yt_dlp/extractor/lynda.py

   1 import re
   2
   3 from .common import InfoExtractor
   4 from ..compat import (
   5     compat_str,
   6     compat_urlparse,
   7 )
   8 from ..utils import (
   9     ExtractorError,
  10     int_or_none,
  11     urlencode_postdata,
  12 )
  13
  14
  15 class LyndaBaseIE(InfoExtractor):
  16     _SIGNIN_URL = 'https://www.lynda.com/signin/lynda'
  17     _PASSWORD_URL = 'https://www.lynda.com/signin/password'
  18     _USER_URL = 'https://www.lynda.com/signin/user'
  19     _ACCOUNT_CREDENTIALS_HINT = 'Use --username and --password options to provide lynda.com account credentials.'
  20     _NETRC_MACHINE = 'lynda'
  21
  22     @staticmethod
  23     def _check_error(json_string, key_or_keys):
  24         keys = [key_or_keys] if isinstance(key_or_keys, compat_str) else key_or_keys
  25         for key in keys:
  26             error = json_string.get(key)
  27             if error:
  28                 raise ExtractorError('Unable to login: %s' % error, expected=True)
  29
  30     def _perform_login_step(self, form_html, fallback_action_url, extra_form_data, note, referrer_url):
  31         action_url = self._search_regex(
  32             r'<form[^>]+action=(["\'])(?P<url>.+?)\1', form_html,
  33             'post url', default=fallback_action_url, group='url')
  34
  35         if not action_url.startswith('http'):
  36             action_url = compat_urlparse.urljoin(self._SIGNIN_URL, action_url)
  37
  38         form_data = self._hidden_inputs(form_html)
  39         form_data.update(extra_form_data)
  40
  41         response = self._download_json(
  42             action_url, None, note,
  43             data=urlencode_postdata(form_data),
  44             headers={
  45                 'Referer': referrer_url,
  46                 'X-Requested-With': 'XMLHttpRequest',
  47             }, expected_status=(418, 500, ))
  48
  49         self._check_error(response, ('email', 'password', 'ErrorMessage'))
  50
  51         return response, action_url
  52
  53     def _perform_login(self, username, password):
  54         # Step 1: download signin page
  55         signin_page = self._download_webpage(
  56             self._SIGNIN_URL, None, 'Downloading signin page')
  57
  58         # Already logged in
  59         if any(re.search(p, signin_page) for p in (
  60                 r'isLoggedIn\s*:\s*true', r'logout\.aspx', r'>Log out<')):
  61             return
  62
  63         # Step 2: submit email
  64         signin_form = self._search_regex(
  65             r'(?s)(<form[^>]+data-form-name=["\']signin["\'][^>]*>.+?</form>)',
  66             signin_page, 'signin form')
  67         signin_page, signin_url = self._login_step(
  68             signin_form, self._PASSWORD_URL, {'email': username},
  69             'Submitting email', self._SIGNIN_URL)
  70
  71         # Step 3: submit password
  72         password_form = signin_page['body']
  73         self._login_step(
  74             password_form, self._USER_URL, {'email': username, 'password': password},
  75             'Submitting password', signin_url)
  76
  77
  78 class LyndaIE(LyndaBaseIE):
  79     IE_NAME = 'lynda'
  80     IE_DESC = 'lynda.com videos'
  81     _VALID_URL = r'''(?x)
  82                     https?://
  83                         (?:www\.)?(?:lynda\.com|educourse\.ga)/
  84                         (?:
  85                             (?:[^/]+/){2,3}(?P<course_id>\d+)|
  86                             player/embed
  87                         )/
  88                         (?P<id>\d+)
  89                     '''
  90
  91     _TIMECODE_REGEX = r'\[(?P<timecode>\d+:\d+:\d+[\.,]\d+)\]'
  92
  93     _TESTS = [{
  94         'url': 'https://www.lynda.com/Bootstrap-tutorials/Using-exercise-files/110885/114408-4.html',
  95         # md5 is unstable
  96         'info_dict': {
  97             'id': '114408',
  98             'ext': 'mp4',
  99             'title': 'Using the exercise files',
 100             'duration': 68
 101         }
 102     }, {
 103         'url': 'https://www.lynda.com/player/embed/133770?tr=foo=1;bar=g;fizz=rt&fs=0',
 104         'only_matching': True,
 105     }, {
 106         'url': 'https://educourse.ga/Bootstrap-tutorials/Using-exercise-files/110885/114408-4.html',
 107         'only_matching': True,
 108     }, {
 109         'url': 'https://www.lynda.com/de/Graphic-Design-tutorials/Willkommen-Grundlagen-guten-Gestaltung/393570/393572-4.html',
 110         'only_matching': True,
 111     }, {
 112         # Status="NotFound", Message="Transcript not found"
 113         'url': 'https://www.lynda.com/ASP-NET-tutorials/What-you-should-know/5034180/2811512-4.html',
