yt_dlp/extractor/itv.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import json
   5
   6 from .common import InfoExtractor
   7 from .brightcove import BrightcoveNewIE
   8
   9 from ..compat import compat_str
  10 from ..utils import (
  11     base_url,
  12     clean_html,
  13     determine_ext,
  14     extract_attributes,
  15     ExtractorError,
  16     get_element_by_class,
  17     JSON_LD_RE,
  18     merge_dicts,
  19     parse_duration,
  20     smuggle_url,
  21     try_get,
  22     url_or_none,
  23     url_basename,
  24     urljoin,
  25 )
  26
  27
  28 class ITVIE(InfoExtractor):
  29     _VALID_URL = r'https?://(?:www\.)?itv\.com/hub/[^/]+/(?P<id>[0-9a-zA-Z]+)'
  30     _GEO_COUNTRIES = ['GB']
  31     _TESTS = [{
  32         'url': 'https://www.itv.com/hub/plebs/2a1873a0002',
  33         'info_dict': {
  34             'id': '2a1873a0002',
  35             'ext': 'mp4',
  36             'title': 'Plebs - The Orgy',
  37             'description': 'md5:4d7159af53ebd5b36e8b3ec82a41fdb4',
  38             'series': 'Plebs',
  39             'season_number': 1,
  40             'episode_number': 1,
  41             'thumbnail': r're:https?://hubimages\.itv\.com/episode/2_1873_0002'
  42         },
  43         'params': {
  44             # m3u8 download
  45             'skip_download': True,
  46         },
  47     }, {
  48         'url': 'https://www.itv.com/hub/the-jonathan-ross-show/2a1166a0209',
  49         'info_dict': {
  50             'id': '2a1166a0209',
  51             'ext': 'mp4',
  52             'title': 'The Jonathan Ross Show - Series 17 - Episode 8',
  53             'description': 'md5:3023dcdd375db1bc9967186cdb3f1399',
  54             'series': 'The Jonathan Ross Show',
  55             'episode_number': 8,
  56             'season_number': 17,
  57             'thumbnail': r're:https?://hubimages\.itv\.com/episode/2_1873_0002'
  58         },
  59         'params': {
  60             # m3u8 download
  61             'skip_download': True,
  62         },
  63     }, {
  64         # unavailable via data-playlist-url
  65         'url': 'https://www.itv.com/hub/through-the-keyhole/2a2271a0033',
  66         'only_matching': True,
  67     }, {
  68         # InvalidVodcrid
  69         'url': 'https://www.itv.com/hub/james-martins-saturday-morning/2a5159a0034',
  70         'only_matching': True,
  71     }, {
  72         # ContentUnavailable
  73         'url': 'https://www.itv.com/hub/whos-doing-the-dishes/2a2898a0024',
  74         'only_matching': True,
  75     }]
  76
  77     def _generate_api_headers(self, hmac):
  78         return merge_dicts({
  79             'Accept': 'application/vnd.itv.vod.playlist.v2+json',
  80             'Content-Type': 'application/json',
  81             'hmac': hmac.upper(),
  82         }, self.geo_verification_headers())
  83
  84     def _call_api(self, video_id, playlist_url, headers, platform_tag, featureset, fatal=True):
  85         return self._download_json(
  86             playlist_url, video_id, data=json.dumps({
  87                 'user': {
  88                     'itvUserId': '',
  89                     'entitlements': [],
  90                     'token': ''
  91                 },
  92                 'device': {
  93                     'manufacturer': 'Safari',
  94                     'model': '5',
  95                     'os': {
  96                         'name': 'Windows NT',
  97                         'version': '6.1',
  98                         'type': 'desktop'
  99                     }
 100                 },
 101                 'client': {
 102                     'version': '4.1',
 103                     'id': 'browser'
 104                 },
 105                 'variantAvailability': {
 106                     'featureset': {
 107                         'min': featureset,
 108                         'max': featureset
 109                     },
 110                     'platformTag': platform_tag
 111                 }
