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

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..compat import compat_str
   7 from ..utils import (
   8     ExtractorError,
   9     int_or_none,
  10     parse_duration,
  11     try_get,
  12     unified_timestamp,
  13     urlencode_postdata,
  14 )
  15
  16
  17 class MedialaanIE(InfoExtractor):
  18     _VALID_URL = r'''(?x)
  19                     https?://
  20                         (?:www\.)?
  21                         (?:
  22                             (?P<site_id>vtm|q2|vtmkzoom)\.be/
  23                             (?:
  24                                 video(?:/[^/]+/id/|/?\?.*?\baid=)|
  25                                 (?:[^/]+/)*
  26                             )
  27                         )
  28                         (?P<id>[^/?#&]+)
  29                     '''
  30     _NETRC_MACHINE = 'medialaan'
  31     _APIKEY = '3_HZ0FtkMW_gOyKlqQzW5_0FHRC7Nd5XpXJZcDdXY4pk5eES2ZWmejRW5egwVm4ug-'
  32     _SITE_TO_APP_ID = {
  33         'vtm': 'vtm_watch',
  34         'q2': 'q2',
  35         'vtmkzoom': 'vtmkzoom',
  36     }
  37     _TESTS = [{
  38         # vod
  39         'url': 'http://vtm.be/video/volledige-afleveringen/id/vtm_20170219_VM0678361_vtmwatch',
  40         'info_dict': {
  41             'id': 'vtm_20170219_VM0678361_vtmwatch',
  42             'ext': 'mp4',
  43             'title': 'Allemaal Chris afl. 6',
  44             'description': 'md5:4be86427521e7b07e0adb0c9c554ddb2',
  45             'timestamp': 1487533280,
  46             'upload_date': '20170219',
  47             'duration': 2562,
  48             'series': 'Allemaal Chris',
  49             'season': 'Allemaal Chris',
  50             'season_number': 1,
  51             'season_id': '256936078124527',
  52             'episode': 'Allemaal Chris afl. 6',
  53             'episode_number': 6,
  54             'episode_id': '256936078591527',
  55         },
  56         'params': {
  57             'skip_download': True,
  58         },
  59         'skip': 'Requires account credentials',
  60     }, {
  61         # clip
  62         'url': 'http://vtm.be/video?aid=168332',
  63         'info_dict': {
  64             'id': '168332',
  65             'ext': 'mp4',
  66             'title': '"Veronique liegt!"',
  67             'description': 'md5:1385e2b743923afe54ba4adc38476155',
  68             'timestamp': 1489002029,
  69             'upload_date': '20170308',
  70             'duration': 96,
  71         },
  72     }, {
  73         # vod
  74         'url': 'http://vtm.be/video/volledige-afleveringen/id/257107153551000',
  75         'only_matching': True,
  76     }, {
  77         # vod
  78         'url': 'http://vtm.be/video?aid=163157',
  79         'only_matching': True,
  80     }, {
  81         # vod
  82         'url': 'http://www.q2.be/video/volledige-afleveringen/id/2be_20170301_VM0684442_q2',
  83         'only_matching': True,
  84     }, {
  85         # clip
  86         'url': 'http://vtmkzoom.be/k3-dansstudio/een-nieuw-seizoen-van-k3-dansstudio',
  87         'only_matching': True,
  88     }]
  89
  90     def _real_initialize(self):
  91         self._logged_in = False
  92
  93     def _login(self):
  94         username, password = self._get_login_info()
  95         if username is None:
  96             self.raise_login_required()
  97
  98         auth_data = {
  99             'APIKey': self._APIKEY,
 100             'sdk': 'js_6.1',
 101             'format': 'json',
 102             'loginID': username,
 103             'password': password,
 104         }
 105
 106         auth_info = self._download_json(
 107             'https://accounts.eu1.gigya.com/accounts.login', None,
 108             note='Logging in', errnote='Unable to log in',
 109             data=urlencode_postdata(auth_data))
 110
 111         error_message = auth_info.get('errorDetails') or auth_info.get('errorMessage')
 112         if error_message:
 113             raise ExtractorError(
 114                 'Unable to login: %s' % error_message, expected=True)
 115
 116         self._uid = auth_info['UID']
 117         self._uid_signature = auth_info['UIDSignature']
 118         self._signature_timestamp = auth_info['signatureTimestamp']
 119
 120         self._logged_in = True
 121
 122     def _real_extract(self, url):
 123         mobj = re.match(self._VALID_URL, url)
 124         video_id, site_id = mobj.group('id', 'site_id')
 125
 126         webpage = self._download_webpage(url, video_id)
 127
 128         config = self._parse_json(
 129             self._search_regex(
 130                 r'videoJSConfig\s*=\s*JSON\.parse\(\'({.+?})\'\);',
 131                 webpage, 'config', default='{}'), video_id,
 132             transform_source=lambda s: s.replace(
 133                 '\\\\', '\\').replace(r'\"', '"').replace(r"\'", "'"))
 134
 135         vod_id = config.get('vodId') or self._search_regex(
 136             (r'\\"vodId\\"\s*:\s*\\"(.+?)\\"',
 137              r'<[^>]+id=["\']vod-(\d+)'),
 138             webpage, 'video_id', default=None)
 139
 140         # clip, no authentication required
 141         if not vod_id:
 142             player = self._parse_json(
 143                 self._search_regex(
 144                     r'vmmaplayer\(({.+?})\);', webpage, 'vmma player',
 145                     default=''),
 146                 video_id, transform_source=lambda s: '[%s]' % s, fatal=False)
