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 )
   9 from ..utils import (
  10     ExtractorError,
  11     float_or_none,
  12     int_or_none,
  13     sanitized_Request,
  14     unescapeHTML,
  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     _NETRC_MACHINE = 'udemy'
  24
  25     _TESTS = [{
  26         'url': 'https://www.udemy.com/java-tutorial/#/lecture/172757',
  27         'md5': '98eda5b657e752cf945d8445e261b5c5',
  28         'info_dict': {
  29             'id': '160614',
  30             'ext': 'mp4',
  31             'title': 'Introduction and Installation',
  32             'description': 'md5:c0d51f6f21ef4ec65f091055a5eef876',
  33             'duration': 579.29,
  34         },
  35         'skip': 'Requires udemy account credentials',
  36     }]
  37
  38     def _enroll_course(self, webpage, course_id):
  39         checkout_url = unescapeHTML(self._search_regex(
  40             r'href=(["\'])(?P<url>https?://(?:www\.)?udemy\.com/payment/checkout/.+?)\1',
  41             webpage, 'checkout url', group='url', default=None))
  42         if checkout_url:
  43             raise ExtractorError(
  44                 'Course %s is not free. You have to pay for it before you can download. '
  45                 'Use this URL to confirm purchase: %s' % (course_id, checkout_url), expected=True)
  46
  47         enroll_url = unescapeHTML(self._search_regex(
  48             r'href=(["\'])(?P<url>https?://(?:www\.)?udemy\.com/course/subscribe/.+?)\1',
  49             webpage, 'enroll url', group='url', default=None))
  50         if enroll_url:
  51             webpage = self._download_webpage(enroll_url, course_id, 'Enrolling in the course')
  52             if '>You have enrolled in' in webpage:
  53                 self.to_screen('%s: Successfully enrolled in the course' % course_id)
  54
  55     def _download_lecture(self, course_id, lecture_id):
  56         return self._download_json(
  57             'https://www.udemy.com/api-2.0/users/me/subscribed-courses/%s/lectures/%s?%s' % (
  58                 course_id, lecture_id, compat_urllib_parse.urlencode({
  59                     'video_only': '',
  60                     'auto_play': '',
  61                     'fields[lecture]': 'title,description,asset',
  62                     'fields[asset]': 'asset_type,stream_url,thumbnail_url,download_urls,data',
  63                     'instructorPreviewMode': 'False',
  64                 })),
  65             lecture_id, 'Downloading lecture JSON')
  66
  67     def _handle_error(self, response):
  68         if not isinstance(response, dict):
  69             return
  70         error = response.get('error')
  71         if error:
  72             error_str = 'Udemy returned error #%s: %s' % (error.get('code'), error.get('message'))
  73             error_data = error.get('data')
  74             if error_data:
  75                 error_str += ' - %s' % error_data.get('formErrors')
  76             raise ExtractorError(error_str, expected=True)
  77
  78     def _download_json(self, url_or_request, video_id, note='Downloading JSON metadata'):
  79         headers = {
  80             'X-Udemy-Snail-Case': 'true',
  81             'X-Requested-With': 'XMLHttpRequest',
  82         }
  83         for cookie in self._downloader.cookiejar:
  84             if cookie.name == 'client_id':
  85                 headers['X-Udemy-Client-Id'] = cookie.value
  86             elif cookie.name == 'access_token':
  87                 headers['X-Udemy-Bearer-Token'] = cookie.value
  88                 headers['X-Udemy-Authorization'] = 'Bearer %s' % cookie.value
  89
  90         if isinstance(url_or_request, compat_urllib_request.Request):
  91             for header, value in headers.items():
  92                 url_or_request.add_header(header, value)
  93         else:
  94             url_or_request = sanitized_Request(url_or_request, headers=headers)
  95
  96         response = super(UdemyIE, self)._download_json(url_or_request, video_id, note)
  97         self._handle_error(response)
  98         return response
  99
 100     def _real_initialize(self):
 101         self._login()
 102
 103     def _login(self):
 104         (username, password) = self._get_login_info()
 105         if username is None:
 106             return
 107
 108         login_popup = self._download_webpage(
 109             self._LOGIN_URL, None, 'Downloading login popup')
 110
 111         def is_logged(webpage):
 112             return any(p in webpage for p in ['href="https://www.udemy.com/user/logout/', '>Logout<'])
 113
 114         # already logged in
 115         if is_logged(login_popup):
 116             return
 117
 118         login_form = self._form_hidden_inputs('login-form', login_popup)
 119
 120         login_form.update({
 121             'email': username.encode('utf-8'),
 122             'password': password.encode('utf-8'),
 123         })
 124
 125         request = sanitized_Request(
 126             self._LOGIN_URL, compat_urllib_parse.urlencode(login_form).encode('utf-8'))
 127         request.add_header('Referer', self._ORIGIN_URL)
 128         request.add_header('Origin', self._ORIGIN_URL)
 129
 130         response = self._download_webpage(
 131             request, None, 'Logging in as %s' % username)
 132
 133         if not is_logged(response):
 134             error = self._html_search_regex(
 135                 r'(?s)<div[^>]+class="form-errors[^"]*">(.+?)</div>',
 136                 response, 'error message', default=None)
 137             if error:
 138                 raise ExtractorError('Unable to login: %s' % error, expected=True)
 139             raise ExtractorError('Unable to log in')
 140
 141     def _real_extract(self, url):
 142         lecture_id = self._match_id(url)
 143
 144         webpage = self._download_webpage(url, lecture_id)
