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