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 .common import InfoExtractor
   7 from ..compat import (
   8     compat_str,
   9     compat_urllib_parse,
  10     compat_urllib_request,
  11 )
  12 from ..utils import (
  13     ExtractorError,
  14     int_or_none,
  15 )
  16
  17
  18 class LyndaIE(InfoExtractor):
  19     IE_NAME = 'lynda'
  20     IE_DESC = 'lynda.com videos'
  21     _VALID_URL = r'https?://www\.lynda\.com/(?:[^/]+/[^/]+/\d+|player/embed)/(\d+)'
  22     _LOGIN_URL = 'https://www.lynda.com/login/login.aspx'
  23     _NETRC_MACHINE = 'lynda'
  24
  25     _SUCCESSFUL_LOGIN_REGEX = r'isLoggedIn: true'
  26     _TIMECODE_REGEX = r'\[(?P<timecode>\d+:\d+:\d+[\.,]\d+)\]'
  27
  28     ACCOUNT_CREDENTIALS_HINT = 'Use --username and --password options to provide lynda.com account credentials.'
  29
  30     _TESTS = [{
  31         'url': 'http://www.lynda.com/Bootstrap-tutorials/Using-exercise-files/110885/114408-4.html',
  32         'md5': 'ecfc6862da89489161fb9cd5f5a6fac1',
  33         'info_dict': {
  34             'id': '114408',
  35             'ext': 'mp4',
  36             'title': 'Using the exercise files',
  37             'duration': 68
  38         }
  39     }, {
  40         'url': 'https://www.lynda.com/player/embed/133770?tr=foo=1;bar=g;fizz=rt&fs=0',
  41         'only_matching': True,
  42     }]
  43
  44     def _real_initialize(self):
  45         self._login()
  46
  47     def _real_extract(self, url):
  48         mobj = re.match(self._VALID_URL, url)
  49         video_id = mobj.group(1)
  50
  51         page = self._download_webpage('http://www.lynda.com/ajax/player?videoId=%s&type=video' % video_id, video_id,
  52                                       'Downloading video JSON')
  53         video_json = json.loads(page)
  54
  55         if 'Status' in video_json:
  56             raise ExtractorError('lynda returned error: %s' % video_json['Message'], expected=True)
  57
  58         if video_json['HasAccess'] is False:
  59             raise ExtractorError(
  60                 'Video %s is only available for members. ' % video_id + self.ACCOUNT_CREDENTIALS_HINT, expected=True)
  61
  62         video_id = compat_str(video_json['ID'])
  63         duration = video_json['DurationInSeconds']
  64         title = video_json['Title']
  65
  66         formats = []
  67
  68         fmts = video_json.get('Formats')
  69         if fmts:
  70             formats.extend([
  71                 {
  72                     'url': fmt['Url'],
  73                     'ext': fmt['Extension'],
  74                     'width': fmt['Width'],
  75                     'height': fmt['Height'],
  76                     'filesize': fmt['FileSize'],
  77                     'format_id': str(fmt['Resolution'])
  78                 } for fmt in fmts])
  79
  80         prioritized_streams = video_json.get('PrioritizedStreams')
  81         if prioritized_streams:
  82             formats.extend([
  83                 {
  84                     'url': video_url,
  85                     'width': int_or_none(format_id),
  86                     'format_id': format_id,
  87                 } for format_id, video_url in prioritized_streams['0'].items()
  88             ])
  89
  90         self._check_formats(formats, video_id)
  91         self._sort_formats(formats)
  92
  93         subtitles = self.extract_subtitles(video_id, page)
  94
  95         return {
  96             'id': video_id,
  97             'title': title,
  98             'duration': duration,
  99             'subtitles': subtitles,
 100             'formats': formats
 101         }
 102
 103     def _login(self):
 104         (username, password) = self._get_login_info()
 105         if username is None:
 106             return
 107
 108         login_form = {
 109             'username': username,
 110             'password': password,
 111             'remember': 'false',
 112             'stayPut': 'false'
 113         }
 114         request = compat_urllib_request.Request(self._LOGIN_URL, compat_urllib_parse.urlencode(login_form))
 115         login_page = self._download_webpage(request, None, 'Logging in as %s' % username)
 116
 117         # Not (yet) logged in
 118         m = re.search(r'loginResultJson = \'(?P<json>[^\']+)\';', login_page)
 119         if m is not None:
 120             response = m.group('json')
 121             response_json = json.loads(response)
 122             state = response_json['state']
 123
 124             if state == 'notlogged':
 125                 raise ExtractorError('Unable to login, incorrect username and/or password', expected=True)
 126
 127             # This is when we get popup:
 128             # > You're already logged in to lynda.com on two devices.
