youtube_dlc/extractor/itv.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import uuid
   5 import xml.etree.ElementTree as etree
   6 import json
   7 import re
   8
   9 from .common import InfoExtractor
  10 from .brightcove import BrightcoveNewIE
  11 from ..compat import (
  12     compat_str,
  13     compat_etree_register_namespace,
  14 )
  15 from ..utils import (
  16     determine_ext,
  17     ExtractorError,
  18     extract_attributes,
  19     int_or_none,
  20     merge_dicts,
  21     parse_duration,
  22     smuggle_url,
  23     try_get,
  24     url_or_none,
  25     xpath_with_ns,
  26     xpath_element,
  27     xpath_text,
  28 )
  29
  30
  31 class ITVIE(InfoExtractor):
  32     _VALID_URL = r'https?://(?:www\.)?itv\.com/hub/[^/]+/(?P<id>[0-9a-zA-Z]+)'
  33     _GEO_COUNTRIES = ['GB']
  34     _TESTS = [{
  35         'url': 'http://www.itv.com/hub/mr-bean-animated-series/2a2936a0053',
  36         'info_dict': {
  37             'id': '2a2936a0053',
  38             'ext': 'flv',
  39             'title': 'Home Movie',
  40         },
  41         'params': {
  42             # rtmp download
  43             'skip_download': True,
  44         },
  45     }, {
  46         # unavailable via data-playlist-url
  47         'url': 'https://www.itv.com/hub/through-the-keyhole/2a2271a0033',
  48         'only_matching': True,
  49     }, {
  50         # InvalidVodcrid
  51         'url': 'https://www.itv.com/hub/james-martins-saturday-morning/2a5159a0034',
  52         'only_matching': True,
  53     }, {
  54         # ContentUnavailable
  55         'url': 'https://www.itv.com/hub/whos-doing-the-dishes/2a2898a0024',
  56         'only_matching': True,
  57     }]
  58
  59     def _real_extract(self, url):
  60         video_id = self._match_id(url)
  61         webpage = self._download_webpage(url, video_id)
  62         params = extract_attributes(self._search_regex(
  63             r'(?s)(<[^>]+id="video"[^>]*>)', webpage, 'params'))
  64
  65         ns_map = {
  66             'soapenv': 'http://schemas.xmlsoap.org/soap/envelope/',
  67             'tem': 'http://tempuri.org/',
  68             'itv': 'http://schemas.datacontract.org/2004/07/Itv.BB.Mercury.Common.Types',
  69             'com': 'http://schemas.itv.com/2009/05/Common',
  70         }
  71         for ns, full_ns in ns_map.items():
  72             compat_etree_register_namespace(ns, full_ns)
  73
  74         def _add_ns(name):
  75             return xpath_with_ns(name, ns_map)
  76
  77         def _add_sub_element(element, name):
  78             return etree.SubElement(element, _add_ns(name))
  79
  80         production_id = (
  81             params.get('data-video-autoplay-id')
  82             or '%s#001' % (
  83                 params.get('data-video-episode-id')
  84                 or video_id.replace('a', '/')))
  85
  86         req_env = etree.Element(_add_ns('soapenv:Envelope'))
  87         _add_sub_element(req_env, 'soapenv:Header')
  88         body = _add_sub_element(req_env, 'soapenv:Body')
  89         get_playlist = _add_sub_element(body, ('tem:GetPlaylist'))
  90         request = _add_sub_element(get_playlist, 'tem:request')
  91         _add_sub_element(request, 'itv:ProductionId').text = production_id
  92         _add_sub_element(request, 'itv:RequestGuid').text = compat_str(uuid.uuid4()).upper()
  93         vodcrid = _add_sub_element(request, 'itv:Vodcrid')
  94         _add_sub_element(vodcrid, 'com:Id')
  95         _add_sub_element(request, 'itv:Partition')
  96         user_info = _add_sub_element(get_playlist, 'tem:userInfo')
  97         _add_sub_element(user_info, 'itv:Broadcaster').text = 'Itv'
  98         _add_sub_element(user_info, 'itv:DM')
  99         _add_sub_element(user_info, 'itv:RevenueScienceValue')
 100         _add_sub_element(user_info, 'itv:SessionId')
 101         _add_sub_element(user_info, 'itv:SsoToken')
 102         _add_sub_element(user_info, 'itv:UserToken')
 103         site_info = _add_sub_element(get_playlist, 'tem:siteInfo')
 104         _add_sub_element(site_info, 'itv:AdvertisingRestriction').text = 'None'
 105         _add_sub_element(site_info, 'itv:AdvertisingSite').text = 'ITV'
 106         _add_sub_element(site_info, 'itv:AdvertisingType').text = 'Any'
 107         _add_sub_element(site_info, 'itv:Area').text = 'ITVPLAYER.VIDEO'
 108         _add_sub_element(site_info, 'itv:Category')
 109         _add_sub_element(site_info, 'itv:Platform').text = 'DotCom'
 110         _add_sub_element(site_info, 'itv:Site').text = 'ItvCom'
 111         device_info = _add_sub_element(get_playlist, 'tem:deviceInfo')
 112         _add_sub_element(device_info, 'itv:ScreenSize').text = 'Big'
 113         player_info = _add_sub_element(get_playlist, 'tem:playerInfo')
 114         _add_sub_element(player_info, 'itv:Version').text = '2'
 115
 116         headers = self.geo_verification_headers()
 117         headers.update({
 118             'Content-Type': 'text/xml; charset=utf-8',
