Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/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     extract_attributes,
  17     xpath_with_ns,
  18     xpath_element,
  19     xpath_text,
  20     int_or_none,
  21     parse_duration,
  22     smuggle_url,
  23     ExtractorError,
  24     determine_ext,
  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': 'http://www.itv.com/hub/mr-bean-animated-series/2a2936a0053',
  33         'info_dict': {
  34             'id': '2a2936a0053',
  35             'ext': 'flv',
  36             'title': 'Home Movie',
  37         },
  38         'params': {
  39             # rtmp download
  40             'skip_download': True,
  41         },
  42     }, {
  43         # unavailable via data-playlist-url
  44         'url': 'https://www.itv.com/hub/through-the-keyhole/2a2271a0033',
  45         'only_matching': True,
  46     }, {
  47         # InvalidVodcrid
  48         'url': 'https://www.itv.com/hub/james-martins-saturday-morning/2a5159a0034',
  49         'only_matching': True,
  50     }, {
  51         # ContentUnavailable
  52         'url': 'https://www.itv.com/hub/whos-doing-the-dishes/2a2898a0024',
  53         'only_matching': True,
  54     }]
  55
  56     def _real_extract(self, url):
  57         video_id = self._match_id(url)
  58         webpage = self._download_webpage(url, video_id)
  59         params = extract_attributes(self._search_regex(
  60             r'(?s)(<[^>]+id="video"[^>]*>)', webpage, 'params'))
  61
  62         ns_map = {
  63             'soapenv': 'http://schemas.xmlsoap.org/soap/envelope/',
  64             'tem': 'http://tempuri.org/',
  65             'itv': 'http://schemas.datacontract.org/2004/07/Itv.BB.Mercury.Common.Types',
  66             'com': 'http://schemas.itv.com/2009/05/Common',
  67         }
  68         for ns, full_ns in ns_map.items():
  69             compat_etree_register_namespace(ns, full_ns)
  70
  71         def _add_ns(name):
  72             return xpath_with_ns(name, ns_map)
  73
  74         def _add_sub_element(element, name):
  75             return etree.SubElement(element, _add_ns(name))
  76
  77         production_id = (
  78             params.get('data-video-autoplay-id') or
  79             '%s#001' % (
  80                 params.get('data-video-episode-id') or
  81                 video_id.replace('a', '/')))
  82
  83         req_env = etree.Element(_add_ns('soapenv:Envelope'))
  84         _add_sub_element(req_env, 'soapenv:Header')
  85         body = _add_sub_element(req_env, 'soapenv:Body')
  86         get_playlist = _add_sub_element(body, ('tem:GetPlaylist'))
  87         request = _add_sub_element(get_playlist, 'tem:request')
  88         _add_sub_element(request, 'itv:ProductionId').text = production_id
  89         _add_sub_element(request, 'itv:RequestGuid').text = compat_str(uuid.uuid4()).upper()
  90         vodcrid = _add_sub_element(request, 'itv:Vodcrid')
  91         _add_sub_element(vodcrid, 'com:Id')
  92         _add_sub_element(request, 'itv:Partition')
  93         user_info = _add_sub_element(get_playlist, 'tem:userInfo')
  94         _add_sub_element(user_info, 'itv:Broadcaster').text = 'Itv'
  95         _add_sub_element(user_info, 'itv:DM')
  96         _add_sub_element(user_info, 'itv:RevenueScienceValue')
  97         _add_sub_element(user_info, 'itv:SessionId')
  98         _add_sub_element(user_info, 'itv:SsoToken')
  99         _add_sub_element(user_info, 'itv:UserToken')
 100         site_info = _add_sub_element(get_playlist, 'tem:siteInfo')
 101         _add_sub_element(site_info, 'itv:AdvertisingRestriction').text = 'None'
 102         _add_sub_element(site_info, 'itv:AdvertisingSite').text = 'ITV'
 103         _add_sub_element(site_info, 'itv:AdvertisingType').text = 'Any'
 104         _add_sub_element(site_info, 'itv:Area').text = 'ITVPLAYER.VIDEO'
 105         _add_sub_element(site_info, 'itv:Category')
 106         _add_sub_element(site_info, 'itv:Platform').text = 'DotCom'
 107         _add_sub_element(site_info, 'itv:Site').text = 'ItvCom'
 108         device_info = _add_sub_element(get_playlist, 'tem:deviceInfo')
 109         _add_sub_element(device_info, 'itv:ScreenSize').text = 'Big'
 110         player_info = _add_sub_element(get_playlist, 'tem:playerInfo')
 111         _add_sub_element(player_info, 'itv:Version').text = '2'
 112
 113         headers = self.geo_verification_headers()
