Raphaël G. Git Repositories - youtubedl/blob - 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         if login_popup == '<div class="run-command close-popup redirect" data-url="https://www.udemy.com/"></div>':
  81             return
  82
  83         login_form = self._form_hidden_inputs('login-form', login_popup)
  84
  85         login_form.update({
  86             'email': username.encode('utf-8'),
  87             'password': password.encode('utf-8'),
  88         })
  89
  90         request = compat_urllib_request.Request(
  91             self._LOGIN_URL, compat_urllib_parse.urlencode(login_form).encode('utf-8'))
  92         request.add_header('Referer', self._ORIGIN_URL)
  93         request.add_header('Origin', self._ORIGIN_URL)
  94
  95         response = self._download_webpage(
  96             request, None, 'Logging in as %s' % username)
  97
  98         if all(logout_pattern not in response
  99                for logout_pattern in ['href="https://www.udemy.com/user/logout/', '>Logout<']):
 100             error = self._html_search_regex(
 101                 r'(?s)<div[^>]+class="form-errors[^"]*">(.+?)</div>',
 102                 response, 'error message', default=None)
 103             if error:
 104                 raise ExtractorError('Unable to login: %s' % error, expected=True)
 105             raise ExtractorError('Unable to log in')
 106
 107     def _real_extract(self, url):
 108         lecture_id = self._match_id(url)
 109
 110         lecture = self._download_json(
 111             'https://www.udemy.com/api-1.1/lectures/%s' % lecture_id,
 112             lecture_id, 'Downloading lecture JSON')
 113
 114         asset_type = lecture.get('assetType') or lecture.get('asset_type')
 115         if asset_type != 'Video':
 116             raise ExtractorError(
 117                 'Lecture %s is not a video' % lecture_id, expected=True)
 118
 119         asset = lecture['asset']
 120
 121         stream_url = asset.get('streamUrl') or asset.get('stream_url')
 122         mobj = re.search(r'(https?://www\.youtube\.com/watch\?v=.*)', stream_url)
 123         if mobj:
 124             return self.url_result(mobj.group(1), 'Youtube')
 125
 126         video_id = asset['id']
 127         thumbnail = asset.get('thumbnailUrl') or asset.get('thumbnail_url')
 128         duration = asset['data']['duration']
 129
 130         download_url = asset.get('downloadUrl') or asset.get('download_url')
 131
 132         video = download_url.get('Video') or download_url.get('video')
 133         video_480p = download_url.get('Video480p') or download_url.get('video_480p')
 134
 135         formats = [
 136             {
 137                 'url': video_480p[0],
 138                 'format_id': '360p',
 139             },
 140             {
 141                 'url': video[0],
 142                 'format_id': '720p',
 143             },
 144         ]
 145
 146         title = lecture['title']
 147         description = lecture['description']
 148
 149         return {
 150             'id': video_id,
 151             'title': title,
 152             'description': description,
 153             'thumbnail': thumbnail,
 154             'duration': duration,
 155             'formats': formats
 156         }
 157
 158
 159 class UdemyCourseIE(UdemyIE):
 160     IE_NAME = 'udemy:course'
 161     _VALID_URL = r'https?://www\.udemy\.com/(?P<coursepath>[\da-z-]+)'
 162     _SUCCESSFULLY_ENROLLED = '>You have enrolled in this course!<'
 163     _ALREADY_ENROLLED = '>You are already taking this course.<'
 164     _TESTS = []
 165
 166     @classmethod
 167     def suitable(cls, url):
 168         return False if UdemyIE.suitable(url) else super(UdemyCourseIE, cls).suitable(url)
 169
 170     def _real_extract(self, url):
 171         mobj = re.match(self._VALID_URL, url)
 172         course_path = mobj.group('coursepath')
 173
 174         response = self._download_json(
 175             'https://www.udemy.com/api-1.1/courses/%s' % course_path,
 176             course_path, 'Downloading course JSON')
 177
 178         course_id = int(response['id'])
 179         course_title = response['title']
 180
 181         webpage = self._download_webpage(
 182             'https://www.udemy.com/course/subscribe/?courseId=%s' % course_id,
 183             course_id, 'Enrolling in the course')
 184
 185         if self._SUCCESSFULLY_ENROLLED in webpage:
 186             self.to_screen('%s: Successfully enrolled in' % course_id)
 187         elif self._ALREADY_ENROLLED in webpage:
 188             self.to_screen('%s: Already enrolled in' % course_id)
 189
 190         response = self._download_json(
 191             'https://www.udemy.com/api-1.1/courses/%s/curriculum' % course_id,
 192             course_id, 'Downloading course curriculum')
 193
 194         entries = [
 195             self.url_result(
 196                 'https://www.udemy.com/%s/#/lecture/%s' % (course_path, asset['id']), 'Udemy')
 197             for asset in response if asset.get('assetType') or asset.get('asset_type') == 'Video'
 198         ]
 199
 200         return self.playlist_result(entries, course_id, course_title)