youtube_dl/extractor/canvas.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4 import json
   5
   6 from .common import InfoExtractor
   7 from .gigya import GigyaBaseIE
   8
   9
  10 from ..compat import compat_HTTPError
  11 from ..utils import (
  12     ExtractorError,
  13     strip_or_none,
  14     float_or_none,
  15     int_or_none,
  16     parse_iso8601,
  17 )
  18
  19
  20 class CanvasIE(InfoExtractor):
  21     _VALID_URL = r'https?://mediazone\.vrt\.be/api/v1/(?P<site_id>canvas|een|ketnet|vrtvideo)/assets/(?P<id>[^/?#&]+)'
  22     _TESTS = [{
  23         'url': 'https://mediazone.vrt.be/api/v1/ketnet/assets/md-ast-4ac54990-ce66-4d00-a8ca-9eac86f4c475',
  24         'md5': '90139b746a0a9bd7bb631283f6e2a64e',
  25         'info_dict': {
  26             'id': 'md-ast-4ac54990-ce66-4d00-a8ca-9eac86f4c475',
  27             'display_id': 'md-ast-4ac54990-ce66-4d00-a8ca-9eac86f4c475',
  28             'ext': 'flv',
  29             'title': 'Nachtwacht: De Greystook',
  30             'description': 'md5:1db3f5dc4c7109c821261e7512975be7',
  31             'thumbnail': r're:^https?://.*\.jpg$',
  32             'duration': 1468.03,
  33         },
  34         'expected_warnings': ['is not a supported codec', 'Unknown MIME type'],
  35     }, {
  36         'url': 'https://mediazone.vrt.be/api/v1/canvas/assets/mz-ast-5e5f90b6-2d72-4c40-82c2-e134f884e93e',
  37         'only_matching': True,
  38     }]
  39
  40     def _real_extract(self, url):
  41         mobj = re.match(self._VALID_URL, url)
  42         site_id, video_id = mobj.group('site_id'), mobj.group('id')
  43
  44         data = self._download_json(
  45             'https://mediazone.vrt.be/api/v1/%s/assets/%s'
  46             % (site_id, video_id), video_id)
  47
  48         title = data['title']
  49         description = data.get('description')
  50
  51         formats = []
  52         for target in data['targetUrls']:
  53             format_url, format_type = target.get('url'), target.get('type')
  54             if not format_url or not format_type:
  55                 continue
  56             if format_type == 'HLS':
  57                 formats.extend(self._extract_m3u8_formats(
  58                     format_url, video_id, 'mp4', entry_protocol='m3u8_native',
  59                     m3u8_id=format_type, fatal=False))
  60             elif format_type == 'HDS':
  61                 formats.extend(self._extract_f4m_formats(
  62                     format_url, video_id, f4m_id=format_type, fatal=False))
  63             elif format_type == 'MPEG_DASH':
  64                 formats.extend(self._extract_mpd_formats(
  65                     format_url, video_id, mpd_id=format_type, fatal=False))
  66             elif format_type == 'HSS':
  67                 formats.extend(self._extract_ism_formats(
  68                     format_url, video_id, ism_id='mss', fatal=False))
  69             else:
  70                 formats.append({
  71                     'format_id': format_type,
  72                     'url': format_url,
  73                 })
  74         self._sort_formats(formats)
  75
  76         subtitles = {}
  77         subtitle_urls = data.get('subtitleUrls')
  78         if isinstance(subtitle_urls, list):
  79             for subtitle in subtitle_urls:
  80                 subtitle_url = subtitle.get('url')
  81                 if subtitle_url and subtitle.get('type') == 'CLOSED':
  82                     subtitles.setdefault('nl', []).append({'url': subtitle_url})
  83
  84         return {
  85             'id': video_id,
  86             'display_id': video_id,
  87             'title': title,
  88             'description': description,
  89             'formats': formats,
  90             'duration': float_or_none(data.get('duration'), 1000),
  91             'thumbnail': data.get('posterImageUrl'),
  92             'subtitles': subtitles,
  93         }
  94
  95
  96 class CanvasEenIE(InfoExtractor):
  97     IE_DESC = 'canvas.be and een.be'
  98     _VALID_URL = r'https?://(?:www\.)?(?P<site_id>canvas|een)\.be/(?:[^/]+/)*(?P<id>[^/?#&]+)'
  99     _TESTS = [{
 100         'url': 'http://www.canvas.be/video/de-afspraak/najaar-2015/de-afspraak-veilt-voor-de-warmste-week',
 101         'md5': 'ed66976748d12350b118455979cca293',
 102         'info_dict': {
 103             'id': 'mz-ast-5e5f90b6-2d72-4c40-82c2-e134f884e93e',
 104             'display_id': 'de-afspraak-veilt-voor-de-warmste-week',
 105             'ext': 'flv',
 106             'title': 'De afspraak veilt voor de Warmste Week',
 107             'description': 'md5:24cb860c320dc2be7358e0e5aa317ba6',
