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 (
  10     compat_urllib_parse_urlparse,
  11     compat_urlparse,
  12 )
  13 from ..utils import (
  14     clean_html,
  15     ExtractorError,
  16     int_or_none,
  17     float_or_none,
  18     parse_duration,
  19     determine_ext,
  20 )
  21
  22
  23 class FranceTVBaseInfoExtractor(InfoExtractor):
  24     def _extract_video(self, video_id, catalogue):
  25         info = self._download_json(
  26             'http://webservices.francetelevisions.fr/tools/getInfosOeuvre/v2/?idDiffusion=%s&catalogue=%s'
  27             % (video_id, catalogue),
  28             video_id, 'Downloading video JSON')
  29
  30         if info.get('status') == 'NOK':
  31             raise ExtractorError(
  32                 '%s returned error: %s' % (self.IE_NAME, info['message']), expected=True)
  33         allowed_countries = info['videos'][0].get('geoblocage')
  34         if allowed_countries:
  35             georestricted = True
  36             geo_info = self._download_json(
  37                 'http://geo.francetv.fr/ws/edgescape.json', video_id,
  38                 'Downloading geo restriction info')
  39             country = geo_info['reponse']['geo_info']['country_code']
  40             if country not in allowed_countries:
  41                 raise ExtractorError(
  42                     'The video is not available from your location',
  43                     expected=True)
  44         else:
  45             georestricted = False
  46
  47         formats = []
  48         for video in info['videos']:
  49             if video['statut'] != 'ONLINE':
  50                 continue
  51             video_url = video['url']
  52             if not video_url:
  53                 continue
  54             format_id = video['format']
  55             ext = determine_ext(video_url)
  56             if ext == 'f4m':
  57                 if georestricted:
  58                     # See https://github.com/rg3/youtube-dl/issues/3963
  59                     # m3u8 urls work fine
  60                     continue
  61                 video_url_parsed = compat_urllib_parse_urlparse(video_url)
  62                 f4m_url = self._download_webpage(
  63                     'http://hdfauth.francetv.fr/esi/urltokengen2.html?url=%s' % video_url_parsed.path,
  64                     video_id, 'Downloading f4m manifest token', fatal=False)
  65                 if f4m_url:
  66                     formats.extend(self._extract_f4m_formats(f4m_url, video_id, 1, format_id))
  67             elif ext == 'm3u8':
  68                 formats.extend(self._extract_m3u8_formats(video_url, video_id, 'mp4', m3u8_id=format_id))
  69             elif video_url.startswith('rtmp'):
  70                 formats.append({
  71                     'url': video_url,
  72                     'format_id': 'rtmp-%s' % format_id,
  73                     'ext': 'flv',
  74                     'preference': 1,
  75                 })
  76             else:
  77                 formats.append({
  78                     'url': video_url,
  79                     'format_id': format_id,
  80                     'preference': -1,
  81                 })
  82         self._sort_formats(formats)
  83
  84         return {
  85             'id': video_id,
  86             'title': info['titre'],
  87             'description': clean_html(info['synopsis']),
  88             'thumbnail': compat_urlparse.urljoin('http://pluzz.francetv.fr', info['image']),
  89             'duration': float_or_none(info.get('real_duration'), 1000) or parse_duration(info['duree']),
  90             'timestamp': int_or_none(info['diffusion']['timestamp']),
  91             'formats': formats,
  92         }
  93
  94
  95 class PluzzIE(FranceTVBaseInfoExtractor):
  96     IE_NAME = 'pluzz.francetv.fr'
  97     _VALID_URL = r'https?://pluzz\.francetv\.fr/videos/(.*?)\.html'
  98
  99     # Can't use tests, videos expire in 7 days
 100
 101     def _real_extract(self, url):
 102         title = re.match(self._VALID_URL, url).group(1)
