youtube_dlc/extractor/khanacademy.py

   1 from __future__ import unicode_literals
   2
   3 import json
   4
   5 from .common import InfoExtractor
   6 from ..utils import (
   7     int_or_none,
   8     parse_iso8601,
   9     try_get,
  10 )
  11
  12
  13 class KhanAcademyBaseIE(InfoExtractor):
  14     _VALID_URL_TEMPL = r'https?://(?:www\.)?khanacademy\.org/(?P<id>(?:[^/]+/){%s}%s[^?#/&]+)'
  15
  16     def _parse_video(self, video):
  17         return {
  18             '_type': 'url_transparent',
  19             'url': video['youtubeId'],
  20             'id': video.get('slug'),
  21             'title': video.get('title'),
  22             'thumbnail': video.get('imageUrl') or video.get('thumbnailUrl'),
  23             'duration': int_or_none(video.get('duration')),
  24             'description': video.get('description'),
  25             'ie_key': 'Youtube',
  26         }
  27
  28     def _real_extract(self, url):
  29         display_id = self._match_id(url)
  30         component_props = self._parse_json(self._download_json(
  31             'https://www.khanacademy.org/api/internal/graphql',
  32             display_id, query={
  33                 'hash': 1604303425,
  34                 'variables': json.dumps({
  35                     'path': display_id,
  36                     'queryParams': '',
  37                 }),
  38             })['data']['contentJson'], display_id)['componentProps']
  39         return self._parse_component_props(component_props)
  40
  41
  42 class KhanAcademyIE(KhanAcademyBaseIE):
  43     IE_NAME = 'khanacademy'
  44     _VALID_URL = KhanAcademyBaseIE._VALID_URL_TEMPL % ('4', 'v/')
  45     _TEST = {
  46         'url': 'https://www.khanacademy.org/computing/computer-science/cryptography/crypt/v/one-time-pad',
  47         'md5': '9c84b7b06f9ebb80d22a5c8dedefb9a0',
  48         'info_dict': {
  49             'id': 'FlIG3TvQCBQ',
  50             'ext': 'mp4',
  51             'title': 'The one-time pad',
  52             'description': 'The perfect cipher',
  53             'duration': 176,
  54             'uploader': 'Brit Cruise',
  55             'uploader_id': 'khanacademy',
  56             'upload_date': '20120411',
  57             'timestamp': 1334170113,
  58             'license': 'cc-by-nc-sa',
  59         },
  60         'add_ie': ['Youtube'],
  61     }
  62
  63     def _parse_component_props(self, component_props):
  64         video = component_props['tutorialPageData']['contentModel']
  65         info = self._parse_video(video)
  66         author_names = video.get('authorNames')
  67         info.update({
  68             'uploader': ', '.join(author_names) if author_names else None,
  69             'timestamp': parse_iso8601(video.get('dateAdded')),
  70             'license': video.get('kaUserLicense'),
  71         })
  72         return info
  73
  74
  75 class KhanAcademyUnitIE(KhanAcademyBaseIE):
  76     IE_NAME = 'khanacademy:unit'
  77     _VALID_URL = (KhanAcademyBaseIE._VALID_URL_TEMPL % ('2', '')) + '/?(?:[?#&]|$)'
  78     _TEST = {
  79         'url': 'https://www.khanacademy.org/computing/computer-science/cryptography',
  80         'info_dict': {
  81             'id': 'cryptography',
  82             'title': 'Cryptography',
  83             'description': 'How have humans protected their secret messages through history? What has changed today?',
  84         },
  85         'playlist_mincount': 31,
  86     }
  87
  88     def _parse_component_props(self, component_props):
  89         curation = component_props['curation']
  90
  91         entries = []
  92         tutorials = try_get(curation, lambda x: x['tabs'][0]['modules'][0]['tutorials'], list) or []
  93         for tutorial_number, tutorial in enumerate(tutorials, 1):
  94             chapter_info = {
  95                 'chapter': tutorial.get('title'),
  96                 'chapter_number': tutorial_number,
  97                 'chapter_id': tutorial.get('id'),
  98             }
  99             for content_item in (tutorial.get('contentItems') or []):
 100                 if content_item.get('kind') == 'Video':
 101                     info = self._parse_video(content_item)
 102                     info.update(chapter_info)
 103                     entries.append(info)
 104
 105         return self.playlist_result(
 106             entries, curation.get('unit'), curation.get('title'),
 107             curation.get('description'))