yt_dlp/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_str,
   8     compat_urlparse,
   9 )
  10 from ..utils import (
  11     ExtractorError,
  12     int_or_none,
  13     urlencode_postdata,
  14 )
  15
  16
  17 class LyndaBaseIE(InfoExtractor):
  18     _SIGNIN_URL = 'https://www.lynda.com/signin/lynda'
  19     _PASSWORD_URL = 'https://www.lynda.com/signin/password'
  20     _USER_URL = 'https://www.lynda.com/signin/user'
  21     _ACCOUNT_CREDENTIALS_HINT = 'Use --username and --password options to provide lynda.com account credentials.'
  22     _NETRC_MACHINE = 'lynda'
  23
  24     @staticmethod
  25     def _check_error(json_string, key_or_keys):
  26         keys = [key_or_keys] if isinstance(key_or_keys, compat_str) else key_or_keys
  27         for key in keys:
  28             error = json_string.get(key)
  29             if error:
  30                 raise ExtractorError('Unable to login: %s' % error, expected=True)
  31
  32     def _perform_login_step(self, form_html, fallback_action_url, extra_form_data, note, referrer_url):
  33         action_url = self._search_regex(
  34             r'<form[^>]+action=(["\'])(?P<url>.+?)\1', form_html,
  35             'post url', default=fallback_action_url, group='url')
  36
  37         if not action_url.startswith('http'):
  38             action_url = compat_urlparse.urljoin(self._SIGNIN_URL, action_url)
  39
  40         form_data = self._hidden_inputs(form_html)
  41         form_data.update(extra_form_data)
  42
  43         response = self._download_json(
  44             action_url, None, note,
  45             data=urlencode_postdata(form_data),
  46             headers={
  47                 'Referer': referrer_url,
  48                 'X-Requested-With': 'XMLHttpRequest',
  49             }, expected_status=(418, 500, ))
  50
  51         self._check_error(response, ('email', 'password', 'ErrorMessage'))
  52
  53         return response, action_url
  54
  55     def _perform_login(self, username, password):
  56         # Step 1: download signin page
  57         signin_page = self._download_webpage(
  58             self._SIGNIN_URL, None, 'Downloading signin page')
  59
  60         # Already logged in
  61         if any(re.search(p, signin_page) for p in (
  62                 r'isLoggedIn\s*:\s*true', r'logout\.aspx', r'>Log out<')):
  63             return
  64
  65         # Step 2: submit email
  66         signin_form = self._search_regex(
  67             r'(?s)(<form[^>]+data-form-name=["\']signin["\'][^>]*>.+?</form>)',
  68             signin_page, 'signin form')
  69         signin_page, signin_url = self._login_step(
  70             signin_form, self._PASSWORD_URL, {'email': username},
  71             'Submitting email', self._SIGNIN_URL)
  72
  73         # Step 3: submit password
  74         password_form = signin_page['body']
  75         self._login_step(
  76             password_form, self._USER_URL, {'email': username, 'password': password},
  77             'Submitting password', signin_url)
  78
  79
  80 class LyndaIE(LyndaBaseIE):
  81     IE_NAME = 'lynda'
  82     IE_DESC = 'lynda.com videos'
  83     _VALID_URL = r'''(?x)
  84                     https?://
  85                         (?:www\.)?(?:lynda\.com|educourse\.ga)/
  86                         (?:
  87                             (?:[^/]+/){2,3}(?P<course_id>\d+)|
  88                             player/embed
  89                         )/
  90                         (?P<id>\d+)
  91                     '''
  92
  93     _TIMECODE_REGEX = r'\[(?P<timecode>\d+:\d+:\d+[\.,]\d+)\]'
  94
  95     _TESTS = [{
  96         'url': 'https://www.lynda.com/Bootstrap-tutorials/Using-exercise-files/110885/114408-4.html',
  97         # md5 is unstable
  98         'info_dict': {
  99             'id': '114408',
 100             'ext': 'mp4',
 101             'title': 'Using the exercise files',
 102             'duration': 68
 103         }
 104     }, {
 105         'url': 'https://www.lynda.com/player/embed/133770?tr=foo=1;bar=g;fizz=rt&fs=0',
 106         'only_matching': True,
 107     }, {
 108         'url': 'https://educourse.ga/Bootstrap-tutorials/Using-exercise-files/110885/114408-4.html',
 109         'only_matching': True,
 110     }, {
 111         'url': 'https://www.lynda.com/de/Graphic-Design-tutorials/Willkommen-Grundlagen-guten-Gestaltung/393570/393572-4.html',
 112         'only_matching': True,
 113     }, {
 114         # Status="NotFound", Message="Transcript not found"
 115         'url': 'https://www.lynda.com/ASP-NET-tutorials/What-you-should-know/5034180/2811512-4.html',
 116         'only_matching': True,
 117     }]
 118