 114         headers.update({
 115             'Content-Type': 'text/xml; charset=utf-8',
 116             'SOAPAction': 'http://tempuri.org/PlaylistService/GetPlaylist',
 117         })
 118
 119         info = self._search_json_ld(webpage, video_id, default={})
 120         formats = []
 121         subtitles = {}
 122
 123         def extract_subtitle(sub_url):
 124             ext = determine_ext(sub_url, 'ttml')
 125             subtitles.setdefault('en', []).append({
 126                 'url': sub_url,
 127                 'ext': 'ttml' if ext == 'xml' else ext,
 128             })
 129
 130         resp_env = self._download_xml(
 131             params['data-playlist-url'], video_id,
 132             headers=headers, data=etree.tostring(req_env))
 133         playlist = xpath_element(resp_env, './/Playlist')
 134         if playlist is None:
 135             fault_code = xpath_text(resp_env, './/faultcode')
 136             fault_string = xpath_text(resp_env, './/faultstring')
 137             if fault_code == 'InvalidGeoRegion':
 138                 self.raise_geo_restricted(
 139                     msg=fault_string, countries=self._GEO_COUNTRIES)
 140             elif fault_code not in (
 141                     'InvalidEntity', 'InvalidVodcrid', 'ContentUnavailable'):
 142                 raise ExtractorError(
 143                     '%s said: %s' % (self.IE_NAME, fault_string), expected=True)
 144             info.update({
 145                 'title': self._og_search_title(webpage),
 146                 'episode_title': params.get('data-video-episode'),
 147                 'series': params.get('data-video-title'),
 148             })
 149         else:
 150             title = xpath_text(playlist, 'EpisodeTitle', default=None)
 151             info.update({
 152                 'title': title,
 153                 'episode_title': title,
 154                 'episode_number': int_or_none(xpath_text(playlist, 'EpisodeNumber')),
 155                 'series': xpath_text(playlist, 'ProgrammeTitle'),
 156                 'duration': parse_duration(xpath_text(playlist, 'Duration')),
 157             })
 158             video_element = xpath_element(playlist, 'VideoEntries/Video', fatal=True)
 159             media_files = xpath_element(video_element, 'MediaFiles', fatal=True)
 160             rtmp_url = media_files.attrib['base']
 161
 162             for media_file in media_files.findall('MediaFile'):
 163                 play_path = xpath_text(media_file, 'URL')
 164                 if not play_path:
 165                     continue
 166                 tbr = int_or_none(media_file.get('bitrate'), 1000)
 167                 f = {
 168                     'format_id': 'rtmp' + ('-%d' % tbr if tbr else ''),
 169                     'play_path': play_path,
 170                     # Providing this swfVfy allows to avoid truncated downloads
 171                     'player_url': 'http://www.itv.com/mercury/Mercury_VideoPlayer.swf',
 172                     'page_url': url,
 173                     'tbr': tbr,
 174                     'ext': 'flv',
 175                 }
 176                 app = self._search_regex(
 177                     'rtmpe?://[^/]+/(.+)$', rtmp_url, 'app', default=None)
 178                 if app:
 179                     f.update({
 180                         'url': rtmp_url.split('?', 1)[0],
 181                         'app': app,
 182                     })
 183                 else:
 184                     f['url'] = rtmp_url
 185                 formats.append(f)
 186
 187             for caption_url in video_element.findall('ClosedCaptioningURIs/URL'):
 188                 if caption_url.text:
 189                     extract_subtitle(caption_url.text)
 190
 191         ios_playlist_url = params.get('data-video-playlist') or params.get('data-video-id')
 192         hmac = params.get('data-video-hmac')
 193         if ios_playlist_url and hmac and re.match(r'https?://', ios_playlist_url):
 194             headers = self.geo_verification_headers()
 195             headers.update({
 196                 'Accept': 'application/vnd.itv.vod.playlist.v2+json',
 197                 'Content-Type': 'application/json',
 198                 'hmac': hmac.upper(),
 199             })
 200             ios_playlist = self._download_json(
 201                 ios_playlist_url, video_id, data=json.dumps({
