yt_dlp/extractor/itv.py

   1 import json
   2
   3 from .common import InfoExtractor
   4 from .brightcove import BrightcoveNewIE
   5
   6 from ..compat import compat_str
   7 from ..utils import (
   8     base_url,
   9     clean_html,
  10     determine_ext,
  11     extract_attributes,
  12     ExtractorError,
  13     get_element_by_class,
  14     JSON_LD_RE,
  15     merge_dicts,
  16     parse_duration,
  17     smuggle_url,
  18     try_get,
  19     url_or_none,
  20     url_basename,
  21     urljoin,
  22 )
  23
  24
  25 class ITVIE(InfoExtractor):
  26     _VALID_URL = r'https?://(?:www\.)?itv\.com/hub/[^/]+/(?P<id>[0-9a-zA-Z]+)'
  27     _GEO_COUNTRIES = ['GB']
  28     _TESTS = [{
  29         'url': 'https://www.itv.com/hub/plebs/2a1873a0002',
  30         'info_dict': {
  31             'id': '2a1873a0002',
  32             'ext': 'mp4',
  33             'title': 'Plebs - The Orgy',
  34             'description': 'md5:4d7159af53ebd5b36e8b3ec82a41fdb4',
  35             'series': 'Plebs',
  36             'season_number': 1,
  37             'episode_number': 1,
  38             'thumbnail': r're:https?://hubimages\.itv\.com/episode/2_1873_0002'
  39         },
  40         'params': {
  41             # m3u8 download
  42             'skip_download': True,
  43         },
  44     }, {
  45         'url': 'https://www.itv.com/hub/the-jonathan-ross-show/2a1166a0209',
  46         'info_dict': {
  47             'id': '2a1166a0209',
  48             'ext': 'mp4',
  49             'title': 'The Jonathan Ross Show - Series 17 - Episode 8',
  50             'description': 'md5:3023dcdd375db1bc9967186cdb3f1399',
  51             'series': 'The Jonathan Ross Show',
  52             'episode_number': 8,
  53             'season_number': 17,
  54             'thumbnail': r're:https?://hubimages\.itv\.com/episode/2_1873_0002'
  55         },
  56         'params': {
  57             # m3u8 download
  58             'skip_download': True,
  59         },
  60     }, {
  61         # unavailable via data-playlist-url
  62         'url': 'https://www.itv.com/hub/through-the-keyhole/2a2271a0033',
  63         'only_matching': True,
  64     }, {
  65         # InvalidVodcrid
  66         'url': 'https://www.itv.com/hub/james-martins-saturday-morning/2a5159a0034',
  67         'only_matching': True,
  68     }, {
  69         # ContentUnavailable
  70         'url': 'https://www.itv.com/hub/whos-doing-the-dishes/2a2898a0024',
  71         'only_matching': True,
  72     }]
  73
  74     def _generate_api_headers(self, hmac):
  75         return merge_dicts({
  76             'Accept': 'application/vnd.itv.vod.playlist.v2+json',
  77             'Content-Type': 'application/json',
  78             'hmac': hmac.upper(),
  79         }, self.geo_verification_headers())
  80
  81     def _call_api(self, video_id, playlist_url, headers, platform_tag, featureset, fatal=True):
  82         return self._download_json(
  83             playlist_url, video_id, data=json.dumps({
  84                 'user': {
  85                     'itvUserId': '',
  86                     'entitlements': [],
  87                     'token': ''
  88                 },
  89                 'device': {
  90                     'manufacturer': 'Safari',
  91                     'model': '5',
  92                     'os': {
  93                         'name': 'Windows NT',
  94                         'version': '6.1',
  95                         'type': 'desktop'
  96                     }
  97                 },
  98                 'client': {
  99                     'version': '4.1',
 100                     'id': 'browser'
 101                 },
 102                 'variantAvailability': {
 103                     'featureset': {
 104                         'min': featureset,
 105                         'max': featureset
 106                     },
 107                     'platformTag': platform_tag
 108                 }
 109             }).encode(), headers=headers, fatal=fatal)
