Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/lynda.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..compat import (
   7     compat_HTTPError,
   8     compat_str,
   9     compat_urlparse,
  10 )
  11 from ..utils import (
  12     ExtractorError,
  13     int_or_none,
  14     urlencode_postdata,
  15 )
  16
  17
  18 class LyndaBaseIE(InfoExtractor):
  19     _SIGNIN_URL = 'https://www.lynda.com/signin'
  20     _PASSWORD_URL = 'https://www.lynda.com/signin/password'
  21     _USER_URL = 'https://www.lynda.com/signin/user'
  22     _ACCOUNT_CREDENTIALS_HINT = 'Use --username and --password options to provide lynda.com account credentials.'
  23     _NETRC_MACHINE = 'lynda'
  24
  25     def _real_initialize(self):
  26         self._login()
  27
  28     @staticmethod
  29     def _check_error(json_string, key_or_keys):
  30         keys = [key_or_keys] if isinstance(key_or_keys, compat_str) else key_or_keys
  31         for key in keys:
  32             error = json_string.get(key)
  33             if error:
  34                 raise ExtractorError('Unable to login: %s' % error, expected=True)
  35
  36     def _login_step(self, form_html, fallback_action_url, extra_form_data, note, referrer_url):
  37         action_url = self._search_regex(
  38             r'<form[^>]+action=(["\'])(?P<url>.+?)\1', form_html,
  39             'post url', default=fallback_action_url, group='url')
  40
  41         if not action_url.startswith('http'):
  42             action_url = compat_urlparse.urljoin(self._SIGNIN_URL, action_url)
  43
  44         form_data = self._hidden_inputs(form_html)
  45         form_data.update(extra_form_data)
  46
  47         try:
  48             response = self._download_json(
  49                 action_url, None, note,
  50                 data=urlencode_postdata(form_data),
  51                 headers={
  52                     'Referer': referrer_url,
  53                     'X-Requested-With': 'XMLHttpRequest',
  54                 })
  55         except ExtractorError as e:
  56             if isinstance(e.cause, compat_HTTPError) and e.cause.code == 500:
  57                 response = self._parse_json(e.cause.read().decode('utf-8'), None)
  58                 self._check_error(response, ('email', 'password'))
  59             raise
  60
  61         self._check_error(response, 'ErrorMessage')
  62
  63         return response, action_url
  64
  65     def _login(self):
  66         username, password = self._get_login_info()
  67         if username is None:
  68             return
  69
  70         # Step 1: download signin page
  71         signin_page = self._download_webpage(
  72             self._SIGNIN_URL, None, 'Downloading signin page')
  73
  74         # Already logged in
  75         if any(re.search(p, signin_page) for p in (
  76                 'isLoggedIn\s*:\s*true', r'logout\.aspx', r'>Log out<')):
  77             return
  78
  79         # Step 2: submit email
  80         signin_form = self._search_regex(
  81             r'(?s)(<form[^>]+data-form-name=["\']signin["\'][^>]*>.+?</form>)',
  82             signin_page, 'signin form')
  83         signin_page, signin_url = self._login_step(
  84             signin_form, self._PASSWORD_URL, {'email': username},
  85             'Submitting email', self._SIGNIN_URL)
  86
  87         # Step 3: submit password
  88         password_form = signin_page['body']
  89         self._login_step(
  90             password_form, self._USER_URL, {'email': username, 'password': password},
  91             'Submitting password', signin_url)
  92
  93
  94 class LyndaIE(LyndaBaseIE):
  95     IE_NAME = 'lynda'
  96     IE_DESC = 'lynda.com videos'
  97     _VALID_URL = r'https?://www\.lynda\.com/(?:[^/]+/[^/]+/\d+|player/embed)/(?P<id>\d+)'
  98
  99     _TIMECODE_REGEX = r'\[(?P<timecode>\d+:\d+:\d+[\.,]\d+)\]'
 100
 101     _TESTS = [{
 102         'url': 'http://www.lynda.com/Bootstrap-tutorials/Using-exercise-files/110885/114408-4.html',
 103         # md5 is unstable
 104         'info_dict': {
 105             'id': '114408',
 106             'ext': 'mp4',
 107             'title': 'Using the exercise files',
 108             'duration': 68
 109         }
 110     }, {
 111         'url': 'https://www.lynda.com/player/embed/133770?tr=foo=1;bar=g;fizz=rt&fs=0',
 112         'only_matching': True,
 113     }]
 114
 115     def _real_extract(self, url):
 116         video_id = self._match_id(url)
 117
 118         video = self._download_json(
 119             'http://www.lynda.com/ajax/player?videoId=%s&type=video' % video_id,
 120             video_id, 'Downloading video JSON')
 121
 122         if 'Status' in video:
 123             raise ExtractorError(
 124                 'lynda returned error: %s' % video['Message'], expected=True)
 125
 126         if video.get('HasAccess') is False:
 127             self.raise_login_required('Video %s is only available for members' % video_id)
 128
 129         video_id = compat_str(video.get('ID') or video_id)
 130         duration = int_or_none(video.get('DurationInSeconds'))
 131         title = video['Title']
 132
 133         formats = []
 134
