yt_dlp/extractor/itv.py

   1 import json
   2
   3 from .brightcove import BrightcoveNewIE
   4 from .common import InfoExtractor
   5 from ..compat import compat_str
   6 from ..utils import (
   7     JSON_LD_RE,
   8     ExtractorError,
   9     base_url,
  10     clean_html,
  11     determine_ext,
  12     extract_attributes,
  13     get_element_by_class,
  14     merge_dicts,
  15     parse_duration,
  16     smuggle_url,
  17     try_get,
  18     url_basename,
  19     url_or_none,
  20     urljoin,
  21 )
  22
  23
  24 class ITVIE(InfoExtractor):
  25     _VALID_URL = r'https?://(?:www\.)?itv\.com/hub/[^/]+/(?P<id>[0-9a-zA-Z]+)'
  26     _GEO_COUNTRIES = ['GB']
  27     _TESTS = [{
  28         'url': 'https://www.itv.com/hub/plebs/2a1873a0002',
  29         'info_dict': {
  30             'id': '2a1873a0002',
  31             'ext': 'mp4',
  32             'title': 'Plebs - The Orgy',
  33             'description': 'md5:4d7159af53ebd5b36e8b3ec82a41fdb4',
  34             'series': 'Plebs',
  35             'season_number': 1,
  36             'episode_number': 1,
  37             'thumbnail': r're:https?://hubimages\.itv\.com/episode/2_1873_0002'
  38         },
  39         'params': {
  40             # m3u8 download
  41             'skip_download': True,
  42         },
  43     }, {
  44         'url': 'https://www.itv.com/hub/the-jonathan-ross-show/2a1166a0209',
  45         'info_dict': {
  46             'id': '2a1166a0209',
  47             'ext': 'mp4',
  48             'title': 'The Jonathan Ross Show - Series 17 - Episode 8',
  49             'description': 'md5:3023dcdd375db1bc9967186cdb3f1399',
  50             'series': 'The Jonathan Ross Show',
  51             'episode_number': 8,
  52             'season_number': 17,
  53             'thumbnail': r're:https?://hubimages\.itv\.com/episode/2_1873_0002'
  54         },
  55         'params': {
  56             # m3u8 download
  57             'skip_download': True,
  58         },
  59     }, {
  60         # unavailable via data-playlist-url
  61         'url': 'https://www.itv.com/hub/through-the-keyhole/2a2271a0033',
  62         'only_matching': True,
  63     }, {
  64         # InvalidVodcrid
  65         'url': 'https://www.itv.com/hub/james-martins-saturday-morning/2a5159a0034',
  66         'only_matching': True,
  67     }, {
  68         # ContentUnavailable
  69         'url': 'https://www.itv.com/hub/whos-doing-the-dishes/2a2898a0024',
  70         'only_matching': True,
  71     }]
  72
  73     def _generate_api_headers(self, hmac):
  74         return merge_dicts({
  75             'Accept': 'application/vnd.itv.vod.playlist.v2+json',
  76             'Content-Type': 'application/json',
  77             'hmac': hmac.upper(),
  78         }, self.geo_verification_headers())
  79
  80     def _call_api(self, video_id, playlist_url, headers, platform_tag, featureset, fatal=True):
  81         return self._download_json(
  82             playlist_url, video_id, data=json.dumps({
  83                 'user': {
  84                     'itvUserId': '',
  85                     'entitlements': [],
  86                     'token': ''
  87                 },
  88                 'device': {
  89                     'manufacturer': 'Safari',
  90                     'model': '5',
  91                     'os': {
  92                         'name': 'Windows NT',
  93                         'version': '6.1',
  94                         'type': 'desktop'
  95                     }
  96                 },
  97                 'client': {
  98                     'version': '4.1',
  99                     'id': 'browser'
 100                 },
 101                 'variantAvailability': {
 102                     'featureset': {
 103                         'min': featureset,
 104                         'max': featureset
 105                     },
 106                     'platformTag': platform_tag
 107                 }
