_ Git - youtube-dl/blob - youtube_dl/extractor/adobepass.py

   1 # -*- coding: utf-8 -*-
   2 from __future__ import unicode_literals
   3
   4 import re
   5 import time
   6 import xml.etree.ElementTree as etree
   7
   8 from .common import InfoExtractor
   9 from ..compat import compat_urlparse
  10 from ..utils import (
  11     unescapeHTML,
  12     urlencode_postdata,
  13     unified_timestamp,
  14     ExtractorError,
  15 )
  16
  17
  18 MSO_INFO = {
  19     'DTV': {
  20         'name': 'DirecTV',
  21         'username_field': 'username',
  22         'password_field': 'password',
  23     },
  24     'Rogers': {
  25         'name': 'Rogers Cable',
  26         'username_field': 'UserName',
  27         'password_field': 'UserPassword',
  28     },
  29 }
  30
  31
  32 class AdobePassIE(InfoExtractor):
  33     _SERVICE_PROVIDER_TEMPLATE = 'https://sp.auth.adobe.com/adobe-services/%s'
  34     _USER_AGENT = 'Mozilla/5.0 (X11; Linux i686; rv:47.0) Gecko/20100101 Firefox/47.0'
  35     _MVPD_CACHE = 'ap-mvpd'
  36
  37     @staticmethod
  38     def _get_mvpd_resource(provider_id, title, guid, rating):
  39         channel = etree.Element('channel')
  40         channel_title = etree.SubElement(channel, 'title')
  41         channel_title.text = provider_id
  42         item = etree.SubElement(channel, 'item')
  43         resource_title = etree.SubElement(item, 'title')
  44         resource_title.text = title
  45         resource_guid = etree.SubElement(item, 'guid')
  46         resource_guid.text = guid
  47         resource_rating = etree.SubElement(item, 'media:rating')
  48         resource_rating.attrib = {'scheme': 'urn:v-chip'}
  49         resource_rating.text = rating
  50         return '<rss version="2.0" xmlns:media="http://search.yahoo.com/mrss/">' + etree.tostring(channel).decode() + '</rss>'
  51
  52     def _extract_mvpd_auth(self, url, video_id, requestor_id, resource):
  53         def xml_text(xml_str, tag):
  54             return self._search_regex(
  55                 '<%s>(.+?)</%s>' % (tag, tag), xml_str, tag)
  56
  57         def is_expired(token, date_ele):
  58             token_expires = unified_timestamp(re.sub(r'[_ ]GMT', '', xml_text(token, date_ele)))
  59             return token_expires and token_expires <= int(time.time())
  60
  61         def post_form(form_page_res, note, data={}):
  62             form_page, urlh = form_page_res
  63             post_url = self._html_search_regex(r'<form[^>]+action=(["\'])(?P<url>.+?)\1', form_page, 'post url', group='url')
  64             if not re.match(r'https?://', post_url):
  65                 post_url = compat_urlparse.urljoin(urlh.geturl(), post_url)
  66             form_data = self._hidden_inputs(form_page)
  67             form_data.update(data)
  68             return self._download_webpage_handle(
  69                 post_url, video_id, note, data=urlencode_postdata(form_data), headers={
  70                     'Content-Type': 'application/x-www-form-urlencoded',
  71                 })
  72
  73         def raise_mvpd_required():
  74             raise ExtractorError(
  75                 'This video is only available for users of participating TV providers. '
  76                 'Use --ap-mso to specify Adobe Pass Multiple-system operator Identifier '
  77                 'and --ap-username and --ap-password or --netrc to provide account credentials.', expected=True)
  78
  79         mvpd_headers = {
  80             'ap_42': 'anonymous',
  81             'ap_11': 'Linux i686',
  82             'ap_z': self._USER_AGENT,
  83             'User-Agent': self._USER_AGENT,
  84         }
  85
  86         guid = xml_text(resource, 'guid') if '<' in resource else resource
  87         count = 0
  88         while count < 2:
  89             requestor_info = self._downloader.cache.load(self._MVPD_CACHE, requestor_id) or {}
  90             authn_token = requestor_info.get('authn_token')
  91             if authn_token and is_expired(authn_token, 'simpleTokenExpires'):
  92                 authn_token = None
  93             if not authn_token:
  94                 # TODO add support for other TV Providers
  95                 mso_id = self._downloader.params.get('ap_mso')
  96                 if not mso_id:
  97                     raise_mvpd_required()
