_ 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 import re
   8
   9 from .common import InfoExtractor
  10 from ..compat import (
  11     compat_str,
  12     compat_etree_register_namespace,
  13 )
  14 from ..utils import (
  15     extract_attributes,
  16     xpath_with_ns,
  17     xpath_element,
  18     xpath_text,
  19     int_or_none,
  20     parse_duration,
  21     ExtractorError,
  22     determine_ext,
  23 )
  24
  25
  26 class ITVIE(InfoExtractor):
  27     _VALID_URL = r'https?://(?:www\.)?itv\.com/hub/[^/]+/(?P<id>[0-9a-zA-Z]+)'
  28     _GEO_COUNTRIES = ['GB']
  29     _TEST = {
  30         'url': 'http://www.itv.com/hub/mr-bean-animated-series/2a2936a0053',
  31         'info_dict': {
  32             'id': '2a2936a0053',
  33             'ext': 'flv',
  34             'title': 'Home Movie',
  35         },
  36         'params': {
  37             # rtmp download
  38             'skip_download': True,
  39         },
  40     }
  41
  42     def _real_extract(self, url):
  43         video_id = self._match_id(url)
  44         webpage = self._download_webpage(url, video_id)
  45         params = extract_attributes(self._search_regex(
  46             r'(?s)(<[^>]+id="video"[^>]*>)', webpage, 'params'))
  47
  48         ns_map = {
  49             'soapenv': 'http://schemas.xmlsoap.org/soap/envelope/',
  50             'tem': 'http://tempuri.org/',
  51             'itv': 'http://schemas.datacontract.org/2004/07/Itv.BB.Mercury.Common.Types',
  52             'com': 'http://schemas.itv.com/2009/05/Common',
  53         }
  54         for ns, full_ns in ns_map.items():
  55             compat_etree_register_namespace(ns, full_ns)
  56
  57         def _add_ns(name):
  58             return xpath_with_ns(name, ns_map)
  59
  60         def _add_sub_element(element, name):
  61             return etree.SubElement(element, _add_ns(name))
  62
  63         production_id = (
  64             params.get('data-video-autoplay-id') or
  65             '%s#001' % (
  66                 params.get('data-video-episode-id') or
  67                 video_id.replace('a', '/')))
  68
  69         req_env = etree.Element(_add_ns('soapenv:Envelope'))
  70         _add_sub_element(req_env, 'soapenv:Header')
  71         body = _add_sub_element(req_env, 'soapenv:Body')
  72         get_playlist = _add_sub_element(body, ('tem:GetPlaylist'))
  73         request = _add_sub_element(get_playlist, 'tem:request')
  74         _add_sub_element(request, 'itv:ProductionId').text = production_id
  75         _add_sub_element(request, 'itv:RequestGuid').text = compat_str(uuid.uuid4()).upper()
  76         vodcrid = _add_sub_element(request, 'itv:Vodcrid')
  77         _add_sub_element(vodcrid, 'com:Id')
  78         _add_sub_element(request, 'itv:Partition')
  79         user_info = _add_sub_element(get_playlist, 'tem:userInfo')
  80         _add_sub_element(user_info, 'itv:Broadcaster').text = 'Itv'
  81         _add_sub_element(user_info, 'itv:DM')
  82         _add_sub_element(user_info, 'itv:RevenueScienceValue')
  83         _add_sub_element(user_info, 'itv:SessionId')
  84         _add_sub_element(user_info, 'itv:SsoToken')
  85         _add_sub_element(user_info, 'itv:UserToken')
  86         site_info = _add_sub_element(get_playlist, 'tem:siteInfo')
  87         _add_sub_element(site_info, 'itv:AdvertisingRestriction').text = 'None'
  88         _add_sub_element(site_info, 'itv:AdvertisingSite').text = 'ITV'
  89         _add_sub_element(site_info, 'itv:AdvertisingType').text = 'Any'
  90         _add_sub_element(site_info, 'itv:Area').text = 'ITVPLAYER.VIDEO'
  91         _add_sub_element(site_info, 'itv:Category')
  92         _add_sub_element(site_info, 'itv:Platform').text = 'DotCom'
  93         _add_sub_element(site_info, 'itv:Site').text = 'ItvCom'
  94         device_info = _add_sub_element(get_playlist, 'tem:deviceInfo')
  95         _add_sub_element(device_info, 'itv:ScreenSize').text = 'Big'
  96         player_info = _add_sub_element(get_playlist, 'tem:playerInfo')
  97         _add_sub_element(player_info, 'itv:Version').text = '2'
  98
  99         headers = self.geo_verification_headers()
 100         headers.update({
 101             'Content-Type': 'text/xml; charset=utf-8',
 102             'SOAPAction': 'http://tempuri.org/PlaylistService/GetPlaylist',
 103         })
 104         resp_env = self._download_xml(
 105             params['data-playlist-url'], video_id,
 106             headers=headers, data=etree.tostring(req_env))
 107         playlist = xpath_element(resp_env, './/Playlist')
 108         if playlist is None:
 109             fault_code = xpath_text(resp_env, './/faultcode')
 110             fault_string = xpath_text(resp_env, './/faultstring')
 111             if fault_code == 'InvalidGeoRegion':
 112                 self.raise_geo_restricted(
 113                     msg=fault_string, countries=self._GEO_COUNTRIES)
 114             raise ExtractorError('%s said: %s' % (self.IE_NAME, fault_string))
 115         title = xpath_text(playlist, 'EpisodeTitle', fatal=True)
