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

   1 from __future__ import unicode_literals
   2
   3 import re
   4 import json
   5
   6 from .subtitles import SubtitlesInfoExtractor
   7 from .common import InfoExtractor
   8 from ..utils import (
   9     compat_urllib_parse,
  10     compat_urllib_request,
  11     ExtractorError,
  12     int_or_none,
  13 )
  14
  15
  16 class LyndaIE(SubtitlesInfoExtractor):
  17     IE_NAME = 'lynda'
  18     IE_DESC = 'lynda.com videos'
  19     _VALID_URL = r'https?://www\.lynda\.com/[^/]+/[^/]+/\d+/(\d+)-\d\.html'
  20     _LOGIN_URL = 'https://www.lynda.com/login/login.aspx'
  21     _NETRC_MACHINE = 'lynda'
  22
  23     _SUCCESSFUL_LOGIN_REGEX = r'isLoggedIn: true'
  24     _TIMECODE_REGEX = r'\[(?P<timecode>\d+:\d+:\d+[\.,]\d+)\]'
  25
  26     ACCOUNT_CREDENTIALS_HINT = 'Use --username and --password options to provide lynda.com account credentials.'
  27
  28     _TEST = {
  29         'url': 'http://www.lynda.com/Bootstrap-tutorials/Using-exercise-files/110885/114408-4.html',
  30         'file': '114408.mp4',
  31         'md5': 'ecfc6862da89489161fb9cd5f5a6fac1',
  32         'info_dict': {
  33             'title': 'Using the exercise files',
  34             'duration': 68
  35         }
  36     }
  37
  38     def _real_initialize(self):
  39         self._login()
  40
  41     def _real_extract(self, url):
  42         mobj = re.match(self._VALID_URL, url)
  43         video_id = mobj.group(1)
  44
  45         page = self._download_webpage('http://www.lynda.com/ajax/player?videoId=%s&type=video' % video_id,
  46                                       video_id, 'Downloading video JSON')
  47         video_json = json.loads(page)
  48
  49         if 'Status' in video_json:
  50             raise ExtractorError('lynda returned error: %s' % video_json['Message'], expected=True)
  51
  52         if video_json['HasAccess'] is False:
  53             raise ExtractorError('Video %s is only available for members. ' % video_id + self.ACCOUNT_CREDENTIALS_HINT, expected=True)
  54
  55         video_id = video_json['ID']
  56         duration = video_json['DurationInSeconds']
  57         title = video_json['Title']
  58
  59         formats = []
  60
  61         fmts = video_json.get('Formats')
  62         if fmts:
  63             formats.extend([
  64                 {
  65                     'url': fmt['Url'],
  66                     'ext': fmt['Extension'],
  67                     'width': fmt['Width'],
  68                     'height': fmt['Height'],
  69                     'filesize': fmt['FileSize'],
  70                     'format_id': str(fmt['Resolution'])
  71                 } for fmt in fmts])
  72
  73         prioritized_streams = video_json.get('PrioritizedStreams')
  74         if prioritized_streams:
  75             formats.extend([
  76                 {
  77                     'url': video_url,
  78                     'width': int_or_none(format_id),
  79                     'format_id': format_id,
  80                 } for format_id, video_url in prioritized_streams['0'].items()
  81             ])
  82
  83         self._sort_formats(formats)
  84
  85         if self._downloader.params.get('listsubtitles', False):
  86             self._list_available_subtitles(video_id, page)
  87             return
  88
  89         subtitles = self._fix_subtitles(self.extract_subtitles(video_id, page))
  90
  91         return {
  92             'id': video_id,
  93             'title': title,
  94             'duration': duration,
  95             'subtitles': subtitles,
  96             'formats': formats
  97         }
  98
  99     def _login(self):
 100         (username, password) = self._get_login_info()
 101         if username is None:
 102             return
 103
 104         login_form = {
 105             'username': username,
 106             'password': password,
 107             'remember': 'false',
 108             'stayPut': 'false'
 109         }
 110         request = compat_urllib_request.Request(self._LOGIN_URL, compat_urllib_parse.urlencode(login_form))
 111         login_page = self._download_webpage(request, None, note='Logging in as %s' % username)
 112
 113         # Not (yet) logged in
 114         m = re.search(r'loginResultJson = \'(?P<json>[^\']+)\';', login_page)
 115         if m is not None:
 116             response = m.group('json')
 117             response_json = json.loads(response)
 118             state = response_json['state']
 119
 120             if state == 'notlogged':
 121                 raise ExtractorError('Unable to login, incorrect username and/or password', expected=True)
 122
 123             # This is when we get popup:
 124             # > You're already logged in to lynda.com on two devices.