 114         'only_matching': True,
 115     }]
 116
 117     def _raise_unavailable(self, video_id):
 118         self.raise_login_required(
 119             'Video %s is only available for members' % video_id)
 120
 121     def _real_extract(self, url):
 122         mobj = self._match_valid_url(url)
 123         video_id = mobj.group('id')
 124         course_id = mobj.group('course_id')
 125
 126         query = {
 127             'videoId': video_id,
 128             'type': 'video',
 129         }
 130
 131         video = self._download_json(
 132             'https://www.lynda.com/ajax/player', video_id,
 133             'Downloading video JSON', fatal=False, query=query)
 134
 135         # Fallback scenario
 136         if not video:
 137             query['courseId'] = course_id
 138
 139             play = self._download_json(
 140                 'https://www.lynda.com/ajax/course/%s/%s/play'
 141                 % (course_id, video_id), video_id, 'Downloading play JSON')
 142
 143             if not play:
 144                 self._raise_unavailable(video_id)
 145
 146             formats = []
 147             for formats_dict in play:
 148                 urls = formats_dict.get('urls')
 149                 if not isinstance(urls, dict):
 150                     continue
 151                 cdn = formats_dict.get('name')
 152                 for format_id, format_url in urls.items():
 153                     if not format_url:
 154                         continue
 155                     formats.append({
 156                         'url': format_url,
 157                         'format_id': '%s-%s' % (cdn, format_id) if cdn else format_id,
 158                         'height': int_or_none(format_id),
 159                     })
 160             self._sort_formats(formats)
 161
 162             conviva = self._download_json(
 163                 'https://www.lynda.com/ajax/player/conviva', video_id,
 164                 'Downloading conviva JSON', query=query)
 165
 166             return {
 167                 'id': video_id,
 168                 'title': conviva['VideoTitle'],
 169                 'description': conviva.get('VideoDescription'),
 170                 'release_year': int_or_none(conviva.get('ReleaseYear')),
 171                 'duration': int_or_none(conviva.get('Duration')),
 172                 'creator': conviva.get('Author'),
 173                 'formats': formats,
 174             }
 175
 176         if 'Status' in video:
 177             raise ExtractorError(
 178                 'lynda returned error: %s' % video['Message'], expected=True)
 179
 180         if video.get('HasAccess') is False:
 181             self._raise_unavailable(video_id)
 182
 183         video_id = compat_str(video.get('ID') or video_id)
 184         duration = int_or_none(video.get('DurationInSeconds'))
 185         title = video['Title']
 186
 187         formats = []
 188
 189         fmts = video.get('Formats')
 190         if fmts:
 191             formats.extend([{
 192                 'url': f['Url'],
 193                 'ext': f.get('Extension'),
 194                 'width': int_or_none(f.get('Width')),
 195                 'height': int_or_none(f.get('Height')),
 196                 'filesize': int_or_none(f.get('FileSize')),
 197                 'format_id': compat_str(f.get('Resolution')) if f.get('Resolution') else None,
 198             } for f in fmts if f.get('Url')])
 199
 200         prioritized_streams = video.get('PrioritizedStreams')
 201         if prioritized_streams:
 202             for prioritized_stream_id, prioritized_stream in prioritized_streams.items():
 203                 formats.extend([{
 204                     'url': video_url,
 205                     'height': int_or_none(format_id),
 206                     'format_id': '%s-%s' % (prioritized_stream_id, format_id),
 207                 } for format_id, video_url in prioritized_stream.items()])
 208
 209         self._check_formats(formats, video_id)
 210         self._sort_formats(formats)
 211
 212         subtitles = self.extract_subtitles(video_id)
 213
 214         return {
 215             'id': video_id,
 216             'title': title,
 217             'duration': duration,
 218             'subtitles': subtitles,
 219             'formats': formats
 220         }
 221
 222     def _fix_subtitles(self, subs):
 223         srt = ''
 224         seq_counter = 0
 225         for pos in range(0, len(subs) - 1):
 226             seq_current = subs[pos]
 227             m_current = re.match(self._TIMECODE_REGEX, seq_current['Timecode'])
 228             if m_current is None:
 229                 continue
 230             seq_next = subs[pos + 1]