 202                     'user': {
 203                         'itvUserId': '',
 204                         'entitlements': [],
 205                         'token': ''
 206                     },
 207                     'device': {
 208                         'manufacturer': 'Safari',
 209                         'model': '5',
 210                         'os': {
 211                             'name': 'Windows NT',
 212                             'version': '6.1',
 213                             'type': 'desktop'
 214                         }
 215                     },
 216                     'client': {
 217                         'version': '4.1',
 218                         'id': 'browser'
 219                     },
 220                     'variantAvailability': {
 221                         'featureset': {
 222                             'min': ['hls', 'aes', 'outband-webvtt'],
 223                             'max': ['hls', 'aes', 'outband-webvtt']
 224                         },
 225                         'platformTag': 'dotcom'
 226                     }
 227                 }).encode(), headers=headers, fatal=False)
 228             if ios_playlist:
 229                 video_data = ios_playlist.get('Playlist', {}).get('Video', {})
 230                 ios_base_url = video_data.get('Base')
 231                 for media_file in video_data.get('MediaFiles', []):
 232                     href = media_file.get('Href')
 233                     if not href:
 234                         continue
 235                     if ios_base_url:
 236                         href = ios_base_url + href
 237                     ext = determine_ext(href)
 238                     if ext == 'm3u8':
 239                         formats.extend(self._extract_m3u8_formats(
 240                             href, video_id, 'mp4', entry_protocol='m3u8_native',
 241                             m3u8_id='hls', fatal=False))
 242                     else:
 243                         formats.append({
 244                             'url': href,
 245                         })
 246                 subs = video_data.get('Subtitles')
 247                 if isinstance(subs, list):
 248                     for sub in subs:
 249                         if not isinstance(sub, dict):
 250                             continue
 251                         href = sub.get('Href')
 252                         if isinstance(href, compat_str):
 253                             extract_subtitle(href)
 254                 if not info.get('duration'):
 255                     info['duration'] = parse_duration(video_data.get('Duration'))
 256
 257         self._sort_formats(formats)
 258
 259         info.update({
 260             'id': video_id,
 261             'formats': formats,
 262             'subtitles': subtitles,
 263         })
 264         return info
 265
 266
 267 class ITVBTCCIE(InfoExtractor):
 268     _VALID_URL = r'https?://(?:www\.)?itv\.com/btcc/(?:[^/]+/)*(?P<id>[^/?#&]+)'
 269     _TEST = {
 270         'url': 'http://www.itv.com/btcc/races/btcc-2018-all-the-action-from-brands-hatch',
 271         'info_dict': {
 272             'id': 'btcc-2018-all-the-action-from-brands-hatch',
 273             'title': 'BTCC 2018: All the action from Brands Hatch',
 274         },
 275         'playlist_mincount': 9,
 276     }
 277     BRIGHTCOVE_URL_TEMPLATE = 'http://players.brightcove.net/1582188683001/HkiHLnNRx_default/index.html?videoId=%s'
 278
 279     def _real_extract(self, url):
 280         playlist_id = self._match_id(url)
 281
 282         webpage = self._download_webpage(url, playlist_id)
 283
 284         entries = [
 285             self.url_result(
 286                 smuggle_url(self.BRIGHTCOVE_URL_TEMPLATE % video_id, {
 287                     # ITV does not like some GB IP ranges, so here are some
 288                     # IP blocks it accepts
 289                     'geo_ip_blocks': [
 290                         '193.113.0.0/16', '54.36.162.0/23', '159.65.16.0/21'
 291                     ],
 292                     'referrer': url,
 293                 }),
 294                 ie=BrightcoveNewIE.ie_key(), video_id=video_id)
 295             for video_id in re.findall(r'data-video-id=["\'](\d+)', webpage)]
 296
 297         title = self._og_search_title(webpage, fatal=False)
 298
 299         return self.playlist_result(entries, playlist_id, title)