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',
  85                             'm3u8' if is_live else 'm3u8_native',
  86                             m3u8_id=format_id, fatal=False))
  87                 elif ext == 'mpd':
  88                     formats.extend(self._extract_mpd_formats(
  89                         video_url, video_id, format_id, fatal=False))
  90                 elif ext == 'ism' or video_url.endswith('.ism/Manifest'):
  91                     pass
  92                 else:
  93                     formats.append({
  94                         'url': video_url,
  95                         'format_id': format_id,
  96                     })
  97         if not formats and data.get('drmProtected'):
  98             self.report_drm(video_id)
  99         self._sort_formats(formats)
 100
 101         thumbnails = [{
 102             'id': type,
 103             'url': thumb_url,
 104         } for type, thumb_url in (asset.get('images') or {}).items()]
 105
 106         return {
 107             'id': video_id,
 108             'url': video_url,
 109             'title': self._live_title(title) if is_live else title,
 110             'description': strip_or_none(asset.get('description')),
 111             'thumbnails': thumbnails,
 112             'timestamp': parse_iso8601(asset.get('live_broadcast_time') or asset.get('update_time')),
 113             'duration': float_or_none(asset.get('accurateDuration') or asset.get('duration')),
 114             'view_count': int_or_none(asset.get('views')),
 115             'categories': asset.get('tags', '').split(','),
 116             'formats': formats,
 117             'is_live': is_live,
 118         }
 119
 120
 121 class TV2ArticleIE(InfoExtractor):
 122     _VALID_URL = r'https?://(?:www\.)?tv2\.no/(?:a|\d{4}/\d{2}/\d{2}(/[^/]+)+)/(?P<id>\d+)'
 123     _TESTS = [{
 124         'url': 'http://www.tv2.no/2015/05/16/nyheter/alesund/krim/pingvin/6930542',
 125         'info_dict': {
 126             'id': '6930542',
 127             'title': 'Russen hetses etter pingvintyveri - innrømmer å ha åpnet luken på buret',
 128             'description': 'De fire siktede nekter fortsatt for å ha stjålet pingvinbabyene, men innrømmer å ha åpnet luken til de små kyllingene.',
 129         },
 130         'playlist_count': 2,
 131     }, {
 132         'url': 'http://www.tv2.no/a/6930542',
 133         'only_matching': True,
 134     }]
 135
 136     def _real_extract(self, url):
 137         playlist_id = self._match_id(url)
 138
 139         webpage = self._download_webpage(url, playlist_id)
 140
 141         # Old embed pattern (looks unused nowadays)
 142         assets = re.findall(r'data-assetid=["\'](\d+)', webpage)
 143
 144         if not assets:
 145             # New embed pattern
 146             for v in re.findall(r'(?s)TV2ContentboxVideo\(({.+?})\)', webpage):
 147                 video = self._parse_json(
 148                     v, playlist_id, transform_source=js_to_json, fatal=False)
 149                 if not video:
 150                     continue
 151                 asset = video.get('assetId')
 152                 if asset:
 153                     assets.append(asset)
 154
 155         entries = [
 156             self.url_result('http://www.tv2.no/v/%s' % asset_id, 'TV2')
 157             for asset_id in assets]
 158
 159         title = remove_end(self._og_search_title(webpage), ' - TV2.no')
 160         description = remove_end(self._og_search_description(webpage), ' - TV2.no')
 161
 162         return self.playlist_result(entries, playlist_id, title, description)
 163
 164
 165 class KatsomoIE(InfoExtractor):
 166     _VALID_URL = r'https?://(?:www\.)?(?:katsomo|mtv(uutiset)?)\.fi/(?:sarja/[0-9a-z-]+-\d+/[0-9a-z-]+-|(?:#!/)?jakso/(?:\d+/[^/]+/)?|video/prog)(?P<id>\d+)'
 167     _TESTS = [{
 168         'url': 'https://www.mtv.fi/sarja/mtv-uutiset-live-33001002003/lahden-pelicans-teki-kovan-ratkaisun-ville-nieminen-pihalle-1181321',
 169         'info_dict': {
 170             'id': '1181321',
 171             'ext': 'mp4',
 172             'title': 'Lahden Pelicans teki kovan ratkaisun – Ville Nieminen pihalle',
 173             'description': 'Päätöksen teki Pelicansin hallitus.',
 174             'timestamp': 1575116484,
 175             'upload_date': '20191130',
 176             'duration': 37.12,
 177             'view_count': int,
 178             'categories': list,
 179         },
 180         'params': {
 181             # m3u8 download
 182             'skip_download': True,
 183         },
 184     }, {
 185         'url': 'http://www.katsomo.fi/#!/jakso/33001005/studio55-fi/658521/jukka-kuoppamaki-tekee-yha-lauluja-vaikka-lentokoneessa',
 186         'only_matching': True,
 187     }, {
 188         'url': 'https://www.mtvuutiset.fi/video/prog1311159',
 189         'only_matching': True,
 190     }, {
 191         'url': 'https://www.katsomo.fi/#!/jakso/1311159',
 192         'only_matching': True,
 193     }]
 194     _API_DOMAIN = 'api.katsomo.fi'
 195     _PROTOCOLS = ('HLS', 'MPD')
 196     _GEO_COUNTRIES = ['FI']
 197
 198     def _real_extract(self, url):
 199         video_id = self._match_id(url)
 200         api_base = 'http://%s/api/web/asset/%s' % (self._API_DOMAIN, video_id)
 201
 202         asset = self._download_json(
 203             api_base + '.json', video_id,
 204             'Downloading metadata JSON')['asset']
 205         title = asset.get('subtitle') or asset['title']
 206         is_live = asset.get('live') is True
 207
 208         formats = []
 209         format_urls = []
 210         for protocol in self._PROTOCOLS:
 211             try:
 212                 data = self._download_json(
 213                     api_base + '/play.json?protocol=%s&videoFormat=SMIL+ISMUSP' % protocol,
