Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/francetv.py

   1 # encoding: utf-8
   2
   3 from __future__ import unicode_literals
   4
   5 import re
   6 import json
   7
   8 from .common import InfoExtractor
   9 from ..compat import compat_urlparse
  10 from ..utils import (
  11     clean_html,
  12     ExtractorError,
  13     int_or_none,
  14     parse_duration,
  15     determine_ext,
  16 )
  17 from .dailymotion import DailymotionCloudIE
  18
  19
  20 class FranceTVBaseInfoExtractor(InfoExtractor):
  21     def _extract_video(self, video_id, catalogue):
  22         info = self._download_json(
  23             'http://webservices.francetelevisions.fr/tools/getInfosOeuvre/v2/?idDiffusion=%s&catalogue=%s'
  24             % (video_id, catalogue),
  25             video_id, 'Downloading video JSON')
  26
  27         if info.get('status') == 'NOK':
  28             raise ExtractorError(
  29                 '%s returned error: %s' % (self.IE_NAME, info['message']), expected=True)
  30         allowed_countries = info['videos'][0].get('geoblocage')
  31         if allowed_countries:
  32             georestricted = True
  33             geo_info = self._download_json(
  34                 'http://geo.francetv.fr/ws/edgescape.json', video_id,
  35                 'Downloading geo restriction info')
  36             country = geo_info['reponse']['geo_info']['country_code']
  37             if country not in allowed_countries:
  38                 raise ExtractorError(
  39                     'The video is not available from your location',
  40                     expected=True)
  41         else:
  42             georestricted = False
  43
  44         formats = []
  45         for video in info['videos']:
  46             if video['statut'] != 'ONLINE':
  47                 continue
  48             video_url = video['url']
  49             if not video_url:
  50                 continue
  51             format_id = video['format']
  52             ext = determine_ext(video_url)
  53             if ext == 'f4m':
  54                 if georestricted:
  55                     # See https://github.com/rg3/youtube-dl/issues/3963
  56                     # m3u8 urls work fine
  57                     continue
  58                 f4m_url = self._download_webpage(
  59                     'http://hdfauth.francetv.fr/esi/TA?url=%s' % video_url,
  60                     video_id, 'Downloading f4m manifest token', fatal=False)
  61                 if f4m_url:
  62                     formats.extend(self._extract_f4m_formats(
  63                         f4m_url + '&hdcore=3.7.0&plugin=aasp-3.7.0.39.44',
  64                         video_id, f4m_id=format_id, fatal=False))
  65             elif ext == 'm3u8':
  66                 formats.extend(self._extract_m3u8_formats(
  67                     video_url, video_id, 'mp4', entry_protocol='m3u8_native',
  68                     m3u8_id=format_id, fatal=False))
  69             elif video_url.startswith('rtmp'):
  70                 formats.append({
  71                     'url': video_url,
  72                     'format_id': 'rtmp-%s' % format_id,
  73                     'ext': 'flv',
  74                 })
  75             else:
  76                 if self._is_valid_url(video_url, video_id, format_id):
  77                     formats.append({
  78                         'url': video_url,
  79                         'format_id': format_id,
  80                     })
  81         self._sort_formats(formats)
  82
  83         title = info['titre']
  84         subtitle = info.get('sous_titre')
  85         if subtitle:
  86             title += ' - %s' % subtitle
  87         title = title.strip()
  88
  89         subtitles = {}
  90         subtitles_list = [{
  91             'url': subformat['url'],
  92             'ext': subformat.get('format'),
  93         } for subformat in info.get('subtitles', []) if subformat.get('url')]
  94         if subtitles_list:
  95             subtitles['fr'] = subtitles_list
  96
  97         return {
  98             'id': video_id,
  99             'title': title,