 108             }).encode(), headers=headers, fatal=fatal)
 109
 110     def _get_subtitles(self, video_id, variants, ios_playlist_url, headers, *args, **kwargs):
 111         subtitles = {}
 112         # Prefer last matching featureset
 113         # See: https://github.com/yt-dlp/yt-dlp/issues/986
 114         platform_tag_subs, featureset_subs = next(
 115             ((platform_tag, featureset)
 116              for platform_tag, featuresets in reversed(list(variants.items())) for featureset in featuresets
 117              if try_get(featureset, lambda x: x[2]) == 'outband-webvtt'),
 118             (None, None))
 119
 120         if platform_tag_subs and featureset_subs:
 121             subs_playlist = self._call_api(
 122                 video_id, ios_playlist_url, headers, platform_tag_subs, featureset_subs, fatal=False)
 123             subs = try_get(subs_playlist, lambda x: x['Playlist']['Video']['Subtitles'], list) or []
 124             for sub in subs:
 125                 if not isinstance(sub, dict):
 126                     continue
 127                 href = url_or_none(sub.get('Href'))
 128                 if not href:
 129                     continue
 130                 subtitles.setdefault('en', []).append({'url': href})
 131         return subtitles
 132
 133     def _real_extract(self, url):
 134         video_id = self._match_id(url)
 135         webpage = self._download_webpage(url, video_id)
 136         params = extract_attributes(self._search_regex(
 137             r'(?s)(<[^>]+id="video"[^>]*>)', webpage, 'params'))
 138         variants = self._parse_json(
 139             try_get(params, lambda x: x['data-video-variants'], compat_str) or '{}',
 140             video_id, fatal=False)
 141         # Prefer last matching featureset
 142         # See: https://github.com/yt-dlp/yt-dlp/issues/986
 143         platform_tag_video, featureset_video = next(
 144             ((platform_tag, featureset)
 145              for platform_tag, featuresets in reversed(list(variants.items())) for featureset in featuresets
 146              if set(try_get(featureset, lambda x: x[:2]) or []) == {'aes', 'hls'}),
 147             (None, None))
 148         if not platform_tag_video or not featureset_video:
 149             raise ExtractorError('No downloads available', expected=True, video_id=video_id)
 150
 151         ios_playlist_url = params.get('data-video-playlist') or params['data-video-id']
 152         headers = self._generate_api_headers(params['data-video-hmac'])
 153         ios_playlist = self._call_api(
 154             video_id, ios_playlist_url, headers, platform_tag_video, featureset_video)
 155
 156         video_data = try_get(ios_playlist, lambda x: x['Playlist']['Video'], dict) or {}
 157         ios_base_url = video_data.get('Base')
 158         formats = []
 159         for media_file in (video_data.get('MediaFiles') or []):
 160             href = media_file.get('Href')
 161             if not href:
 162                 continue
 163             if ios_base_url:
 164                 href = ios_base_url + href
 165             ext = determine_ext(href)
 166             if ext == 'm3u8':
 167                 formats.extend(self._extract_m3u8_formats(
 168                     href, video_id, 'mp4', entry_protocol='m3u8_native',
 169                     m3u8_id='hls', fatal=False))
 170             else:
 171                 formats.append({
 172                     'url': href,
 173                 })
 174         info = self._search_json_ld(webpage, video_id, default={})
 175         if not info:
 176             json_ld = self._parse_json(self._search_regex(
 177                 JSON_LD_RE, webpage, 'JSON-LD', '{}',
 178                 group='json_ld'), video_id, fatal=False)
 179             if json_ld and json_ld.get('@type') == 'BreadcrumbList':
 180                 for ile in (json_ld.get('itemListElement:') or []):
 181                     item = ile.get('item:') or {}
 182                     if item.get('@type') == 'TVEpisode':
