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                 r'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|educourse\.ga)/(?:[^/]+/[^/]+/(?P<course_id>\d+)|player/embed)/(?P<id>\d+)'
  98
  99     _TIMECODE_REGEX = r'\[(?P<timecode>\d+:\d+:\d+[\.,]\d+)\]'
 100
 101     _TESTS = [{
 102         'url': 'https://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         'url': 'https://educourse.ga/Bootstrap-tutorials/Using-exercise-files/110885/114408-4.html',
 115         'only_matching': True,
 116     }]
 117
 118     def _raise_unavailable(self, video_id):
 119         self.raise_login_required(
 120             'Video %s is only available for members' % video_id)
 121
 122     def _real_extract(self, url):
 123         mobj = re.match(self._VALID_URL, url)
 124         video_id = mobj.group('id')
 125         course_id = mobj.group('course_id')
 126
 127         query = {
 128             'videoId': video_id,
 129             'type': 'video',
 130         }
 131
 132         video = self._download_json(
 133             'https://www.lynda.com/ajax/player', video_id,
 134             'Downloading video JSON', fatal=False, query=query)
 135
 136         # Fallback scenario
 137         if not video:
 138             query['courseId'] = course_id
 139
 140             play = self._download_json(
 141                 'https://www.lynda.com/ajax/course/%s/%s/play'
 142                 % (course_id, video_id), video_id, 'Downloading play JSON')
 143
 144             if not play:
 145                 self._raise_unavailable(video_id)
 146
 147             formats = []
 148             for formats_dict in play:
 149                 urls = formats_dict.get('urls')
 150                 if not isinstance(urls, dict):
 151                     continue
 152                 cdn = formats_dict.get('name')
 153                 for format_id, format_url in urls.items():
 154                     if not format_url:
 155                         continue
 156                     formats.append({
 157                         'url': format_url,
 158                         'format_id': '%s-%s' % (cdn, format_id) if cdn else format_id,
 159                         'height': int_or_none(format_id),
 160                     })
 161             self._sort_formats(formats)
 162
 163             conviva = self._download_json(
 164                 'https://www.lynda.com/ajax/player/conviva', video_id,
 165                 'Downloading conviva JSON', query=query)
 166
 167             return {
 168                 'id': video_id,
 169                 'title': conviva['VideoTitle'],
 170                 'description': conviva.get('VideoDescription'),
 171                 'release_year': int_or_none(conviva.get('ReleaseYear')),
 172                 'duration': int_or_none(conviva.get('Duration')),
 173                 'creator': conviva.get('Author'),
 174                 'formats': formats,
 175             }
 176
 177         if 'Status' in video:
 178             raise ExtractorError(
 179                 'lynda returned error: %s' % video['Message'], expected=True)
 180
 181         if video.get('HasAccess') is False:
 182             self._raise_unavailable(video_id)
 183
 184         video_id = compat_str(video.get('ID') or video_id)
 185         duration = int_or_none(video.get('DurationInSeconds'))
 186         title = video['Title']
 187
 188         formats = []
 189
 190         fmts = video.get('Formats')
 191         if fmts:
 192             formats.extend([{
 193                 'url': f['Url'],
 194                 'ext': f.get('Extension'),
 195                 'width': int_or_none(f.get('Width')),
 196                 'height': int_or_none(f.get('Height')),
 197                 'filesize': int_or_none(f.get('FileSize')),
 198                 'format_id': compat_str(f.get('Resolution')) if f.get('Resolution') else None,
 199             } for f in fmts if f.get('Url')])
 200
 201         prioritized_streams = video.get('PrioritizedStreams')
 202         if prioritized_streams:
 203             for prioritized_stream_id, prioritized_stream in prioritized_streams.items():
 204                 formats.extend([{
 205                     'url': video_url,
 206                     'height': int_or_none(format_id),
 207                     'format_id': '%s-%s' % (prioritized_stream_id, format_id),
 208                 } for format_id, video_url in prioritized_stream.items()])
 209
 210         self._check_formats(formats, video_id)
 211         self._sort_formats(formats)
 212
 213         subtitles = self.extract_subtitles(video_id)
 214
 215         return {