  98                 username, password = self._get_login_info('ap_username', 'ap_password', mso_id)
  99                 if not username or not password:
 100                     raise_mvpd_required()
 101                 mso_info = MSO_INFO[mso_id]
 102
 103                 provider_redirect_page_res = self._download_webpage_handle(
 104                     self._SERVICE_PROVIDER_TEMPLATE % 'authenticate/saml', video_id,
 105                     'Downloading Provider Redirect Page', query={
 106                         'noflash': 'true',
 107                         'mso_id': mso_id,
 108                         'requestor_id': requestor_id,
 109                         'no_iframe': 'false',
 110                         'domain_name': 'adobe.com',
 111                         'redirect_url': url,
 112                     })
 113                 provider_login_page_res = post_form(
 114                     provider_redirect_page_res, 'Downloading Provider Login Page')
 115                 mvpd_confirm_page_res = post_form(provider_login_page_res, 'Logging in', {
 116                     mso_info['username_field']: username,
 117                     mso_info['password_field']: password,
 118                 })
 119                 if mso_id == 'DTV':
 120                     post_form(mvpd_confirm_page_res, 'Confirming Login')
 121
 122                 session = self._download_webpage(
 123                     self._SERVICE_PROVIDER_TEMPLATE % 'session', video_id,
 124                     'Retrieving Session', data=urlencode_postdata({
 125                         '_method': 'GET',
 126                         'requestor_id': requestor_id,
 127                     }), headers=mvpd_headers)
 128                 if '<pendingLogout' in session:
 129                     self._downloader.cache.store(self._MVPD_CACHE, requestor_id, {})
 130                     count += 1
 131                     continue
 132                 authn_token = unescapeHTML(xml_text(session, 'authnToken'))
 133                 requestor_info['authn_token'] = authn_token
 134                 self._downloader.cache.store(self._MVPD_CACHE, requestor_id, requestor_info)
 135
 136             authz_token = requestor_info.get(guid)
 137             if authz_token and is_expired(authz_token, 'simpleTokenTTL'):
 138                 authz_token = None
 139             if not authz_token:
 140                 authorize = self._download_webpage(
 141                     self._SERVICE_PROVIDER_TEMPLATE % 'authorize', video_id,
 142                     'Retrieving Authorization Token', data=urlencode_postdata({
 143                         'resource_id': resource,
 144                         'requestor_id': requestor_id,
 145                         'authentication_token': authn_token,
 146                         'mso_id': xml_text(authn_token, 'simpleTokenMsoID'),
 147                         'userMeta': '1',
 148                     }), headers=mvpd_headers)
 149                 if '<pendingLogout' in authorize:
 150                     self._downloader.cache.store(self._MVPD_CACHE, requestor_id, {})
 151                     count += 1
 152                     continue
 153                 authz_token = unescapeHTML(xml_text(authorize, 'authzToken'))
 154                 requestor_info[guid] = authz_token
 155                 self._downloader.cache.store(self._MVPD_CACHE, requestor_id, requestor_info)
 156
 157             mvpd_headers.update({
 158                 'ap_19': xml_text(authn_token, 'simpleSamlNameID'),
 159                 'ap_23': xml_text(authn_token, 'simpleSamlSessionIndex'),
 160             })
 161
 162             short_authorize = self._download_webpage(
 163                 self._SERVICE_PROVIDER_TEMPLATE % 'shortAuthorize',
 164                 video_id, 'Retrieving Media Token', data=urlencode_postdata({
 165                     'authz_token': authz_token,
 166                     'requestor_id': requestor_id,
 167                     'session_guid': xml_text(authn_token, 'simpleTokenAuthenticationGuid'),
 168                     'hashed_guid': 'false',
 169                 }), headers=mvpd_headers)
 170             if '<pendingLogout' in short_authorize:
 171                 self._downloader.cache.store(self._MVPD_CACHE, requestor_id, {})
 172                 count += 1
 173                 continue
 174             return short_authorize