 100             'description': clean_html(info['synopsis']),
 101             'thumbnail': compat_urlparse.urljoin('http://pluzz.francetv.fr', info['image']),
 102             'duration': int_or_none(info.get('real_duration')) or parse_duration(info['duree']),
 103             'timestamp': int_or_none(info['diffusion']['timestamp']),
 104             'formats': formats,
 105             'subtitles': subtitles,
 106         }
 107
 108
 109 class PluzzIE(FranceTVBaseInfoExtractor):
 110     IE_NAME = 'pluzz.francetv.fr'
 111     _VALID_URL = r'https?://(?:m\.)?pluzz\.francetv\.fr/videos/(?P<id>.+?)\.html'
 112
 113     # Can't use tests, videos expire in 7 days
 114
 115     def _real_extract(self, url):
 116         display_id = self._match_id(url)
 117
 118         webpage = self._download_webpage(url, display_id)
 119
 120         video_id = self._html_search_meta(
 121             'id_video', webpage, 'video id', default=None)
 122         if not video_id:
 123             video_id = self._search_regex(
 124                 r'data-diffusion=["\'](\d+)', webpage, 'video id')
 125
 126         return self._extract_video(video_id, 'Pluzz')
 127
 128
 129 class FranceTvInfoIE(FranceTVBaseInfoExtractor):
 130     IE_NAME = 'francetvinfo.fr'
 131     _VALID_URL = r'https?://(?:www|mobile|france3-regions)\.francetvinfo\.fr/.*/(?P<title>.+)\.html'
 132
 133     _TESTS = [{
 134         'url': 'http://www.francetvinfo.fr/replay-jt/france-3/soir-3/jt-grand-soir-3-lundi-26-aout-2013_393427.html',
 135         'info_dict': {
 136             'id': '84981923',
 137             'ext': 'mp4',
 138             'title': 'Soir 3',
 139             'upload_date': '20130826',
 140             'timestamp': 1377548400,
 141             'subtitles': {
 142                 'fr': 'mincount:2',
 143             },
 144         },
 145         'params': {
 146             # m3u8 downloads
 147             'skip_download': True,
 148         },
 149     }, {
 150         'url': 'http://www.francetvinfo.fr/elections/europeennes/direct-europeennes-regardez-le-debat-entre-les-candidats-a-la-presidence-de-la-commission_600639.html',
 151         'info_dict': {
 152             'id': 'EV_20019',
 153             'ext': 'mp4',
 154             'title': 'Débat des candidats à la Commission européenne',
 155             'description': 'Débat des candidats à la Commission européenne',
 156         },
 157         'params': {
 158             'skip_download': 'HLS (reqires ffmpeg)'
 159         },
 160         'skip': 'Ce direct est terminé et sera disponible en rattrapage dans quelques minutes.',
 161     }, {
 162         'url': 'http://www.francetvinfo.fr/economie/entreprises/les-entreprises-familiales-le-secret-de-la-reussite_933271.html',
 163         'md5': 'f485bda6e185e7d15dbc69b72bae993e',
 164         'info_dict': {
 165             'id': 'NI_173343',
 166             'ext': 'mp4',
 167             'title': 'Les entreprises familiales : le secret de la réussite',
 168             'thumbnail': 're:^https?://.*\.jpe?g$',
 169             'timestamp': 1433273139,
 170             'upload_date': '20150602',
 171         },
 172         'params': {
 173             # m3u8 downloads
 174             'skip_download': True,
 175         },
 176     }, {
 177         'url': 'http://france3-regions.francetvinfo.fr/bretagne/cotes-d-armor/thalassa-echappee-breizh-ce-venredi-dans-les-cotes-d-armor-954961.html',
 178         'md5': 'f485bda6e185e7d15dbc69b72bae993e',
 179         'info_dict': {
 180             'id': 'NI_657393',
 181             'ext': 'mp4',
 182             'title': 'Olivier Monthus, réalisateur de "Bretagne, le choix de l’Armor"',
 183             'description': 'md5:a3264114c9d29aeca11ced113c37b16c',
 184             'thumbnail': 're:^https?://.*\.jpe?g$',
 185             'timestamp': 1458300695,
 186             'upload_date': '20160318',
 187         },
 188         'params': {
 189             'skip_download': True,
 190         },
 191     }]
 192
 193     def _real_extract(self, url):
 194         mobj = re.match(self._VALID_URL, url)
 195         page_title = mobj.group('title')
