Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/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 )
  20 from .dailymotion import DailymotionIE
  21
  22
  23 class FranceTVBaseInfoExtractor(InfoExtractor):
  24     def _make_url_result(self, video_or_full_id, catalog=None):
  25         full_id = 'francetv:%s' % video_or_full_id
  26         if '@' not in video_or_full_id and catalog:
  27             full_id += '@%s' % catalog
  28         return self.url_result(
  29             full_id, ie=FranceTVIE.ie_key(),
  30             video_id=video_or_full_id.split('@')[0])
  31
  32
  33 class FranceTVIE(InfoExtractor):
  34     _VALID_URL = r'''(?x)
  35                     (?:
  36                         https?://
  37                             sivideo\.webservices\.francetelevisions\.fr/tools/getInfosOeuvre/v2/\?
  38                             .*?\bidDiffusion=[^&]+|
  39                         (?:
  40                             https?://videos\.francetv\.fr/video/|
  41                             francetv:
  42                         )
  43                         (?P<id>[^@]+)(?:@(?P<catalog>.+))?
  44                     )
  45                     '''
  46
  47     _TESTS = [{
  48         # without catalog
  49         'url': 'https://sivideo.webservices.francetelevisions.fr/tools/getInfosOeuvre/v2/?idDiffusion=162311093&callback=_jsonp_loader_callback_request_0',
  50         'md5': 'c2248a8de38c4e65ea8fae7b5df2d84f',
  51         'info_dict': {
  52             'id': '162311093',
  53             'ext': 'mp4',
  54             'title': '13h15, le dimanche... - Les mystères de Jésus',
  55             'description': 'md5:75efe8d4c0a8205e5904498ffe1e1a42',
  56             'timestamp': 1502623500,
  57             'upload_date': '20170813',
  58         },
  59     }, {
  60         # with catalog
  61         'url': 'https://sivideo.webservices.francetelevisions.fr/tools/getInfosOeuvre/v2/?idDiffusion=NI_1004933&catalogue=Zouzous&callback=_jsonp_loader_callback_request_4',
  62         'only_matching': True,
  63     }, {
  64         'url': 'http://videos.francetv.fr/video/NI_657393@Regions',
  65         'only_matching': True,
  66     }, {
  67         'url': 'francetv:162311093',
  68         'only_matching': True,
  69     }, {
  70         'url': 'francetv:NI_1004933@Zouzous',
  71         'only_matching': True,
  72     }, {
  73         'url': 'francetv:NI_983319@Info-web',
  74         'only_matching': True,
  75     }, {
  76         'url': 'francetv:NI_983319',
  77         'only_matching': True,
  78     }, {
  79         'url': 'francetv:NI_657393@Regions',
  80         'only_matching': True,
  81     }, {
  82         # france-3 live
  83         'url': 'francetv:SIM_France3',
  84         'only_matching': True,
  85     }]
  86
  87     def _extract_video(self, video_id, catalogue=None):
  88         # Videos are identified by idDiffusion so catalogue part is optional.
  89         # However when provided, some extra formats may be returned so we pass
  90         # it if available.