 108             'thumbnail': r're:^https?://.*\.jpg$',
 109             'duration': 49.02,
 110         },
 111         'expected_warnings': ['is not a supported codec'],
 112     }, {
 113         # with subtitles
 114         'url': 'http://www.canvas.be/video/panorama/2016/pieter-0167',
 115         'info_dict': {
 116             'id': 'mz-ast-5240ff21-2d30-4101-bba6-92b5ec67c625',
 117             'display_id': 'pieter-0167',
 118             'ext': 'mp4',
 119             'title': 'Pieter 0167',
 120             'description': 'md5:943cd30f48a5d29ba02c3a104dc4ec4e',
 121             'thumbnail': r're:^https?://.*\.jpg$',
 122             'duration': 2553.08,
 123             'subtitles': {
 124                 'nl': [{
 125                     'ext': 'vtt',
 126                 }],
 127             },
 128         },
 129         'params': {
 130             'skip_download': True,
 131         },
 132         'skip': 'Pagina niet gevonden',
 133     }, {
 134         'url': 'https://www.een.be/sorry-voor-alles/herbekijk-sorry-voor-alles',
 135         'info_dict': {
 136             'id': 'mz-ast-11a587f8-b921-4266-82e2-0bce3e80d07f',
 137             'display_id': 'herbekijk-sorry-voor-alles',
 138             'ext': 'mp4',
 139             'title': 'Herbekijk Sorry voor alles',
 140             'description': 'md5:8bb2805df8164e5eb95d6a7a29dc0dd3',
 141             'thumbnail': r're:^https?://.*\.jpg$',
 142             'duration': 3788.06,
 143         },
 144         'params': {
 145             'skip_download': True,
 146         },
 147         'skip': 'Episode no longer available',
 148     }, {
 149         'url': 'https://www.canvas.be/check-point/najaar-2016/de-politie-uw-vriend',
 150         'only_matching': True,
 151     }]
 152
 153     def _real_extract(self, url):
 154         mobj = re.match(self._VALID_URL, url)
 155         site_id, display_id = mobj.group('site_id'), mobj.group('id')
 156
 157         webpage = self._download_webpage(url, display_id)
 158
 159         title = strip_or_none(self._search_regex(
 160             r'<h1[^>]+class="video__body__header__title"[^>]*>(.+?)</h1>',
 161             webpage, 'title', default=None) or self._og_search_title(
 162             webpage, default=None))
 163
 164         video_id = self._html_search_regex(
 165             r'data-video=(["\'])(?P<id>(?:(?!\1).)+)\1', webpage, 'video id',
 166             group='id')
 167
 168         return {
 169             '_type': 'url_transparent',
 170             'url': 'https://mediazone.vrt.be/api/v1/%s/assets/%s' % (site_id, video_id),
 171             'ie_key': CanvasIE.ie_key(),
 172             'id': video_id,
 173             'display_id': display_id,
 174             'title': title,
 175             'description': self._og_search_description(webpage),
 176         }
 177
 178
 179 class VrtNUIE(GigyaBaseIE):
 180     IE_DESC = 'VrtNU.be'
 181     _VALID_URL = r'https?://(?:www\.)?vrt\.be/(?P<site_id>vrtnu)/(?:[^/]+/)*(?P<id>[^/?#&]+)'
 182     _TESTS = [{
 183         'url': 'https://www.vrt.be/vrtnu/a-z/postbus-x/1/postbus-x-s1a1/',
 184         'info_dict': {
 185             'id': 'pbs-pub-2e2d8c27-df26-45c9-9dc6-90c78153044d$vid-90c932b1-e21d-4fb8-99b1-db7b49cf74de',
 186             'ext': 'flv',
 187             'title': 'De zwarte weduwe',
 188             'description': 'md5:d90c21dced7db869a85db89a623998d4',
 189             'duration': 1457.04,
 190             'thumbnail': r're:^https?://.*\.jpg$',
 191             'season': '1',
 192             'season_number': 1,
 193             'episode_number': 1,
 194         },
 195         # 'skip': 'This video is only available for registered users'
 196     }]
 197     _NETRC_MACHINE = 'vrtnu'
 198     _APIKEY = '3_0Z2HujMtiWq_pkAjgnS2Md2E11a1AwZjYiBETtwNE-EoEHDINgtnvcAOpNgmrVGy'
 199     _CONTEXT_ID = 'R3595707040'
 200
 201     def _real_initialize(self):
 202         self._login()
 203
 204     def _login(self):
 205         username, password = self._get_login_info()
 206         if username is None:
 207             self.raise_login_required()
 208
 209         auth_data = {
 210             'APIKey': self._APIKEY,
 211             'targetEnv': 'jssdk',
 212             'loginID': username,
 213             'password': password,
 214             'authMode': 'cookie',
 215         }
 216
 217         auth_info = self._gigya_login(auth_data)
 218
 219         # Sometimes authentication fails for no good reason, retry
 220         login_attempt = 1
 221         while login_attempt <= 3:
 222             try:
 223                 # When requesting a token, no actual token is returned, but the
 224                 # necessary cookies are set.
