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
   8 from .common import InfoExtractor
   9 from ..compat import (
  10     compat_str,
  11     compat_etree_register_namespace,
  12 )
  13 from ..utils import (
  14     extract_attributes,
  15     xpath_with_ns,
  16     xpath_element,
  17     xpath_text,
  18     int_or_none,
  19     parse_duration,
  20     ExtractorError,
  21     determine_ext,
  22 )
  23
  24
  25 class ITVIE(InfoExtractor):
  26     _VALID_URL = r'https?://(?:www\.)?itv\.com/hub/[^/]+/(?P<id>[0-9a-zA-Z]+)'
  27     _GEO_COUNTRIES = ['GB']
  28     _TEST = {
  29         'url': 'http://www.itv.com/hub/mr-bean-animated-series/2a2936a0053',
  30         'info_dict': {
  31             'id': '2a2936a0053',
  32             'ext': 'flv',
  33             'title': 'Home Movie',
  34         },
  35         'params': {
  36             # rtmp download
  37             'skip_download': True,
  38         },
  39     }
  40
  41     def _real_extract(self, url):
  42         video_id = self._match_id(url)
  43         webpage = self._download_webpage(url, video_id)
  44         params = extract_attributes(self._search_regex(
  45             r'(?s)(<[^>]+id="video"[^>]*>)', webpage, 'params'))
  46
  47         ns_map = {
  48             'soapenv': 'http://schemas.xmlsoap.org/soap/envelope/',
  49             'tem': 'http://tempuri.org/',
  50             'itv': 'http://schemas.datacontract.org/2004/07/Itv.BB.Mercury.Common.Types',
  51             'com': 'http://schemas.itv.com/2009/05/Common',
  52         }
  53         for ns, full_ns in ns_map.items():
  54             compat_etree_register_namespace(ns, full_ns)
  55
  56         def _add_ns(name):
  57             return xpath_with_ns(name, ns_map)
  58
  59         def _add_sub_element(element, name):
  60             return etree.SubElement(element, _add_ns(name))
  61
  62         production_id = (
  63             params.get('data-video-autoplay-id') or
  64             '%s#001' % (
  65                 params.get('data-video-episode-id') or
  66                 video_id.replace('a', '/')))
  67
  68         req_env = etree.Element(_add_ns('soapenv:Envelope'))
  69         _add_sub_element(req_env, 'soapenv:Header')
  70         body = _add_sub_element(req_env, 'soapenv:Body')
  71         get_playlist = _add_sub_element(body, ('tem:GetPlaylist'))
  72         request = _add_sub_element(get_playlist, 'tem:request')
  73         _add_sub_element(request, 'itv:ProductionId').text = production_id
  74         _add_sub_element(request, 'itv:RequestGuid').text = compat_str(uuid.uuid4()).upper()
  75         vodcrid = _add_sub_element(request, 'itv:Vodcrid')
  76         _add_sub_element(vodcrid, 'com:Id')
  77         _add_sub_element(request, 'itv:Partition')
  78         user_info = _add_sub_element(get_playlist, 'tem:userInfo')
  79         _add_sub_element(user_info, 'itv:Broadcaster').text = 'Itv'
  80         _add_sub_element(user_info, 'itv:DM')
  81         _add_sub_element(user_info, 'itv:RevenueScienceValue')
  82         _add_sub_element(user_info, 'itv:SessionId')
  83         _add_sub_element(user_info, 'itv:SsoToken')
  84         _add_sub_element(user_info, 'itv:UserToken')
  85         site_info = _add_sub_element(get_playlist, 'tem:siteInfo')
  86         _add_sub_element(site_info, 'itv:AdvertisingRestriction').text = 'None'
  87         _add_sub_element(site_info, 'itv:AdvertisingSite').text = 'ITV'
  88         _add_sub_element(site_info, 'itv:AdvertisingType').text = 'Any'
  89         _add_sub_element(site_info, 'itv:Area').text = 'ITVPLAYER.VIDEO'
  90         _add_sub_element(site_info, 'itv:Category')
  91         _add_sub_element(site_info, 'itv:Platform').text = 'DotCom'
  92         _add_sub_element(site_info, 'itv:Site').text = 'ItvCom'
  93         device_info = _add_sub_element(get_playlist, 'tem:deviceInfo')
  94         _add_sub_element(device_info, 'itv:ScreenSize').text = 'Big'
  95         player_info = _add_sub_element(get_playlist, 'tem:playerInfo')
  96         _add_sub_element(player_info, 'itv:Version').text = '2'
  97
  98         headers = self.geo_verification_headers()
  99         headers.update({
 100             'Content-Type': 'text/xml; charset=utf-8',
 101             'SOAPAction': 'http://tempuri.org/PlaylistService/GetPlaylist',
 102         })
 103         resp_env = self._download_xml(
 104             params['data-playlist-url'], video_id,
 105             headers=headers, data=etree.tostring(req_env))
 106         playlist = xpath_element(resp_env, './/Playlist')
 107         if playlist is None:
 108             fault_code = xpath_text(resp_env, './/faultcode')
 109             fault_string = xpath_text(resp_env, './/faultstring')
 110             if fault_code == 'InvalidGeoRegion':
 111                 self.raise_geo_restricted(
 112                     msg=fault_string, countries=self._GEO_COUNTRIES)
 113             raise ExtractorError('%s said: %s' % (self.IE_NAME, fault_string))
 114         title = xpath_text(playlist, 'EpisodeTitle', fatal=True)