  91         info = self._download_json(
  92             'https://sivideo.webservices.francetelevisions.fr/tools/getInfosOeuvre/v2/',
  93             video_id, 'Downloading video JSON', query={
  94                 'idDiffusion': video_id,
  95                 'catalogue': catalogue or '',
  96             })
  97
  98         if info.get('status') == 'NOK':
  99             raise ExtractorError(
 100                 '%s returned error: %s' % (self.IE_NAME, info['message']),
 101                 expected=True)
 102         allowed_countries = info['videos'][0].get('geoblocage')
 103         if allowed_countries:
 104             georestricted = True
 105             geo_info = self._download_json(
 106                 'http://geo.francetv.fr/ws/edgescape.json', video_id,
 107                 'Downloading geo restriction info')
 108             country = geo_info['reponse']['geo_info']['country_code']
 109             if country not in allowed_countries:
 110                 raise ExtractorError(
 111                     'The video is not available from your location',
 112                     expected=True)
 113         else:
 114             georestricted = False
 115
 116         def sign(manifest_url, manifest_id):
 117             for host in ('hdfauthftv-a.akamaihd.net', 'hdfauth.francetv.fr'):
 118                 signed_url = self._download_webpage(
 119                     'https://%s/esi/TA' % host, video_id,
 120                     'Downloading signed %s manifest URL' % manifest_id,
 121                     fatal=False, query={
 122                         'url': manifest_url,
 123                     })
 124                 if (signed_url and isinstance(signed_url, compat_str) and
 125                         re.search(r'^(?:https?:)?//', signed_url)):
 126                     return signed_url
 127             return manifest_url
 128
 129         is_live = None
 130
 131         formats = []
 132         for video in info['videos']:
 133             if video['statut'] != 'ONLINE':
 134                 continue
 135             video_url = video['url']
 136             if not video_url:
 137                 continue
 138             if is_live is None:
 139                 is_live = (try_get(
 140                     video, lambda x: x['plages_ouverture'][0]['direct'],
 141                     bool) is True) or '/live.francetv.fr/' in video_url
 142             format_id = video['format']
 143             ext = determine_ext(video_url)
 144             if ext == 'f4m':
 145                 if georestricted:
 146                     # See https://github.com/rg3/youtube-dl/issues/3963
 147                     # m3u8 urls work fine
 148                     continue
 149                 formats.extend(self._extract_f4m_formats(
 150                     sign(video_url, format_id) + '&hdcore=3.7.0&plugin=aasp-3.7.0.39.44',
 151                     video_id, f4m_id=format_id, fatal=False))
 152             elif ext == 'm3u8':
 153                 formats.extend(self._extract_m3u8_formats(
 154                     sign(video_url, format_id), video_id, 'mp4',
 155                     entry_protocol='m3u8_native', m3u8_id=format_id,
 156                     fatal=False))
 157             elif video_url.startswith('rtmp'):
 158                 formats.append({
 159                     'url': video_url,
 160                     'format_id': 'rtmp-%s' % format_id,
 161                     'ext': 'flv',
 162                 })
 163             else:
 164                 if self._is_valid_url(video_url, video_id, format_id):
 165                     formats.append({
 166                         'url': video_url,
 167                         'format_id': format_id,
 168                     })
 169         self._sort_formats(formats)
 170
 171         title = info['titre']
 172         subtitle = info.get('sous_titre')
 173         if subtitle:
 174             title += ' - %s' % subtitle
 175         title = title.strip()
 176
 177         subtitles = {}
 178         subtitles_list = [{
