Raphaël G. Git Repositories - youtubedl/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 ..compat import (
   9     compat_str,
  10     compat_urllib_parse,
  11     compat_urllib_request,
  12 )
  13 from ..utils import (
  14     ExtractorError,
  15     int_or_none,
  16 )
  17
  18
  19 class LyndaIE(SubtitlesInfoExtractor):
  20     IE_NAME = 'lynda'
  21     IE_DESC = 'lynda.com videos'
  22     _VALID_URL = r'https?://www\.lynda\.com/[^/]+/[^/]+/\d+/(\d+)-\d\.html'
  23     _LOGIN_URL = 'https://www.lynda.com/login/login.aspx'
  24     _NETRC_MACHINE = 'lynda'
  25
  26     _SUCCESSFUL_LOGIN_REGEX = r'isLoggedIn: true'
  27     _TIMECODE_REGEX = r'\[(?P<timecode>\d+:\d+:\d+[\.,]\d+)\]'
  28
  29     ACCOUNT_CREDENTIALS_HINT = 'Use --username and --password options to provide lynda.com account credentials.'
  30
  31     _TEST = {
  32         'url': 'http://www.lynda.com/Bootstrap-tutorials/Using-exercise-files/110885/114408-4.html',
  33         'md5': 'ecfc6862da89489161fb9cd5f5a6fac1',
  34         'info_dict': {
  35             'id': '114408',
  36             'ext': 'mp4',
  37             'title': 'Using the exercise files',
  38             'duration': 68
  39         }
  40     }
  41
  42     def _real_initialize(self):
  43         self._login()
  44
  45     def _real_extract(self, url):
  46         mobj = re.match(self._VALID_URL, url)
  47         video_id = mobj.group(1)
  48
  49         page = self._download_webpage('http://www.lynda.com/ajax/player?videoId=%s&type=video' % video_id, video_id,
  50                                       'Downloading video JSON')
  51         video_json = json.loads(page)
  52
  53         if 'Status' in video_json:
  54             raise ExtractorError('lynda returned error: %s' % video_json['Message'], expected=True)
  55
  56         if video_json['HasAccess'] is False:
  57             raise ExtractorError(
  58                 'Video %s is only available for members. ' % video_id + self.ACCOUNT_CREDENTIALS_HINT, expected=True)
  59
  60         video_id = compat_str(video_json['ID'])
  61         duration = video_json['DurationInSeconds']
  62         title = video_json['Title']
  63
  64         formats = []
  65
  66         fmts = video_json.get('Formats')
  67         if fmts:
  68             formats.extend([
  69                 {
  70                     'url': fmt['Url'],
  71                     'ext': fmt['Extension'],
  72                     'width': fmt['Width'],
  73                     'height': fmt['Height'],
  74                     'filesize': fmt['FileSize'],
  75                     'format_id': str(fmt['Resolution'])
  76                 } for fmt in fmts])
  77
  78         prioritized_streams = video_json.get('PrioritizedStreams')
  79         if prioritized_streams:
  80             formats.extend([
  81                 {
  82                     'url': video_url,
  83                     'width': int_or_none(format_id),
  84                     'format_id': format_id,
  85                 } for format_id, video_url in prioritized_streams['0'].items()
  86             ])
  87
  88         self._check_formats(formats, video_id)
  89         self._sort_formats(formats)
  90
  91         if self._downloader.params.get('listsubtitles', False):
  92             self._list_available_subtitles(video_id, page)
  93             return
  94
  95         subtitles = self._fix_subtitles(self.extract_subtitles(video_id, page))
  96
  97         return {
  98             'id': video_id,
  99             'title': title,
 100             'duration': duration,
 101             'subtitles': subtitles,
 102             'formats': formats
 103         }
 104
 105     def _login(self):
 106         (username, password) = self._get_login_info()
 107         if username is None:
 108             return
 109
 110         login_form = {
 111             'username': username,
 112             'password': password,
 113             'remember': 'false',
 114             'stayPut': 'false'
 115         }
 116         request = compat_urllib_request.Request(self._LOGIN_URL, compat_urllib_parse.urlencode(login_form))
 117         login_page = self._download_webpage(request, None, 'Logging in as %s' % username)
 118
 119         # Not (yet) logged in
 120         m = re.search(r'loginResultJson = \'(?P<json>[^\']+)\';', login_page)
 121         if m is not None:
 122             response = m.group('json')
 123             response_json = json.loads(response)
 124             state = response_json['state']
 125
 126             if state == 'notlogged':
 127                 raise ExtractorError('Unable to login, incorrect username and/or password', expected=True)
 128
 129             # This is when we get popup:
 130             # > You're already logged in to lynda.com on two devices.
