_ Git - youtube-dl/blob - youtube_dl/extractor/udemy.py

   1 from __future__ import unicode_literals
   2
   3 from .common import InfoExtractor
   4 from ..compat import (
   5     compat_HTTPError,
   6     compat_urllib_parse,
   7     compat_urllib_request,
   8     compat_urlparse,
   9 )
  10 from ..utils import (
  11     ExtractorError,
  12     float_or_none,
  13     int_or_none,
  14     sanitized_Request,
  15 )
  16
  17
  18 class UdemyIE(InfoExtractor):
  19     IE_NAME = 'udemy'
  20     _VALID_URL = r'https?://www\.udemy\.com/(?:[^#]+#/lecture/|lecture/view/?\?lectureId=)(?P<id>\d+)'
  21     _LOGIN_URL = 'https://www.udemy.com/join/login-popup/?displayType=ajax&showSkipButton=1'
  22     _ORIGIN_URL = 'https://www.udemy.com'
  23     _SUCCESSFULLY_ENROLLED = '>You have enrolled in this course!<'
  24     _ALREADY_ENROLLED = '>You are already taking this course.<'
  25     _NETRC_MACHINE = 'udemy'
  26
  27     _TESTS = [{
  28         'url': 'https://www.udemy.com/java-tutorial/#/lecture/172757',
  29         'md5': '98eda5b657e752cf945d8445e261b5c5',
  30         'info_dict': {
  31             'id': '160614',
  32             'ext': 'mp4',
  33             'title': 'Introduction and Installation',
  34             'description': 'md5:c0d51f6f21ef4ec65f091055a5eef876',
  35             'duration': 579.29,
  36         },
  37         'skip': 'Requires udemy account credentials',
  38     }]
  39
  40     def _enroll_course(self, webpage, course_id):
  41         enroll_url = self._search_regex(
  42             r'href=(["\'])(?P<url>https?://(?:www\.)?udemy\.com/course/subscribe/.+?)\1',
  43             webpage, 'enroll url', group='url',
  44             default='https://www.udemy.com/course/subscribe/?courseId=%s' % course_id)
  45         webpage = self._download_webpage(enroll_url, course_id, 'Enrolling in the course')
  46         if self._SUCCESSFULLY_ENROLLED in webpage:
  47             self.to_screen('%s: Successfully enrolled in' % course_id)
  48         elif self._ALREADY_ENROLLED in webpage:
  49             self.to_screen('%s: Already enrolled in' % course_id)
  50
  51     def _download_lecture(self, course_id, lecture_id):
  52         return self._download_json(
  53             'https://www.udemy.com/api-2.0/users/me/subscribed-courses/%s/lectures/%s?%s' % (
  54                 course_id, lecture_id, compat_urllib_parse.urlencode({
  55                     'video_only': '',
  56                     'auto_play': '',
  57                     'fields[lecture]': 'title,description,asset',
  58                     'fields[asset]': 'asset_type,stream_url,thumbnail_url,download_urls,data',
  59                     'instructorPreviewMode': 'False',
  60                 })),
  61             lecture_id, 'Downloading lecture JSON', fatal=False)
  62
  63     def _handle_error(self, response):
  64         if not isinstance(response, dict):
  65             return
  66         error = response.get('error')
  67         if error:
  68             error_str = 'Udemy returned error #%s: %s' % (error.get('code'), error.get('message'))
  69             error_data = error.get('data')
  70             if error_data:
  71                 error_str += ' - %s' % error_data.get('formErrors')
  72             raise ExtractorError(error_str, expected=True)
  73
  74     def _download_json(self, url_or_request, video_id, note='Downloading JSON metadata', *args, **kwargs):
  75         headers = {
  76             'X-Udemy-Snail-Case': 'true',
  77             'X-Requested-With': 'XMLHttpRequest',
  78         }
  79         for cookie in self._downloader.cookiejar:
  80             if cookie.name == 'client_id':
  81                 headers['X-Udemy-Client-Id'] = cookie.value
  82             elif cookie.name == 'access_token':
  83                 headers['X-Udemy-Bearer-Token'] = cookie.value
  84                 headers['X-Udemy-Authorization'] = 'Bearer %s' % cookie.value
  85
  86         if isinstance(url_or_request, compat_urllib_request.Request):
  87             for header, value in headers.items():
  88                 url_or_request.add_header(header, value)
  89         else:
  90             url_or_request = sanitized_Request(url_or_request, headers=headers)
  91
  92         response = super(UdemyIE, self)._download_json(url_or_request, video_id, note, *args, **kwargs)
  93         self._handle_error(response)
  94         return response
  95
  96     def _real_initialize(self):
  97         self._login()
  98
  99     def _login(self):
 100         (username, password) = self._get_login_info()
 101         if username is None:
 102             return
 103
 104         login_popup = self._download_webpage(
 105             self._LOGIN_URL, None, 'Downloading login popup')
 106
 107         def is_logged(webpage):
 108             return any(p in webpage for p in ['href="https://www.udemy.com/user/logout/', '>Logout<'])
 109
 110         # already logged in
 111         if is_logged(login_popup):
 112             return
 113
 114         login_form = self._form_hidden_inputs('login-form', login_popup)
 115
 116         login_form.update({
 117             'email': username.encode('utf-8'),
 118             'password': password.encode('utf-8'),
 119         })
 120
 121         request = sanitized_Request(
 122             self._LOGIN_URL, compat_urllib_parse.urlencode(login_form).encode('utf-8'))
 123         request.add_header('Referer', self._ORIGIN_URL)
 124         request.add_header('Origin', self._ORIGIN_URL)
 125
 126         response = self._download_webpage(
 127             request, None, 'Logging in as %s' % username)
