yt_dlp/extractor/itv.py

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