 196         webpage = self._download_webpage(url, page_title)
 197
 198         dmcloud_url = DailymotionCloudIE._extract_dmcloud_url(webpage)
 199         if dmcloud_url:
 200             return self.url_result(dmcloud_url, 'DailymotionCloud')
 201
 202         video_id, catalogue = self._search_regex(
 203             (r'id-video=([^@]+@[^"]+)',
 204              r'<a[^>]+href="(?:https?:)?//videos\.francetv\.fr/video/([^@]+@[^"]+)"'),
 205             webpage, 'video id').split('@')
 206         return self._extract_video(video_id, catalogue)
 207
 208
 209 class FranceTVIE(FranceTVBaseInfoExtractor):
 210     IE_NAME = 'francetv'
 211     IE_DESC = 'France 2, 3, 4, 5 and Ô'
 212     _VALID_URL = r'''(?x)
 213                     https?://
 214                         (?:
 215                             (?:www\.)?france[2345o]\.fr/
 216                                 (?:
 217                                     emissions/[^/]+/(?:videos|diffusions)|
 218                                     emission/[^/]+|
 219                                     videos|
 220                                     jt
 221                                 )
 222                             /|
 223                             embed\.francetv\.fr/\?ue=
 224                         )
 225                         (?P<id>[^/?]+)
 226                     '''
 227
 228     _TESTS = [
 229         # france2
 230         {
 231             'url': 'http://www.france2.fr/emissions/13h15-le-samedi-le-dimanche/videos/75540104',
 232             'md5': 'c03fc87cb85429ffd55df32b9fc05523',
 233             'info_dict': {
 234                 'id': '109169362',
 235                 'ext': 'flv',
 236                 'title': '13h15, le dimanche...',
 237                 'description': 'md5:9a0932bb465f22d377a449be9d1a0ff7',
 238                 'upload_date': '20140914',
 239                 'timestamp': 1410693600,
 240             },
 241         },
 242         # france3
 243         {
 244             'url': 'http://www.france3.fr/emissions/pieces-a-conviction/diffusions/13-11-2013_145575',
 245             'md5': '679bb8f8921f8623bd658fa2f8364da0',
 246             'info_dict': {
 247                 'id': '000702326_CAPP_PicesconvictionExtrait313022013_120220131722_Au',
 248                 'ext': 'mp4',
 249                 'title': 'Le scandale du prix des médicaments',
 250                 'description': 'md5:1384089fbee2f04fc6c9de025ee2e9ce',
 251                 'upload_date': '20131113',
 252                 'timestamp': 1384380000,
 253             },
 254         },
 255         # france4
 256         {
 257             'url': 'http://www.france4.fr/emissions/hero-corp/videos/rhozet_herocorp_bonus_1_20131106_1923_06112013172108_F4',
 258             'md5': 'a182bf8d2c43d88d46ec48fbdd260c1c',
 259             'info_dict': {
 260                 'id': 'rhozet_herocorp_bonus_1_20131106_1923_06112013172108_F4',
 261                 'ext': 'mp4',
 262                 'title': 'Hero Corp Making of - Extrait 1',
 263                 'description': 'md5:c87d54871b1790679aec1197e73d650a',
 264                 'upload_date': '20131106',
 265                 'timestamp': 1383766500,
 266             },
 267         },
 268         # france5
 269         {
 270             'url': 'http://www.france5.fr/emissions/c-a-dire/videos/quels_sont_les_enjeux_de_cette_rentree_politique__31-08-2015_908948?onglet=tous&page=1',
 271             'md5': 'f6c577df3806e26471b3d21631241fd0',
 272             'info_dict': {
 273                 'id': '123327454',
 274                 'ext': 'flv',
 275                 'title': 'C à dire ?! - Quels sont les enjeux de cette rentrée politique ?',
 276                 'description': 'md5:4a0d5cb5dce89d353522a84462bae5a4',
 277                 'upload_date': '20150831',
 278                 'timestamp': 1441035120,
 279             },
 280         },
 281         # franceo
 282         {
 283             'url': 'http://www.franceo.fr/jt/info-soir/18-07-2015',
 284             'md5': '47d5816d3b24351cdce512ad7ab31da8',
 285             'info_dict': {
