youtube_dl/extractor/udemy.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..compat import (
   7     compat_urllib_parse,
   8     compat_urllib_request,
   9 )
  10 from ..utils import (
  11     ExtractorError,
  12 )
  13
  14
  15 class UdemyIE(InfoExtractor):
  16     IE_NAME = 'udemy'
  17     _VALID_URL = r'https?://www\.udemy\.com/(?:[^#]+#/lecture/|lecture/view/?\?lectureId=)(?P<id>\d+)'
  18     _LOGIN_URL = 'https://www.udemy.com/join/login-popup/?displayType=ajax&showSkipButton=1'
  19     _ORIGIN_URL = 'https://www.udemy.com'
  20     _NETRC_MACHINE = 'udemy'
  21
  22     _TESTS = [{
  23         'url': 'https://www.udemy.com/java-tutorial/#/lecture/172757',
  24         'md5': '98eda5b657e752cf945d8445e261b5c5',
  25         'info_dict': {
  26             'id': '160614',
  27             'ext': 'mp4',
  28             'title': 'Introduction and Installation',
  29             'description': 'md5:c0d51f6f21ef4ec65f091055a5eef876',
  30             'duration': 579.29,
  31         },
  32         'skip': 'Requires udemy account credentials',
  33     }]
  34
  35     def _handle_error(self, response):
  36         if not isinstance(response, dict):
  37             return
  38         error = response.get('error')
  39         if error:
  40             error_str = 'Udemy returned error #%s: %s' % (error.get('code'), error.get('message'))
  41             error_data = error.get('data')
  42             if error_data:
  43                 error_str += ' - %s' % error_data.get('formErrors')
  44             raise ExtractorError(error_str, expected=True)
  45
  46     def _download_json(self, url_or_request, video_id, note='Downloading JSON metadata'):
  47         headers = {
  48             'X-Udemy-Snail-Case': 'true',
  49             'X-Requested-With': 'XMLHttpRequest',
  50         }
  51         for cookie in self._downloader.cookiejar:
  52             if cookie.name == 'client_id':
  53                 headers['X-Udemy-Client-Id'] = cookie.value
  54             elif cookie.name == 'access_token':
  55                 headers['X-Udemy-Bearer-Token'] = cookie.value
  56
  57         if isinstance(url_or_request, compat_urllib_request.Request):
  58             for header, value in headers.items():
  59                 url_or_request.add_header(header, value)
  60         else:
  61             url_or_request = compat_urllib_request.Request(url_or_request, headers=headers)
  62
  63         response = super(UdemyIE, self)._download_json(url_or_request, video_id, note)
  64         self._handle_error(response)
  65         return response
  66
  67     def _real_initialize(self):
  68         self._login()
  69
  70     def _login(self):
  71         (username, password) = self._get_login_info()
  72         if username is None:
  73             raise ExtractorError(
  74                 'Udemy account is required, use --username and --password options to provide account credentials.',
  75                 expected=True)
  76
  77         login_popup = self._download_webpage(
  78             self._LOGIN_URL, None, 'Downloading login popup')
  79
  80         def is_logged(webpage):
  81             return any(p in webpage for p in ['href="https://www.udemy.com/user/logout/', '>Logout<'])
  82
  83         # already logged in
  84         if is_logged(login_popup):
  85             return
  86
  87         login_form = self._form_hidden_inputs('login-form', login_popup)
  88
  89         login_form.update({
  90             'email': username.encode('utf-8'),
  91             'password': password.encode('utf-8'),
  92         })
  93
  94         request = compat_urllib_request.Request(
  95             self._LOGIN_URL, compat_urllib_parse.urlencode(login_form).encode('utf-8'))
  96         request.add_header('Referer', self._ORIGIN_URL)
  97         request.add_header('Origin', self._ORIGIN_URL)
  98
  99         response = self._download_webpage(
 100             request, None, 'Logging in as %s' % username)
 101
 102         if not is_logged(response):
 103             error = self._html_search_regex(