 119     def _raise_unavailable(self, video_id):
 120         self.raise_login_required(
 121             'Video %s is only available for members' % video_id)
 122
 123     def _real_extract(self, url):
 124         mobj = self._match_valid_url(url)
 125         video_id = mobj.group('id')
 126         course_id = mobj.group('course_id')
 127
 128         query = {
 129             'videoId': video_id,
 130             'type': 'video',
 131         }
 132
 133         video = self._download_json(
 134             'https://www.lynda.com/ajax/player', video_id,
 135             'Downloading video JSON', fatal=False, query=query)
 136
 137         # Fallback scenario
 138         if not video:
 139             query['courseId'] = course_id
 140
 141             play = self._download_json(
 142                 'https://www.lynda.com/ajax/course/%s/%s/play'
 143                 % (course_id, video_id), video_id, 'Downloading play JSON')
 144
 145             if not play:
 146                 self._raise_unavailable(video_id)
 147
 148             formats = []
 149             for formats_dict in play:
 150                 urls = formats_dict.get('urls')
 151                 if not isinstance(urls, dict):
 152                     continue
 153                 cdn = formats_dict.get('name')
 154                 for format_id, format_url in urls.items():
 155                     if not format_url:
 156                         continue
 157                     formats.append({
 158                         'url': format_url,
 159                         'format_id': '%s-%s' % (cdn, format_id) if cdn else format_id,
 160                         'height': int_or_none(format_id),
 161                     })
 162             self._sort_formats(formats)
 163
 164             conviva = self._download_json(
 165                 'https://www.lynda.com/ajax/player/conviva', video_id,
 166                 'Downloading conviva JSON', query=query)
 167
 168             return {
 169                 'id': video_id,
 170                 'title': conviva['VideoTitle'],
 171                 'description': conviva.get('VideoDescription'),
 172                 'release_year': int_or_none(conviva.get('ReleaseYear')),
 173                 'duration': int_or_none(conviva.get('Duration')),
 174                 'creator': conviva.get('Author'),
 175                 'formats': formats,
 176             }
 177
 178         if 'Status' in video:
 179             raise ExtractorError(
 180                 'lynda returned error: %s' % video['Message'], expected=True)
 181
 182         if video.get('HasAccess') is False:
 183             self._raise_unavailable(video_id)
 184
 185         video_id = compat_str(video.get('ID') or video_id)
 186         duration = int_or_none(video.get('DurationInSeconds'))
 187         title = video['Title']
 188
 189         formats = []
 190
 191         fmts = video.get('Formats')
 192         if fmts:
 193             formats.extend([{
 194                 'url': f['Url'],
 195                 'ext': f.get('Extension'),
 196                 'width': int_or_none(f.get('Width')),
 197                 'height': int_or_none(f.get('Height')),
 198                 'filesize': int_or_none(f.get('FileSize')),
 199                 'format_id': compat_str(f.get('Resolution')) if f.get('Resolution') else None,
 200             } for f in fmts if f.get('Url')])
 201
 202         prioritized_streams = video.get('PrioritizedStreams')
 203         if prioritized_streams:
 204             for prioritized_stream_id, prioritized_stream in prioritized_streams.items():
 205                 formats.extend([{
 206                     'url': video_url,
 207                     'height': int_or_none(format_id),
 208                     'format_id': '%s-%s' % (prioritized_stream_id, format_id),
 209                 } for format_id, video_url in prioritized_stream.items()])
 210
 211         self._check_formats(formats, video_id)
 212         self._sort_formats(formats)
 213
 214         subtitles = self.extract_subtitles(video_id)
 215
 216         return {
 217             'id': video_id,
 218             'title': title,
 219             'duration': duration,
 220             'subtitles': subtitles,
 221             'formats': formats
 222         }
 223
 224     def _fix_subtitles(self, subs):
 225         srt = ''
 226         seq_counter = 0
 227         for pos in range(0, len(subs) - 1):
 228             seq_current = subs[pos]
 229             m_current = re.match(self._TIMECODE_REGEX, seq_current['Timecode'])
 230             if m_current is None:
 231                 continue
 232             seq_next = subs[pos + 1]
