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 )
  20
  21
  22 class FranceTVBaseInfoExtractor(InfoExtractor):
  23     def _extract_video(self, video_id, catalogue):
  24         info = self._download_json(
  25             'http://webservices.francetelevisions.fr/tools/getInfosOeuvre/v2/?idDiffusion=%s&catalogue=%s'
  26             % (video_id, catalogue),
  27             video_id, 'Downloading video JSON')
  28
  29         if info.get('status') == 'NOK':
  30             raise ExtractorError(
  31                 '%s returned error: %s' % (self.IE_NAME, info['message']), expected=True)
  32         allowed_countries = info['videos'][0].get('geoblocage')
  33         if allowed_countries:
  34             georestricted = True
  35             geo_info = self._download_json(
  36                 'http://geo.francetv.fr/ws/edgescape.json', video_id,
  37                 'Downloading geo restriction info')
  38             country = geo_info['reponse']['geo_info']['country_code']
  39             if country not in allowed_countries:
  40                 raise ExtractorError(
  41                     'The video is not available from your location',
  42                     expected=True)
  43         else:
  44             georestricted = False
  45
  46         formats = []
  47         for video in info['videos']:
  48             if video['statut'] != 'ONLINE':
  49                 continue
  50             video_url = video['url']
  51             if not video_url:
  52                 continue
  53             format_id = video['format']
  54             if video_url.endswith('.f4m'):
  55                 if georestricted:
  56                     # See https://github.com/rg3/youtube-dl/issues/3963
  57                     # m3u8 urls work fine
  58                     continue
  59                 video_url_parsed = compat_urllib_parse_urlparse(video_url)
  60                 f4m_url = self._download_webpage(
  61                     'http://hdfauth.francetv.fr/esi/urltokengen2.html?url=%s' % video_url_parsed.path,
  62                     video_id, 'Downloading f4m manifest token', fatal=False)
  63                 if f4m_url:
  64                     f4m_formats = self._extract_f4m_formats(f4m_url, video_id)
  65                     for f4m_format in f4m_formats:
  66                         f4m_format['preference'] = 1
  67                     formats.extend(f4m_formats)
  68             elif video_url.endswith('.m3u8'):
  69                 formats.extend(self._extract_m3u8_formats(video_url, video_id, 'mp4'))
  70             elif video_url.startswith('rtmp'):
  71                 formats.append({
  72                     'url': video_url,
  73                     'format_id': 'rtmp-%s' % format_id,
  74                     'ext': 'flv',
  75                     'preference': 1,
  76                 })
  77             else:
  78                 formats.append({
  79                     'url': video_url,
  80                     'format_id': format_id,
  81                     'preference': -1,
  82                 })
  83         self._sort_formats(formats)
  84
  85         return {
  86             'id': video_id,
  87             'title': info['titre'],
  88             'description': clean_html(info['synopsis']),
  89             'thumbnail': compat_urlparse.urljoin('http://pluzz.francetv.fr', info['image']),
  90             'duration': float_or_none(info.get('real_duration'), 1000) or parse_duration(info['duree']),
  91             'timestamp': int_or_none(info['diffusion']['timestamp']),
  92             'formats': formats,
  93         }
  94
  95
  96 class PluzzIE(FranceTVBaseInfoExtractor):
  97     IE_NAME = 'pluzz.francetv.fr'
  98     _VALID_URL = r'https?://pluzz\.francetv\.fr/videos/(.*?)\.html'
  99
 100     # Can't use tests, videos expire in 7 days
 101
 102     def _real_extract(self, url):
 103         title = re.match(self._VALID_URL, url).group(1)