 116         video_element = xpath_element(playlist, 'VideoEntries/Video', fatal=True)
 117         media_files = xpath_element(video_element, 'MediaFiles', fatal=True)
 118         rtmp_url = media_files.attrib['base']
 119
 120         formats = []
 121         for media_file in media_files.findall('MediaFile'):
 122             play_path = xpath_text(media_file, 'URL')
 123             if not play_path:
 124                 continue
 125             tbr = int_or_none(media_file.get('bitrate'), 1000)
 126             f = {
 127                 'format_id': 'rtmp' + ('-%d' % tbr if tbr else ''),
 128                 'play_path': play_path,
 129                 # Providing this swfVfy allows to avoid truncated downloads
 130                 'player_url': 'http://www.itv.com/mercury/Mercury_VideoPlayer.swf',
 131                 'page_url': url,
 132                 'tbr': tbr,
 133                 'ext': 'flv',
 134             }
 135             app = self._search_regex(
 136                 'rtmpe?://[^/]+/(.+)$', rtmp_url, 'app', default=None)
 137             if app:
 138                 f.update({
 139                     'url': rtmp_url.split('?', 1)[0],
 140                     'app': app,
 141                 })
 142             else:
 143                 f['url'] = rtmp_url
 144             formats.append(f)
 145
 146         ios_playlist_url = params.get('data-video-playlist') or params.get('data-video-id')
 147         hmac = params.get('data-video-hmac')
 148         if ios_playlist_url and hmac and re.match(r'https?://', ios_playlist_url):
 149             headers = self.geo_verification_headers()
 150             headers.update({
 151                 'Accept': 'application/vnd.itv.vod.playlist.v2+json',
 152                 'Content-Type': 'application/json',
 153                 'hmac': hmac.upper(),
 154             })
 155             ios_playlist = self._download_json(
 156                 ios_playlist_url, video_id, data=json.dumps({
 157                     'user': {
 158                         'itvUserId': '',
 159                         'entitlements': [],
 160                         'token': ''
 161                     },
 162                     'device': {
 163                         'manufacturer': 'Safari',
 164                         'model': '5',
 165                         'os': {
 166                             'name': 'Windows NT',
 167                             'version': '6.1',
 168                             'type': 'desktop'
 169                         }
 170                     },
 171                     'client': {
 172                         'version': '4.1',
 173                         'id': 'browser'
 174                     },
 175                     'variantAvailability': {
 176                         'featureset': {
 177                             'min': ['hls', 'aes', 'outband-webvtt'],
 178                             'max': ['hls', 'aes', 'outband-webvtt']
 179                         },
 180                         'platformTag': 'dotcom'
 181                     }
 182                 }).encode(), headers=headers, fatal=False)
 183             if ios_playlist:
 184                 video_data = ios_playlist.get('Playlist', {}).get('Video', {})
 185                 ios_base_url = video_data.get('Base')
 186                 for media_file in video_data.get('MediaFiles', []):
 187                     href = media_file.get('Href')
 188                     if not href:
 189                         continue
 190                     if ios_base_url:
 191                         href = ios_base_url + href
 192                     ext = determine_ext(href)
 193                     if ext == 'm3u8':
 194                         formats.extend(self._extract_m3u8_formats(
 195                             href, video_id, 'mp4', entry_protocol='m3u8_native',
 196                             m3u8_id='hls', fatal=False))
 197                     else:
 198                         formats.append({
 199                             'url': href,
 200                         })
 201         self._sort_formats(formats)
 202
 203         subtitles = {}
 204         for caption_url in video_element.findall('ClosedCaptioningURIs/URL'):
 205             if not caption_url.text:
 206                 continue
 207             ext = determine_ext(caption_url.text, 'ttml')
 208             subtitles.setdefault('en', []).append({
 209                 'url': caption_url.text,
 210                 'ext': 'ttml' if ext == 'xml' else ext,
 211             })
 212
 213         info = self._search_json_ld(webpage, video_id, default={})
 214         info.update({
 215             'id': video_id,
 216             'title': title,
 217             'formats': formats,
 218             'subtitles': subtitles,
 219             'episode_title': title,
 220             'episode_number': int_or_none(xpath_text(playlist, 'EpisodeNumber')),
 221             'series': xpath_text(playlist, 'ProgrammeTitle'),
 222             'duartion': parse_duration(xpath_text(playlist, 'Duration')),
 223         })
 224         return info