-les-cotes-d-armor-954961.html',
 365         'only_matching': True,
 366     }, {
 367         # Dailymotion embed
 368         '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',
 369         'md5': 'ee7f1828f25a648addc90cb2687b1f12',
 370         'info_dict': {
 371             'id': 'x4iiko0',
 372             'ext': 'mp4',
 373             'title': 'NDDL, référendum, Brexit : Cécile Duflot répond à Patrick Cohen',
 374             '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',
 375             'timestamp': 1467011958,
 376             'upload_date': '20160627',
 377             'uploader': 'France Inter',
 378             'uploader_id': 'x2q2ez',
 379         },
 380         'add_ie': ['Dailymotion'],
 381     }, {
 382         'url': 'http://france3-regions.francetvinfo.fr/limousin/emissions/jt-1213-limousin',
 383         'only_matching': True,
 384     }, {
 385         # "<figure id=" pattern (#28792)
 386         'url': 'https://www.francetvinfo.fr/culture/patrimoine/incendie-de-notre-dame-de-paris/notre-dame-de-paris-de-l-incendie-de-la-cathedrale-a-sa-reconstruction_4372291.html',
 387         'only_matching': True,
 388     }]
 389
 390     def _real_extract(self, url):
 391         display_id = self._match_id(url)
 392
 393         webpage = self._download_webpage(url, display_id)
 394
 395         dailymotion_urls = tuple(DailymotionIE._extract_embed_urls(url, webpage))
 396         if dailymotion_urls:
 397             return self.playlist_result([
 398                 self.url_result(dailymotion_url, DailymotionIE.ie_key())
 399                 for dailymotion_url in dailymotion_urls])
 400
 401         video_id = self._search_regex(
 402             (r'player\.load[^;]+src:\s*["\']([^"\']+)',
 403              r'id-video=([^@]+@[^"]+)',
 404              r'<a[^>]+href="(?:https?:)?//videos\.francetv\.fr/video/([^@]+@[^"]+)"',
 405              r'(?:data-id|<figure[^<]+\bid)=["\']([\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})'),
 406             webpage, 'video id')
 407
 408         return self._make_url_result(video_id)