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