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 LyndaBaseIE(InfoExtractor):
  19     _LOGIN_URL = 'https://www.lynda.com/login/login.aspx'
  20     _SUCCESSFUL_LOGIN_REGEX = r'isLoggedIn: true'
  21     _ACCOUNT_CREDENTIALS_HINT = 'Use --username and --password options to provide lynda.com account credentials.'
  22     _NETRC_MACHINE = 'lynda'
  23
  24     def _real_initialize(self):
  25         self._login()
  26
  27     def _login(self):
  28         (username, password) = self._get_login_info()
  29         if username is None:
  30             return
  31
  32         login_form = {
  33             'username': username.encode('utf-8'),
  34             'password': password.encode('utf-8'),
  35             'remember': 'false',
  36             'stayPut': 'false'
  37         }
  38         request = compat_urllib_request.Request(
  39             self._LOGIN_URL, compat_urllib_parse.urlencode(login_form).encode('utf-8'))
  40         login_page = self._download_webpage(
  41             request, None, 'Logging in as %s' % username)
  42
  43         # Not (yet) logged in
  44         m = re.search(r'loginResultJson = \'(?P<json>[^\']+)\';', login_page)
  45         if m is not None:
  46             response = m.group('json')
  47             response_json = json.loads(response)
  48             state = response_json['state']
  49
  50             if state == 'notlogged':
  51                 raise ExtractorError(
  52                     'Unable to login, incorrect username and/or password',
  53                     expected=True)
  54
  55             # This is when we get popup:
  56             # > You're already logged in to lynda.com on two devices.
  57             # > If you log in here, we'll log you out of another device.
  58             # So, we need to confirm this.
  59             if state == 'conflicted':
  60                 confirm_form = {
  61                     'username': '',
  62                     'password': '',
  63                     'resolve': 'true',
  64                     'remember': 'false',
  65                     'stayPut': 'false',
  66                 }
  67                 request = compat_urllib_request.Request(
  68                     self._LOGIN_URL, compat_urllib_parse.urlencode(confirm_form).encode('utf-8'))
  69                 login_page = self._download_webpage(
  70                     request, None,
  71                     'Confirming log in and log out from another device')
  72
  73         if re.search(self._SUCCESSFUL_LOGIN_REGEX, login_page) is None:
  74             raise ExtractorError('Unable to log in')
  75
  76
  77 class LyndaIE(LyndaBaseIE):
  78     IE_NAME = 'lynda'
  79     IE_DESC = 'lynda.com videos'
  80     _VALID_URL = r'https?://www\.lynda\.com/(?:[^/]+/[^/]+/\d+|player/embed)/(?P<id>\d+)'
  81     _NETRC_MACHINE = 'lynda'
  82
  83     _TIMECODE_REGEX = r'\[(?P<timecode>\d+:\d+:\d+[\.,]\d+)\]'
  84
  85     _TESTS = [{
  86         'url': 'http://www.lynda.com/Bootstrap-tutorials/Using-exercise-files/110885/114408-4.html',
  87         'md5': 'ecfc6862da89489161fb9cd5f5a6fac1',
  88         'info_dict': {
  89             'id': '114408',
  90             'ext': 'mp4',
  91             'title': 'Using the exercise files',
  92             'duration': 68
  93         }
  94     }, {
  95         'url': 'https://www.lynda.com/player/embed/133770?tr=foo=1;bar=g;fizz=rt&fs=0',
  96         'only_matching': True,
  97     }]
  98
  99     def _real_extract(self, url):
 100         video_id = self._match_id(url)
 101
 102         page = self._download_webpage(
 103             'http://www.lynda.com/ajax/player?videoId=%s&type=video' % video_id,
 104             video_id, 'Downloading video JSON')
 105         video_json = json.loads(page)
 106
 107         if 'Status' in video_json:
 108             raise ExtractorError(
 109                 'lynda returned error: %s' % video_json['Message'], expected=True)
 110
 111         if video_json['HasAccess'] is False:
 112             raise ExtractorError(
 113                 'Video %s is only available for members. '
 114                 % video_id + self._ACCOUNT_CREDENTIALS_HINT, expected=True)
 115
 116         video_id = compat_str(video_json['ID'])
 117         duration = video_json['DurationInSeconds']
 118         title = video_json['Title']
 119
 120         formats = []
 121
 122         fmts = video_json.get('Formats')
 123         if fmts:
 124             formats.extend([
 125                 {
 126                     'url': fmt['Url'],
