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