yt_dlp/extractor/francetv.py

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