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