 233             m_next = re.match(self._TIMECODE_REGEX, seq_next['Timecode'])
 234             if m_next is None:
 235                 continue
 236             appear_time = m_current.group('timecode')
 237             disappear_time = m_next.group('timecode')
 238             text = seq_current['Caption'].strip()
 239             if text:
 240                 seq_counter += 1
 241                 srt += '%s\r\n%s --> %s\r\n%s\r\n\r\n' % (seq_counter, appear_time, disappear_time, text)
 242         if srt:
 243             return srt
 244
 245     def _get_subtitles(self, video_id):
 246         url = 'https://www.lynda.com/ajax/player?videoId=%s&type=transcript' % video_id
 247         subs = self._download_webpage(
 248             url, video_id, 'Downloading subtitles JSON', fatal=False)
 249         if not subs or 'Status="NotFound"' in subs:
 250             return {}
 251         subs = self._parse_json(subs, video_id, fatal=False)
 252         if not subs:
 253             return {}
 254         fixed_subs = self._fix_subtitles(subs)
 255         if fixed_subs:
 256             return {'en': [{'ext': 'srt', 'data': fixed_subs}]}
 257         return {}
 258
 259
 260 class LyndaCourseIE(LyndaBaseIE):
 261     IE_NAME = 'lynda:course'
 262     IE_DESC = 'lynda.com online courses'
 263
 264     # Course link equals to welcome/introduction video link of same course
 265     # We will recognize it as course link
 266     _VALID_URL = r'https?://(?:www|m)\.(?:lynda\.com|educourse\.ga)/(?P<coursepath>(?:[^/]+/){2,3}(?P<courseid>\d+))-2\.html'
 267
 268     _TESTS = [{
 269         'url': 'https://www.lynda.com/Graphic-Design-tutorials/Grundlagen-guten-Gestaltung/393570-2.html',
 270         'only_matching': True,
 271     }, {
 272         'url': 'https://www.lynda.com/de/Graphic-Design-tutorials/Grundlagen-guten-Gestaltung/393570-2.html',
 273         'only_matching': True,
 274     }]
 275
 276     def _real_extract(self, url):
 277         mobj = self._match_valid_url(url)
 278         course_path = mobj.group('coursepath')
 279         course_id = mobj.group('courseid')
 280
 281         item_template = 'https://www.lynda.com/%s/%%s-4.html' % course_path
 282
 283         course = self._download_json(
 284             'https://www.lynda.com/ajax/player?courseId=%s&type=course' % course_id,
 285             course_id, 'Downloading course JSON', fatal=False)
 286
 287         if not course:
 288             webpage = self._download_webpage(url, course_id)
 289             entries = [
 290                 self.url_result(
 291                     item_template % video_id, ie=LyndaIE.ie_key(),
 292                     video_id=video_id)
 293                 for video_id in re.findall(
 294                     r'data-video-id=["\'](\d+)', webpage)]
 295             return self.playlist_result(
 296                 entries, course_id,
 297                 self._og_search_title(webpage, fatal=False),
 298                 self._og_search_description(webpage))
 299
 300         if course.get('Status') == 'NotFound':
 301             raise ExtractorError(
 302                 'Course %s does not exist' % course_id, expected=True)
 303
 304         unaccessible_videos = 0
 305         entries = []
 306
 307         # Might want to extract videos right here from video['Formats'] as it seems 'Formats' is not provided
 308         # by single video API anymore
 309
 310         for chapter in course['Chapters']:
 311             for video in chapter.get('Videos', []):
 312                 if video.get('HasAccess') is False:
 313                     unaccessible_videos += 1
 314                     continue
 315                 video_id = video.get('ID')
 316                 if video_id:
 317                     entries.append({
 318                         '_type': 'url_transparent',
 319                         'url': item_template % video_id,
 320                         'ie_key': LyndaIE.ie_key(),
 321                         'chapter': chapter.get('Title'),
 322                         'chapter_number': int_or_none(chapter.get('ChapterIndex')),
 323                         'chapter_id': compat_str(chapter.get('ID')),
 324                     })
 325
 326         if unaccessible_videos > 0:
 327             self.report_warning(
 328                 '%s videos are only available for members (or paid members) and will not be downloaded. '
 329                 % unaccessible_videos + self._ACCOUNT_CREDENTIALS_HINT)
 330
 331         course_title = course.get('Title')
 332         course_description = course.get('Description')
 333
 334         return self.playlist_result(entries, course_id, course_title, course_description)