 179             'url': subformat['url'],
 180             'ext': subformat.get('format'),
 181         } for subformat in info.get('subtitles', []) if subformat.get('url')]
 182         if subtitles_list:
 183             subtitles['fr'] = subtitles_list
 184
 185         return {
 186             'id': video_id,
 187             'title': self._live_title(title) if is_live else title,
 188             'description': clean_html(info['synopsis']),
 189             'thumbnail': compat_urlparse.urljoin('http://pluzz.francetv.fr', info['image']),
 190             'duration': int_or_none(info.get('real_duration')) or parse_duration(info['duree']),
 191             'timestamp': int_or_none(info['diffusion']['timestamp']),
 192             'is_live': is_live,
 193             'formats': formats,
 194             'subtitles': subtitles,
 195         }
 196
 197     def _real_extract(self, url):
 198         mobj = re.match(self._VALID_URL, url)
 199         video_id = mobj.group('id')
 200         catalog = mobj.group('catalog')
 201
 202         if not video_id:
 203             qs = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
 204             video_id = qs.get('idDiffusion', [None])[0]
 205             catalog = qs.get('catalogue', [None])[0]
 206             if not video_id:
 207                 raise ExtractorError('Invalid URL', expected=True)
 208
 209         return self._extract_video(video_id, catalog)
 210
 211
 212 class FranceTVSiteIE(FranceTVBaseInfoExtractor):
 213     _VALID_URL = r'https?://(?:(?:www\.)?france\.tv|mobile\.france\.tv)/(?:[^/]+/)*(?P<id>[^/]+)\.html'
 214
 215     _TESTS = [{
 216         'url': 'https://www.france.tv/france-2/13h15-le-dimanche/140921-les-mysteres-de-jesus.html',
 217         'info_dict': {
 218             'id': '162311093',
 219             'ext': 'mp4',
 220             'title': '13h15, le dimanche... - Les mystères de Jésus',
 221             'description': 'md5:75efe8d4c0a8205e5904498ffe1e1a42',
 222             'timestamp': 1502623500,
 223             'upload_date': '20170813',
 224         },
 225         'params': {
 226             'skip_download': True,
 227         },
 228         'add_ie': [FranceTVIE.ie_key()],
 229     }, {
 230         # france3
 231         'url': 'https://www.france.tv/france-3/des-chiffres-et-des-lettres/139063-emission-du-mardi-9-mai-2017.html',
 232         'only_matching': True,
 233     }, {
 234         # france4
 235         'url': 'https://www.france.tv/france-4/hero-corp/saison-1/134151-apres-le-calme.html',
 236         'only_matching': True,
 237     }, {
 238         # france5
 239         'url': 'https://www.france.tv/france-5/c-a-dire/saison-10/137013-c-a-dire.html',
 240         'only_matching': True,
 241     }, {
 242         # franceo
 243         'url': 'https://www.france.tv/france-o/archipels/132249-mon-ancetre-l-esclave.html',
 244         'only_matching': True,
 245     }, {
 246         # france2 live
 247         'url': 'https://www.france.tv/france-2/direct.html',
 248         'only_matching': True,
 249     }, {
 250         'url': 'https://www.france.tv/documentaires/histoire/136517-argentine-les-500-bebes-voles-de-la-dictature.html',
 251         'only_matching': True,
 252     }, {
 253         'url': 'https://www.france.tv/jeux-et-divertissements/divertissements/133965-le-web-contre-attaque.html',
 254         'only_matching': True,
 255     }, {
 256         'url': 'https://mobile.france.tv/france-5/c-dans-l-air/137347-emission-du-vendredi-12-mai-2017.html',
 257         'only_matching': True,
 258     }, {
 259         'url': 'https://www.france.tv/142749-rouge-sang.html',
 260         'only_matching': True,
 261     }, {
 262         # france-3 live
 263         'url': 'https://www.france.tv/france-3/direct.html',
 264         'only_matching': True,
 265     }]
 266
 267     def _real_extract(self, url):
 268         display_id = self._match_id(url)
 269
 270         webpage = self._download_webpage(url, display_id)
 271
 272         catalogue = None
 273         video_id = self._search_regex(