 110
 111     def _get_subtitles(self, video_id, variants, ios_playlist_url, headers, *args, **kwargs):
 112         subtitles = {}
 113         # Prefer last matching featureset
 114         # See: https://github.com/yt-dlp/yt-dlp/issues/986
 115         platform_tag_subs, featureset_subs = next(
 116             ((platform_tag, featureset)
 117              for platform_tag, featuresets in reversed(list(variants.items())) for featureset in featuresets
 118              if try_get(featureset, lambda x: x[2]) == 'outband-webvtt'),
 119             (None, None))
 120
 121         if platform_tag_subs and featureset_subs:
 122             subs_playlist = self._call_api(
 123                 video_id, ios_playlist_url, headers, platform_tag_subs, featureset_subs, fatal=False)
 124             subs = try_get(subs_playlist, lambda x: x['Playlist']['Video']['Subtitles'], list) or []
 125             for sub in subs:
 126                 if not isinstance(sub, dict):
 127                     continue
 128                 href = url_or_none(sub.get('Href'))
 129                 if not href:
 130                     continue
 131                 subtitles.setdefault('en', []).append({'url': href})
 132         return subtitles
 133
 134     def _real_extract(self, url):
 135         video_id = self._match_id(url)
 136         webpage = self._download_webpage(url, video_id)
 137         params = extract_attributes(self._search_regex(
 138             r'(?s)(<[^>]+id="video"[^>]*>)', webpage, 'params'))
 139         variants = self._parse_json(
 140             try_get(params, lambda x: x['data-video-variants'], compat_str) or '{}',
 141             video_id, fatal=False)
 142         # Prefer last matching featureset
 143         # See: https://github.com/yt-dlp/yt-dlp/issues/986
 144         platform_tag_video, featureset_video = next(
 145             ((platform_tag, featureset)
 146              for platform_tag, featuresets in reversed(list(variants.items())) for featureset in featuresets
 147              if set(try_get(featureset, lambda x: x[:2]) or []) == {'aes', 'hls'}),
 148             (None, None))
 149         if not platform_tag_video or not featureset_video:
 150             raise ExtractorError('No downloads available', expected=True, video_id=video_id)
 151
 152         ios_playlist_url = params.get('data-video-playlist') or params['data-video-id']
 153         headers = self._generate_api_headers(params['data-video-hmac'])
 154         ios_playlist = self._call_api(
 155             video_id, ios_playlist_url, headers, platform_tag_video, featureset_video)
 156
 157         video_data = try_get(ios_playlist, lambda x: x['Playlist']['Video'], dict) or {}
 158         ios_base_url = video_data.get('Base')
 159         formats = []
 160         for media_file in (video_data.get('MediaFiles') or []):
 161             href = media_file.get('Href')
 162             if not href:
 163                 continue
 164             if ios_base_url:
 165                 href = ios_base_url + href
 166             ext = determine_ext(href)
 167             if ext == 'm3u8':
 168                 formats.extend(self._extract_m3u8_formats(
 169                     href, video_id, 'mp4', entry_protocol='m3u8_native',
 170                     m3u8_id='hls', fatal=False))
 171             else:
 172                 formats.append({
 173                     'url': href,
 174                 })
 175         self._sort_formats(formats)
 176         info = self._search_json_ld(webpage, video_id, default={})
 177         if not info:
 178             json_ld = self._parse_json(self._search_regex(
 179                 JSON_LD_RE, webpage, 'JSON-LD', '{}',
 180                 group='json_ld'), video_id, fatal=False)
 181             if json_ld and json_ld.get('@type') == 'BreadcrumbList':
 182                 for ile in (json_ld.get('itemListElement:') or []):
 183                     item = ile.get('item:') or {}