 131             # > If you log in here, we'll log you out of another device.
 132             # So, we need to confirm this.
 133             if state == 'conflicted':
 134                 confirm_form = {
 135                     'username': '',
 136                     'password': '',
 137                     'resolve': 'true',
 138                     'remember': 'false',
 139                     'stayPut': 'false',
 140                 }
 141                 request = compat_urllib_request.Request(self._LOGIN_URL, compat_urllib_parse.urlencode(confirm_form))
 142                 login_page = self._download_webpage(request, None, 'Confirming log in and log out from another device')
 143
 144         if re.search(self._SUCCESSFUL_LOGIN_REGEX, login_page) is None:
 145             raise ExtractorError('Unable to log in')
 146
 147     def _fix_subtitles(self, subtitles):
 148         if subtitles is None:
 149             return subtitles  # subtitles not requested
 150
 151         fixed_subtitles = {}
 152         for k, v in subtitles.items():
 153             subs = json.loads(v)
 154             if len(subs) == 0:
 155                 continue
 156             srt = ''
 157             for pos in range(0, len(subs) - 1):
 158                 seq_current = subs[pos]
 159                 m_current = re.match(self._TIMECODE_REGEX, seq_current['Timecode'])
 160                 if m_current is None:
 161                     continue
 162                 seq_next = subs[pos + 1]
 163                 m_next = re.match(self._TIMECODE_REGEX, seq_next['Timecode'])
 164                 if m_next is None:
 165                     continue
 166                 appear_time = m_current.group('timecode')
 167                 disappear_time = m_next.group('timecode')
 168                 text = seq_current['Caption']
 169                 srt += '%s\r\n%s --> %s\r\n%s' % (str(pos), appear_time, disappear_time, text)
 170             if srt:
 171                 fixed_subtitles[k] = srt
 172         return fixed_subtitles
 173
 174     def _get_available_subtitles(self, video_id, webpage):
 175         url = 'http://www.lynda.com/ajax/player?videoId=%s&type=transcript' % video_id
 176         sub = self._download_webpage(url, None, False)
 177         sub_json = json.loads(sub)
 178         return {'en': url} if len(sub_json) > 0 else {}
 179
 180
 181 class LyndaCourseIE(InfoExtractor):
 182     IE_NAME = 'lynda:course'
 183     IE_DESC = 'lynda.com online courses'
 184
 185     # Course link equals to welcome/introduction video link of same course
 186     # We will recognize it as course link
 187     _VALID_URL = r'https?://(?:www|m)\.lynda\.com/(?P<coursepath>[^/]+/[^/]+/(?P<courseid>\d+))-\d\.html'
 188
 189     def _real_extract(self, url):
 190         mobj = re.match(self._VALID_URL, url)
 191         course_path = mobj.group('coursepath')
 192         course_id = mobj.group('courseid')
 193
 194         page = self._download_webpage('http://www.lynda.com/ajax/player?courseId=%s&type=course' % course_id,
 195                                       course_id, 'Downloading course JSON')
 196         course_json = json.loads(page)
 197
 198         if 'Status' in course_json and course_json['Status'] == 'NotFound':
 199             raise ExtractorError('Course %s does not exist' % course_id, expected=True)
 200
 201         unaccessible_videos = 0
 202         videos = []
 203         (username, _) = self._get_login_info()
 204
 205         # Might want to extract videos right here from video['Formats'] as it seems 'Formats' is not provided
 206         # by single video API anymore
 207
 208         for chapter in course_json['Chapters']:
 209             for video in chapter['Videos']:
 210                 if username is None and video['HasAccess'] is False:
 211                     unaccessible_videos += 1
 212                     continue
 213                 videos.append(video['ID'])
 214
 215         if unaccessible_videos > 0:
 216             self._downloader.report_warning('%s videos are only available for members and will not be downloaded. '
 217                                             % unaccessible_videos + LyndaIE.ACCOUNT_CREDENTIALS_HINT)
 218
 219         entries = [
 220             self.url_result('http://www.lynda.com/%s/%s-4.html' %
 221                             (course_path, video_id),
 222                             'Lynda')
 223             for video_id in videos]
 224
 225         course_title = course_json['Title']
 226
 227         return self.playlist_result(entries, course_id, course_title)