 274             r'data-main-video=(["\'])(?P<id>(?:(?!\1).)+)\1',
 275             webpage, 'video id', default=None, group='id')
 276
 277         if not video_id:
 278             video_id, catalogue = self._html_search_regex(
 279                 r'(?:href=|player\.setVideo\(\s*)"http://videos?\.francetv\.fr/video/([^@]+@[^"]+)"',
 280                 webpage, 'video ID').split('@')
 281
 282         return self._make_url_result(video_id, catalogue)
 283
 284
 285 class FranceTVEmbedIE(FranceTVBaseInfoExtractor):
 286     _VALID_URL = r'https?://embed\.francetv\.fr/*\?.*?\bue=(?P<id>[^&]+)'
 287
 288     _TESTS = [{
 289         'url': 'http://embed.francetv.fr/?ue=7fd581a2ccf59d2fc5719c5c13cf6961',
 290         'info_dict': {
 291             'id': 'NI_983319',
 292             'ext': 'mp4',
 293             'title': 'Le Pen Reims',
 294             'upload_date': '20170505',
 295             'timestamp': 1493981780,
 296             'duration': 16,
 297         },
 298         'params': {
 299             'skip_download': True,
 300         },
 301         'add_ie': [FranceTVIE.ie_key()],
 302     }]
 303
 304     def _real_extract(self, url):
 305         video_id = self._match_id(url)
 306
 307         video = self._download_json(
 308             'http://api-embed.webservices.francetelevisions.fr/key/%s' % video_id,
 309             video_id)
 310
 311         return self._make_url_result(video['video_id'], video.get('catalog'))
 312
 313
 314 class FranceTVInfoIE(FranceTVBaseInfoExtractor):
 315     IE_NAME = 'francetvinfo.fr'
 316     _VALID_URL = r'https?://(?:www|mobile|france3-regions)\.francetvinfo\.fr/(?:[^/]+/)*(?P<id>[^/?#&.]+)'
 317
 318     _TESTS = [{
 319         'url': 'http://www.francetvinfo.fr/replay-jt/france-3/soir-3/jt-grand-soir-3-lundi-26-aout-2013_393427.html',
 320         'info_dict': {
 321             'id': '84981923',
 322             'ext': 'mp4',
 323             'title': 'Soir 3',
 324             'upload_date': '20130826',
 325             'timestamp': 1377548400,
 326             'subtitles': {
 327                 'fr': 'mincount:2',
 328             },
 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
 363     def _real_extract(self, url):
 364         display_id = self._match_id(url)
 365
 366         webpage = self._download_webpage(url, display_id)
 367
 368         dailymotion_urls = DailymotionIE._extract_urls(webpage)
 369         if dailymotion_urls:
 370             return self.playlist_result([
 371                 self.url_result(dailymotion_url, DailymotionIE.ie_key())
 372                 for dailymotion_url in dailymotion_urls])
 373
 374         video_id, catalogue = self._search_regex(
 375             (r'id-video=([^@]+@[^"]+)',
 376              r'<a[^>]+href="(?:https?:)?//videos\.francetv\.fr/video/([^@]+@[^"]+)"'),
 377             webpage, 'video id').split('@')
 378
 379         return self._make_url_result(video_id, catalogue)
 380
 381
 382 class FranceTVInfoSportIE(FranceTVBaseInfoExtractor):
 383     IE_NAME = 'sport.francetvinfo.fr'
 384     _VALID_URL = r'https?://sport\.francetvinfo\.fr/(?:[^/]+/)*(?P<id>[^/?#&]+)'
 385     _TESTS = [{
 386         'url': 'https://sport.francetvinfo.fr/les-jeux-olympiques/retour-sur-les-meilleurs-moments-de-pyeongchang-2018',
 387         'info_dict': {
 388             'id': '6e49080e-3f45-11e8-b459-000d3a2439ea',
 389             'ext': 'mp4',
 390             'title': 'Retour sur les meilleurs moments de Pyeongchang 2018',
 391             'timestamp': 1523639962,
 392             'upload_date': '20180413',
 393         },
 394         'params': {
 395             'skip_download': True,
 396         },
 397         'add_ie': [FranceTVIE.ie_key()],
 398     }]
 399
 400     def _real_extract(self, url):
 401         display_id = self._match_id(url)
 402         webpage = self._download_webpage(url, display_id)
 403         video_id = self._search_regex(r'data-video="([^"]+)"', webpage, 'video_id')
 404         return self._make_url_result(video_id, 'Sport-web')
 405
 406
 407 class GenerationWhatIE(InfoExtractor):
