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