Raphaël G. Git Repositories - youtubedl/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 ..utils import (
  10     unescapeHTML,
  11     urlencode_postdata,
  12     unified_timestamp,
  13 )
  14
  15
  16 class AdobePassIE(InfoExtractor):
  17     _SERVICE_PROVIDER_TEMPLATE = 'https://sp.auth.adobe.com/adobe-services/%s'
  18     _USER_AGENT = 'Mozilla/5.0 (X11; Linux i686; rv:47.0) Gecko/20100101 Firefox/47.0'
  19
  20     @staticmethod
  21     def _get_mvpd_resource(provider_id, title, guid, rating):
  22         channel = etree.Element('channel')
  23         channel_title = etree.SubElement(channel, 'title')
  24         channel_title.text = provider_id
  25         item = etree.SubElement(channel, 'item')
  26         resource_title = etree.SubElement(item, 'title')
  27         resource_title.text = title
  28         resource_guid = etree.SubElement(item, 'guid')
  29         resource_guid.text = guid
  30         resource_rating = etree.SubElement(item, 'media:rating')
  31         resource_rating.attrib = {'scheme': 'urn:v-chip'}
  32         resource_rating.text = rating
  33         return '<rss version="2.0" xmlns:media="http://search.yahoo.com/mrss/">' + etree.tostring(channel).decode() + '</rss>'
  34
  35     def _extract_mvpd_auth(self, url, video_id, requestor_id, resource):
  36         def xml_text(xml_str, tag):
  37             return self._search_regex(
  38                 '<%s>(.+?)</%s>' % (tag, tag), xml_str, tag)
  39
  40         mvpd_headers = {
  41             'ap_42': 'anonymous',
  42             'ap_11': 'Linux i686',
  43             'ap_z': self._USER_AGENT,
  44             'User-Agent': self._USER_AGENT,
  45         }
  46
  47         guid = xml_text(resource, 'guid')
  48         requestor_info = self._downloader.cache.load('mvpd', requestor_id) or {}
  49         authn_token = requestor_info.get('authn_token')
  50         if authn_token:
  51             token_expires = unified_timestamp(re.sub(r'[_ ]GMT', '', xml_text(authn_token, 'simpleTokenExpires')))
  52             if token_expires and token_expires <= int(time.time()):
  53                 authn_token = None
  54                 requestor_info = {}
  55         if not authn_token:
  56             # TODO add support for other TV Providers
  57             mso_id = 'DTV'
  58             username, password = self._get_netrc_login_info(mso_id)
  59             if not username or not password:
  60                 return ''
  61
  62             def post_form(form_page, note, data={}):
  63                 post_url = self._html_search_regex(r'<form[^>]+action=(["\'])(?P<url>.+?)\1', form_page, 'post url', group='url')
  64                 return self._download_webpage(
  65                     post_url, video_id, note, data=urlencode_postdata(data or self._hidden_inputs(form_page)), headers={
  66                         'Content-Type': 'application/x-www-form-urlencoded',
  67                     })
  68
  69             provider_redirect_page = self._download_webpage(
  70                 self._SERVICE_PROVIDER_TEMPLATE % 'authenticate/saml', video_id,
  71                 'Downloading Provider Redirect Page', query={
  72                     'noflash': 'true',
  73                     'mso_id': mso_id,
  74                     'requestor_id': requestor_id,
  75                     'no_iframe': 'false',
  76                     'domain_name': 'adobe.com',
  77                     'redirect_url': url,
  78                 })
  79             provider_login_page = post_form(
  80                 provider_redirect_page, 'Downloading Provider Login Page')
  81             mvpd_confirm_page = post_form(provider_login_page, 'Logging in', {
  82                 'username': username,
  83                 'password': password,
  84             })
  85             post_form(mvpd_confirm_page, 'Confirming Login')
  86
  87             session = self._download_webpage(
  88                 self._SERVICE_PROVIDER_TEMPLATE % 'session', video_id,
  89                 'Retrieving Session', data=urlencode_postdata({
  90                     '_method': 'GET',
  91                     'requestor_id': requestor_id,
  92                 }), headers=mvpd_headers)
  93             if '<pendingLogout' in session:
  94                 self._downloader.cache.store('mvpd', requestor_id, {})
  95                 return self._extract_mvpd_auth(url, video_id, requestor_id, resource)
  96             authn_token = unescapeHTML(xml_text(session, 'authnToken'))
  97             requestor_info['authn_token'] = authn_token
  98             self._downloader.cache.store('mvpd', requestor_id, requestor_info)
  99
 100         authz_token = requestor_info.get(guid)
 101         if not authz_token:
 102             authorize = self._download_webpage(
 103                 self._SERVICE_PROVIDER_TEMPLATE % 'authorize', video_id,
 104                 'Retrieving Authorization Token', data=urlencode_postdata({
 105                     'resource_id': resource,
 106                     'requestor_id': requestor_id,
 107                     'authentication_token': authn_token,
 108                     'mso_id': xml_text(authn_token, 'simpleTokenMsoID'),
 109                     'userMeta': '1',
 110                 }), headers=mvpd_headers)
 111             if '<pendingLogout' in authorize:
 112                 self._downloader.cache.store('mvpd', requestor_id, {})
 113                 return self._extract_mvpd_auth(url, video_id, requestor_id, resource)
 114             authz_token = unescapeHTML(xml_text(authorize, 'authzToken'))
 115             requestor_info[guid] = authz_token
 116             self._downloader.cache.store('mvpd', requestor_id, requestor_info)
 117
 118         mvpd_headers.update({
 119             'ap_19': xml_text(authn_token, 'simpleSamlNameID'),
 120             'ap_23': xml_text(authn_token, 'simpleSamlSessionIndex'),
 121         })
 122
 123         short_authorize = self._download_webpage(
 124             self._SERVICE_PROVIDER_TEMPLATE % 'shortAuthorize',
 125             video_id, 'Retrieving Media Token', data=urlencode_postdata({
 126                 'authz_token': authz_token,
 127                 'requestor_id': requestor_id,
 128                 'session_guid': xml_text(authn_token, 'simpleTokenAuthenticationGuid'),
 129                 'hashed_guid': 'false',
 130             }), headers=mvpd_headers)
 131         if '<pendingLogout' in short_authorize:
 132             self._downloader.cache.store('mvpd', requestor_id, {})
 133             return self._extract_mvpd_auth(url, video_id, requestor_id, resource)
 134         return short_authorize