youtube_dl/extractor/mitele.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import uuid
   5
   6 from .common import InfoExtractor
   7 from .ooyala import OoyalaIE
   8 from ..compat import (
   9     compat_str,
  10     compat_urllib_parse_urlencode,
  11     compat_urlparse,
  12 )
  13 from ..utils import (
  14     int_or_none,
  15     extract_attributes,
  16     determine_ext,
  17     smuggle_url,
  18     parse_duration,
  19 )
  20
  21
  22 class MiTeleBaseIE(InfoExtractor):
  23     def _get_player_info(self, url, webpage):
  24         player_data = extract_attributes(self._search_regex(
  25             r'(?s)(<ms-video-player.+?</ms-video-player>)',
  26             webpage, 'ms video player'))
  27         video_id = player_data['data-media-id']
  28         if player_data.get('data-cms-id') == 'ooyala':
  29             return self.url_result(
  30                 'ooyala:%s' % video_id, ie=OoyalaIE.ie_key(), video_id=video_id)
  31         config_url = compat_urlparse.urljoin(url, player_data['data-config'])
  32         config = self._download_json(
  33             config_url, video_id, 'Downloading config JSON')
  34         mmc_url = config['services']['mmc']
  35
  36         duration = None
  37         formats = []
  38         for m_url in (mmc_url, mmc_url.replace('/flash.json', '/html5.json')):
  39             mmc = self._download_json(
  40                 m_url, video_id, 'Downloading mmc JSON')
  41             if not duration:
  42                 duration = int_or_none(mmc.get('duration'))
  43             for location in mmc['locations']:
  44                 gat = self._proto_relative_url(location.get('gat'), 'http:')
  45                 bas = location.get('bas')
  46                 loc = location.get('loc')
  47                 ogn = location.get('ogn')
  48                 if None in (gat, bas, loc, ogn):
  49                     continue
  50                 token_data = {
  51                     'bas': bas,
  52                     'icd': loc,
  53                     'ogn': ogn,
  54                     'sta': '0',
  55                 }
  56                 media = self._download_json(
  57                     '%s/?%s' % (gat, compat_urllib_parse_urlencode(token_data)),
  58                     video_id, 'Downloading %s JSON' % location['loc'])
  59                 file_ = media.get('file')
  60                 if not file_:
  61                     continue
  62                 ext = determine_ext(file_)
  63                 if ext == 'f4m':
  64                     formats.extend(self._extract_f4m_formats(
  65                         file_ + '&hdcore=3.2.0&plugin=aasp-3.2.0.77.18',
  66                         video_id, f4m_id='hds', fatal=False))
  67                 elif ext == 'm3u8':
  68                     formats.extend(self._extract_m3u8_formats(
  69                         file_, video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False))
  70         self._sort_formats(formats)
  71
  72         return {
  73             'id': video_id,
  74             'formats': formats,
  75             'thumbnail': player_data.get('data-poster') or config.get('poster', {}).get('imageUrl'),
  76             'duration': duration,
  77         }
  78
  79
  80 class MiTeleIE(InfoExtractor):
  81     IE_DESC = 'mitele.es'
  82     _VALID_URL = r'https?://(?:www\.)?mitele\.es/(?:[^/]+/)+(?P<id>[^/]+)/player'
  83
  84     _TESTS = [{
  85         'url': 'http://www.mitele.es/programas-tv/diario-de/57b0dfb9c715da65618b4afa/player',
  86         'info_dict': {
  87             'id': '57b0dfb9c715da65618b4afa',
  88             'ext': 'mp4',
  89             'title': 'Tor, la web invisible',
  90             'description': 'md5:3b6fce7eaa41b2d97358726378d9369f',
  91             'series': 'Diario de',
  92             'season': 'La redacción',
  93             'season_number': 14,
  94             'season_id': 'diario_de_t14_11981',
  95             'episode': 'Programa 144',
  96             'episode_number': 3,
  97             'thumbnail': r're:(?i)^https?://.*\.jpg$',
  98             'duration': 2913,
  99         },
 100         'add_ie': ['Ooyala'],
 101     }, {
 102         # no explicit title
 103         'url': 'http://www.mitele.es/programas-tv/cuarto-milenio/57b0de3dc915da14058b4876/player',
 104         'info_dict': {
 105             'id': '57b0de3dc915da14058b4876',
 106             'ext': 'mp4',
 107             'title': 'Cuarto Milenio Temporada 6 Programa 226',
