yt_dlp/extractor/khanacademy.py

   1 import json
   2
   3 from .common import InfoExtractor
   4 from ..utils import (
   5     int_or_none,
   6     parse_iso8601,
   7     try_get,
   8 )
   9
  10
  11 class KhanAcademyBaseIE(InfoExtractor):
  12     _VALID_URL_TEMPL = r'https?://(?:www\.)?khanacademy\.org/(?P<id>(?:[^/]+/){%s}%s[^?#/&]+)'
  13
  14     def _parse_video(self, video):
  15         return {
  16             '_type': 'url_transparent',
  17             'url': video['youtubeId'],
  18             'id': video.get('slug'),
  19             'title': video.get('title'),
  20             'thumbnail': video.get('imageUrl') or video.get('thumbnailUrl'),
  21             'duration': int_or_none(video.get('duration')),
  22             'description': video.get('description'),
  23             'ie_key': 'Youtube',
  24         }
  25
  26     def _real_extract(self, url):
  27         display_id = self._match_id(url)
  28         content = self._download_json(
  29             'https://www.khanacademy.org/api/internal/graphql/FetchContentData',
  30             display_id, query={
  31                 'fastly_cacheable': 'persist_until_publish',
  32                 'hash': '4134764944',
  33                 'lang': 'en',
  34                 'variables': json.dumps({
  35                     'path': display_id,
  36                     'queryParams': 'lang=en',
  37                     'isModal': False,
  38                     'followRedirects': True,
  39                     'countryCode': 'US',
  40                 }),
  41             })['data']['contentJson']
  42         return self._parse_component_props(self._parse_json(content, display_id)['componentProps'])
  43
  44
  45 class KhanAcademyIE(KhanAcademyBaseIE):
  46     IE_NAME = 'khanacademy'
  47     _VALID_URL = KhanAcademyBaseIE._VALID_URL_TEMPL % ('4', 'v/')
  48     _TEST = {
  49         'url': 'https://www.khanacademy.org/computing/computer-science/cryptography/crypt/v/one-time-pad',
  50         'md5': '9c84b7b06f9ebb80d22a5c8dedefb9a0',
  51         'info_dict': {
  52             'id': 'FlIG3TvQCBQ',
  53             'ext': 'mp4',
  54             'title': 'The one-time pad',
  55             'description': 'The perfect cipher',
  56             'duration': 176,
  57             'uploader': 'Brit Cruise',
  58             'uploader_id': 'khanacademy',
  59             'upload_date': '20120411',
  60             'timestamp': 1334170113,
  61             'license': 'cc-by-nc-sa',
  62         },
  63         'add_ie': ['Youtube'],
  64     }
  65
  66     def _parse_component_props(self, component_props):
  67         video = component_props['tutorialPageData']['contentModel']
  68         info = self._parse_video(video)
  69         author_names = video.get('authorNames')
  70         info.update({
  71             'uploader': ', '.join(author_names) if author_names else None,
  72             'timestamp': parse_iso8601(video.get('dateAdded')),
  73             'license': video.get('kaUserLicense'),
  74         })
  75         return info
  76
  77
  78 class KhanAcademyUnitIE(KhanAcademyBaseIE):
  79     IE_NAME = 'khanacademy:unit'
  80     _VALID_URL = (KhanAcademyBaseIE._VALID_URL_TEMPL % ('2', '')) + '/?(?:[?#&]|$)'
  81     _TEST = {
  82         'url': 'https://www.khanacademy.org/computing/computer-science/cryptography',
  83         'info_dict': {
  84             'id': 'cryptography',
  85             'title': 'Cryptography',
  86             'description': 'How have humans protected their secret messages through history? What has changed today?',
  87         },
  88         'playlist_mincount': 31,
  89     }
  90
  91     def _parse_component_props(self, component_props):
  92         curation = component_props['curation']
  93
  94         entries = []
  95         tutorials = try_get(curation, lambda x: x['tabs'][0]['modules'][0]['tutorials'], list) or []
  96         for tutorial_number, tutorial in enumerate(tutorials, 1):
  97             chapter_info = {
  98                 'chapter': tutorial.get('title'),
  99                 'chapter_number': tutorial_number,
 100                 'chapter_id': tutorial.get('id'),
 101             }
 102             for content_item in (tutorial.get('contentItems') or []):
 103                 if content_item.get('kind') == 'Video':
 104                     info = self._parse_video(content_item)
 105                     info.update(chapter_info)
 106                     entries.append(info)
 107
 108         return self.playlist_result(
 109             entries, curation.get('unit'), curation.get('title'),
 110             curation.get('description'))