 129             # > If you log in here, we'll log you out of another device.
 130             # So, we need to confirm this.
 131             if state == 'conflicted':
 132                 confirm_form = {
 133                     'username': '',
 134                     'password': '',
 135                     'resolve': 'true',
 136                     'remember': 'false',
 137                     'stayPut': 'false',
 138                 }
 139                 request = compat_urllib_request.Request(self._LOGIN_URL, compat_urllib_parse.urlencode(confirm_form))
 140                 login_page = self._download_webpage(request, None, 'Confirming log in and log out from another device')
 141
 142         if re.search(self._SUCCESSFUL_LOGIN_REGEX, login_page) is None:
 143             raise ExtractorError('Unable to log in')
 144
 145     def _fix_subtitles(self, subs):
 146         srt = ''
 147         for pos in range(0, len(subs) - 1):
 148             seq_current = subs[pos]
 149             m_current = re.match(self._TIMECODE_REGEX, seq_current['Timecode'])
 150             if m_current is None:
 151                 continue
 152             seq_next = subs[pos + 1]
 153             m_next = re.match(self._TIMECODE_REGEX, seq_next['Timecode'])
 154             if m_next is None:
 155                 continue
 156             appear_time = m_current.group('timecode')
 157             disappear_time = m_next.group('timecode')
 158             text = seq_current['Caption'].lstrip()
 159             srt += '%s\r\n%s --> %s\r\n%s' % (str(pos), appear_time, disappear_time, text)
 160         if srt:
 161             return srt
 162
 163     def _get_subtitles(self, video_id, webpage):
 164         url = 'http://www.lynda.com/ajax/player?videoId=%s&type=transcript' % video_id
 165         subs = self._download_json(url, None, False)
 166         if subs:
 167             return {'en': [{'ext': 'srt', 'data': self._fix_subtitles(subs)}]}
 168         else:
 169             return {}
 170
 171
 172 class LyndaCourseIE(InfoExtractor):
 173     IE_NAME = 'lynda:course'
 174     IE_DESC = 'lynda.com online courses'
 175
 176     # Course link equals to welcome/introduction video link of same course
 177     # We will recognize it as course link
 178     _VALID_URL = r'https?://(?:www|m)\.lynda\.com/(?P<coursepath>[^/]+/[^/]+/(?P<courseid>\d+))-\d\.html'
 179
 180     def _real_extract(self, url):
 181         mobj = re.match(self._VALID_URL, url)
 182         course_path = mobj.group('coursepath')
 183         course_id = mobj.group('courseid')
 184
 185         page = self._download_webpage('http://www.lynda.com/ajax/player?courseId=%s&type=course' % course_id,
 186                                       course_id, 'Downloading course JSON')
 187         course_json = json.loads(page)
 188
 189         if 'Status' in course_json and course_json['Status'] == 'NotFound':
 190             raise ExtractorError('Course %s does not exist' % course_id, expected=True)
 191
 192         unaccessible_videos = 0
 193         videos = []
 194         (username, _) = self._get_login_info()
 195
 196         # Might want to extract videos right here from video['Formats'] as it seems 'Formats' is not provided
 197         # by single video API anymore
 198
 199         for chapter in course_json['Chapters']:
 200             for video in chapter['Videos']:
 201                 if username is None and video['HasAccess'] is False:
 202                     unaccessible_videos += 1
 203                     continue
 204                 videos.append(video['ID'])
 205
 206         if unaccessible_videos > 0:
 207             self._downloader.report_warning('%s videos are only available for members and will not be downloaded. '
 208                                             % unaccessible_videos + LyndaIE.ACCOUNT_CREDENTIALS_HINT)
 209
 210         entries = [
 211             self.url_result('http://www.lynda.com/%s/%s-4.html' %
 212                             (course_path, video_id),
 213                             'Lynda')
 214             for video_id in videos]
 215
 216         course_title = course_json['Title']
 217
 218         return self.playlist_result(entries, course_id, course_title)