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_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     def _real_initialize(self):
  25         self._login()
  26
  27     @staticmethod
  28     def _check_error(json_string, key_or_keys):
  29         keys = [key_or_keys] if isinstance(key_or_keys, compat_str) else key_or_keys
  30         for key in keys:
  31             error = json_string.get(key)
  32             if error:
  33                 raise ExtractorError('Unable to login: %s' % error, expected=True)
  34
  35     def _login_step(self, form_html, fallback_action_url, extra_form_data, note, referrer_url):
  36         action_url = self._search_regex(
  37             r'<form[^>]+action=(["\'])(?P<url>.+?)\1', form_html,
  38             'post url', default=fallback_action_url, group='url')
  39
  40         if not action_url.startswith('http'):
  41             action_url = compat_urlparse.urljoin(self._SIGNIN_URL, action_url)
  42
  43         form_data = self._hidden_inputs(form_html)
  44         form_data.update(extra_form_data)
  45
  46         response = self._download_json(
  47             action_url, None, note,
  48             data=urlencode_postdata(form_data),
  49             headers={
  50                 'Referer': referrer_url,
  51                 'X-Requested-With': 'XMLHttpRequest',
  52             }, expected_status=(418, 500, ))
  53
  54         self._check_error(response, ('email', 'password', 'ErrorMessage'))
  55
  56         return response, action_url
  57
  58     def _login(self):
  59         username, password = self._get_login_info()
  60         if username is None:
  61             return
  62
  63         # Step 1: download signin page
  64         signin_page = self._download_webpage(
  65             self._SIGNIN_URL, None, 'Downloading signin page')
  66
  67         # Already logged in
  68         if any(re.search(p, signin_page) for p in (
  69                 r'isLoggedIn\s*:\s*true', r'logout\.aspx', r'>Log out<')):
  70             return
  71
  72         # Step 2: submit email
  73         signin_form = self._search_regex(
  74             r'(?s)(<form[^>]+data-form-name=["\']signin["\'][^>]*>.+?</form>)',
  75             signin_page, 'signin form')
  76         signin_page, signin_url = self._login_step(
  77             signin_form, self._PASSWORD_URL, {'email': username},
  78             'Submitting email', self._SIGNIN_URL)
  79
  80         # Step 3: submit password
  81         password_form = signin_page['body']
  82         self._login_step(
  83             password_form, self._USER_URL, {'email': username, 'password': password},
  84             'Submitting password', signin_url)
  85
  86
  87 class LyndaIE(LyndaBaseIE):
  88     IE_NAME = 'lynda'
  89     IE_DESC = 'lynda.com videos'
  90     _VALID_URL = r'''(?x)
  91                     https?://
  92                         (?:www\.)?(?:lynda\.com|educourse\.ga)/
  93                         (?:
  94                             (?:[^/]+/){2,3}(?P<course_id>\d+)|
  95                             player/embed
  96                         )/
  97                         (?P<id>\d+)
  98                     '''
  99
 100     _TIMECODE_REGEX = r'\[(?P<timecode>\d+:\d+:\d+[\.,]\d+)\]'
 101
 102     _TESTS = [{
 103         'url': 'https://www.lynda.com/Bootstrap-tutorials/Using-exercise-files/110885/114408-4.html',
 104         # md5 is unstable
 105         'info_dict': {
 106             'id': '114408',
 107             'ext': 'mp4',
 108             'title': 'Using the exercise files',
 109             'duration': 68
 110         }
 111     }, {
 112         'url': 'https://www.lynda.com/player/embed/133770?tr=foo=1;bar=g;fizz=rt&fs=0',
 113         'only_matching': True,
 114     }, {
 115         'url': 'https://educourse.ga/Bootstrap-tutorials/Using-exercise-files/110885/114408-4.html',
 116         'only_matching': True,
 117     }, {
 118         'url': 'https://www.lynda.com/de/Graphic-Design-tutorials/Willkommen-Grundlagen-guten-Gestaltung/393570/393572-4.html',
 119         'only_matching': True,
 120     }, {
 121         # Status="NotFound", Message="Transcript not found"
 122         'url': 'https://www.lynda.com/ASP-NET-tutorials/What-you-should-know/5034180/2811512-4.html',
 123         'only_matching': True,
 124     }]
 125
 126     def _raise_unavailable(self, video_id):
 127         self.raise_login_required(
 128             'Video %s is only available for members' % video_id)
 129
 130     def _real_extract(self, url):
 131         mobj = re.match(self._VALID_URL, url)