 147             if player:
 148                 video = player[-1]
 149                 info = {
 150                     'id': video_id,
 151                     'url': video['videoUrl'],
 152                     'title': video['title'],
 153                     'thumbnail': video.get('imageUrl'),
 154                     'timestamp': int_or_none(video.get('createdDate')),
 155                     'duration': int_or_none(video.get('duration')),
 156                 }
 157             else:
 158                 info = self._parse_html5_media_entries(
 159                     url, webpage, video_id, m3u8_id='hls')[0]
 160                 info.update({
 161                     'id': video_id,
 162                     'title': self._html_search_meta('description', webpage),
 163                     'duration': parse_duration(self._html_search_meta('duration', webpage)),
 164                 })
 165         # vod, authentication required
 166         else:
 167             if not self._logged_in:
 168                 self._login()
 169
 170             settings = self._parse_json(
 171                 self._search_regex(
 172                     r'jQuery\.extend\(Drupal\.settings\s*,\s*({.+?})\);',
 173                     webpage, 'drupal settings', default='{}'),
 174                 video_id)
 175
 176             def get(container, item):
 177                 return try_get(
 178                     settings, lambda x: x[container][item],
 179                     compat_str) or self._search_regex(
 180                     r'"%s"\s*:\s*"([^"]+)' % item, webpage, item,
 181                     default=None)
 182
 183             app_id = get('vod', 'app_id') or self._SITE_TO_APP_ID.get(site_id, 'vtm_watch')
 184             sso = get('vod', 'gigyaDatabase') or 'vtm-sso'
 185
 186             data = self._download_json(
 187                 'http://vod.medialaan.io/api/1.0/item/%s/video' % vod_id,
 188                 video_id, query={
 189                     'app_id': app_id,
 190                     'user_network': sso,
 191                     'UID': self._uid,
 192                     'UIDSignature': self._uid_signature,
 193                     'signatureTimestamp': self._signature_timestamp,
 194                 })
 195
 196             formats = self._extract_m3u8_formats(
 197                 data['response']['uri'], video_id, entry_protocol='m3u8_native',
 198                 ext='mp4', m3u8_id='hls')
 199
 200             self._sort_formats(formats)
 201
 202             info = {
 203                 'id': vod_id,
 204                 'formats': formats,
 205             }
 206
 207             api_key = get('vod', 'apiKey')
 208             channel = get('medialaanGigya', 'channel')
 209
 210             if api_key:
 211                 videos = self._download_json(
 212                     'http://vod.medialaan.io/vod/v2/videos', video_id, fatal=False,
 213                     query={
 214                         'channels': channel,
 215                         'ids': vod_id,
 216                         'limit': 1,
 217                         'apikey': api_key,
 218                     })
 219                 if videos:
 220                     video = try_get(
 221                         videos, lambda x: x['response']['videos'][0], dict)
 222                     if video:
 223                         def get(container, item, expected_type=None):
 224                             return try_get(
 225                                 video, lambda x: x[container][item], expected_type)
 226
 227                         def get_string(container, item):
 228                             return get(container, item, compat_str)
 229
 230                         info.update({
 231                             'series': get_string('program', 'title'),
 232                             'season': get_string('season', 'title'),
 233                             'season_number': int_or_none(get('season', 'number')),
 234                             'season_id': get_string('season', 'id'),
 235                             'episode': get_string('episode', 'title'),
 236                             'episode_number': int_or_none(get('episode', 'number')),
 237                             'episode_id': get_string('episode', 'id'),
 238                             'duration': int_or_none(
 239                                 video.get('duration')) or int_or_none(
 240                                 video.get('durationMillis'), scale=1000),
 241                             'title': get_string('episode', 'title'),
 242                             'description': get_string('episode', 'text'),
 243                             'timestamp': unified_timestamp(get_string(
 244                                 'publication', 'begin')),
 245                         })
 246
 247             if not info.get('title'):
 248                 info['title'] = try_get(
 249                     config, lambda x: x['videoConfig']['title'],
 250                     compat_str) or self._html_search_regex(
 251                     r'\\"title\\"\s*:\s*\\"(.+?)\\"', webpage, 'title',
 252                     default=None) or self._og_search_title(webpage)
 253
 254         if not info.get('description'):
 255             info['description'] = self._html_search_regex(
 256                 r'<div[^>]+class="field-item\s+even">\s*<p>(.+?)</p>',
 257                 webpage, 'description', default=None)
 258
 259         return info