 103         webpage = self._download_webpage(url, title)
 104         video_id = self._search_regex(
 105             r'data-diffusion="(\d+)"', webpage, 'ID')
 106         return self._extract_video(video_id, 'Pluzz')
 107
 108
 109 class FranceTvInfoIE(FranceTVBaseInfoExtractor):
 110     IE_NAME = 'francetvinfo.fr'
 111     _VALID_URL = r'https?://(?:www|mobile)\.francetvinfo\.fr/.*/(?P<title>.+)\.html'
 112
 113     _TESTS = [{
 114         'url': 'http://www.francetvinfo.fr/replay-jt/france-3/soir-3/jt-grand-soir-3-lundi-26-aout-2013_393427.html',
 115         'info_dict': {
 116             'id': '84981923',
 117             'ext': 'flv',
 118             'title': 'Soir 3',
 119             'upload_date': '20130826',
 120             'timestamp': 1377548400,
 121         },
 122     }, {
 123         'url': 'http://www.francetvinfo.fr/elections/europeennes/direct-europeennes-regardez-le-debat-entre-les-candidats-a-la-presidence-de-la-commission_600639.html',
 124         'info_dict': {
 125             'id': 'EV_20019',
 126             'ext': 'mp4',
 127             'title': 'Débat des candidats à la Commission européenne',
 128             'description': 'Débat des candidats à la Commission européenne',
 129         },
 130         'params': {
 131             'skip_download': 'HLS (reqires ffmpeg)'
 132         },
 133         'skip': 'Ce direct est terminé et sera disponible en rattrapage dans quelques minutes.',
 134     }]
 135
 136     def _real_extract(self, url):
 137         mobj = re.match(self._VALID_URL, url)
 138         page_title = mobj.group('title')
 139         webpage = self._download_webpage(url, page_title)
 140         video_id, catalogue = self._search_regex(
 141             r'id-video=([^@]+@[^"]+)', webpage, 'video id').split('@')
 142         return self._extract_video(video_id, catalogue)
 143
 144
 145 class FranceTVIE(FranceTVBaseInfoExtractor):
 146     IE_NAME = 'francetv'
 147     IE_DESC = 'France 2, 3, 4, 5 and Ô'
 148     _VALID_URL = r'''(?x)https?://www\.france[2345o]\.fr/
 149         (?:
 150             emissions/.*?/(videos|emissions)/(?P<id>[^/?]+)
 151         |   (emissions?|jt)/(?P<key>[^/?]+)
 152         )'''
 153
 154     _TESTS = [
 155         # france2
 156         {
 157             'url': 'http://www.france2.fr/emissions/13h15-le-samedi-le-dimanche/videos/75540104',
 158             'md5': 'c03fc87cb85429ffd55df32b9fc05523',
 159             'info_dict': {
 160                 'id': '109169362',
 161                 'ext': 'flv',
 162                 'title': '13h15, le dimanche...',
 163                 'description': 'md5:9a0932bb465f22d377a449be9d1a0ff7',
 164                 'upload_date': '20140914',
 165                 'timestamp': 1410693600,
 166             },
 167         },
 168         # france3
 169         {
 170             'url': 'http://www.france3.fr/emissions/pieces-a-conviction/diffusions/13-11-2013_145575',
 171             'md5': '679bb8f8921f8623bd658fa2f8364da0',
 172             'info_dict': {
 173                 'id': '000702326_CAPP_PicesconvictionExtrait313022013_120220131722_Au',
 174                 'ext': 'mp4',
 175                 'title': 'Le scandale du prix des médicaments',
 176                 'description': 'md5:1384089fbee2f04fc6c9de025ee2e9ce',
 177                 'upload_date': '20131113',
 178                 'timestamp': 1384380000,
 179             },
 180         },
 181         # france4
 182         {
 183             'url': 'http://www.france4.fr/emissions/hero-corp/videos/rhozet_herocorp_bonus_1_20131106_1923_06112013172108_F4',
 184             'md5': 'a182bf8d2c43d88d46ec48fbdd260c1c',
 185             'info_dict': {
 186                 'id': 'rhozet_herocorp_bonus_1_20131106_1923_06112013172108_F4',
 187                 'ext': 'mp4',
 188                 'title': 'Hero Corp Making of - Extrait 1',
 189                 'description': 'md5:c87d54871b1790679aec1197e73d650a',
 190                 'upload_date': '20131106',
 191                 'timestamp': 1383766500,
