youtube_dl/extractor/udemy.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..utils import (
   7     compat_urllib_parse,
   8     compat_urllib_request,
   9     ExtractorError,
  10 )
  11
  12
  13 class UdemyIE(InfoExtractor):
  14     IE_NAME = 'udemy'
  15     _VALID_URL = r'https?://www\.udemy\.com/(?:[^#]+#/lecture/|lecture/view/?\?lectureId=)(?P<id>\d+)'
  16     _LOGIN_URL = 'https://www.udemy.com/join/login-submit/'
  17     _NETRC_MACHINE = 'udemy'
  18
  19     _TESTS = [{
  20         'url': 'https://www.udemy.com/java-tutorial/#/lecture/172757',
  21         'md5': '98eda5b657e752cf945d8445e261b5c5',
  22         'info_dict': {
  23             'id': '160614',
  24             'ext': 'mp4',
  25             'title': 'Introduction and Installation',
  26             'description': 'md5:c0d51f6f21ef4ec65f091055a5eef876',
  27             'duration': 579.29,
  28         },
  29         'skip': 'Requires udemy account credentials',
  30     }]
  31
  32     def _handle_error(self, response):
  33         if not isinstance(response, dict):
  34             return
  35         error = response.get('error')
  36         if error:
  37             error_str = 'Udemy returned error #%s: %s' % (error.get('code'), error.get('message'))
  38             error_data = error.get('data')
  39             if error_data:
  40                 error_str += ' - %s' % error_data.get('formErrors')
  41             raise ExtractorError(error_str, expected=True)
  42
  43     def _download_json(self, url, video_id, note='Downloading JSON metadata'):
  44         response = super(UdemyIE, self)._download_json(url, video_id, note)
  45         self._handle_error(response)
  46         return response
  47
  48     def _real_initialize(self):
  49         self._login()
  50
  51     def _login(self):
  52         (username, password) = self._get_login_info()
  53         if username is None:
  54             raise ExtractorError(
  55                 'Udemy account is required, use --username and --password options to provide account credentials.',
  56                 expected=True)
  57
  58         login_popup = self._download_webpage(
  59             'https://www.udemy.com/join/login-popup?displayType=ajax&showSkipButton=1', None,
  60             'Downloading login popup')
  61
  62         if login_popup == '<div class="run-command close-popup redirect" data-url="https://www.udemy.com/"></div>':
  63             return
  64
  65         csrf = self._html_search_regex(r'<input type="hidden" name="csrf" value="(.+?)"', login_popup, 'csrf token')
  66
  67         login_form = {
  68             'email': username,
  69             'password': password,
  70             'csrf': csrf,
  71             'displayType': 'json',
  72             'isSubmitted': '1',
  73         }
  74         request = compat_urllib_request.Request(self._LOGIN_URL, compat_urllib_parse.urlencode(login_form))
  75         response = self._download_json(request, None, 'Logging in as %s' % username)
  76
  77         if 'returnUrl' not in response:
  78             raise ExtractorError('Unable to log in')
  79
  80     def _real_extract(self, url):
  81         mobj = re.match(self._VALID_URL, url)
  82         lecture_id = mobj.group('id')
  83
  84         lecture = self._download_json(
  85             'https://www.udemy.com/api-1.1/lectures/%s' % lecture_id, lecture_id, 'Downloading lecture JSON')
  86
  87         if lecture['assetType'] != 'Video':
  88             raise ExtractorError('Lecture %s is not a video' % lecture_id, expected=True)
  89
  90         asset = lecture['asset']
  91
  92         stream_url = asset['streamUrl']
  93         mobj = re.search(r'(https?://www\.youtube\.com/watch\?v=.*)', stream_url)
  94         if mobj:
  95             return self.url_result(mobj.group(1), 'Youtube')
  96
  97         video_id = asset['id']
  98         thumbnail = asset['thumbnailUrl']
  99         duration = asset['data']['duration']
 100
 101         download_url = asset['downloadUrl']
 102
 103         formats = [
 104             {
 105                 'url': download_url['Video480p'][0],
 106                 'format_id': '360p',
 107             },
 108             {
 109                 'url': download_url['Video'][0],
 110                 'format_id': '720p',
 111             },
 112         ]
 113
 114         title = lecture['title']
 115         description = lecture['description']
 116
 117         return {
 118             'id': video_id,
 119             'title': title,
 120             'description': description,
 121             'thumbnail': thumbnail,
 122             'duration': duration,
 123             'formats': formats
 124         }
 125
 126
 127 class UdemyCourseIE(UdemyIE):
 128     IE_NAME = 'udemy:course'
 129     _VALID_URL = r'https?://www\.udemy\.com/(?P<coursepath>[\da-z-]+)'
 130     _SUCCESSFULLY_ENROLLED = '>You have enrolled in this course!<'
 131     _ALREADY_ENROLLED = '>You are already taking this course.<'
 132     _TESTS = []
 133
 134     @classmethod
 135     def suitable(cls, url):
 136         return False if UdemyIE.suitable(url) else super(UdemyCourseIE, cls).suitable(url)
 137
 138     def _real_extract(self, url):
 139         mobj = re.match(self._VALID_URL, url)
 140         course_path = mobj.group('coursepath')
 141
 142         response = self._download_json(
 143             'https://www.udemy.com/api-1.1/courses/%s' % course_path, course_path, 'Downloading course JSON')
 144
 145         course_id = int(response['id'])
 146         course_title = response['title']
 147
 148         webpage = self._download_webpage(
 149             'https://www.udemy.com/course/subscribe/?courseId=%s' % course_id, course_id, 'Enrolling in the course')
 150
 151         if self._SUCCESSFULLY_ENROLLED in webpage:
 152             self.to_screen('%s: Successfully enrolled in' % course_id)
 153         elif self._ALREADY_ENROLLED in webpage:
 154             self.to_screen('%s: Already enrolled in' % course_id)
 155
 156         response = self._download_json('https://www.udemy.com/api-1.1/courses/%s/curriculum' % course_id,
 157             course_id, 'Downloading course curriculum')
 158
 159         entries = [
 160             self.url_result('https://www.udemy.com/%s/#/lecture/%s' % (course_path, asset['id']), 'Udemy')
 161             for asset in response if asset.get('assetType') == 'Video'
 162         ]
 163
 164         return self.playlist_result(entries, course_id, course_title)