 132         video_id = mobj.group('id')
 133         course_id = mobj.group('course_id')
 134
 135         query = {
 136             'videoId': video_id,
 137             'type': 'video',
 138         }
 139
 140         video = self._download_json(
 141             'https://www.lynda.com/ajax/player', video_id,
 142             'Downloading video JSON', fatal=False, query=query)
 143
 144         # Fallback scenario
 145         if not video:
 146             query['courseId'] = course_id
 147
 148             play = self._download_json(
 149                 'https://www.lynda.com/ajax/course/%s/%s/play'
 150                 % (course_id, video_id), video_id, 'Downloading play JSON')
 151
 152             if not play:
 153                 self._raise_unavailable(video_id)
 154
 155             formats = []
 156             for formats_dict in play:
 157                 urls = formats_dict.get('urls')
 158                 if not isinstance(urls, dict):
 159                     continue
 160                 cdn = formats_dict.get('name')
 161                 for format_id, format_url in urls.items():
 162                     if not format_url:
 163                         continue
 164                     formats.append({
 165                         'url': format_url,
 166                         'format_id': '%s-%s' % (cdn, format_id) if cdn else format_id,
 167                         'height': int_or_none(format_id),
 168                     })
 169             self._sort_formats(formats)
 170
 171             conviva = self._download_json(
 172                 'https://www.lynda.com/ajax/player/conviva', video_id,
 173                 'Downloading conviva JSON', query=query)
 174
 175             return {
 176                 'id': video_id,
 177                 'title': conviva['VideoTitle'],
 178                 'description': conviva.get('VideoDescription'),
 179                 'release_year': int_or_none(conviva.get('ReleaseYear')),
 180                 'duration': int_or_none(conviva.get('Duration')),
 181                 'creator': conviva.get('Author'),
 182                 'formats': formats,
 183             }
 184
 185         if 'Status' in video:
 186             raise ExtractorError(
 187                 'lynda returned error: %s' % video['Message'], expected=True)
 188
 189         if video.get('HasAccess') is False:
 190             self._raise_unavailable(video_id)
 191
 192         video_id = compat_str(video.get('ID') or video_id)
 193         duration = int_or_none(video.get('DurationInSeconds'))
 194         title = video['Title']
 195
 196         formats = []
 197
 198         fmts = video.get('Formats')
 199         if fmts:
 200             formats.extend([{
 201                 'url': f['Url'],
 202                 'ext': f.get('Extension'),
 203                 'width': int_or_none(f.get('Width')),
 204                 'height': int_or_none(f.get('Height')),
 205                 'filesize': int_or_none(f.get('FileSize')),
 206                 'format_id': compat_str(f.get('Resolution')) if f.get('Resolution') else None,
 207             } for f in fmts if f.get('Url')])
 208
 209         prioritized_streams = video.get('PrioritizedStreams')
 210         if prioritized_streams:
 211             for prioritized_stream_id, prioritized_stream in prioritized_streams.items():
 212                 formats.extend([{
 213                     'url': video_url,
 214                     'height': int_or_none(format_id),
 215                     'format_id': '%s-%s' % (prioritized_stream_id, format_id),
 216                 } for format_id, video_url in prioritized_stream.items()])
 217
 218         self._check_formats(formats, video_id)
 219         self._sort_formats(formats)
 220
 221         subtitles = self.extract_subtitles(video_id)
 222
 223         return {
 224             'id': video_id,
 225             'title': title,
 226             'duration': duration,
 227             'subtitles': subtitles,
 228             'formats': formats
 229         }
 230
 231     def _fix_subtitles(self, subs):
 232         srt = ''
 233         seq_counter = 0
 234         for pos in range(0, len(subs) - 1):
 235             seq_current = subs[pos]
 236             m_current = re.match(self._TIMECODE_REGEX, seq_current['Timecode'])
 237             if m_current is None:
 238                 continue
 239             seq_next = subs[pos + 1]
 240             m_next = re.match(self._TIMECODE_REGEX, seq_next['Timecode'])
 241             if m_next is None:
 242                 continue
