yt_dlp/extractor/francetv.py

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