 104         webpage = self._download_webpage(url, title)
 105         video_id = self._search_regex(
 106             r'data-diffusion="(\d+)"', webpage, 'ID')
 107         return self._extract_video(video_id, 'Pluzz')
 108
 109
 110 class FranceTvInfoIE(FranceTVBaseInfoExtractor):
 111     IE_NAME = 'francetvinfo.fr'
 112     _VALID_URL = r'https?://(?:www|mobile)\.francetvinfo\.fr/.*/(?P<title>.+)\.html'
 113
 114     _TESTS = [{
 115         'url': 'http://www.francetvinfo.fr/replay-jt/france-3/soir-3/jt-grand-soir-3-lundi-26-aout-2013_393427.html',
 116         'info_dict': {
 117             'id': '84981923',
 118             'ext': 'flv',
 119             'title': 'Soir 3',
 120             'upload_date': '20130826',
 121             'timestamp': 1377548400,
 122         },
 123     }, {
 124         'url': 'http://www.francetvinfo.fr/elections/europeennes/direct-europeennes-regardez-le-debat-entre-les-candidats-a-la-presidence-de-la-commission_600639.html',
 125         'info_dict': {
 126             'id': 'EV_20019',
 127             'ext': 'mp4',
 128             'title': 'Débat des candidats à la Commission européenne',
 129             'description': 'Débat des candidats à la Commission européenne',
 130         },
 131         'params': {
 132             'skip_download': 'HLS (reqires ffmpeg)'
 133         },
 134         'skip': 'Ce direct est terminé et sera disponible en rattrapage dans quelques minutes.',
 135     }]
 136
 137     def _real_extract(self, url):
 138         mobj = re.match(self._VALID_URL, url)
 139         page_title = mobj.group('title')
 140         webpage = self._download_webpage(url, page_title)
 141         video_id, catalogue = self._search_regex(
 142             r'id-video=([^@]+@[^"]+)', webpage, 'video id').split('@')
 143         return self._extract_video(video_id, catalogue)
 144
 145
 146 class FranceTVIE(FranceTVBaseInfoExtractor):
 147     IE_NAME = 'francetv'
 148     IE_DESC = 'France 2, 3, 4, 5 and Ô'
 149     _VALID_URL = r'''(?x)https?://www\.france[2345o]\.fr/
 150         (?:
 151             emissions/.*?/(videos|emissions)/(?P<id>[^/?]+)
 152         |   (emissions?|jt)/(?P<key>[^/?]+)
 153         )'''
 154
 155     _TESTS = [
 156         # france2
 157         {
 158             'url': 'http://www.france2.fr/emissions/13h15-le-samedi-le-dimanche/videos/75540104',
 159             'md5': 'c03fc87cb85429ffd55df32b9fc05523',
 160             'info_dict': {
 161                 'id': '109169362',
 162                 'ext': 'flv',
 163                 'title': '13h15, le dimanche...',
 164                 'description': 'md5:9a0932bb465f22d377a449be9d1a0ff7',
 165                 'upload_date': '20140914',
 166                 'timestamp': 1410693600,
 167             },
 168         },
 169         # france3
 170         {
 171             'url': 'http://www.france3.fr/emissions/pieces-a-conviction/diffusions/13-11-2013_145575',
 172             'md5': '679bb8f8921f8623bd658fa2f8364da0',
 173             'info_dict': {
 174                 'id': '000702326_CAPP_PicesconvictionExtrait313022013_120220131722_Au',
 175                 'ext': 'mp4',
 176                 'title': 'Le scandale du prix des médicaments',
 177                 'description': 'md5:1384089fbee2f04fc6c9de025ee2e9ce',
 178                 'upload_date': '20131113',
 179                 'timestamp': 1384380000,
 180             },
 181         },
 182         # france4
 183         {
 184             'url': 'http://www.france4.fr/emissions/hero-corp/videos/rhozet_herocorp_bonus_1_20131106_1923_06112013172108_F4',
 185             'md5': 'a182bf8d2c43d88d46ec48fbdd260c1c',
 186             'info_dict': {
 187                 'id': 'rhozet_herocorp_bonus_1_20131106_1923_06112013172108_F4',
 188                 'ext': 'mp4',
 189                 'title': 'Hero Corp Making of - Extrait 1',
 190                 'description': 'md5:c87d54871b1790679aec1197e73d650a',