 115         video_element = xpath_element(playlist, 'VideoEntries/Video', fatal=True)
 116         media_files = xpath_element(video_element, 'MediaFiles', fatal=True)
 117         rtmp_url = media_files.attrib['base']
 118
 119         formats = []
 120         for media_file in media_files.findall('MediaFile'):
 121             play_path = xpath_text(media_file, 'URL')
 122             if not play_path:
 123                 continue
 124             tbr = int_or_none(media_file.get('bitrate'), 1000)
 125             f = {
 126                 'format_id': 'rtmp' + ('-%d' % tbr if tbr else ''),
 127                 'play_path': play_path,
 128                 # Providing this swfVfy allows to avoid truncated downloads
 129                 'player_url': 'http://www.itv.com/mercury/Mercury_VideoPlayer.swf',
 130                 'page_url': url,
 131                 'tbr': tbr,
 132                 'ext': 'flv',
 133             }
 134             app = self._search_regex(
 135                 'rtmpe?://[^/]+/(.+)$', rtmp_url, 'app', default=None)
 136             if app:
 137                 f.update({
 138                     'url': rtmp_url.split('?', 1)[0],
 139                     'app': app,
 140                 })
 141             else:
 142                 f['url'] = rtmp_url
 143             formats.append(f)
 144
 145         ios_playlist_url = params.get('data-video-playlist')
 146         hmac = params.get('data-video-hmac')
 147         if ios_playlist_url and hmac:
 148             headers = self.geo_verification_headers()
 149             headers.update({
 150                 'Accept': 'application/vnd.itv.vod.playlist.v2+json',
 151                 'Content-Type': 'application/json',
 152                 'hmac': hmac.upper(),
 153             })
 154             ios_playlist = self._download_json(
 155                 ios_playlist_url, video_id, data=json.dumps({
 156                     'user': {
 157                         'itvUserId': '',
 158                         'entitlements': [],
 159                         'token': ''
 160                     },
 161                     'device': {
 162                         'manufacturer': 'Apple',
 163                         'model': 'iPad',
 164                         'os': {
 165                             'name': 'iPhone OS',
 166                             'version': '9.3',
 167                             'type': 'ios'
 168                         }
 169                     },
 170                     'client': {
 171                         'version': '4.1',
 172                         'id': 'browser'
 173                     },
 174                     'variantAvailability': {
 175                         'featureset': {
 176                             'min': ['hls', 'aes'],
 177                             'max': ['hls', 'aes']
 178                         },
 179                         'platformTag': 'mobile'
 180                     }
 181                 }).encode(), headers=headers, fatal=False)
 182             if ios_playlist:
 183                 video_data = ios_playlist.get('Playlist', {}).get('Video', {})
 184                 ios_base_url = video_data.get('Base')
 185                 for media_file in video_data.get('MediaFiles', []):
 186                     href = media_file.get('Href')
 187                     if not href:
 188                         continue
 189                     if ios_base_url:
 190                         href = ios_base_url + href
 191                     ext = determine_ext(href)
 192                     if ext == 'm3u8':
 193                         formats.extend(self._extract_m3u8_formats(
 194                             href, video_id, 'mp4', entry_protocol='m3u8_native',
 195                             m3u8_id='hls', fatal=False))
 196                     else:
 197                         formats.append({
 198                             'url': href,
 199                         })
 200         self._sort_formats(formats)
 201
 202         subtitles = {}
 203         for caption_url in video_element.findall('ClosedCaptioningURIs/URL'):
 204             if not caption_url.text:
 205                 continue
 206             ext = determine_ext(caption_url.text, 'ttml')
 207             subtitles.setdefault('en', []).append({
 208                 'url': caption_url.text,
 209                 'ext': 'ttml' if ext == 'xml' else ext,
 210             })
 211
 212         info = self._search_json_ld(webpage, video_id, default={})
 213         info.update({
 214             'id': video_id,
 215             'title': title,
 216             'formats': formats,
 217             'subtitles': subtitles,
 218             'episode_title': title,
 219             'episode_number': int_or_none(xpath_text(playlist, 'EpisodeNumber')),
 220             'series': xpath_text(playlist, 'ProgrammeTitle'),
 221             'duartion': parse_duration(xpath_text(playlist, 'Duration')),
 222         })
 223         return info