 108             'description': 'md5:5ff132013f0cd968ffbf1f5f3538a65f',
 109             'series': 'Cuarto Milenio',
 110             'season': 'Temporada 6',
 111             'season_number': 6,
 112             'season_id': 'cuarto_milenio_t06_12715',
 113             'episode': 'Programa 226',
 114             'episode_number': 24,
 115             'thumbnail': r're:(?i)^https?://.*\.jpg$',
 116             'duration': 7313,
 117         },
 118         'params': {
 119             'skip_download': True,
 120         },
 121         'add_ie': ['Ooyala'],
 122     }, {
 123         'url': 'http://www.mitele.es/series-online/la-que-se-avecina/57aac5c1c915da951a8b45ed/player',
 124         'only_matching': True,
 125     }]
 126
 127     def _real_extract(self, url):
 128         video_id = self._match_id(url)
 129         webpage = self._download_webpage(url, video_id)
 130
 131         gigya_url = self._search_regex(
 132             r'<gigya-api>[^>]*</gigya-api>[^>]*<script\s+src="([^"]*)">[^>]*</script>',
 133             webpage, 'gigya', default=None)
 134         gigya_sc = self._download_webpage(
 135             compat_urlparse.urljoin('http://www.mitele.es/', gigya_url),
 136             video_id, 'Downloading gigya script')
 137
 138         # Get a appKey/uuid for getting the session key
 139         appKey_var = self._search_regex(
 140             r'value\s*\(\s*["\']appGridApplicationKey["\']\s*,\s*([0-9a-f]+)',
 141             gigya_sc, 'appKey variable')
 142         appKey = self._search_regex(
 143             r'var\s+%s\s*=\s*["\']([0-9a-f]+)' % appKey_var, gigya_sc, 'appKey')
 144
 145         session_json = self._download_json(
 146             'https://appgrid-api.cloud.accedo.tv/session',
 147             video_id, 'Downloading session keys', query={
 148                 'appKey': appKey,
 149                 'uuid': compat_str(uuid.uuid4()),
 150             })
 151
 152         paths = self._download_json(
 153             'https://appgrid-api.cloud.accedo.tv/metadata/general_configuration,%20web_configuration',
 154             video_id, 'Downloading paths JSON',
 155             query={'sessionKey': compat_str(session_json['sessionKey'])})
 156
 157         ooyala_s = paths['general_configuration']['api_configuration']['ooyala_search']
 158         source = self._download_json(
 159             'http://%s%s%s/docs/%s' % (
 160                 ooyala_s['base_url'], ooyala_s['full_path'],
 161                 ooyala_s['provider_id'], video_id),
 162             video_id, 'Downloading data JSON', query={
 163                 'include_titles': 'Series,Season',
 164                 'product_name': 'test',
 165                 'format': 'full',
 166             })['hits']['hits'][0]['_source']
 167
 168         embedCode = source['offers'][0]['embed_codes'][0]
 169         titles = source['localizable_titles'][0]
 170
 171         title = titles.get('title_medium') or titles['title_long']
 172
 173         description = titles.get('summary_long') or titles.get('summary_medium')
 174
 175         def get(key1, key2):
 176             value1 = source.get(key1)
 177             if not value1 or not isinstance(value1, list):
 178                 return
 179             if not isinstance(value1[0], dict):
 180                 return
 181             return value1[0].get(key2)
 182
 183         series = get('localizable_titles_series', 'title_medium')
 184
 185         season = get('localizable_titles_season', 'title_medium')
 186         season_number = int_or_none(source.get('season_number'))
 187         season_id = source.get('season_id')
 188
 189         episode = titles.get('title_sort_name')
 190         episode_number = int_or_none(source.get('episode_number'))
 191
 192         duration = parse_duration(get('videos', 'duration'))
 193
 194         return {
 195             '_type': 'url_transparent',
 196             # for some reason only HLS is supported
 197             'url': smuggle_url('ooyala:' + embedCode, {'supportedformats': 'm3u8,dash'}),
 198             'id': video_id,
 199             'title': title,
 200             'description': description,
 201             'series': series,
 202             'season': season,
 203             'season_number': season_number,
 204             'season_id': season_id,
 205             'episode': episode,
 206             'episode_number': episode_number,
 207             'duration': duration,
 208             'thumbnail': get('images', 'url'),
 209         }