 112             }).encode(), headers=headers, fatal=fatal)
 113
 114     def _get_subtitles(self, video_id, variants, ios_playlist_url, headers, *args, **kwargs):
 115         subtitles = {}
 116         # Prefer last matching featureset
 117         # See: https://github.com/yt-dlp/yt-dlp/issues/986
 118         platform_tag_subs, featureset_subs = next(
 119             ((platform_tag, featureset)
 120              for platform_tag, featuresets in reversed(list(variants.items())) for featureset in featuresets
 121              if try_get(featureset, lambda x: x[2]) == 'outband-webvtt'),
 122             (None, None))
 123
 124         if platform_tag_subs and featureset_subs:
 125             subs_playlist = self._call_api(
 126                 video_id, ios_playlist_url, headers, platform_tag_subs, featureset_subs, fatal=False)
 127             subs = try_get(subs_playlist, lambda x: x['Playlist']['Video']['Subtitles'], list) or []
 128             for sub in subs:
 129                 if not isinstance(sub, dict):
 130                     continue
 131                 href = url_or_none(sub.get('Href'))
 132                 if not href:
 133                     continue
 134                 subtitles.setdefault('en', []).append({'url': href})
 135         return subtitles
 136
 137     def _real_extract(self, url):
 138         video_id = self._match_id(url)
 139         webpage = self._download_webpage(url, video_id)
 140         params = extract_attributes(self._search_regex(
 141             r'(?s)(<[^>]+id="video"[^>]*>)', webpage, 'params'))
 142         variants = self._parse_json(
 143             try_get(params, lambda x: x['data-video-variants'], compat_str) or '{}',
 144             video_id, fatal=False)
 145         # Prefer last matching featureset
 146         # See: https://github.com/yt-dlp/yt-dlp/issues/986
 147         platform_tag_video, featureset_video = next(
 148             ((platform_tag, featureset)
 149              for platform_tag, featuresets in reversed(list(variants.items())) for featureset in featuresets
 150              if set(try_get(featureset, lambda x: x[:2]) or []) == {'aes', 'hls'}),
 151             (None, None))
 152         if not platform_tag_video or not featureset_video:
 153             raise ExtractorError('No downloads available', expected=True, video_id=video_id)
 154
 155         ios_playlist_url = params.get('data-video-playlist') or params['data-video-id']
 156         headers = self._generate_api_headers(params['data-video-hmac'])
 157         ios_playlist = self._call_api(
 158             video_id, ios_playlist_url, headers, platform_tag_video, featureset_video)
 159
 160         video_data = try_get(ios_playlist, lambda x: x['Playlist']['Video'], dict) or {}
 161         ios_base_url = video_data.get('Base')
 162         formats = []
 163         for media_file in (video_data.get('MediaFiles') or []):
 164             href = media_file.get('Href')
 165             if not href:
 166                 continue
 167             if ios_base_url:
 168                 href = ios_base_url + href
 169             ext = determine_ext(href)
 170             if ext == 'm3u8':
 171                 formats.extend(self._extract_m3u8_formats(
 172                     href, video_id, 'mp4', entry_protocol='m3u8_native',
 173                     m3u8_id='hls', fatal=False))
 174             else:
 175                 formats.append({
 176                     'url': href,
 177                 })
 178         self._sort_formats(formats)
 179         info = self._search_json_ld(webpage, video_id, default={})
 180         if not info:
 181             json_ld = self._parse_json(self._search_regex(
 182                 JSON_LD_RE, webpage, 'JSON-LD', '{}',
 183                 group='json_ld'), video_id, fatal=False)
 184             if json_ld and json_ld.get('@type') == 'BreadcrumbList':
 185                 for ile in (json_ld.get('itemListElement:') or []):
 186                     item = ile.get('item:') or {}