 184                     if item.get('@type') == 'TVEpisode':
 185                         item['@context'] = 'http://schema.org'
 186                         info = self._json_ld(item, video_id, fatal=False) or {}
 187                         break
 188
 189         thumbnails = []
 190         thumbnail_url = try_get(params, lambda x: x['data-video-posterframe'], compat_str)
 191         if thumbnail_url:
 192             thumbnails.extend([{
 193                 'url': thumbnail_url.format(width=1920, height=1080, quality=100, blur=0, bg='false'),
 194                 'width': 1920,
 195                 'height': 1080,
 196             }, {
 197                 'url': urljoin(base_url(thumbnail_url), url_basename(thumbnail_url)),
 198                 'preference': -2
 199             }])
 200
 201         thumbnail_url = self._html_search_meta(['og:image', 'twitter:image'], webpage, default=None)
 202         if thumbnail_url:
 203             thumbnails.append({
 204                 'url': thumbnail_url,
 205             })
 206         self._remove_duplicate_formats(thumbnails)
 207
 208         return merge_dicts({
 209             'id': video_id,
 210             'title': self._html_search_meta(['og:title', 'twitter:title'], webpage),
 211             'formats': formats,
 212             'subtitles': self.extract_subtitles(video_id, variants, ios_playlist_url, headers),
 213             'duration': parse_duration(video_data.get('Duration')),
 214             'description': clean_html(get_element_by_class('episode-info__synopsis', webpage)),
 215             'thumbnails': thumbnails
 216         }, info)
 217
 218
 219 class ITVBTCCIE(InfoExtractor):
 220     _VALID_URL = r'https?://(?:www\.)?itv\.com/(?:news|btcc)/(?:[^/]+/)*(?P<id>[^/?#&]+)'
 221     _TESTS = [{
 222         'url': 'https://www.itv.com/btcc/articles/btcc-2019-brands-hatch-gp-race-action',
 223         'info_dict': {
 224             'id': 'btcc-2019-brands-hatch-gp-race-action',
 225             'title': 'BTCC 2019: Brands Hatch GP race action',
 226         },
 227         'playlist_count': 12,
 228     }, {
 229         'url': 'https://www.itv.com/news/2021-10-27/i-have-to-protect-the-country-says-rishi-sunak-as-uk-faces-interest-rate-hike',
 230         'info_dict': {
 231             'id': 'i-have-to-protect-the-country-says-rishi-sunak-as-uk-faces-interest-rate-hike',
 232             'title': 'md5:6ef054dd9f069330db3dcc66cb772d32'
 233         },
 234         'playlist_count': 4
 235     }]
 236     BRIGHTCOVE_URL_TEMPLATE = 'http://players.brightcove.net/%s/%s_default/index.html?videoId=%s'
 237
 238     def _real_extract(self, url):
 239         playlist_id = self._match_id(url)
 240
 241         webpage = self._download_webpage(url, playlist_id)
 242
 243         json_map = try_get(
 244             self._search_nextjs_data(webpage, playlist_id),
 245             lambda x: x['props']['pageProps']['article']['body']['content']) or []
 246
 247         entries = []
 248         for video in json_map:
 249             if not any(video['data'].get(attr) == 'Brightcove' for attr in ('name', 'type')):
 250                 continue
 251             video_id = video['data']['id']
 252             account_id = video['data']['accountId']
 253             player_id = video['data']['playerId']
 254             entries.append(self.url_result(
 255                 smuggle_url(self.BRIGHTCOVE_URL_TEMPLATE % (account_id, player_id, video_id), {
 256                     # ITV does not like some GB IP ranges, so here are some
 257                     # IP blocks it accepts
 258                     'geo_ip_blocks': [
 259                         '193.113.0.0/16', '54.36.162.0/23', '159.65.16.0/21'
 260                     ],
 261                     'referrer': url,
 262                 }),
 263                 ie=BrightcoveNewIE.ie_key(), video_id=video_id))
 264
 265         title = self._og_search_title(webpage, fatal=False)
 266
 267         return self.playlist_result(entries, playlist_id, title)