 408     IE_NAME = 'france2.fr:generation-what'
 409     _VALID_URL = r'https?://generation-what\.francetv\.fr/[^/]+/video/(?P<id>[^/?#&]+)'
 410
 411     _TESTS = [{
 412         'url': 'http://generation-what.francetv.fr/portrait/video/present-arms',
 413         'info_dict': {
 414             'id': 'wtvKYUG45iw',
 415             'ext': 'mp4',
 416             'title': 'Generation What - Garde à vous - FRA',
 417             'uploader': 'Generation What',
 418             'uploader_id': 'UCHH9p1eetWCgt4kXBYCb3_w',
 419             'upload_date': '20160411',
 420         },
 421         'params': {
 422             'skip_download': True,
 423         },
 424         'add_ie': ['Youtube'],
 425     }, {
 426         'url': 'http://generation-what.francetv.fr/europe/video/present-arms',
 427         'only_matching': True,
 428     }]
 429
 430     def _real_extract(self, url):
 431         display_id = self._match_id(url)
 432
 433         webpage = self._download_webpage(url, display_id)
 434
 435         youtube_id = self._search_regex(
 436             r"window\.videoURL\s*=\s*'([0-9A-Za-z_-]{11})';",
 437             webpage, 'youtube id')
 438
 439         return self.url_result(youtube_id, ie='Youtube', video_id=youtube_id)
 440
 441
 442 class CultureboxIE(FranceTVBaseInfoExtractor):
 443     _VALID_URL = r'https?://(?:m\.)?culturebox\.francetvinfo\.fr/(?:[^/]+/)*(?P<id>[^/?#&]+)'
 444
 445     _TESTS = [{
 446         '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',
 447         'info_dict': {
 448             'id': 'EV_134885',
 449             'ext': 'mp4',
 450             'title': 'Cantates BWV 4, 106 et 131 de Bach par Raphaël Pichon 5/7',
 451             'description': 'md5:19c44af004b88219f4daa50fa9a351d4',
 452             'upload_date': '20180206',
 453             'timestamp': 1517945220,
 454             'duration': 5981,
 455         },
 456         'params': {
 457             'skip_download': True,
 458         },
 459         'add_ie': [FranceTVIE.ie_key()],
 460     }]
 461
 462     def _real_extract(self, url):
 463         display_id = self._match_id(url)
 464
 465         webpage = self._download_webpage(url, display_id)
 466
 467         if ">Ce live n'est plus disponible en replay<" in webpage:
 468             raise ExtractorError(
 469                 'Video %s is not available' % display_id, expected=True)
 470
 471         video_id, catalogue = self._search_regex(
 472             r'["\'>]https?://videos\.francetv\.fr/video/([^@]+@.+?)["\'<]',
 473             webpage, 'video id').split('@')
 474
 475         return self._make_url_result(video_id, catalogue)
 476
 477
 478 class FranceTVJeunesseIE(FranceTVBaseInfoExtractor):
 479     _VALID_URL = r'(?P<url>https?://(?:www\.)?(?:zouzous|ludo)\.fr/heros/(?P<id>[^/?#&]+))'
 480
 481     _TESTS = [{
 482         'url': 'https://www.zouzous.fr/heros/simon',
 483         'info_dict': {
 484             'id': 'simon',
 485         },
 486         'playlist_count': 9,
 487     }, {
 488         'url': 'https://www.ludo.fr/heros/ninjago',
 489         'info_dict': {
 490             'id': 'ninjago',
 491         },
 492         'playlist_count': 10,
 493     }, {
 494         'url': 'https://www.zouzous.fr/heros/simon?abc',
 495         'only_matching': True,
 496     }]
 497
 498     def _real_extract(self, url):
 499         mobj = re.match(self._VALID_URL, url)
 500         playlist_id = mobj.group('id')
 501
 502         playlist = self._download_json(
 503             '%s/%s' % (mobj.group('url'), 'playlist'), playlist_id)
 504
 505         if not playlist.get('count'):
 506             raise ExtractorError(
 507                 '%s is not available' % playlist_id, expected=True)
 508
 509         entries = []
 510         for item in playlist['items']:
 511             identity = item.get('identity')
 512             if identity and isinstance(identity, compat_str):
 513                 entries.append(self._make_url_result(identity))
 514
 515         return self.playlist_result(entries, playlist_id)