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_or_request, video_id, note='Downloading JSON metadata'):
  44         headers = {
  45             'X-Udemy-Snail-Case': 'true',
  46             'X-Requested-With': 'XMLHttpRequest',
  47         }
  48         for cookie in self._downloader.cookiejar:
  49             if cookie.name == 'client_id':
  50                 headers['X-Udemy-Client-Id'] = cookie.value
  51             elif cookie.name == 'access_token':
  52                 headers['X-Udemy-Bearer-Token'] = cookie.value
  53
  54         if isinstance(url_or_request, compat_urllib_request.Request):
  55             for header, value in headers.items():
  56                 url_or_request.add_header(header, value)
  57         else:
  58             url_or_request = compat_urllib_request.Request(url_or_request, headers=headers)
  59
  60         response = super(UdemyIE, self)._download_json(url_or_request, video_id, note)
  61         self._handle_error(response)
  62         return response
  63
  64     def _real_initialize(self):
  65         self._login()
  66
  67     def _login(self):
  68         (username, password) = self._get_login_info()
  69         if username is None:
  70             raise ExtractorError(
  71                 'Udemy account is required, use --username and --password options to provide account credentials.',
  72                 expected=True)
  73
  74         login_popup = self._download_webpage(
  75             'https://www.udemy.com/join/login-popup?displayType=ajax&showSkipButton=1', None,
  76             'Downloading login popup')
  77
  78         if login_popup == '<div class="run-command close-popup redirect" data-url="https://www.udemy.com/"></div>':
  79             return
  80
  81         csrf = self._html_search_regex(
  82             r'<input type="hidden" name="csrf" value="(.+?)"',
  83             login_popup, 'csrf token')
  84
  85         login_form = {
  86             'email': username,
  87             'password': password,
  88             'csrf': csrf,
  89             'displayType': 'json',
  90             'isSubmitted': '1',
  91         }
  92         request = compat_urllib_request.Request(
  93             self._LOGIN_URL, compat_urllib_parse.urlencode(login_form).encode('utf-8'))
  94         response = self._download_json(
  95             request, None, 'Logging in as %s' % username)
  96
  97         if 'returnUrl' not in response:
  98             raise ExtractorError('Unable to log in')
  99
 100     def _real_extract(self, url):
 101         lecture_id = self._match_id(url)
 102
 103         lecture = self._download_json(
 104             'https://www.udemy.com/api-1.1/lectures/%s' % lecture_id,
 105             lecture_id, 'Downloading lecture JSON')
 106
 107         asset_type = lecture.get('assetType') or lecture.get('asset_type')
 108         if asset_type != 'Video':
 109             raise ExtractorError(
 110                 'Lecture %s is not a video' % lecture_id, expected=True)
 111
 112         asset = lecture['asset']
 113
 114         stream_url = asset.get('streamUrl') or asset.get('stream_url')
 115         mobj = re.search(r'(https?://www\.youtube\.com/watch\?v=.*)', stream_url)
 116         if mobj:
 117             return self.url_result(mobj.group(1), 'Youtube')
 118
 119         video_id = asset['id']
 120         thumbnail = asset.get('thumbnailUrl') or asset.get('thumbnail_url')
 121         duration = asset['data']['duration']
 122
 123         download_url = asset.get('downloadUrl') or asset.get('download_url')
 124
 125         video = download_url.get('Video') or download_url.get('video')
 126         video_480p = download_url.get('Video480p') or download_url.get('video_480p')
 127
 128         formats = [
 129             {
 130                 'url': video_480p[0],
 131                 'format_id': '360p',
 132             },
 133             {
 134                 'url': video[0],
 135                 'format_id': '720p',
 136             },
 137         ]
 138
 139         title = lecture['title']
 140         description = lecture['description']
 141
 142         return {
 143             'id': video_id,
 144             'title': title,
 145             'description': description,
 146             'thumbnail': thumbnail,
 147             'duration': duration,
 148             'formats': formats
 149         }
 150
 151
 152 class UdemyCourseIE(UdemyIE):
 153     IE_NAME = 'udemy:course'
 154     _VALID_URL = r'https?://www\.udemy\.com/(?P<coursepath>[\da-z-]+)'
 155     _SUCCESSFULLY_ENROLLED = '>You have enrolled in this course!<'
 156     _ALREADY_ENROLLED = '>You are already taking this course.<'
 157     _TESTS = []
 158
 159     @classmethod
 160     def suitable(cls, url):
 161         return False if UdemyIE.suitable(url) else super(UdemyCourseIE, cls).suitable(url)
 162
 163     def _real_extract(self, url):
 164         mobj = re.match(self._VALID_URL, url)
 165         course_path = mobj.group('coursepath')
 166
 167         response = self._download_json(
 168             'https://www.udemy.com/api-1.1/courses/%s' % course_path,
 169             course_path, 'Downloading course JSON')
 170
 171         course_id = int(response['id'])
 172         course_title = response['title']
 173
 174         webpage = self._download_webpage(
 175             'https://www.udemy.com/course/subscribe/?courseId=%s' % course_id,
 176             course_id, 'Enrolling in the course')
 177
 178         if self._SUCCESSFULLY_ENROLLED in webpage:
 179             self.to_screen('%s: Successfully enrolled in' % course_id)
 180         elif self._ALREADY_ENROLLED in webpage:
 181             self.to_screen('%s: Already enrolled in' % course_id)
 182
 183         response = self._download_json(
 184             'https://www.udemy.com/api-1.1/courses/%s/curriculum' % course_id,
 185             course_id, 'Downloading course curriculum')
 186
 187         entries = [
 188             self.url_result(
 189                 'https://www.udemy.com/%s/#/lecture/%s' % (course_path, asset['id']), 'Udemy')
 190             for asset in response if asset.get('assetType') or asset.get('asset_type') == 'Video'
 191         ]
 192
 193         return self.playlist_result(entries, course_id, course_title)