yt_dlp/extractor/tv2.py

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