_ Git - youtube-dl/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     int_or_none,
  13     sanitized_Request,
  14 )
  15
  16
  17 class UdemyIE(InfoExtractor):
  18     IE_NAME = 'udemy'
  19     _VALID_URL = r'https?://www\.udemy\.com/(?:[^#]+#/lecture/|lecture/view/?\?lectureId=)(?P<id>\d+)'
  20     _LOGIN_URL = 'https://www.udemy.com/join/login-popup/?displayType=ajax&showSkipButton=1'
  21     _ORIGIN_URL = 'https://www.udemy.com'
  22     _NETRC_MACHINE = 'udemy'
  23
  24     _TESTS = [{
  25         'url': 'https://www.udemy.com/java-tutorial/#/lecture/172757',
  26         'md5': '98eda5b657e752cf945d8445e261b5c5',
  27         'info_dict': {
  28             'id': '160614',
  29             'ext': 'mp4',
  30             'title': 'Introduction and Installation',
  31             'description': 'md5:c0d51f6f21ef4ec65f091055a5eef876',
  32             'duration': 579.29,
  33         },
  34         'skip': 'Requires udemy account credentials',
  35     }]
  36
  37     def _handle_error(self, response):
  38         if not isinstance(response, dict):
  39             return
  40         error = response.get('error')
  41         if error:
  42             error_str = 'Udemy returned error #%s: %s' % (error.get('code'), error.get('message'))
  43             error_data = error.get('data')
  44             if error_data:
  45                 error_str += ' - %s' % error_data.get('formErrors')
  46             raise ExtractorError(error_str, expected=True)
  47
  48     def _download_json(self, url_or_request, video_id, note='Downloading JSON metadata'):
  49         headers = {
  50             'X-Udemy-Snail-Case': 'true',
  51             'X-Requested-With': 'XMLHttpRequest',
  52         }
  53         for cookie in self._downloader.cookiejar:
  54             if cookie.name == 'client_id':
  55                 headers['X-Udemy-Client-Id'] = cookie.value
  56             elif cookie.name == 'access_token':
  57                 headers['X-Udemy-Bearer-Token'] = cookie.value
  58
  59         if isinstance(url_or_request, compat_urllib_request.Request):
  60             for header, value in headers.items():
  61                 url_or_request.add_header(header, value)
  62         else:
  63             url_or_request = sanitized_Request(url_or_request, headers=headers)
  64
  65         response = super(UdemyIE, self)._download_json(url_or_request, video_id, note)
  66         self._handle_error(response)
  67         return response
  68
  69     def _real_initialize(self):
  70         self._login()
  71
  72     def _login(self):
  73         (username, password) = self._get_login_info()
  74         if username is None:
  75             return
  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 = sanitized_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 = int_or_none(asset.get('data', {}).get('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             'url': video_480p[0],
 140             'format_id': 'download-360p',
 141         }, {
 142             'url': video[0],
 143             'format_id': 'download-720p',
 144         }]
 145
 146         # Some videos also contain formats in asset['data']['outputs'] (e.g.
 147         # https://www.udemy.com/ios9-swift/learn/#/lecture/3383208)
 148         outputs = asset.get('data', {}).get('outputs')
 149         if isinstance(outputs, dict):
 150             for format_id, f in outputs.items():
 151                 video_url = f.get('url')
 152                 if video_url:
 153                     formats.append({
 154                         'url': video_url,
 155                         'format_id': '%sp' % (f.get('labe1l') or format_id),
 156                         'width': int_or_none(f.get('width')),
 157                         'height': int_or_none(f.get('height')),
 158                         'vbr': int_or_none(f.get('video_bitrate_in_kbps')),
 159                         'vcodec': f.get('video_codec'),
 160                         'fps': int_or_none(f.get('frame_rate')),
 161                         'abr': int_or_none(f.get('audio_bitrate_in_kbps')),
 162                         'acodec': f.get('audio_codec'),
 163                         'asr': int_or_none(f.get('audio_sample_rate')),
 164                         'tbr': int_or_none(f.get('total_bitrate_in_kbps')),
 165                         'filesize': int_or_none(f.get('file_size_in_bytes')),
 166                     })
 167
 168         self._sort_formats(formats)
 169
 170         title = lecture['title']
 171         description = lecture.get('description')
 172
 173         return {
 174             'id': video_id,
 175             'title': title,
 176             'description': description,
 177             'thumbnail': thumbnail,
 178             'duration': duration,
 179             'formats': formats
 180         }
 181
 182
 183 class UdemyCourseIE(UdemyIE):
 184     IE_NAME = 'udemy:course'
 185     _VALID_URL = r'https?://www\.udemy\.com/(?P<coursepath>[\da-z-]+)'
 186     _SUCCESSFULLY_ENROLLED = '>You have enrolled in this course!<'
 187     _ALREADY_ENROLLED = '>You are already taking this course.<'
 188     _TESTS = []
 189
 190     @classmethod
 191     def suitable(cls, url):
 192         return False if UdemyIE.suitable(url) else super(UdemyCourseIE, cls).suitable(url)
 193
 194     def _real_extract(self, url):
 195         mobj = re.match(self._VALID_URL, url)
 196         course_path = mobj.group('coursepath')
 197
 198         response = self._download_json(
 199             'https://www.udemy.com/api-1.1/courses/%s' % course_path,
 200             course_path, 'Downloading course JSON')
 201
 202         course_id = int(response['id'])
 203         course_title = response['title']
 204
 205         webpage = self._download_webpage(
 206             'https://www.udemy.com/course/subscribe/?courseId=%s' % course_id,
 207             course_id, 'Enrolling in the course')
 208
 209         if self._SUCCESSFULLY_ENROLLED in webpage:
 210             self.to_screen('%s: Successfully enrolled in' % course_id)
 211         elif self._ALREADY_ENROLLED in webpage:
 212             self.to_screen('%s: Already enrolled in' % course_id)
 213
 214         response = self._download_json(
 215             'https://www.udemy.com/api-1.1/courses/%s/curriculum' % course_id,
 216             course_id, 'Downloading course curriculum')
 217
 218         entries = [
 219             self.url_result(
 220                 'https://www.udemy.com/%s/#/lecture/%s' % (course_path, asset['id']), 'Udemy')
 221             for asset in response if asset.get('assetType') or asset.get('asset_type') == 'Video'
 222         ]
 223
 224         return self.playlist_result(entries, course_id, course_title)