 119             'SOAPAction': 'http://tempuri.org/PlaylistService/GetPlaylist',
 120         })
 121
 122         info = self._search_json_ld(webpage, video_id, default={})
 123         formats = []
 124         subtitles = {}
 125
 126         def extract_subtitle(sub_url):
 127             ext = determine_ext(sub_url, 'ttml')
 128             subtitles.setdefault('en', []).append({
 129                 'url': sub_url,
 130                 'ext': 'ttml' if ext == 'xml' else ext,
 131             })
 132
 133         resp_env = self._download_xml(
 134             params['data-playlist-url'], video_id,
 135             headers=headers, data=etree.tostring(req_env), fatal=False)
 136         if resp_env:
 137             playlist = xpath_element(resp_env, './/Playlist')
 138             if playlist is None:
 139                 fault_code = xpath_text(resp_env, './/faultcode')
 140                 fault_string = xpath_text(resp_env, './/faultstring')
 141                 if fault_code == 'InvalidGeoRegion':
 142                     self.raise_geo_restricted(
 143                         msg=fault_string, countries=self._GEO_COUNTRIES)
 144                 elif fault_code not in (
 145                         'InvalidEntity', 'InvalidVodcrid', 'ContentUnavailable'):
 146                     raise ExtractorError(
 147                         '%s said: %s' % (self.IE_NAME, fault_string), expected=True)
 148                 info.update({
 149                     'title': self._og_search_title(webpage),
 150                     'episode_title': params.get('data-video-episode'),
 151                     'series': params.get('data-video-title'),
 152                 })
 153             else:
 154                 title = xpath_text(playlist, 'EpisodeTitle', default=None)
 155                 info.update({
 156                     'title': title,
 157                     'episode_title': title,
 158                     'episode_number': int_or_none(xpath_text(playlist, 'EpisodeNumber')),
 159                     'series': xpath_text(playlist, 'ProgrammeTitle'),
 160                     'duration': parse_duration(xpath_text(playlist, 'Duration')),
 161                 })
 162                 video_element = xpath_element(playlist, 'VideoEntries/Video', fatal=True)
 163                 media_files = xpath_element(video_element, 'MediaFiles', fatal=True)
 164                 rtmp_url = media_files.attrib['base']
 165
 166                 for media_file in media_files.findall('MediaFile'):
 167                     play_path = xpath_text(media_file, 'URL')
 168                     if not play_path:
 169                         continue
 170                     tbr = int_or_none(media_file.get('bitrate'), 1000)
 171                     f = {
 172                         'format_id': 'rtmp' + ('-%d' % tbr if tbr else ''),
 173                         'play_path': play_path,
 174                         # Providing this swfVfy allows to avoid truncated downloads
 175                         'player_url': 'http://www.itv.com/mercury/Mercury_VideoPlayer.swf',
 176                         'page_url': url,
 177                         'tbr': tbr,
 178                         'ext': 'flv',
 179                     }
 180                     app = self._search_regex(
 181                         'rtmpe?://[^/]+/(.+)$', rtmp_url, 'app', default=None)
 182                     if app:
 183                         f.update({
 184                             'url': rtmp_url.split('?', 1)[0],
 185                             'app': app,
 186                         })
 187                     else:
 188                         f['url'] = rtmp_url
 189                     formats.append(f)
 190
 191                 for caption_url in video_element.findall('ClosedCaptioningURIs/URL'):
 192                     if caption_url.text:
 193                         extract_subtitle(caption_url.text)
 194
 195         ios_playlist_url = params.get('data-video-playlist') or params.get('data-video-id')
 196         hmac = params.get('data-video-hmac')
 197         if ios_playlist_url and hmac and re.match(r'https?://', ios_playlist_url):
 198             headers = self.geo_verification_headers()
 199             headers.update({
 200                 'Accept': 'application/vnd.itv.vod.playlist.v2+json',
 201                 'Content-Type': 'application/json',
 202                 'hmac': hmac.upper(),
 203             })
 204             ios_playlist = self._download_json(
 205                 ios_playlist_url, video_id, data=json.dumps({
 206                     'user': {
 207                         'itvUserId': '',
 208                         'entitlements': [],
 209                         'token': ''
 210                     },
 211                     'device': {
 212                         'manufacturer': 'Safari',
 213                         'model': '5',
 214                         'os': {
 215                             'name': 'Windows NT',