 145
 146         course_id = self._search_regex(
 147             (r'data-course-id=["\'](\d+)', r'&quot;id&quot;\s*:\s*(\d+)'),
 148             webpage, 'course id')
 149
 150         try:
 151             lecture = self._download_lecture(course_id, lecture_id)
 152         except ExtractorError as e:
 153             # Error could possibly mean we are not enrolled in the course
 154             if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403:
 155                 self._enroll_course(webpage, course_id)
 156                 lecture = self._download_lecture(course_id, lecture_id)
 157             else:
 158                 raise
 159
 160         title = lecture['title']
 161         description = lecture.get('description')
 162
 163         asset = lecture['asset']
 164
 165         asset_type = asset.get('assetType') or asset.get('asset_type')
 166         if asset_type != 'Video':
 167             raise ExtractorError(
 168                 'Lecture %s is not a video' % lecture_id, expected=True)
 169
 170         stream_url = asset.get('streamUrl') or asset.get('stream_url')
 171         if stream_url:
 172             youtube_url = self._search_regex(
 173                 r'(https?://www\.youtube\.com/watch\?v=.*)', stream_url, 'youtube URL', default=None)
 174             if youtube_url:
 175                 return self.url_result(youtube_url, 'Youtube')
 176
 177         video_id = asset['id']
 178         thumbnail = asset.get('thumbnailUrl') or asset.get('thumbnail_url')
 179         duration = float_or_none(asset.get('data', {}).get('duration'))
 180         outputs = asset.get('data', {}).get('outputs', {})
 181
 182         formats = []
 183         for format_ in asset.get('download_urls', {}).get('Video', []):
 184             video_url = format_.get('file')
 185             if not video_url:
 186                 continue
 187             format_id = format_.get('label')
 188             f = {
 189                 'url': format_['file'],
 190                 'height': int_or_none(format_id),
 191             }
 192             if format_id:
 193                 # Some videos contain additional metadata (e.g.
 194                 # https://www.udemy.com/ios9-swift/learn/#/lecture/3383208)
 195                 output = outputs.get(format_id)
 196                 if isinstance(output, dict):
 197                     f.update({
 198                         'format_id': '%sp' % (output.get('label') or format_id),
 199                         'width': int_or_none(output.get('width')),
 200                         'height': int_or_none(output.get('height')),
 201                         'vbr': int_or_none(output.get('video_bitrate_in_kbps')),
 202                         'vcodec': output.get('video_codec'),
 203                         'fps': int_or_none(output.get('frame_rate')),
 204                         'abr': int_or_none(output.get('audio_bitrate_in_kbps')),
 205                         'acodec': output.get('audio_codec'),
 206                         'asr': int_or_none(output.get('audio_sample_rate')),
 207                         'tbr': int_or_none(output.get('total_bitrate_in_kbps')),
 208                         'filesize': int_or_none(output.get('file_size_in_bytes')),
 209                     })
 210                 else:
 211                     f['format_id'] = '%sp' % format_id
 212             formats.append(f)
 213
 214         self._sort_formats(formats)
 215
 216         return {
 217             'id': video_id,
 218             'title': title,
 219             'description': description,
 220             'thumbnail': thumbnail,
 221             'duration': duration,
 222             'formats': formats
 223         }
 224
 225
 226 class UdemyCourseIE(UdemyIE):
 227     IE_NAME = 'udemy:course'
 228     _VALID_URL = r'https?://www\.udemy\.com/(?P<id>[\da-z-]+)'
 229     _TESTS = []
 230
 231     @classmethod
 232     def suitable(cls, url):
 233         return False if UdemyIE.suitable(url) else super(UdemyCourseIE, cls).suitable(url)
 234
 235     def _real_extract(self, url):
 236         course_path = self._match_id(url)
 237
 238         webpage = self._download_webpage(url, course_path)
 239
 240         response = self._download_json(
 241             'https://www.udemy.com/api-1.1/courses/%s' % course_path,
 242             course_path, 'Downloading course JSON')
 243
 244         course_id = response['id']
 245         course_title = response.get('title')
 246
 247         self._enroll_course(webpage, course_id)
 248
 249         response = self._download_json(
 250             'https://www.udemy.com/api-1.1/courses/%s/curriculum' % course_id,
 251             course_id, 'Downloading course curriculum')
 252
 253         entries = []
 254         chapter, chapter_number = None, None
 255         for asset in response:
 256             asset_type = asset.get('assetType') or asset.get('asset_type')
 257             if asset_type == 'Video':
 258                 asset_id = asset.get('id')
 259                 if asset_id:
 260                     entry = {
 261                         '_type': 'url_transparent',
 262                         'url': 'https://www.udemy.com/%s/#/lecture/%s' % (course_path, asset['id']),
 263                         'ie_key': UdemyIE.ie_key(),
 264                     }
 265                     if chapter_number:
 266                         entry['chapter_number'] = chapter_number
 267                     if chapter:
 268                         entry['chapter'] = chapter
 269                     entries.append(entry)
 270             elif asset.get('type') == 'chapter':
 271                 chapter_number = asset.get('index') or asset.get('object_index')
 272                 chapter = asset.get('title')
 273
 274         return self.playlist_result(entries, course_id, course_title)