 191                 'upload_date': '20131106',
 192                 'timestamp': 1383766500,
 193             },
 194         },
 195         # france5
 196         {
 197             'url': 'http://www.france5.fr/emissions/c-a-dire/videos/92837968',
 198             'md5': '78f0f4064f9074438e660785bbf2c5d9',
 199             'info_dict': {
 200                 'id': '108961659',
 201                 'ext': 'flv',
 202                 'title': 'C à dire ?!',
 203                 'description': 'md5:1a4aeab476eb657bf57c4ff122129f81',
 204                 'upload_date': '20140915',
 205                 'timestamp': 1410795000,
 206             },
 207         },
 208         # franceo
 209         {
 210             'url': 'http://www.franceo.fr/jt/info-afrique/04-12-2013',
 211             'md5': '52f0bfe202848b15915a2f39aaa8981b',
 212             'info_dict': {
 213                 'id': '108634970',
 214                 'ext': 'flv',
 215                 'title': 'Infô Afrique',
 216                 'description': 'md5:ebf346da789428841bee0fd2a935ea55',
 217                 'upload_date': '20140915',
 218                 'timestamp': 1410822000,
 219             },
 220         },
 221     ]
 222
 223     def _real_extract(self, url):
 224         mobj = re.match(self._VALID_URL, url)
 225         webpage = self._download_webpage(url, mobj.group('key') or mobj.group('id'))
 226         video_id, catalogue = self._html_search_regex(
 227             r'href="http://videos\.francetv\.fr/video/([^@]+@[^"]+)"',
 228             webpage, 'video ID').split('@')
 229         return self._extract_video(video_id, catalogue)
 230
 231
 232 class GenerationQuoiIE(InfoExtractor):
 233     IE_NAME = 'france2.fr:generation-quoi'
 234     _VALID_URL = r'https?://generation-quoi\.france2\.fr/portrait/(?P<id>[^/?#]+)'
 235
 236     _TEST = {
 237         'url': 'http://generation-quoi.france2.fr/portrait/garde-a-vous',
 238         'info_dict': {
 239             'id': 'k7FJX8VBcvvLmX4wA5Q',
 240             'ext': 'mp4',
 241             'title': 'Génération Quoi - Garde à Vous',
 242             'uploader': 'Génération Quoi',
 243         },
 244         'params': {
 245             # It uses Dailymotion
 246             'skip_download': True,
 247         },
 248     }
 249
 250     def _real_extract(self, url):
 251         display_id = self._match_id(url)
 252         info_url = compat_urlparse.urljoin(url, '/medias/video/%s.json' % display_id)
 253         info_json = self._download_webpage(info_url, display_id)
 254         info = json.loads(info_json)
 255         return self.url_result('http://www.dailymotion.com/video/%s' % info['id'],
 256                                ie='Dailymotion')
 257
 258
 259 class CultureboxIE(FranceTVBaseInfoExtractor):
 260     IE_NAME = 'culturebox.francetvinfo.fr'
 261     _VALID_URL = r'https?://(?:m\.)?culturebox\.francetvinfo\.fr/(?P<name>.*?)(\?|$)'
 262
 263     _TEST = {
 264         'url': 'http://culturebox.francetvinfo.fr/live/musique/musique-classique/le-livre-vermeil-de-montserrat-a-la-cathedrale-delne-214511',
 265         'info_dict': {
 266             'id': 'EV_50111',
 267             'ext': 'mp4',
 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         'params': {
 275             'skip_download': True,
 276         },
 277     }
 278
 279     def _real_extract(self, url):
 280         mobj = re.match(self._VALID_URL, url)
 281         name = mobj.group('name')
 282
 283         webpage = self._download_webpage(url, name)
 284
 285         if ">Ce live n'est plus disponible en replay<" in webpage:
 286             raise ExtractorError('Video %s is not available' % name, expected=True)
 287
 288         video_id, catalogue = self._search_regex(
 289             r'"http://videos\.francetv\.fr/video/([^@]+@[^"]+)"', webpage, 'video id').split('@')
 290
 291         return self._extract_video(video_id, catalogue)