 128
 129         if not is_logged(response):
 130             error = self._html_search_regex(
 131                 r'(?s)<div[^>]+class="form-errors[^"]*">(.+?)</div>',
 132                 response, 'error message', default=None)
 133             if error:
 134                 raise ExtractorError('Unable to login: %s' % error, expected=True)
 135             raise ExtractorError('Unable to log in')
 136
 137     def _real_extract(self, url):
 138         lecture_id = self._match_id(url)
 139
 140         webpage = self._download_webpage(url, lecture_id)
 141
 142         course_id = self._search_regex(
 143             r'data-course-id=["\'](\d+)', webpage, 'course id')
 144
 145         try:
 146             lecture = self._download_lecture(course_id, lecture_id)
 147         except ExtractorError as e:
 148             # Error could possibly mean we are not enrolled in the course
 149             if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403:
 150                 self._enroll_course(webpage, course_id)
 151                 lecture_id = self._download_lecture(course_id, lecture_id)
 152             else:
 153                 raise
 154
 155         title = lecture['title']
 156         description = lecture.get('description')
 157
 158         asset = lecture['asset']
 159
 160         asset_type = asset.get('assetType') or asset.get('asset_type')
 161         if asset_type != 'Video':
 162             raise ExtractorError(
 163                 'Lecture %s is not a video' % lecture_id, expected=True)
 164
 165         stream_url = asset.get('streamUrl') or asset.get('stream_url')
 166         if stream_url:
 167             youtube_url = self._search_regex(
 168                 r'(https?://www\.youtube\.com/watch\?v=.*)', stream_url, 'youtube URL', default=None)
 169             if youtube_url:
 170                 return self.url_result(youtube_url, 'Youtube')
 171
 172         video_id = asset['id']
 173         thumbnail = asset.get('thumbnailUrl') or asset.get('thumbnail_url')
 174         duration = float_or_none(asset.get('data', {}).get('duration'))
 175         outputs = asset.get('data', {}).get('outputs', {})
 176
 177         formats = []
 178         for format_ in asset.get('download_urls', {}).get('Video', []):
 179             video_url = format_.get('file')
 180             if not video_url:
 181                 continue
 182             format_id = format_.get('label')
 183             f = {
 184                 'url': format_['file'],
 185                 'height': int_or_none(format_id),
 186             }
 187             if format_id:
 188                 # Some videos contain additional metadata (e.g.
 189                 # https://www.udemy.com/ios9-swift/learn/#/lecture/3383208)
 190                 output = outputs.get(format_id)
 191                 if isinstance(output, dict):
 192                     f.update({
 193                         'format_id': '%sp' % (output.get('label') or format_id),
 194                         'width': int_or_none(output.get('width')),
 195                         'height': int_or_none(output.get('height')),
 196                         'vbr': int_or_none(output.get('video_bitrate_in_kbps')),
 197                         'vcodec': output.get('video_codec'),
 198                         'fps': int_or_none(output.get('frame_rate')),
 199                         'abr': int_or_none(output.get('audio_bitrate_in_kbps')),
 200                         'acodec': output.get('audio_codec'),
 201                         'asr': int_or_none(output.get('audio_sample_rate')),
 202                         'tbr': int_or_none(output.get('total_bitrate_in_kbps')),
 203                         'filesize': int_or_none(output.get('file_size_in_bytes')),
 204                     })
 205                 else:
 206                     f['format_id'] = '%sp' % format_id
 207             formats.append(f)
 208
 209         self._sort_formats(formats)
 210
 211         return {
 212             'id': video_id,
 213             'title': title,
 214             'description': description,
 215             'thumbnail': thumbnail,
 216             'duration': duration,
 217             'formats': formats
 218         }
 219
 220
 221 class UdemyCourseIE(UdemyIE):
 222     IE_NAME = 'udemy:course'
 223     _VALID_URL = r'https?://www\.udemy\.com/(?P<id>[\da-z-]+)'
 224     _TESTS = []
 225
 226     @classmethod
 227     def suitable(cls, url):
 228         return False if UdemyIE.suitable(url) else super(UdemyCourseIE, cls).suitable(url)
 229
 230     def _real_extract(self, url):
 231         course_path = self._match_id(url)
 232
 233         webpage = self._download_webpage(url, course_path)
 234
 235         response = self._download_json(
 236             'https://www.udemy.com/api-1.1/courses/%s' % course_path,
 237             course_path, 'Downloading course JSON')
 238
 239         course_id = response['id']
 240         course_title = response.get('title')
 241
 242         self._enroll_course(webpage, course_id)
 243
 244         response = self._download_json(
 245             'https://www.udemy.com/api-1.1/courses/%s/curriculum' % course_id,
 246             course_id, 'Downloading course curriculum')
 247
 248         entries = [
 249             self.url_result(
 250                 'https://www.udemy.com/%s/#/lecture/%s' % (course_path, asset['id']), 'Udemy')
 251             for asset in response if asset.get('assetType') or asset.get('asset_type') == 'Video'
 252         ]
 253
 254         return self.playlist_result(entries, course_id, course_title)