yt_dlp/extractor/ciscolive.py

   1 import itertools
   2
   3 from .common import InfoExtractor
   4 from ..utils import (
   5     clean_html,
   6     float_or_none,
   7     int_or_none,
   8     parse_qs,
   9     try_get,
  10     urlencode_postdata,
  11 )
  12
  13
  14 class CiscoLiveBaseIE(InfoExtractor):
  15     # These appear to be constant across all Cisco Live presentations
  16     # and are not tied to any user session or event
  17     RAINFOCUS_API_URL = 'https://events.rainfocus.com/api/%s'
  18     RAINFOCUS_API_PROFILE_ID = 'Na3vqYdAlJFSxhYTYQGuMbpafMqftalz'
  19     RAINFOCUS_WIDGET_ID = 'n6l4Lo05R8fiy3RpUBm447dZN8uNWoye'
  20     BRIGHTCOVE_URL_TEMPLATE = 'http://players.brightcove.net/5647924234001/SyK2FdqjM_default/index.html?videoId=%s'
  21
  22     HEADERS = {
  23         'Origin': 'https://ciscolive.cisco.com',
  24         'rfApiProfileId': RAINFOCUS_API_PROFILE_ID,
  25         'rfWidgetId': RAINFOCUS_WIDGET_ID,
  26     }
  27
  28     def _call_api(self, ep, rf_id, query, referrer, note=None):
  29         headers = self.HEADERS.copy()
  30         headers['Referer'] = referrer
  31         return self._download_json(
  32             self.RAINFOCUS_API_URL % ep, rf_id, note=note,
  33             data=urlencode_postdata(query), headers=headers)
  34
  35     def _parse_rf_item(self, rf_item):
  36         event_name = rf_item.get('eventName')
  37         title = rf_item['title']
  38         description = clean_html(rf_item.get('abstract'))
  39         presenter_name = try_get(rf_item, lambda x: x['participants'][0]['fullName'])
  40         bc_id = rf_item['videos'][0]['url']
  41         bc_url = self.BRIGHTCOVE_URL_TEMPLATE % bc_id
  42         duration = float_or_none(try_get(rf_item, lambda x: x['times'][0]['length']))
  43         location = try_get(rf_item, lambda x: x['times'][0]['room'])
  44
  45         if duration:
  46             duration = duration * 60
  47
  48         return {
  49             '_type': 'url_transparent',
  50             'url': bc_url,
  51             'ie_key': 'BrightcoveNew',
  52             'title': title,
  53             'description': description,
  54             'duration': duration,
  55             'creator': presenter_name,
  56             'location': location,
  57             'series': event_name,
  58         }
  59
  60
  61 class CiscoLiveSessionIE(CiscoLiveBaseIE):
  62     _VALID_URL = r'https?://(?:www\.)?ciscolive(?:\.cisco)?\.com/[^#]*#/session/(?P<id>[^/?&]+)'
  63     _TESTS = [{
  64         'url': 'https://ciscolive.cisco.com/on-demand-library/?#/session/1423353499155001FoSs',
  65         'md5': 'c98acf395ed9c9f766941c70f5352e22',
  66         'info_dict': {
  67             'id': '5803694304001',
  68             'ext': 'mp4',
  69             'title': '13 Smart Automations to Monitor Your Cisco IOS Network',
  70             'description': 'md5:ec4a436019e09a918dec17714803f7cc',
  71             'timestamp': 1530305395,
  72             'upload_date': '20180629',
  73             'uploader_id': '5647924234001',
  74             'location': '16B Mezz.',
  75         },
  76     }, {
  77         'url': 'https://www.ciscolive.com/global/on-demand-library.html?search.event=ciscoliveemea2019#/session/15361595531500013WOU',
  78         'only_matching': True,
  79     }, {
  80         'url': 'https://www.ciscolive.com/global/on-demand-library.html?#/session/1490051371645001kNaS',
  81         'only_matching': True,
  82     }]
  83
  84     def _real_extract(self, url):
  85         rf_id = self._match_id(url)
  86         rf_result = self._call_api('session', rf_id, {'id': rf_id}, url)
  87         return self._parse_rf_item(rf_result['items'][0])
  88
  89
  90 class CiscoLiveSearchIE(CiscoLiveBaseIE):
  91     _VALID_URL = r'https?://(?:www\.)?ciscolive(?:\.cisco)?\.com/(?:global/)?on-demand-library(?:\.html|/)'
  92     _TESTS = [{
  93         'url': 'https://ciscolive.cisco.com/on-demand-library/?search.event=ciscoliveus2018&search.technicallevel=scpsSkillLevel_aintroductory&search.focus=scpsSessionFocus_designAndDeployment#/',
  94         'info_dict': {
  95             'title': 'Search query',
  96         },
  97         'playlist_count': 5,
  98     }, {
  99         'url': 'https://ciscolive.cisco.com/on-demand-library/?search.technology=scpsTechnology_applicationDevelopment&search.technology=scpsTechnology_ipv6&search.focus=scpsSessionFocus_troubleshootingTroubleshooting#/',
 100         'only_matching': True,
 101     }, {
 102         'url': 'https://www.ciscolive.com/global/on-demand-library.html?search.technicallevel=scpsSkillLevel_aintroductory&search.event=ciscoliveemea2019&search.technology=scpsTechnology_dataCenter&search.focus=scpsSessionFocus_bestPractices#/',
 103         'only_matching': True,
 104     }]
 105
 106     @classmethod
 107     def suitable(cls, url):
 108         return False if CiscoLiveSessionIE.suitable(url) else super(CiscoLiveSearchIE, cls).suitable(url)
 109
 110     @staticmethod
 111     def _check_bc_id_exists(rf_item):
 112         return int_or_none(try_get(rf_item, lambda x: x['videos'][0]['url'])) is not None
 113
 114     def _entries(self, query, url):
 115         query['size'] = 50
 116         query['from'] = 0
 117         for page_num in itertools.count(1):
 118             results = self._call_api(
 119                 'search', None, query, url,
 120                 'Downloading search JSON page %d' % page_num)
 121             sl = try_get(results, lambda x: x['sectionList'][0], dict)
 122             if sl:
 123                 results = sl
 124             items = results.get('items')
 125             if not items or not isinstance(items, list):
 126                 break
 127             for item in items:
 128                 if not isinstance(item, dict):
 129                     continue
 130                 if not self._check_bc_id_exists(item):
 131                     continue
 132                 yield self._parse_rf_item(item)
 133             size = int_or_none(results.get('size'))
 134             if size is not None:
 135                 query['size'] = size
 136             total = int_or_none(results.get('total'))
 137             if total is not None and query['from'] + query['size'] > total:
 138                 break
 139             query['from'] += query['size']
 140
 141     def _real_extract(self, url):
 142         query = parse_qs(url)
 143         query['type'] = 'session'
 144         return self.playlist_result(
 145             self._entries(query, url), playlist_title='Search query')