 192             },
 193         },
 194         # france5
 195         {
 196             'url': 'http://www.france5.fr/emissions/c-a-dire/videos/92837968',
 197             'md5': '78f0f4064f9074438e660785bbf2c5d9',
 198             'info_dict': {
 199                 'id': '108961659',
 200                 'ext': 'flv',
 201                 'title': 'C à dire ?!',
 202                 'description': 'md5:1a4aeab476eb657bf57c4ff122129f81',
 203                 'upload_date': '20140915',
 204                 'timestamp': 1410795000,
 205             },
 206         },
 207         # franceo
 208         {
 209             'url': 'http://www.franceo.fr/jt/info-afrique/04-12-2013',
 210             'md5': '52f0bfe202848b15915a2f39aaa8981b',
 211             'info_dict': {
 212                 'id': '108634970',
 213                 'ext': 'flv',
 214                 'title': 'Infô Afrique',
 215                 'description': 'md5:ebf346da789428841bee0fd2a935ea55',
 216                 'upload_date': '20140915',
 217                 'timestamp': 1410822000,
 218             },
 219         },
 220     ]
 221
 222     def _real_extract(self, url):
 223         mobj = re.match(self._VALID_URL, url)
 224         webpage = self._download_webpage(url, mobj.group('key') or mobj.group('id'))
 225         video_id, catalogue = self._html_search_regex(
 226             r'href="http://videos\.francetv\.fr/video/([^@]+@[^"]+)"',
 227             webpage, 'video ID').split('@')
 228         return self._extract_video(video_id, catalogue)
 229
 230
 231 class GenerationQuoiIE(InfoExtractor):
 232     IE_NAME = 'france2.fr:generation-quoi'
 233     _VALID_URL = r'https?://generation-quoi\.france2\.fr/portrait/(?P<id>[^/?#]+)'
 234
 235     _TEST = {
 236         'url': 'http://generation-quoi.france2.fr/portrait/garde-a-vous',
 237         'info_dict': {
 238             'id': 'k7FJX8VBcvvLmX4wA5Q',
 239             'ext': 'mp4',
 240             'title': 'Génération Quoi - Garde à Vous',
 241             'uploader': 'Génération Quoi',
 242         },
 243         'params': {
 244             # It uses Dailymotion
 245             'skip_download': True,
 246         },
 247     }
 248
 249     def _real_extract(self, url):
 250         display_id = self._match_id(url)
 251         info_url = compat_urlparse.urljoin(url, '/medias/video/%s.json' % display_id)
 252         info_json = self._download_webpage(info_url, display_id)
 253         info = json.loads(info_json)
 254         return self.url_result('http://www.dailymotion.com/video/%s' % info['id'],
 255                                ie='Dailymotion')
 256
 257
 258 class CultureboxIE(FranceTVBaseInfoExtractor):
 259     IE_NAME = 'culturebox.francetvinfo.fr'
 260     _VALID_URL = r'https?://(?:m\.)?culturebox\.francetvinfo\.fr/(?P<name>.*?)(\?|$)'
 261
 262     _TEST = {
 263         'url': 'http://culturebox.francetvinfo.fr/live/musique/musique-classique/le-livre-vermeil-de-montserrat-a-la-cathedrale-delne-214511',
 264         'md5': '9b88dc156781c4dbebd4c3e066e0b1d6',
 265         'info_dict': {
 266             'id': 'EV_50111',
 267             'ext': 'flv',
 268             'title': "Le Livre Vermeil de Montserrat à la Cathédrale d'Elne",
 269             'description': 'md5:f8a4ad202e8fe533e2c493cc12e739d9',
 270             'upload_date': '20150320',
 271             'timestamp': 1426892400,
 272             'duration': 2760.9,
 273         },
 274     }
 275
 276     def _real_extract(self, url):
 277         mobj = re.match(self._VALID_URL, url)
 278         name = mobj.group('name')
 279
 280         webpage = self._download_webpage(url, name)
 281
 282         if ">Ce live n'est plus disponible en replay<" in webpage:
 283             raise ExtractorError('Video %s is not available' % name, expected=True)
 284
 285         video_id, catalogue = self._search_regex(
 286             r'"http://videos\.francetv\.fr/video/([^@]+@[^"]+)"', webpage, 'video id').split('@')
 287
 288         return self._extract_video(video_id, catalogue)