 125             # > If you log in here, we'll log you out of another device.
 126             # So, we need to confirm this.
 127             if state == 'conflicted':
 128                 confirm_form = {
 129                     'username': '',
 130                     'password': '',
 131                     'resolve': 'true',
 132                     'remember': 'false',
 133                     'stayPut': 'false',
 134                 }
 135                 request = compat_urllib_request.Request(self._LOGIN_URL, compat_urllib_parse.urlencode(confirm_form))
 136                 login_page = self._download_webpage(request, None, note='Confirming log in and log out from another device')
 137
 138         if re.search(self._SUCCESSFUL_LOGIN_REGEX, login_page) is None:
 139             raise ExtractorError('Unable to log in')
 140
 141     def _fix_subtitles(self, subtitles):
 142         if subtitles is None:
 143             return subtitles  # subtitles not requested
 144
 145         fixed_subtitles = {}
 146         for k, v in subtitles.items():
 147             subs = json.loads(v)
 148             if len(subs) == 0:
 149                 continue
 150             srt = ''
 151             for pos in range(0, len(subs) - 1):
 152                 seq_current = subs[pos]
 153                 m_current = re.match(self._TIMECODE_REGEX, seq_current['Timecode'])
 154                 if m_current is None:
 155                     continue
 156                 seq_next = subs[pos + 1]
 157                 m_next = re.match(self._TIMECODE_REGEX, seq_next['Timecode'])
 158                 if m_next is None:
 159                     continue
 160                 appear_time = m_current.group('timecode')
 161                 disappear_time = m_next.group('timecode')
 162                 text = seq_current['Caption']
 163                 srt += '%s\r\n%s --> %s\r\n%s' % (str(pos), appear_time, disappear_time, text)
 164             if srt:
 165                 fixed_subtitles[k] = srt
 166         return fixed_subtitles
 167
 168     def _get_available_subtitles(self, video_id, webpage):
 169         url = 'http://www.lynda.com/ajax/player?videoId=%s&type=transcript' % video_id
 170         sub = self._download_webpage(url, None, note=False)
 171         sub_json = json.loads(sub)
 172         return {'en': url} if len(sub_json) > 0 else {}
 173
 174
 175 class LyndaCourseIE(InfoExtractor):
 176     IE_NAME = 'lynda:course'
 177     IE_DESC = 'lynda.com online courses'
 178
 179     # Course link equals to welcome/introduction video link of same course
 180     # We will recognize it as course link
 181     _VALID_URL = r'https?://(?:www|m)\.lynda\.com/(?P<coursepath>[^/]+/[^/]+/(?P<courseid>\d+))-\d\.html'
 182
 183     def _real_extract(self, url):
 184         mobj = re.match(self._VALID_URL, url)
 185         course_path = mobj.group('coursepath')
 186         course_id = mobj.group('courseid')
 187
 188         page = self._download_webpage('http://www.lynda.com/ajax/player?courseId=%s&type=course' % course_id,
 189                                       course_id, 'Downloading course JSON')
 190         course_json = json.loads(page)
 191
 192         if 'Status' in course_json and course_json['Status'] == 'NotFound':
 193             raise ExtractorError('Course %s does not exist' % course_id, expected=True)
 194
 195         unaccessible_videos = 0
 196         videos = []
 197         (username, _) = self._get_login_info()
 198
 199         # Might want to extract videos right here from video['Formats'] as it seems 'Formats' is not provided
 200         # by single video API anymore
 201
 202         for chapter in course_json['Chapters']:
 203             for video in chapter['Videos']:
 204                 if username is None and video['HasAccess'] is False:
 205                     unaccessible_videos += 1
 206                     continue
 207                 videos.append(video['ID'])
 208
 209         if unaccessible_videos > 0:
 210             self._downloader.report_warning('%s videos are only available for members and will not be downloaded. '
 211                                             % unaccessible_videos + LyndaIE.ACCOUNT_CREDENTIALS_HINT)
 212
 213         entries = [
 214             self.url_result('http://www.lynda.com/%s/%s-4.html' %
 215                             (course_path, video_id),
 216                             'Lynda')
 217             for video_id in videos]
 218
 219         course_title = course_json['Title']
 220
 221         return self.playlist_result(entries, course_id, course_title)