 187                     if item.get('@type') == 'TVEpisode':
 188                         item['@context'] = 'http://schema.org'
 189                         info = self._json_ld(item, video_id, fatal=False) or {}
 190                         break
 191
 192         thumbnails = []
 193         thumbnail_url = try_get(params, lambda x: x['data-video-posterframe'], compat_str)
 194         if thumbnail_url:
 195             thumbnails.extend([{
 196                 'url': thumbnail_url.format(width=1920, height=1080, quality=100, blur=0, bg='false'),
 197                 'width': 1920,
 198                 'height': 1080,
 199             }, {
 200                 'url': urljoin(base_url(thumbnail_url), url_basename(thumbnail_url)),
 201                 'preference': -2
 202             }])
 203
 204         thumbnail_url = self._html_search_meta(['og:image', 'twitter:image'], webpage, default=None)
 205         if thumbnail_url:
 206             thumbnails.append({
 207                 'url': thumbnail_url,
 208             })
 209         self._remove_duplicate_formats(thumbnails)
 210
 211         return merge_dicts({
 212             'id': video_id,
 213             'title': self._html_search_meta(['og:title', 'twitter:title'], webpage),
 214             'formats': formats,
 215             'subtitles': self.extract_subtitles(video_id, variants, ios_playlist_url, headers),
 216             'duration': parse_duration(video_data.get('Duration')),
 217             'description': clean_html(get_element_by_class('episode-info__synopsis', webpage)),
 218             'thumbnails': thumbnails
 219         }, info)
 220
 221
 222 class ITVBTCCIE(InfoExtractor):
 223     _VALID_URL = r'https?://(?:www\.)?itv\.com/(?:news|btcc)/(?:[^/]+/)*(?P<id>[^/?#&]+)'
 224     _TESTS = [{
 225         'url': 'https://www.itv.com/btcc/articles/btcc-2019-brands-hatch-gp-race-action',
 226         'info_dict': {
 227             'id': 'btcc-2019-brands-hatch-gp-race-action',
 228             'title': 'BTCC 2019: Brands Hatch GP race action',
 229         },
 230         'playlist_count': 12,
 231     }, {
 232         'url': 'https://www.itv.com/news/2021-10-27/i-have-to-protect-the-country-says-rishi-sunak-as-uk-faces-interest-rate-hike',
 233         'info_dict': {
 234             'id': 'i-have-to-protect-the-country-says-rishi-sunak-as-uk-faces-interest-rate-hike',
 235             'title': 'md5:6ef054dd9f069330db3dcc66cb772d32'
 236         },
 237         'playlist_count': 4
 238     }]
 239     BRIGHTCOVE_URL_TEMPLATE = 'http://players.brightcove.net/%s/%s_default/index.html?videoId=%s'
 240
 241     def _real_extract(self, url):
 242         playlist_id = self._match_id(url)
 243
 244         webpage = self._download_webpage(url, playlist_id)
 245
 246         json_map = try_get(
 247             self._search_nextjs_data(webpage, playlist_id),
 248             lambda x: x['props']['pageProps']['article']['body']['content']) or []
 249
 250         entries = []
 251         for video in json_map:
 252             if not any(video['data'].get(attr) == 'Brightcove' for attr in ('name', 'type')):
 253                 continue
 254             video_id = video['data']['id']
 255             account_id = video['data']['accountId']
 256             player_id = video['data']['playerId']
 257             entries.append(self.url_result(
 258                 smuggle_url(self.BRIGHTCOVE_URL_TEMPLATE % (account_id, player_id, video_id), {
 259                     # ITV does not like some GB IP ranges, so here are some
 260                     # IP blocks it accepts
 261                     'geo_ip_blocks': [
 262                         '193.113.0.0/16', '54.36.162.0/23', '159.65.16.0/21'
 263                     ],
 264                     'referrer': url,
 265                 }),
 266                 ie=BrightcoveNewIE.ie_key(), video_id=video_id))
 267
 268         title = self._og_search_title(webpage, fatal=False)
 269
 270         return self.playlist_result(entries, playlist_id, title)