yt_dlp/extractor/tv2.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..compat import compat_HTTPError
   8 from ..utils import (
   9     determine_ext,
  10     ExtractorError,
  11     int_or_none,
  12     float_or_none,
  13     js_to_json,
  14     parse_iso8601,
  15     remove_end,
  16     strip_or_none,
  17     try_get,
  18 )
  19
  20
  21 class TV2IE(InfoExtractor):
  22     _VALID_URL = r'https?://(?:www\.)?tv2\.no/v\d*/(?P<id>\d+)'
  23     _TESTS = [{
  24         'url': 'http://www.tv2.no/v/916509/',
  25         'info_dict': {
  26             'id': '916509',
  27             'ext': 'mp4',
  28             'title': 'Se Frode Gryttens hyllest av Steven Gerrard',
  29             'description': 'TV 2 Sportens huspoet tar avskjed med Liverpools kaptein Steven Gerrard.',
  30             'timestamp': 1431715610,
  31             'upload_date': '20150515',
  32             'duration': 157,
  33             'view_count': int,
  34             'categories': list,
  35         },
  36     }, {
  37         'url': 'http://www.tv2.no/v2/916509',
  38         'only_matching': True,
  39     }]
  40     _PROTOCOLS = ('HLS', 'DASH')
  41     _GEO_COUNTRIES = ['NO']
  42
  43     def _real_extract(self, url):
  44         video_id = self._match_id(url)
  45         asset = self._download_json('https://sumo.tv2.no/rest/assets/' + video_id, video_id,
  46                                     'Downloading metadata JSON')
  47         title = asset['title']
  48         is_live = asset.get('live') is True
  49
  50         formats = []
  51         format_urls = []
  52         for protocol in self._PROTOCOLS:
  53             try:
  54                 data = self._download_json('https://api.sumo.tv2.no/play/%s?stream=%s' % (video_id, protocol),
  55                                            video_id, 'Downloading playabck JSON',
  56                                            headers={'content-type': 'application/json'},
  57                                            data='{"device":{"id":"1-1-1","name":"Nettleser (HTML)"}}'.encode())['playback']
  58             except ExtractorError as e:
  59                 if isinstance(e.cause, compat_HTTPError) and e.cause.code == 401:
  60                     error = self._parse_json(e.cause.read().decode(), video_id)['error']
  61                     error_code = error.get('code')
  62                     if error_code == 'ASSET_PLAYBACK_INVALID_GEO_LOCATION':
  63                         self.raise_geo_restricted(countries=self._GEO_COUNTRIES)
  64                     elif error_code == 'SESSION_NOT_AUTHENTICATED':
  65                         self.raise_login_required()
  66                     raise ExtractorError(error['description'])
  67                 raise
  68             items = data.get('streams', [])
  69             for item in items:
  70                 video_url = item.get('url')
  71                 if not video_url or video_url in format_urls:
  72                     continue
  73                 format_id = '%s-%s' % (protocol.lower(), item.get('type'))
  74                 if not self._is_valid_url(video_url, video_id, format_id):
  75                     continue
  76                 format_urls.append(video_url)
  77                 ext = determine_ext(video_url)
  78                 if ext == 'f4m':
  79                     formats.extend(self._extract_f4m_formats(
  80                         video_url, video_id, f4m_id=format_id, fatal=False))
  81                 elif ext == 'm3u8':
  82                     if not data.get('drmProtected'):
  83                         formats.extend(self._extract_m3u8_formats(
  84                             video_url, video_id, 'mp4', live=is_live, m3u8_id=format_id, fatal=False))
  85                 elif ext == 'mpd':
  86                     formats.extend(self._extract_mpd_formats(
  87                         video_url, video_id, format_id, fatal=False))
  88                 elif ext == 'ism' or video_url.endswith('.ism/Manifest'):
  89                     pass
  90                 else:
  91                     formats.append({
  92                         'url': video_url,
  93                         'format_id': format_id,
  94                     })
  95         if not formats and data.get('drmProtected'):
  96             self.report_drm(video_id)
  97         self._sort_formats(formats)
  98
  99         thumbnails = [{
 100             'id': type,
 101             'url': thumb_url,
 102         } for type, thumb_url in (asset.get('images') or {}).items()]
 103
 104         return {
 105             'id': video_id,
 106             'url': video_url,
 107             'title': title,
 108             'description': strip_or_none(asset.get('description')),
 109             'thumbnails': thumbnails,
 110             'timestamp': parse_iso8601(asset.get('live_broadcast_time') or asset.get('update_time')),
 111             'duration': float_or_none(asset.get('accurateDuration') or asset.get('duration')),