 214                     video_id, 'Downloading play JSON')['playback']
 215             except ExtractorError as e:
 216                 if isinstance(e.cause, compat_HTTPError) and e.cause.code == 401:
 217                     error = self._parse_json(e.cause.read().decode(), video_id)['error']
 218                     error_code = error.get('code')
 219                     if error_code == 'ASSET_PLAYBACK_INVALID_GEO_LOCATION':
 220                         self.raise_geo_restricted(countries=self._GEO_COUNTRIES)
 221                     elif error_code == 'SESSION_NOT_AUTHENTICATED':
 222                         self.raise_login_required()
 223                     raise ExtractorError(error['description'])
 224                 raise
 225             items = try_get(data, lambda x: x['items']['item'])
 226             if not items:
 227                 continue
 228             if not isinstance(items, list):
 229                 items = [items]
 230             for item in items:
 231                 if not isinstance(item, dict):
 232                     continue
 233                 video_url = item.get('url')
 234                 if not video_url or video_url in format_urls:
 235                     continue
 236                 format_id = '%s-%s' % (protocol.lower(), item.get('mediaFormat'))
 237                 if not self._is_valid_url(video_url, video_id, format_id):
 238                     continue
 239                 format_urls.append(video_url)
 240                 ext = determine_ext(video_url)
 241                 if ext == 'f4m':
 242                     formats.extend(self._extract_f4m_formats(
 243                         video_url, video_id, f4m_id=format_id, fatal=False))
 244                 elif ext == 'm3u8':
 245                     if not data.get('drmProtected'):
 246                         formats.extend(self._extract_m3u8_formats(
 247                             video_url, video_id, 'mp4',
 248                             'm3u8' if is_live else 'm3u8_native',
 249                             m3u8_id=format_id, fatal=False))
 250                 elif ext == 'mpd':
 251                     formats.extend(self._extract_mpd_formats(
 252                         video_url, video_id, format_id, fatal=False))
 253                 elif ext == 'ism' or video_url.endswith('.ism/Manifest'):
 254                     pass
 255                 else:
 256                     formats.append({
 257                         'url': video_url,
 258                         'format_id': format_id,
 259                         'tbr': int_or_none(item.get('bitrate')),
 260                         'filesize': int_or_none(item.get('fileSize')),
 261                     })
 262         if not formats and data.get('drmProtected'):
 263             self.report_drm(video_id)
 264         self._sort_formats(formats)
 265
 266         thumbnails = [{
 267             'id': thumbnail.get('@type'),
 268             'url': thumbnail.get('url'),
 269         } for _, thumbnail in (asset.get('imageVersions') or {}).items()]
 270
 271         return {
 272             'id': video_id,
 273             'url': video_url,
 274             'title': self._live_title(title) if is_live else title,
 275             'description': strip_or_none(asset.get('description')),
 276             'thumbnails': thumbnails,
 277             'timestamp': parse_iso8601(asset.get('createTime')),
 278             'duration': float_or_none(asset.get('accurateDuration') or asset.get('duration')),
 279             'view_count': int_or_none(asset.get('views')),
 280             'categories': asset.get('keywords', '').split(','),
 281             'formats': formats,
 282             'is_live': is_live,
 283         }
 284
 285
 286 class MTVUutisetArticleIE(InfoExtractor):
 287     _VALID_URL = r'https?://(?:www\.)mtvuutiset\.fi/artikkeli/[^/]+/(?P<id>\d+)'
 288     _TESTS = [{
 289         'url': 'https://www.mtvuutiset.fi/artikkeli/tallaisia-vaurioita-viking-amorellassa-on-useamman-osaston-alla-vetta/7931384',
 290         'info_dict': {
 291             'id': '1311159',
 292             'ext': 'mp4',
 293             'title': 'Viking Amorellan matkustajien evakuointi on alkanut – tältä operaatio näyttää laivalla',
 294             'description': 'Viking Amorellan matkustajien evakuointi on alkanut – tältä operaatio näyttää laivalla',
 295             'timestamp': 1600608966,
 296             'upload_date': '20200920',
 297             'duration': 153.7886666,
 298             'view_count': int,
 299             'categories': list,
 300         },
 301         'params': {
 302             # m3u8 download
 303             'skip_download': True,
 304         },
 305     }, {
 306         # multiple Youtube embeds
 307         'url': 'https://www.mtvuutiset.fi/artikkeli/50-vuotta-subarun-vastaiskua/6070962',
 308         'only_matching': True,
 309     }]
 310
 311     def _real_extract(self, url):
 312         article_id = self._match_id(url)
 313         article = self._download_json(
 314             'http://api.mtvuutiset.fi/mtvuutiset/api/json/' + article_id,
 315             article_id)
 316
 317         def entries():
 318             for video in (article.get('videos') or []):
 319                 video_type = video.get('videotype')
 320                 video_url = video.get('url')
 321                 if not (video_url and video_type in ('katsomo', 'youtube')):
 322                     continue
 323                 yield self.url_result(
 324                     video_url, video_type.capitalize(), video.get('video_id'))
 325
 326         return self.playlist_result(
 327             entries(), article_id, article.get('title'), article.get('description'))