 243             appear_time = m_current.group('timecode')
 244             disappear_time = m_next.group('timecode')
 245             text = seq_current['Caption'].strip()
 246             if text:
 247                 seq_counter += 1
 248                 srt += '%s\r\n%s --> %s\r\n%s\r\n\r\n' % (seq_counter, appear_time, disappear_time, text)
 249         if srt:
 250             return srt
 251
 252     def _get_subtitles(self, video_id):
 253         url = 'https://www.lynda.com/ajax/player?videoId=%s&type=transcript' % video_id
 254         subs = self._download_webpage(
 255             url, video_id, 'Downloading subtitles JSON', fatal=False)
 256         if not subs or 'Status="NotFound"' in subs:
 257             return {}
 258         subs = self._parse_json(subs, video_id, fatal=False)
 259         if not subs:
 260             return {}
 261         fixed_subs = self._fix_subtitles(subs)
 262         if fixed_subs:
 263             return {'en': [{'ext': 'srt', 'data': fixed_subs}]}
 264         return {}
 265
 266
 267 class LyndaCourseIE(LyndaBaseIE):
 268     IE_NAME = 'lynda:course'
 269     IE_DESC = 'lynda.com online courses'
 270
 271     # Course link equals to welcome/introduction video link of same course
 272     # We will recognize it as course link
 273     _VALID_URL = r'https?://(?:www|m)\.(?:lynda\.com|educourse\.ga)/(?P<coursepath>(?:[^/]+/){2,3}(?P<courseid>\d+))-2\.html'
 274
 275     _TESTS = [{
 276         'url': 'https://www.lynda.com/Graphic-Design-tutorials/Grundlagen-guten-Gestaltung/393570-2.html',
 277         'only_matching': True,
 278     }, {
 279         'url': 'https://www.lynda.com/de/Graphic-Design-tutorials/Grundlagen-guten-Gestaltung/393570-2.html',
 280         'only_matching': True,
 281     }]
 282
 283     def _real_extract(self, url):
 284         mobj = re.match(self._VALID_URL, url)
 285         course_path = mobj.group('coursepath')
 286         course_id = mobj.group('courseid')
 287
 288         item_template = 'https://www.lynda.com/%s/%%s-4.html' % course_path
 289
 290         course = self._download_json(
 291             'https://www.lynda.com/ajax/player?courseId=%s&type=course' % course_id,
 292             course_id, 'Downloading course JSON', fatal=False)
 293
 294         if not course:
 295             webpage = self._download_webpage(url, course_id)
 296             entries = [
 297                 self.url_result(
 298                     item_template % video_id, ie=LyndaIE.ie_key(),
 299                     video_id=video_id)
 300                 for video_id in re.findall(
 301                     r'data-video-id=["\'](\d+)', webpage)]
 302             return self.playlist_result(
 303                 entries, course_id,
 304                 self._og_search_title(webpage, fatal=False),
 305                 self._og_search_description(webpage))
 306
 307         if course.get('Status') == 'NotFound':
 308             raise ExtractorError(
 309                 'Course %s does not exist' % course_id, expected=True)
 310
 311         unaccessible_videos = 0
 312         entries = []
 313
 314         # Might want to extract videos right here from video['Formats'] as it seems 'Formats' is not provided
 315         # by single video API anymore
 316
 317         for chapter in course['Chapters']:
 318             for video in chapter.get('Videos', []):
 319                 if video.get('HasAccess') is False:
 320                     unaccessible_videos += 1
 321                     continue
 322                 video_id = video.get('ID')
 323                 if video_id:
 324                     entries.append({
 325                         '_type': 'url_transparent',
 326                         'url': item_template % video_id,
 327                         'ie_key': LyndaIE.ie_key(),
 328                         'chapter': chapter.get('Title'),
 329                         'chapter_number': int_or_none(chapter.get('ChapterIndex')),
 330                         'chapter_id': compat_str(chapter.get('ID')),
 331                     })
 332
 333         if unaccessible_videos > 0:
 334             self._downloader.report_warning(
 335                 '%s videos are only available for members (or paid members) and will not be downloaded. '
 336                 % unaccessible_videos + self._ACCOUNT_CREDENTIALS_HINT)
 337
 338         course_title = course.get('Title')
 339         course_description = course.get('Description')
 340
 341         return self.playlist_result(entries, course_id, course_title, course_description)