 112             'view_count': int_or_none(asset.get('views')),
 113             'categories': asset.get('tags', '').split(','),
 114             'formats': formats,
 115             'is_live': is_live,
 116         }
 117
 118
 119 class TV2ArticleIE(InfoExtractor):
 120     _VALID_URL = r'https?://(?:www\.)?tv2\.no/(?:a|\d{4}/\d{2}/\d{2}(/[^/]+)+)/(?P<id>\d+)'
 121     _TESTS = [{
 122         'url': 'http://www.tv2.no/2015/05/16/nyheter/alesund/krim/pingvin/6930542',
 123         'info_dict': {
 124             'id': '6930542',
 125             'title': 'Russen hetses etter pingvintyveri - innrømmer å ha åpnet luken på buret',
 126             'description': 'De fire siktede nekter fortsatt for å ha stjålet pingvinbabyene, men innrømmer å ha åpnet luken til de små kyllingene.',
 127         },
 128         'playlist_count': 2,
 129     }, {
 130         'url': 'http://www.tv2.no/a/6930542',
 131         'only_matching': True,
 132     }]
 133
 134     def _real_extract(self, url):
 135         playlist_id = self._match_id(url)
 136
 137         webpage = self._download_webpage(url, playlist_id)
 138
 139         # Old embed pattern (looks unused nowadays)
 140         assets = re.findall(r'data-assetid=["\'](\d+)', webpage)
 141
 142         if not assets:
 143             # New embed pattern
 144             for v in re.findall(r'(?s)TV2ContentboxVideo\(({.+?})\)', webpage):
 145                 video = self._parse_json(
 146                     v, playlist_id, transform_source=js_to_json, fatal=False)
 147                 if not video:
 148                     continue
 149                 asset = video.get('assetId')
 150                 if asset:
 151                     assets.append(asset)
 152
 153         entries = [
 154             self.url_result('http://www.tv2.no/v/%s' % asset_id, 'TV2')
 155             for asset_id in assets]
 156
 157         title = remove_end(self._og_search_title(webpage), ' - TV2.no')
 158         description = remove_end(self._og_search_description(webpage), ' - TV2.no')
 159
 160         return self.playlist_result(entries, playlist_id, title, description)
 161
 162
 163 class KatsomoIE(InfoExtractor):
 164     _VALID_URL = r'https?://(?:www\.)?(?:katsomo|mtv(uutiset)?)\.fi/(?:sarja/[0-9a-z-]+-\d+/[0-9a-z-]+-|(?:#!/)?jakso/(?:\d+/[^/]+/)?|video/prog)(?P<id>\d+)'
 165     _TESTS = [{
 166         'url': 'https://www.mtv.fi/sarja/mtv-uutiset-live-33001002003/lahden-pelicans-teki-kovan-ratkaisun-ville-nieminen-pihalle-1181321',
 167         'info_dict': {
 168             'id': '1181321',
 169             'ext': 'mp4',
 170             'title': 'Lahden Pelicans teki kovan ratkaisun – Ville Nieminen pihalle',
 171             'description': 'Päätöksen teki Pelicansin hallitus.',
 172             'timestamp': 1575116484,
 173             'upload_date': '20191130',
 174             'duration': 37.12,
 175             'view_count': int,
 176             'categories': list,
 177         },
 178         'params': {
 179             # m3u8 download
 180             'skip_download': True,
 181         },
 182     }, {
 183         'url': 'http://www.katsomo.fi/#!/jakso/33001005/studio55-fi/658521/jukka-kuoppamaki-tekee-yha-lauluja-vaikka-lentokoneessa',
 184         'only_matching': True,
 185     }, {
 186         'url': 'https://www.mtvuutiset.fi/video/prog1311159',
 187         'only_matching': True,
 188     }, {
 189         'url': 'https://www.katsomo.fi/#!/jakso/1311159',
 190         'only_matching': True,
 191     }]
 192     _API_DOMAIN = 'api.katsomo.fi'
 193     _PROTOCOLS = ('HLS', 'MPD')
 194     _GEO_COUNTRIES = ['FI']
 195
 196     def _real_extract(self, url):
 197         video_id = self._match_id(url)
 198         api_base = 'http://%s/api/web/asset/%s' % (self._API_DOMAIN, video_id)
 199
 200         asset = self._download_json(
 201             api_base + '.json', video_id,
 202             'Downloading metadata JSON')['asset']
 203         title = asset.get('subtitle') or asset['title']
 204         is_live = asset.get('live') is True
 205
 206         formats = []
 207         format_urls = []
 208         for protocol in self._PROTOCOLS:
 209             try:
 210                 data = self._download_json(
 211                     api_base + '/play.json?protocol=%s&videoFormat=SMIL+ISMUSP' % protocol,
 212                     video_id, 'Downloading play JSON')['playback']
 213             except ExtractorError as e:
 214                 if isinstance(e.cause, compat_HTTPError) and e.cause.code == 401:
 215                     error = self._parse_json(e.cause.read().decode(), video_id)['error']
 216                     error_code = error.get('code')
 217                     if error_code == 'ASSET_PLAYBACK_INVALID_GEO_LOCATION':
 218                         self.raise_geo_restricted(countries=self._GEO_COUNTRIES)
 219                     elif error_code == 'SESSION_NOT_AUTHENTICATED':