 127                     'ext': fmt['Extension'],
 128                     'width': fmt['Width'],
 129                     'height': fmt['Height'],
 130                     'filesize': fmt['FileSize'],
 131                     'format_id': str(fmt['Resolution'])
 132                 } for fmt in fmts])
 133
 134         prioritized_streams = video_json.get('PrioritizedStreams')
 135         if prioritized_streams:
 136             formats.extend([
 137                 {
 138                     'url': video_url,
 139                     'width': int_or_none(format_id),
 140                     'format_id': format_id,
 141                 } for format_id, video_url in prioritized_streams['0'].items()
 142             ])
 143
 144         self._check_formats(formats, video_id)
 145         self._sort_formats(formats)
 146
 147         subtitles = self.extract_subtitles(video_id, page)
 148
 149         return {
 150             'id': video_id,
 151             'title': title,
 152             'duration': duration,
 153             'subtitles': subtitles,
 154             'formats': formats
 155         }
 156
 157     def _fix_subtitles(self, subs):
 158         srt = ''
 159         seq_counter = 0
 160         for pos in range(0, len(subs) - 1):
 161             seq_current = subs[pos]
 162             m_current = re.match(self._TIMECODE_REGEX, seq_current['Timecode'])
 163             if m_current is None:
 164                 continue
 165             seq_next = subs[pos + 1]
 166             m_next = re.match(self._TIMECODE_REGEX, seq_next['Timecode'])
 167             if m_next is None:
 168                 continue
 169             appear_time = m_current.group('timecode')
 170             disappear_time = m_next.group('timecode')
 171             text = seq_current['Caption'].strip()
 172             if text:
 173                 seq_counter += 1
 174                 srt += '%s\r\n%s --> %s\r\n%s\r\n\r\n' % (seq_counter, appear_time, disappear_time, text)
 175         if srt:
 176             return srt
 177
 178     def _get_subtitles(self, video_id, webpage):
 179         url = 'http://www.lynda.com/ajax/player?videoId=%s&type=transcript' % video_id
 180         subs = self._download_json(url, None, False)
 181         if subs:
 182             return {'en': [{'ext': 'srt', 'data': self._fix_subtitles(subs)}]}
 183         else:
 184             return {}
 185
 186
 187 class LyndaCourseIE(LyndaBaseIE):
 188     IE_NAME = 'lynda:course'
 189     IE_DESC = 'lynda.com online courses'
 190
 191     # Course link equals to welcome/introduction video link of same course
 192     # We will recognize it as course link
 193     _VALID_URL = r'https?://(?:www|m)\.lynda\.com/(?P<coursepath>[^/]+/[^/]+/(?P<courseid>\d+))-\d\.html'
 194
 195     def _real_extract(self, url):
 196         mobj = re.match(self._VALID_URL, url)
 197         course_path = mobj.group('coursepath')
 198         course_id = mobj.group('courseid')
 199
 200         page = self._download_webpage(
 201             'http://www.lynda.com/ajax/player?courseId=%s&type=course' % course_id,
 202             course_id, 'Downloading course JSON')
 203         course_json = json.loads(page)
 204
 205         if 'Status' in course_json and course_json['Status'] == 'NotFound':
 206             raise ExtractorError(
 207                 'Course %s does not exist' % course_id, expected=True)
 208
 209         unaccessible_videos = 0
 210         videos = []
 211
 212         # Might want to extract videos right here from video['Formats'] as it seems 'Formats' is not provided
 213         # by single video API anymore
 214
 215         for chapter in course_json['Chapters']:
 216             for video in chapter['Videos']:
 217                 if video['HasAccess'] is False:
 218                     unaccessible_videos += 1
 219                     continue
 220                 videos.append(video['ID'])
 221
 222         if unaccessible_videos > 0:
 223             self._downloader.report_warning(
 224                 '%s videos are only available for members (or paid members) and will not be downloaded. '
 225                 % unaccessible_videos + self._ACCOUNT_CREDENTIALS_HINT)
 226
 227         entries = [
 228             self.url_result(
 229                 'http://www.lynda.com/%s/%s-4.html' % (course_path, video_id),
 230                 'Lynda')
 231             for video_id in videos]
 232
 233         course_title = course_json['Title']
 234
 235         return self.playlist_result(entries, course_id, course_title)