 216                             'version': '6.1',
 217                             'type': 'desktop'
 218                         }
 219                     },
 220                     'client': {
 221                         'version': '4.1',
 222                         'id': 'browser'
 223                     },
 224                     'variantAvailability': {
 225                         'featureset': {
 226                             'min': ['hls', 'aes', 'outband-webvtt'],
 227                             'max': ['hls', 'aes', 'outband-webvtt']
 228                         },
 229                         'platformTag': 'dotcom'
 230                     }
 231                 }).encode(), headers=headers, fatal=False)
 232             if ios_playlist:
 233                 video_data = ios_playlist.get('Playlist', {}).get('Video', {})
 234                 ios_base_url = video_data.get('Base')
 235                 for media_file in video_data.get('MediaFiles', []):
 236                     href = media_file.get('Href')
 237                     if not href:
 238                         continue
 239                     if ios_base_url:
 240                         href = ios_base_url + href
 241                     ext = determine_ext(href)
 242                     if ext == 'm3u8':
 243                         formats.extend(self._extract_m3u8_formats(
 244                             href, video_id, 'mp4', entry_protocol='m3u8_native',
 245                             m3u8_id='hls', fatal=False))
 246                     else:
 247                         formats.append({
 248                             'url': href,
 249                         })
 250                 subs = video_data.get('Subtitles')
 251                 if isinstance(subs, list):
 252                     for sub in subs:
 253                         if not isinstance(sub, dict):
 254                             continue
 255                         href = url_or_none(sub.get('Href'))
 256                         if href:
 257                             extract_subtitle(href)
 258                 if not info.get('duration'):
 259                     info['duration'] = parse_duration(video_data.get('Duration'))
 260
 261         self._sort_formats(formats)
 262
 263         info.update({
 264             'id': video_id,
 265             'formats': formats,
 266             'subtitles': subtitles,
 267         })
 268
 269         webpage_info = self._search_json_ld(webpage, video_id, default={})
 270         if not webpage_info.get('title'):
 271             webpage_info['title'] = self._html_search_regex(
 272                 r'(?s)<h\d+[^>]+\bclass=["\'][^>]*episode-title["\'][^>]*>([^<]+)<',
 273                 webpage, 'title', default=None) or self._og_search_title(
 274                 webpage, default=None) or self._html_search_meta(
 275                 'twitter:title', webpage, 'title',
 276                 default=None) or webpage_info['episode']
 277
 278         return merge_dicts(info, webpage_info)
 279
 280
 281 class ITVBTCCIE(InfoExtractor):
 282     _VALID_URL = r'https?://(?:www\.)?itv\.com/btcc/(?:[^/]+/)*(?P<id>[^/?#&]+)'
 283     _TEST = {
 284         'url': 'https://www.itv.com/btcc/articles/btcc-2019-brands-hatch-gp-race-action',
 285         'info_dict': {
 286             'id': 'btcc-2019-brands-hatch-gp-race-action',
 287             'title': 'BTCC 2019: Brands Hatch GP race action',
 288         },
 289         'playlist_count': 12,
 290     }
 291     BRIGHTCOVE_URL_TEMPLATE = 'http://players.brightcove.net/1582188683001/HkiHLnNRx_default/index.html?videoId=%s'
 292
 293     def _real_extract(self, url):
 294         playlist_id = self._match_id(url)
 295
 296         webpage = self._download_webpage(url, playlist_id)
 297
 298         json_map = try_get(self._parse_json(self._html_search_regex(
 299             '(?s)<script[^>]+id=[\'"]__NEXT_DATA__[^>]*>([^<]+)</script>', webpage, 'json_map'), playlist_id),
 300             lambda x: x['props']['pageProps']['article']['body']['content']) or []
 301
 302         # Discard empty objects
 303         video_ids = []
 304         for video in json_map:
 305             if video['data'].get('id'):
 306                 video_ids.append(video['data']['id'])
 307
 308         entries = [
 309             self.url_result(
 310                 smuggle_url(self.BRIGHTCOVE_URL_TEMPLATE % video_id, {
 311                     # ITV does not like some GB IP ranges, so here are some
 312                     # IP blocks it accepts
 313                     'geo_ip_blocks': [
 314                         '193.113.0.0/16', '54.36.162.0/23', '159.65.16.0/21'
 315                     ],
 316                     'referrer': url,
 317                 }),
 318                 ie=BrightcoveNewIE.ie_key(), video_id=video_id)
 319             for video_id in video_ids]
 320
 321         title = self._og_search_title(webpage, fatal=False)
 322
 323         return self.playlist_result(entries, playlist_id, title)