 216             'id': video_id,
 217             'title': title,
 218             'duration': duration,
 219             'subtitles': subtitles,
 220             'formats': formats
 221         }
 222
 223     def _fix_subtitles(self, subs):
 224         srt = ''
 225         seq_counter = 0
 226         for pos in range(0, len(subs) - 1):
 227             seq_current = subs[pos]
 228             m_current = re.match(self._TIMECODE_REGEX, seq_current['Timecode'])
 229             if m_current is None:
 230                 continue
 231             seq_next = subs[pos + 1]
 232             m_next = re.match(self._TIMECODE_REGEX, seq_next['Timecode'])
 233             if m_next is None:
 234                 continue
 235             appear_time = m_current.group('timecode')
 236             disappear_time = m_next.group('timecode')
 237             text = seq_current['Caption'].strip()
 238             if text:
 239                 seq_counter += 1
 240                 srt += '%s\r\n%s --> %s\r\n%s\r\n\r\n' % (seq_counter, appear_time, disappear_time, text)
 241         if srt:
 242             return srt
 243
 244     def _get_subtitles(self, video_id):
 245         url = 'https://www.lynda.com/ajax/player?videoId=%s&type=transcript' % video_id
 246         subs = self._download_json(url, None, False)
 247         if subs:
 248             return {'en': [{'ext': 'srt', 'data': self._fix_subtitles(subs)}]}
 249         else:
 250             return {}
 251
 252
 253 class LyndaCourseIE(LyndaBaseIE):
 254     IE_NAME = 'lynda:course'
 255     IE_DESC = 'lynda.com online courses'
 256
 257     # Course link equals to welcome/introduction video link of same course
 258     # We will recognize it as course link
 259     _VALID_URL = r'https?://(?:www|m)\.(?:lynda\.com|educourse\.ga)/(?P<coursepath>[^/]+/[^/]+/(?P<courseid>\d+))-\d\.html'
 260
 261     def _real_extract(self, url):
 262         mobj = re.match(self._VALID_URL, url)
 263         course_path = mobj.group('coursepath')
 264         course_id = mobj.group('courseid')
 265
 266         item_template = 'https://www.lynda.com/%s/%%s-4.html' % course_path
 267
 268         course = self._download_json(
 269             'https://www.lynda.com/ajax/player?courseId=%s&type=course' % course_id,
 270             course_id, 'Downloading course JSON', fatal=False)
 271
 272         if not course:
 273             webpage = self._download_webpage(url, course_id)
 274             entries = [
 275                 self.url_result(
 276                     item_template % video_id, ie=LyndaIE.ie_key(),
 277                     video_id=video_id)
 278                 for video_id in re.findall(
 279                     r'data-video-id=["\'](\d+)', webpage)]
 280             return self.playlist_result(
 281                 entries, course_id,
 282                 self._og_search_title(webpage, fatal=False),
 283                 self._og_search_description(webpage))
 284
 285         if course.get('Status') == 'NotFound':
 286             raise ExtractorError(
 287                 'Course %s does not exist' % course_id, expected=True)
 288
 289         unaccessible_videos = 0
 290         entries = []
 291
 292         # Might want to extract videos right here from video['Formats'] as it seems 'Formats' is not provided
 293         # by single video API anymore
 294
 295         for chapter in course['Chapters']:
 296             for video in chapter.get('Videos', []):
 297                 if video.get('HasAccess') is False:
 298                     unaccessible_videos += 1
 299                     continue
 300                 video_id = video.get('ID')
 301                 if video_id:
 302                     entries.append({
 303                         '_type': 'url_transparent',
 304                         'url': item_template % video_id,
 305                         'ie_key': LyndaIE.ie_key(),
 306                         'chapter': chapter.get('Title'),
 307                         'chapter_number': int_or_none(chapter.get('ChapterIndex')),
 308                         'chapter_id': compat_str(chapter.get('ID')),
 309                     })
 310
 311         if unaccessible_videos > 0:
 312             self._downloader.report_warning(
 313                 '%s videos are only available for members (or paid members) and will not be downloaded. '
 314                 % unaccessible_videos + self._ACCOUNT_CREDENTIALS_HINT)
 315
 316         course_title = course.get('Title')
 317         course_description = course.get('Description')
 318
 319         return self.playlist_result(entries, course_id, course_title, course_description)