 220                         self.raise_login_required()
 221                     raise ExtractorError(error['description'])
 222                 raise
 223             items = try_get(data, lambda x: x['items']['item'])
 224             if not items:
 225                 continue
 226             if not isinstance(items, list):
 227                 items = [items]
 228             for item in items:
 229                 if not isinstance(item, dict):
 230                     continue
 231                 video_url = item.get('url')
 232                 if not video_url or video_url in format_urls:
 233                     continue
 234                 format_id = '%s-%s' % (protocol.lower(), item.get('mediaFormat'))
 235                 if not self._is_valid_url(video_url, video_id, format_id):
 236                     continue
 237                 format_urls.append(video_url)
 238                 ext = determine_ext(video_url)
 239                 if ext == 'f4m':
 240                     formats.extend(self._extract_f4m_formats(
 241                         video_url, video_id, f4m_id=format_id, fatal=False))
 242                 elif ext == 'm3u8':
 243                     if not data.get('drmProtected'):
 244                         formats.extend(self._extract_m3u8_formats(
 245                             video_url, video_id, 'mp4', live=is_live, m3u8_id=format_id, fatal=False))
 246                 elif ext == 'mpd':
 247                     formats.extend(self._extract_mpd_formats(
 248                         video_url, video_id, format_id, fatal=False))
 249                 elif ext == 'ism' or video_url.endswith('.ism/Manifest'):
 250                     pass
 251                 else:
 252                     formats.append({
 253                         'url': video_url,
 254                         'format_id': format_id,
 255                         'tbr': int_or_none(item.get('bitrate')),
 256                         'filesize': int_or_none(item.get('fileSize')),
 257                     })
 258         if not formats and data.get('drmProtected'):
 259             self.report_drm(video_id)
 260         self._sort_formats(formats)
 261
 262         thumbnails = [{
 263             'id': thumbnail.get('@type'),
 264             'url': thumbnail.get('url'),
 265         } for _, thumbnail in (asset.get('imageVersions') or {}).items()]
 266
 267         return {
 268             'id': video_id,
 269             'url': video_url,
 270             'title': title,
 271             'description': strip_or_none(asset.get('description')),
 272             'thumbnails': thumbnails,
 273             'timestamp': parse_iso8601(asset.get('createTime')),
 274             'duration': float_or_none(asset.get('accurateDuration') or asset.get('duration')),
 275             'view_count': int_or_none(asset.get('views')),
 276             'categories': asset.get('keywords', '').split(','),
 277             'formats': formats,
 278             'is_live': is_live,
 279         }
 280
 281
 282 class MTVUutisetArticleIE(InfoExtractor):
 283     _VALID_URL = r'https?://(?:www\.)mtvuutiset\.fi/artikkeli/[^/]+/(?P<id>\d+)'
 284     _TESTS = [{
 285         'url': 'https://www.mtvuutiset.fi/artikkeli/tallaisia-vaurioita-viking-amorellassa-on-useamman-osaston-alla-vetta/7931384',
 286         'info_dict': {
 287             'id': '1311159',
 288             'ext': 'mp4',
 289             'title': 'Viking Amorellan matkustajien evakuointi on alkanut – tältä operaatio näyttää laivalla',
 290             'description': 'Viking Amorellan matkustajien evakuointi on alkanut – tältä operaatio näyttää laivalla',
 291             'timestamp': 1600608966,
 292             'upload_date': '20200920',
 293             'duration': 153.7886666,
 294             'view_count': int,
 295             'categories': list,
 296         },
 297         'params': {
 298             # m3u8 download
 299             'skip_download': True,
 300         },
 301     }, {
 302         # multiple Youtube embeds
 303         'url': 'https://www.mtvuutiset.fi/artikkeli/50-vuotta-subarun-vastaiskua/6070962',
 304         'only_matching': True,
 305     }]
 306
 307     def _real_extract(self, url):
 308         article_id = self._match_id(url)
 309         article = self._download_json(
 310             'http://api.mtvuutiset.fi/mtvuutiset/api/json/' + article_id,
 311             article_id)
 312
 313         def entries():
 314             for video in (article.get('videos') or []):
 315                 video_type = video.get('videotype')
 316                 video_url = video.get('url')
 317                 if not (video_url and video_type in ('katsomo', 'youtube')):
 318                     continue
 319                 yield self.url_result(
 320                     video_url, video_type.capitalize(), video.get('video_id'))
 321
 322         return self.playlist_result(
 323             entries(), article_id, article.get('title'), article.get('description'))