 104                 r'(?s)<div[^>]+class="form-errors[^"]*">(.+?)</div>',
 105                 response, 'error message', default=None)
 106             if error:
 107                 raise ExtractorError('Unable to login: %s' % error, expected=True)
 108             raise ExtractorError('Unable to log in')
 109
 110     def _real_extract(self, url):
 111         lecture_id = self._match_id(url)
 112
 113         lecture = self._download_json(
 114             'https://www.udemy.com/api-1.1/lectures/%s' % lecture_id,
 115             lecture_id, 'Downloading lecture JSON')
 116
 117         asset_type = lecture.get('assetType') or lecture.get('asset_type')
 118         if asset_type != 'Video':
 119             raise ExtractorError(
 120                 'Lecture %s is not a video' % lecture_id, expected=True)
 121
 122         asset = lecture['asset']
 123
 124         stream_url = asset.get('streamUrl') or asset.get('stream_url')
 125         mobj = re.search(r'(https?://www\.youtube\.com/watch\?v=.*)', stream_url)
 126         if mobj:
 127             return self.url_result(mobj.group(1), 'Youtube')
 128
 129         video_id = asset['id']
 130         thumbnail = asset.get('thumbnailUrl') or asset.get('thumbnail_url')
 131         duration = asset['data']['duration']
 132
 133         download_url = asset.get('downloadUrl') or asset.get('download_url')
 134
 135         video = download_url.get('Video') or download_url.get('video')
 136         video_480p = download_url.get('Video480p') or download_url.get('video_480p')
 137
 138         formats = [
 139             {
 140                 'url': video_480p[0],
 141                 'format_id': '360p',
 142             },
 143             {
 144                 'url': video[0],
 145                 'format_id': '720p',
 146             },
 147         ]
 148
 149         title = lecture['title']
 150         description = lecture['description']
 151
 152         return {
 153             'id': video_id,
 154             'title': title,
 155             'description': description,
 156             'thumbnail': thumbnail,
 157             'duration': duration,
 158             'formats': formats
 159         }
 160
 161
 162 class UdemyCourseIE(UdemyIE):
 163     IE_NAME = 'udemy:course'
 164     _VALID_URL = r'https?://www\.udemy\.com/(?P<coursepath>[\da-z-]+)'
 165     _SUCCESSFULLY_ENROLLED = '>You have enrolled in this course!<'
 166     _ALREADY_ENROLLED = '>You are already taking this course.<'
 167     _TESTS = []
 168
 169     @classmethod
 170     def suitable(cls, url):
 171         return False if UdemyIE.suitable(url) else super(UdemyCourseIE, cls).suitable(url)
 172
 173     def _real_extract(self, url):
 174         mobj = re.match(self._VALID_URL, url)
 175         course_path = mobj.group('coursepath')
 176
 177         response = self._download_json(
 178             'https://www.udemy.com/api-1.1/courses/%s' % course_path,
 179             course_path, 'Downloading course JSON')
 180
 181         course_id = int(response['id'])
 182         course_title = response['title']
 183
 184         webpage = self._download_webpage(
 185             'https://www.udemy.com/course/subscribe/?courseId=%s' % course_id,
 186             course_id, 'Enrolling in the course')
 187
 188         if self._SUCCESSFULLY_ENROLLED in webpage:
 189             self.to_screen('%s: Successfully enrolled in' % course_id)
 190         elif self._ALREADY_ENROLLED in webpage:
 191             self.to_screen('%s: Already enrolled in' % course_id)
 192
 193         response = self._download_json(
 194             'https://www.udemy.com/api-1.1/courses/%s/curriculum' % course_id,
 195             course_id, 'Downloading course curriculum')
 196
 197         entries = [
 198             self.url_result(
 199                 'https://www.udemy.com/%s/#/lecture/%s' % (course_path, asset['id']), 'Udemy')
 200             for asset in response if asset.get('assetType') or asset.get('asset_type') == 'Video'
 201         ]
 202
 203         return self.playlist_result(entries, course_id, course_title)