 135         fmts = video.get('Formats')
 136         if fmts:
 137             formats.extend([{
 138                 'url': f['Url'],
 139                 'ext': f.get('Extension'),
 140                 'width': int_or_none(f.get('Width')),
 141                 'height': int_or_none(f.get('Height')),
 142                 'filesize': int_or_none(f.get('FileSize')),
 143                 'format_id': compat_str(f.get('Resolution')) if f.get('Resolution') else None,
 144             } for f in fmts if f.get('Url')])
 145
 146         prioritized_streams = video.get('PrioritizedStreams')
 147         if prioritized_streams:
 148             for prioritized_stream_id, prioritized_stream in prioritized_streams.items():
 149                 formats.extend([{
 150                     'url': video_url,
 151                     'width': int_or_none(format_id),
 152                     'format_id': '%s-%s' % (prioritized_stream_id, format_id),
 153                 } for format_id, video_url in prioritized_stream.items()])
 154
 155         self._check_formats(formats, video_id)
 156         self._sort_formats(formats)
 157
 158         subtitles = self.extract_subtitles(video_id)
 159
 160         return {
 161             'id': video_id,
 162             'title': title,
 163             'duration': duration,
 164             'subtitles': subtitles,
 165             'formats': formats
 166         }
 167
 168     def _fix_subtitles(self, subs):
 169         srt = ''
 170         seq_counter = 0
 171         for pos in range(0, len(subs) - 1):
 172             seq_current = subs[pos]
 173             m_current = re.match(self._TIMECODE_REGEX, seq_current['Timecode'])
 174             if m_current is None:
 175                 continue
 176             seq_next = subs[pos + 1]
 177             m_next = re.match(self._TIMECODE_REGEX, seq_next['Timecode'])
 178             if m_next is None:
 179                 continue
 180             appear_time = m_current.group('timecode')
 181             disappear_time = m_next.group('timecode')
 182             text = seq_current['Caption'].strip()
 183             if text:
 184                 seq_counter += 1
 185                 srt += '%s\r\n%s --> %s\r\n%s\r\n\r\n' % (seq_counter, appear_time, disappear_time, text)
 186         if srt:
 187             return srt
 188
 189     def _get_subtitles(self, video_id):
 190         url = 'http://www.lynda.com/ajax/player?videoId=%s&type=transcript' % video_id
 191         subs = self._download_json(url, None, False)
 192         if subs:
 193             return {'en': [{'ext': 'srt', 'data': self._fix_subtitles(subs)}]}
 194         else:
 195             return {}
 196
 197
 198 class LyndaCourseIE(LyndaBaseIE):
 199     IE_NAME = 'lynda:course'
 200     IE_DESC = 'lynda.com online courses'
 201
 202     # Course link equals to welcome/introduction video link of same course
 203     # We will recognize it as course link
 204     _VALID_URL = r'https?://(?:www|m)\.lynda\.com/(?P<coursepath>[^/]+/[^/]+/(?P<courseid>\d+))-\d\.html'
 205
 206     def _real_extract(self, url):
 207         mobj = re.match(self._VALID_URL, url)
 208         course_path = mobj.group('coursepath')
 209         course_id = mobj.group('courseid')
 210
 211         course = self._download_json(
 212             'http://www.lynda.com/ajax/player?courseId=%s&type=course' % course_id,
 213             course_id, 'Downloading course JSON')
 214
 215         if course.get('Status') == 'NotFound':
 216             raise ExtractorError(
 217                 'Course %s does not exist' % course_id, expected=True)
 218
 219         unaccessible_videos = 0
 220         entries = []
 221
 222         # Might want to extract videos right here from video['Formats'] as it seems 'Formats' is not provided
 223         # by single video API anymore
 224
 225         for chapter in course['Chapters']:
 226             for video in chapter.get('Videos', []):
 227                 if video.get('HasAccess') is False:
 228                     unaccessible_videos += 1
 229                     continue
 230                 video_id = video.get('ID')
 231                 if video_id:
 232                     entries.append({
 233                         '_type': 'url_transparent',
 234                         'url': 'http://www.lynda.com/%s/%s-4.html' % (course_path, video_id),
 235                         'ie_key': LyndaIE.ie_key(),
 236                         'chapter': chapter.get('Title'),
 237                         'chapter_number': int_or_none(chapter.get('ChapterIndex')),
 238                         'chapter_id': compat_str(chapter.get('ID')),
 239                     })
 240
 241         if unaccessible_videos > 0:
 242             self._downloader.report_warning(
 243                 '%s videos are only available for members (or paid members) and will not be downloaded. '
 244                 % unaccessible_videos + self._ACCOUNT_CREDENTIALS_HINT)
 245
 246         course_title = course.get('Title')
 247         course_description = course.get('Description')
 248
 249         return self.playlist_result(entries, course_id, course_title, course_description)