 231             m_next = re.match(self._TIMECODE_REGEX, seq_next['Timecode'])
 232             if m_next is None:
 233                 continue
 234             appear_time = m_current.group('timecode')
 235             disappear_time = m_next.group('timecode')
 236             text = seq_current['Caption'].strip()
 237             if text:
 238                 seq_counter += 1
 239                 srt += '%s\r\n%s --> %s\r\n%s\r\n\r\n' % (seq_counter, appear_time, disappear_time, text)
 240         if srt:
 241             return srt
 242
 243     def _get_subtitles(self, video_id):
 244         url = 'https://www.lynda.com/ajax/player?videoId=%s&type=transcript' % video_id
 245         subs = self._download_webpage(
 246             url, video_id, 'Downloading subtitles JSON', fatal=False)
 247         if not subs or 'Status="NotFound"' in subs:
 248             return {}
 249         subs = self._parse_json(subs, video_id, fatal=False)
 250         if not subs:
 251             return {}
 252         fixed_subs = self._fix_subtitles(subs)
 253         if fixed_subs:
 254             return {'en': [{'ext': 'srt', 'data': fixed_subs}]}
 255         return {}
 256
 257
 258 class LyndaCourseIE(LyndaBaseIE):
 259     IE_NAME = 'lynda:course'
 260     IE_DESC = 'lynda.com online courses'
 261
 262     # Course link equals to welcome/introduction video link of same course
 263     # We will recognize it as course link
 264     _VALID_URL = r'https?://(?:www|m)\.(?:lynda\.com|educourse\.ga)/(?P<coursepath>(?:[^/]+/){2,3}(?P<courseid>\d+))-2\.html'
 265
 266     _TESTS = [{
 267         'url': 'https://www.lynda.com/Graphic-Design-tutorials/Grundlagen-guten-Gestaltung/393570-2.html',
 268         'only_matching': True,
 269     }, {
 270         'url': 'https://www.lynda.com/de/Graphic-Design-tutorials/Grundlagen-guten-Gestaltung/393570-2.html',
 271         'only_matching': True,
 272     }]
 273
 274     def _real_extract(self, url):
 275         mobj = self._match_valid_url(url)
 276         course_path = mobj.group('coursepath')
 277         course_id = mobj.group('courseid')
 278
 279         item_template = 'https://www.lynda.com/%s/%%s-4.html' % course_path
 280
 281         course = self._download_json(
 282             'https://www.lynda.com/ajax/player?courseId=%s&type=course' % course_id,
 283             course_id, 'Downloading course JSON', fatal=False)
 284
 285         if not course:
 286             webpage = self._download_webpage(url, course_id)
 287             entries = [
 288                 self.url_result(
 289                     item_template % video_id, ie=LyndaIE.ie_key(),
 290                     video_id=video_id)
 291                 for video_id in re.findall(
 292                     r'data-video-id=["\'](\d+)', webpage)]
 293             return self.playlist_result(
 294                 entries, course_id,
 295                 self._og_search_title(webpage, fatal=False),
 296                 self._og_search_description(webpage))
 297
 298         if course.get('Status') == 'NotFound':
 299             raise ExtractorError(
 300                 'Course %s does not exist' % course_id, expected=True)
 301
 302         unaccessible_videos = 0
 303         entries = []
 304
 305         # Might want to extract videos right here from video['Formats'] as it seems 'Formats' is not provided
 306         # by single video API anymore
 307
 308         for chapter in course['Chapters']:
 309             for video in chapter.get('Videos', []):
 310                 if video.get('HasAccess') is False:
 311                     unaccessible_videos += 1
 312                     continue
 313                 video_id = video.get('ID')
 314                 if video_id:
 315                     entries.append({
 316                         '_type': 'url_transparent',
 317                         'url': item_template % video_id,
 318                         'ie_key': LyndaIE.ie_key(),
 319                         'chapter': chapter.get('Title'),
 320                         'chapter_number': int_or_none(chapter.get('ChapterIndex')),
 321                         'chapter_id': compat_str(chapter.get('ID')),
 322                     })
 323
 324         if unaccessible_videos > 0:
 325             self.report_warning(
 326                 '%s videos are only available for members (or paid members) and will not be downloaded. '
 327                 % unaccessible_videos + self._ACCOUNT_CREDENTIALS_HINT)
 328
 329         course_title = course.get('Title')
 330         course_description = course.get('Description')
 331
 332         return self.playlist_result(entries, course_id, course_title, course_description)