 225                 self._request_webpage(
 226                     'https://token.vrt.be',
 227                     None, note='Requesting a token', errnote='Could not get a token',
 228                     headers={
 229                         'Content-Type': 'application/json',
 230                         'Referer': 'https://www.vrt.be/vrtnu/',
 231                     },
 232                     data=json.dumps({
 233                         'uid': auth_info['UID'],
 234                         'uidsig': auth_info['UIDSignature'],
 235                         'ts': auth_info['signatureTimestamp'],
 236                         'email': auth_info['profile']['email'],
 237                     }).encode('utf-8'))
 238             except ExtractorError as e:
 239                 if isinstance(e.cause, compat_HTTPError) and e.cause.code == 401:
 240                     login_attempt += 1
 241                     self.report_warning('Authentication failed')
 242                     self._sleep(1, None, msg_template='Waiting for %(timeout)s seconds before trying again')
 243                 else:
 244                     raise e
 245             else:
 246                 break
 247
 248     def _real_extract(self, url):
 249         display_id = self._match_id(url)
 250
 251         webpage = self._download_webpage(url, display_id)
 252
 253         title = self._html_search_regex(
 254             r'(?ms)<h1 class="content__heading">(.+?)</h1>',
 255             webpage, 'title').strip()
 256
 257         description = self._html_search_regex(
 258             r'(?ms)<div class="content__description">(.+?)</div>',
 259             webpage, 'description', default=None)
 260
 261         season = self._html_search_regex(
 262             [r'''(?xms)<div\ class="tabs__tab\ tabs__tab--active">\s*
 263                     <span>seizoen\ (.+?)</span>\s*
 264                 </div>''',
 265              r'<option value="seizoen (\d{1,3})" data-href="[^"]+?" selected>'],
 266             webpage, 'season', default=None)
 267
 268         season_number = int_or_none(season)
 269
 270         episode_number = int_or_none(self._html_search_regex(
 271             r'''(?xms)<div\ class="content__episode">\s*
 272                     <abbr\ title="aflevering">afl</abbr>\s*<span>(\d+)</span>
 273                 </div>''',
 274             webpage, 'episode_number', default=None))
 275
 276         release_date = parse_iso8601(self._html_search_regex(
 277             r'(?ms)<div class="content__broadcastdate">\s*<time\ datetime="(.+?)"',
 278             webpage, 'release_date', default=None))
 279
 280         # If there's a ? or a # in the URL, remove them and everything after
 281         clean_url = url.split('?')[0].split('#')[0].strip('/')
 282         securevideo_url = clean_url + '.mssecurevideo.json'
 283
 284         json = self._download_json(securevideo_url, display_id)
 285         # We are dealing with a '../<show>.relevant' URL
 286         redirect_url = json.get('url')
 287         if redirect_url:
 288             return self.url_result('https:' + redirect_url)
 289         else:
 290             # There is only one entry, but with an unknown key, so just get
 291             # the first one
 292             video_id = list(json.values())[0].get('videoid')
 293
 294         return {
 295             '_type': 'url_transparent',
 296             'url': 'https://mediazone.vrt.be/api/v1/vrtvideo/assets/%s' % video_id,
 297             'ie_key': CanvasIE.ie_key(),
 298             'id': video_id,
 299             'display_id': display_id,
 300             'title': title,
 301             'description': description,
 302             'season': season,
 303             'season_number': season_number,
 304             'episode_number': episode_number,
 305             'release_date': release_date,
 306         }