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\.|nieuws\.)?
  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         # http/s redirect
  90         'url': 'https://vtmkzoom.be/video?aid=45724',
  91         'info_dict': {
  92             'id': '257136373657000',
  93             'ext': 'mp4',
  94             'title': 'K3 Dansstudio Ushuaia afl.6',
  95         },
  96         'params': {
  97             'skip_download': True,
  98         },
  99         'skip': 'Requires account credentials',
 100     }, {
 101         # nieuws.vtm.be
 102         'url': 'https://nieuws.vtm.be/stadion/stadion/genk-nog-moeilijk-programma',
 103         'only_matching': True,
 104     }]
 105
 106     def _real_initialize(self):
 107         self._logged_in = False
 108
 109     def _login(self):
 110         username, password = self._get_login_info()
 111         if username is None:
 112             self.raise_login_required()
 113
 114         auth_data = {
 115             'APIKey': self._APIKEY,
 116             'sdk': 'js_6.1',
 117             'format': 'json',
 118             'loginID': username,
 119             'password': password,
 120         }
 121
 122         auth_info = self._download_json(
 123             'https://accounts.eu1.gigya.com/accounts.login', None,
 124             note='Logging in', errnote='Unable to log in',
 125             data=urlencode_postdata(auth_data))
 126
 127         error_message = auth_info.get('errorDetails') or auth_info.get('errorMessage')
 128         if error_message:
 129             raise ExtractorError(
 130                 'Unable to login: %s' % error_message, expected=True)
 131
 132         self._uid = auth_info['UID']
 133         self._uid_signature = auth_info['UIDSignature']
 134         self._signature_timestamp = auth_info['signatureTimestamp']
 135
 136         self._logged_in = True
 137
 138     def _real_extract(self, url):
 139         mobj = re.match(self._VALID_URL, url)
 140         video_id, site_id = mobj.group('id', 'site_id')
 141
 142         webpage = self._download_webpage(url, video_id)
 143
 144         config = self._parse_json(
 145             self._search_regex(
 146                 r'videoJSConfig\s*=\s*JSON\.parse\(\'({.+?})\'\);',
 147                 webpage, 'config', default='{}'), video_id,
 148             transform_source=lambda s: s.replace(
 149                 '\\\\', '\\').replace(r'\"', '"').replace(r"\'", "'"))
 150
 151         vod_id = config.get('vodId') or self._search_regex(
 152             (r'\\"vodId\\"\s*:\s*\\"(.+?)\\"',
 153              r'<[^>]+id=["\']vod-(\d+)'),
 154             webpage, 'video_id', default=None)
 155
 156         # clip, no authentication required
 157         if not vod_id:
 158             player = self._parse_json(
 159                 self._search_regex(
 160                     r'vmmaplayer\(({.+?})\);', webpage, 'vmma player',
 161                     default=''),
 162                 video_id, transform_source=lambda s: '[%s]' % s, fatal=False)
 163             if player:
 164                 video = player[-1]
 165                 if video['videoUrl'] in ('http', 'https'):
 166                     return self.url_result(video['url'], MedialaanIE.ie_key())
 167                 info = {
 168                     'id': video_id,
 169                     'url': video['videoUrl'],
 170                     'title': video['title'],
 171                     'thumbnail': video.get('imageUrl'),
 172                     'timestamp': int_or_none(video.get('createdDate')),
 173                     'duration': int_or_none(video.get('duration')),
 174                 }
 175             else:
 176                 info = self._parse_html5_media_entries(
 177                     url, webpage, video_id, m3u8_id='hls')[0]
 178                 info.update({
 179                     'id': video_id,
 180                     'title': self._html_search_meta('description', webpage),
 181                     'duration': parse_duration(self._html_search_meta('duration', webpage)),
 182                 })
 183         # vod, authentication required
 184         else:
 185             if not self._logged_in:
 186                 self._login()
 187
 188             settings = self._parse_json(
 189                 self._search_regex(
 190                     r'jQuery\.extend\(Drupal\.settings\s*,\s*({.+?})\);',
 191                     webpage, 'drupal settings', default='{}'),
 192                 video_id)
 193
 194             def get(container, item):
 195                 return try_get(
 196                     settings, lambda x: x[container][item],
 197                     compat_str) or self._search_regex(
 198                     r'"%s"\s*:\s*"([^"]+)' % item, webpage, item,
 199                     default=None)
 200
 201             app_id = get('vod', 'app_id') or self._SITE_TO_APP_ID.get(site_id, 'vtm_watch')
 202             sso = get('vod', 'gigyaDatabase') or 'vtm-sso'
 203
 204             data = self._download_json(
 205                 'http://vod.medialaan.io/api/1.0/item/%s/video' % vod_id,
 206                 video_id, query={
 207                     'app_id': app_id,
 208                     'user_network': sso,
 209                     'UID': self._uid,
 210                     'UIDSignature': self._uid_signature,
 211                     'signatureTimestamp': self._signature_timestamp,
 212                 })
 213
 214             formats = self._extract_m3u8_formats(
 215                 data['response']['uri'], video_id, entry_protocol='m3u8_native',
 216                 ext='mp4', m3u8_id='hls')
 217
 218             self._sort_formats(formats)
 219
 220             info = {
 221                 'id': vod_id,
 222                 'formats': formats,
 223             }
 224
 225             api_key = get('vod', 'apiKey')
 226             channel = get('medialaanGigya', 'channel')
 227
 228             if api_key:
 229                 videos = self._download_json(
 230                     'http://vod.medialaan.io/vod/v2/videos', video_id, fatal=False,
 231                     query={
 232                         'channels': channel,
 233                         'ids': vod_id,
 234                         'limit': 1,
 235                         'apikey': api_key,
 236                     })
 237                 if videos:
 238                     video = try_get(
 239                         videos, lambda x: x['response']['videos'][0], dict)
 240                     if video:
 241                         def get(container, item, expected_type=None):
 242                             return try_get(
 243                                 video, lambda x: x[container][item], expected_type)
 244
 245                         def get_string(container, item):
 246                             return get(container, item, compat_str)
 247
 248                         info.update({
 249                             'series': get_string('program', 'title'),
 250                             'season': get_string('season', 'title'),
 251                             'season_number': int_or_none(get('season', 'number')),
 252                             'season_id': get_string('season', 'id'),
 253                             'episode': get_string('episode', 'title'),
 254                             'episode_number': int_or_none(get('episode', 'number')),
 255                             'episode_id': get_string('episode', 'id'),
 256                             'duration': int_or_none(
 257                                 video.get('duration')) or int_or_none(
 258                                 video.get('durationMillis'), scale=1000),
 259                             'title': get_string('episode', 'title'),
 260                             'description': get_string('episode', 'text'),
 261                             'timestamp': unified_timestamp(get_string(
 262                                 'publication', 'begin')),
 263                         })
 264
 265             if not info.get('title'):
 266                 info['title'] = try_get(
 267                     config, lambda x: x['videoConfig']['title'],
 268                     compat_str) or self._html_search_regex(
 269                     r'\\"title\\"\s*:\s*\\"(.+?)\\"', webpage, 'title',
 270                     default=None) or self._og_search_title(webpage)
 271
 272         if not info.get('description'):
 273             info['description'] = self._html_search_regex(
 274                 r'<div[^>]+class="field-item\s+even">\s*<p>(.+?)</p>',
 275                 webpage, 'description', default=None)
 276
 277         return info