 286                 'id': '125377621',
 287                 'ext': 'flv',
 288                 'title': 'Infô soir',
 289                 'description': 'md5:01b8c6915a3d93d8bbbd692651714309',
 290                 'upload_date': '20150718',
 291                 'timestamp': 1437241200,
 292                 'duration': 414,
 293             },
 294         },
 295         {
 296             # francetv embed
 297             'url': 'http://embed.francetv.fr/?ue=8d7d3da1e3047c42ade5a5d7dfd3fc87',
 298             'info_dict': {
 299                 'id': 'EV_30231',
 300                 'ext': 'flv',
 301                 'title': 'Alcaline, le concert avec Calogero',
 302                 'description': 'md5:61f08036dcc8f47e9cfc33aed08ffaff',
 303                 'upload_date': '20150226',
 304                 'timestamp': 1424989860,
 305                 'duration': 5400,
 306             },
 307         },
 308         {
 309             'url': 'http://www.france4.fr/emission/highlander/diffusion-du-17-07-2015-04h05',
 310             'only_matching': True,
 311         },
 312         {
 313             'url': 'http://www.franceo.fr/videos/125377617',
 314             'only_matching': True,
 315         }
 316     ]
 317
 318     def _real_extract(self, url):
 319         video_id = self._match_id(url)
 320         webpage = self._download_webpage(url, video_id)
 321         video_id, catalogue = self._html_search_regex(
 322             r'(?:href=|player\.setVideo\(\s*)"http://videos?\.francetv\.fr/video/([^@]+@[^"]+)"',
 323             webpage, 'video ID').split('@')
 324         return self._extract_video(video_id, catalogue)
 325
 326
 327 class GenerationQuoiIE(InfoExtractor):
 328     IE_NAME = 'france2.fr:generation-quoi'
 329     _VALID_URL = r'https?://generation-quoi\.france2\.fr/portrait/(?P<id>[^/?#]+)'
 330
 331     _TEST = {
 332         'url': 'http://generation-quoi.france2.fr/portrait/garde-a-vous',
 333         'info_dict': {
 334             'id': 'k7FJX8VBcvvLmX4wA5Q',
 335             'ext': 'mp4',
 336             'title': 'Génération Quoi - Garde à Vous',
 337             'uploader': 'Génération Quoi',
 338         },
 339         'params': {
 340             # It uses Dailymotion
 341             'skip_download': True,
 342         },
 343     }
 344
 345     def _real_extract(self, url):
 346         display_id = self._match_id(url)
 347         info_url = compat_urlparse.urljoin(url, '/medias/video/%s.json' % display_id)
 348         info_json = self._download_webpage(info_url, display_id)
 349         info = json.loads(info_json)
 350         return self.url_result('http://www.dailymotion.com/video/%s' % info['id'],
 351                                ie='Dailymotion')
 352
 353
 354 class CultureboxIE(FranceTVBaseInfoExtractor):
 355     IE_NAME = 'culturebox.francetvinfo.fr'
 356     _VALID_URL = r'https?://(?:m\.)?culturebox\.francetvinfo\.fr/(?P<name>.*?)(\?|$)'
 357
 358     _TEST = {
 359         'url': 'http://culturebox.francetvinfo.fr/live/musique/musique-classique/le-livre-vermeil-de-montserrat-a-la-cathedrale-delne-214511',
 360         'md5': '9b88dc156781c4dbebd4c3e066e0b1d6',
 361         'info_dict': {
 362             'id': 'EV_50111',
 363             'ext': 'flv',
 364             'title': "Le Livre Vermeil de Montserrat à la Cathédrale d'Elne",
 365             'description': 'md5:f8a4ad202e8fe533e2c493cc12e739d9',
 366             'upload_date': '20150320',
 367             'timestamp': 1426892400,
 368             'duration': 2760.9,
 369         },
 370     }
 371
 372     def _real_extract(self, url):
 373         mobj = re.match(self._VALID_URL, url)
 374         name = mobj.group('name')
 375
 376         webpage = self._download_webpage(url, name)
 377
 378         if ">Ce live n'est plus disponible en replay<" in webpage:
 379             raise ExtractorError('Video %s is not available' % name, expected=True)
 380
 381         video_id, catalogue = self._search_regex(
 382             r'"http://videos\.francetv\.fr/video/([^@]+@[^"]+)"', webpage, 'video id').split('@')
 383
 384         return self._extract_video(video_id, catalogue)