 183                         item['@context'] = 'http://schema.org'
 184                         info = self._json_ld(item, video_id, fatal=False) or {}
 185                         break
 186
 187         thumbnails = []
 188         thumbnail_url = try_get(params, lambda x: x['data-video-posterframe'], compat_str)
 189         if thumbnail_url:
 190             thumbnails.extend([{
 191                 'url': thumbnail_url.format(width=1920, height=1080, quality=100, blur=0, bg='false'),
 192                 'width': 1920,
 193                 'height': 1080,
 194             }, {
 195                 'url': urljoin(base_url(thumbnail_url), url_basename(thumbnail_url)),
 196                 'preference': -2
 197             }])
 198
 199         thumbnail_url = self._html_search_meta(['og:image', 'twitter:image'], webpage, default=None)
 200         if thumbnail_url:
 201             thumbnails.append({
 202                 'url': thumbnail_url,
 203             })
 204         self._remove_duplicate_formats(thumbnails)
 205
 206         return merge_dicts({
 207             'id': video_id,
 208             'title': self._html_search_meta(['og:title', 'twitter:title'], webpage),
 209             'formats': formats,
 210             'subtitles': self.extract_subtitles(video_id, variants, ios_playlist_url, headers),
 211             'duration': parse_duration(video_data.get('Duration')),
 212             'description': clean_html(get_element_by_class('episode-info__synopsis', webpage)),
 213             'thumbnails': thumbnails
 214         }, info)
 215
 216
 217 class ITVBTCCIE(InfoExtractor):
 218     _VALID_URL = r'https?://(?:www\.)?itv\.com/(?:news|btcc)/(?:[^/]+/)*(?P<id>[^/?#&]+)'
 219     _TESTS = [{
 220         'url': 'https://www.itv.com/btcc/articles/btcc-2019-brands-hatch-gp-race-action',
 221         'info_dict': {
 222             'id': 'btcc-2019-brands-hatch-gp-race-action',
 223             'title': 'BTCC 2019: Brands Hatch GP race action',
 224         },
 225         'playlist_count': 12,
 226     }, {
 227         'url': 'https://www.itv.com/news/2021-10-27/i-have-to-protect-the-country-says-rishi-sunak-as-uk-faces-interest-rate-hike',
 228         'info_dict': {
 229             'id': 'i-have-to-protect-the-country-says-rishi-sunak-as-uk-faces-interest-rate-hike',
 230             'title': 'md5:6ef054dd9f069330db3dcc66cb772d32'
 231         },
 232         'playlist_count': 4
 233     }]
 234     BRIGHTCOVE_URL_TEMPLATE = 'http://players.brightcove.net/%s/%s_default/index.html?videoId=%s'
 235
 236     def _real_extract(self, url):
 237         playlist_id = self._match_id(url)
 238
 239         webpage = self._download_webpage(url, playlist_id)
 240
 241         json_map = try_get(
 242             self._search_nextjs_data(webpage, playlist_id),
 243             lambda x: x['props']['pageProps']['article']['body']['content']) or []
 244
 245         entries = []
 246         for video in json_map:
 247             if not any(video['data'].get(attr) == 'Brightcove' for attr in ('name', 'type')):
 248                 continue
 249             video_id = video['data']['id']
 250             account_id = video['data']['accountId']
 251             player_id = video['data']['playerId']
 252             entries.append(self.url_result(
 253                 smuggle_url(self.BRIGHTCOVE_URL_TEMPLATE % (account_id, player_id, video_id), {
 254                     # ITV does not like some GB IP ranges, so here are some
 255                     # IP blocks it accepts
 256                     'geo_ip_blocks': [
 257                         '193.113.0.0/16', '54.36.162.0/23', '159.65.16.0/21'
 258                     ],
 259                     'referrer': url,
 260                 }),
 261                 ie=BrightcoveNewIE.ie_key(), video_id=video_id))
 262
 263         title = self._og_search_title(webpage, fatal=False)
 264
 265         return self.playlist_result(entries, playlist_id, title)