youtube_dlc/extractor/itv.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import json
   5 import re
   6
   7 from .common import InfoExtractor
   8 from .brightcove import BrightcoveNewIE
   9 from ..utils import (
  10     clean_html,
  11     determine_ext,
  12     extract_attributes,
  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 )
  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/liar/2a4547a0012',
  28         'info_dict': {
  29             'id': '2a4547a0012',
  30             'ext': 'mp4',
  31             'title': 'Liar - Series 2 - Episode 6',
  32             'description': 'md5:d0f91536569dec79ea184f0a44cca089',
  33             'series': 'Liar',
  34             'season_number': 2,
  35             'episode_number': 6,
  36         },
  37         'params': {
  38             # m3u8 download
  39             'skip_download': True,
  40         },
  41     }, {
  42         # unavailable via data-playlist-url
  43         'url': 'https://www.itv.com/hub/through-the-keyhole/2a2271a0033',
  44         'only_matching': True,
  45     }, {
  46         # InvalidVodcrid
  47         'url': 'https://www.itv.com/hub/james-martins-saturday-morning/2a5159a0034',
  48         'only_matching': True,
  49     }, {
  50         # ContentUnavailable
  51         'url': 'https://www.itv.com/hub/whos-doing-the-dishes/2a2898a0024',
  52         'only_matching': True,
  53     }]
  54
  55     def _real_extract(self, url):
  56         video_id = self._match_id(url)
  57         webpage = self._download_webpage(url, video_id)
  58         params = extract_attributes(self._search_regex(
  59             r'(?s)(<[^>]+id="video"[^>]*>)', webpage, 'params'))
  60
  61         ios_playlist_url = params.get('data-video-playlist') or params['data-video-id']
  62         hmac = params['data-video-hmac']
  63         headers = self.geo_verification_headers()
  64         headers.update({
  65             'Accept': 'application/vnd.itv.vod.playlist.v2+json',
  66             'Content-Type': 'application/json',
  67             'hmac': hmac.upper(),
  68         })
  69         ios_playlist = self._download_json(
  70             ios_playlist_url, video_id, data=json.dumps({
  71                 'user': {
  72                     'itvUserId': '',
  73                     'entitlements': [],
  74                     'token': ''
  75                 },
  76                 'device': {
  77                     'manufacturer': 'Safari',
  78                     'model': '5',
  79                     'os': {
  80                         'name': 'Windows NT',
  81                         'version': '6.1',
  82                         'type': 'desktop'
  83                     }
  84                 },
  85                 'client': {
  86                     'version': '4.1',
  87                     'id': 'browser'
  88                 },
  89                 'variantAvailability': {
  90                     'featureset': {
  91                         'min': ['hls', 'aes', 'outband-webvtt'],
  92                         'max': ['hls', 'aes', 'outband-webvtt']
  93                     },
  94                     'platformTag': 'dotcom'
  95                 }
  96             }).encode(), headers=headers)
  97         video_data = ios_playlist['Playlist']['Video']
  98         ios_base_url = video_data.get('Base')
  99
 100         formats = []
 101         for media_file in (video_data.get('MediaFiles') or []):
 102             href = media_file.get('Href')
 103             if not href:
 104                 continue
 105             if ios_base_url:
 106                 href = ios_base_url + href
 107             ext = determine_ext(href)
 108             if ext == 'm3u8':
 109                 formats.extend(self._extract_m3u8_formats(
 110                     href, video_id, 'mp4', entry_protocol='m3u8_native',
 111                     m3u8_id='hls', fatal=False))
 112             else:
 113                 formats.append({
 114                     'url': href,
 115                 })
 116         self._sort_formats(formats)
 117
 118         subtitles = {}
 119         subs = video_data.get('Subtitles') or []
 120         for sub in subs:
 121             if not isinstance(sub, dict):
 122                 continue
 123             href = url_or_none(sub.get('Href'))
 124             if not href:
 125                 continue
 126             subtitles.setdefault('en', []).append({
 127                 'url': href,
 128                 'ext': determine_ext(href, 'vtt'),
 129             })
 130
 131         info = self._search_json_ld(webpage, video_id, default={})
 132         if not info:
 133             json_ld = self._parse_json(self._search_regex(
 134                 JSON_LD_RE, webpage, 'JSON-LD', '{}',
 135                 group='json_ld'), video_id, fatal=False)
 136             if json_ld and json_ld.get('@type') == 'BreadcrumbList':
 137                 for ile in (json_ld.get('itemListElement:') or []):
 138                     item = ile.get('item:') or {}
 139                     if item.get('@type') == 'TVEpisode':
 140                         item['@context'] = 'http://schema.org'
 141                         info = self._json_ld(item, video_id, fatal=False) or {}
 142                         break
 143
 144         return merge_dicts({
 145             'id': video_id,
 146             'title': self._html_search_meta(['og:title', 'twitter:title'], webpage),
 147             'formats': formats,
 148             'subtitles': subtitles,
 149             'duration': parse_duration(video_data.get('Duration')),
 150             'description': clean_html(get_element_by_class('episode-info__synopsis', webpage)),
 151         }, info)
 152
 153
 154 class ITVBTCCIE(InfoExtractor):
 155     _VALID_URL = r'https?://(?:www\.)?itv\.com/btcc/(?:[^/]+/)*(?P<id>[^/?#&]+)'
 156     _TEST = {
 157         'url': 'https://www.itv.com/btcc/articles/btcc-2019-brands-hatch-gp-race-action',
 158         'info_dict': {
 159             'id': 'btcc-2019-brands-hatch-gp-race-action',
 160             'title': 'BTCC 2019: Brands Hatch GP race action',
 161         },
 162         'playlist_count': 12,
 163     }
 164     BRIGHTCOVE_URL_TEMPLATE = 'http://players.brightcove.net/1582188683001/HkiHLnNRx_default/index.html?videoId=%s'
 165
 166     def _real_extract(self, url):
 167         playlist_id = self._match_id(url)
 168
 169         webpage = self._download_webpage(url, playlist_id)
 170
 171         json_map = try_get(self._parse_json(self._html_search_regex(
 172             '(?s)<script[^>]+id=[\'"]__NEXT_DATA__[^>]*>([^<]+)</script>', webpage, 'json_map'), playlist_id),
 173             lambda x: x['props']['pageProps']['article']['body']['content']) or []
 174
 175         # Discard empty objects
 176         video_ids = []
 177         for video in json_map:
 178             if video['data'].get('id'):
 179                 video_ids.append(video['data']['id'])
 180
 181         entries = [
 182             self.url_result(
 183                 smuggle_url(self.BRIGHTCOVE_URL_TEMPLATE % video_id, {
 184                     # ITV does not like some GB IP ranges, so here are some
 185                     # IP blocks it accepts
 186                     'geo_ip_blocks': [
 187                         '193.113.0.0/16', '54.36.162.0/23', '159.65.16.0/21'
 188                     ],
 189                     'referrer': url,
 190                 }),
 191                 ie=BrightcoveNewIE.ie_key(), video_id=video_id)
 192             for video_id in video_ids]
 193
 194         title = self._og_search_title(webpage, fatal=False)
 195
 196         return self.playlist_result(entries, playlist_id, title)