_ Git - youtube-dl/blob - youtube_dl/extractor/pluralsight.py

   1 from __future__ import unicode_literals
   2
   3 import collections
   4 import json
   5 import os
   6 import random
   7
   8 from .common import InfoExtractor
   9 from ..compat import (
  10     compat_str,
  11     compat_urlparse,
  12 )
  13 from ..utils import (
  14     dict_get,
  15     ExtractorError,
  16     float_or_none,
  17     int_or_none,
  18     parse_duration,
  19     qualities,
  20     srt_subtitles_timecode,
  21     urlencode_postdata,
  22 )
  23
  24
  25 class PluralsightBaseIE(InfoExtractor):
  26     _API_BASE = 'https://app.pluralsight.com'
  27
  28
  29 class PluralsightIE(PluralsightBaseIE):
  30     IE_NAME = 'pluralsight'
  31     _VALID_URL = r'https?://(?:(?:www|app)\.)?pluralsight\.com/(?:training/)?player\?'
  32     _LOGIN_URL = 'https://app.pluralsight.com/id/'
  33
  34     _NETRC_MACHINE = 'pluralsight'
  35
  36     _TESTS = [{
  37         'url': 'http://www.pluralsight.com/training/player?author=mike-mckeown&name=hosting-sql-server-windows-azure-iaas-m7-mgmt&mode=live&clip=3&course=hosting-sql-server-windows-azure-iaas',
  38         'md5': '4d458cf5cf4c593788672419a8dd4cf8',
  39         'info_dict': {
  40             'id': 'hosting-sql-server-windows-azure-iaas-m7-mgmt-04',
  41             'ext': 'mp4',
  42             'title': 'Management of SQL Server - Demo Monitoring',
  43             'duration': 338,
  44         },
  45         'skip': 'Requires pluralsight account credentials',
  46     }, {
  47         'url': 'https://app.pluralsight.com/training/player?course=angularjs-get-started&author=scott-allen&name=angularjs-get-started-m1-introduction&clip=0&mode=live',
  48         'only_matching': True,
  49     }, {
  50         # available without pluralsight account
  51         'url': 'http://app.pluralsight.com/training/player?author=scott-allen&name=angularjs-get-started-m1-introduction&mode=live&clip=0&course=angularjs-get-started',
  52         'only_matching': True,
  53     }, {
  54         'url': 'https://app.pluralsight.com/player?course=ccna-intro-networking&author=ross-bagurdes&name=ccna-intro-networking-m06&clip=0',
  55         'only_matching': True,
  56     }]
  57
  58     def _real_initialize(self):
  59         self._login()
  60
  61     def _login(self):
  62         (username, password) = self._get_login_info()
  63         if username is None:
  64             return
  65
  66         login_page = self._download_webpage(
  67             self._LOGIN_URL, None, 'Downloading login page')
  68
  69         login_form = self._hidden_inputs(login_page)
  70
  71         login_form.update({
  72             'Username': username,
  73             'Password': password,
  74         })
  75
  76         post_url = self._search_regex(
  77             r'<form[^>]+action=(["\'])(?P<url>.+?)\1', login_page,
  78             'post url', default=self._LOGIN_URL, group='url')
  79
  80         if not post_url.startswith('http'):
  81             post_url = compat_urlparse.urljoin(self._LOGIN_URL, post_url)
  82
  83         response = self._download_webpage(
  84             post_url, None, 'Logging in as %s' % username,
  85             data=urlencode_postdata(login_form),
  86             headers={'Content-Type': 'application/x-www-form-urlencoded'})
  87
  88         error = self._search_regex(
  89             r'<span[^>]+class="field-validation-error"[^>]*>([^<]+)</span>',
  90             response, 'error message', default=None)
  91         if error:
  92             raise ExtractorError('Unable to login: %s' % error, expected=True)
  93
  94         if all(p not in response for p in ('__INITIAL_STATE__', '"currentUser"')):
  95             BLOCKED = 'Your account has been blocked due to suspicious activity'
  96             if BLOCKED in response:
  97                 raise ExtractorError(
  98                     'Unable to login: %s' % BLOCKED, expected=True)
  99             raise ExtractorError('Unable to log in')
 100
 101     def _get_subtitles(self, author, clip_id, lang, name, duration, video_id):
 102         captions_post = {
 103             'a': author,
 104             'cn': clip_id,
 105             'lc': lang,
 106             'm': name,
 107         }
 108         captions = self._download_json(
 109             '%s/player/retrieve-captions' % self._API_BASE, video_id,
 110             'Downloading captions JSON', 'Unable to download captions JSON',
 111             fatal=False, data=json.dumps(captions_post).encode('utf-8'),
 112             headers={'Content-Type': 'application/json;charset=utf-8'})
 113         if captions:
 114             return {
 115                 lang: [{
 116                     'ext': 'json',
 117                     'data': json.dumps(captions),
 118                 }, {
 119                     'ext': 'srt',
 120                     'data': self._convert_subtitles(duration, captions),
 121                 }]
 122             }
 123
 124     @staticmethod
 125     def _convert_subtitles(duration, subs):
 126         srt = ''
 127         TIME_OFFSET_KEYS = ('displayTimeOffset', 'DisplayTimeOffset')
 128         TEXT_KEYS = ('text', 'Text')
 129         for num, current in enumerate(subs):
 130             current = subs[num]
 131             start, text = (
 132                 float_or_none(dict_get(current, TIME_OFFSET_KEYS)),
 133                 dict_get(current, TEXT_KEYS))
 134             if start is None or text is None:
 135                 continue
 136             end = duration if num == len(subs) - 1 else float_or_none(
 137                 dict_get(subs[num + 1], TIME_OFFSET_KEYS))
 138             if end is None:
 139                 continue
 140             srt += os.linesep.join(
 141                 (
 142                     '%d' % num,
 143                     '%s --> %s' % (
 144                         srt_subtitles_timecode(start),
 145                         srt_subtitles_timecode(end)),
 146                     text,
 147                     os.linesep,
 148                 ))
 149         return srt
 150
 151     def _real_extract(self, url):
 152         qs = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
 153
 154         author = qs.get('author', [None])[0]
 155         name = qs.get('name', [None])[0]
 156         clip_id = qs.get('clip', [None])[0]
 157         course_name = qs.get('course', [None])[0]
 158
 159         if any(not f for f in (author, name, clip_id, course_name,)):
 160             raise ExtractorError('Invalid URL', expected=True)
 161
 162         display_id = '%s-%s' % (name, clip_id)
 163
 164         course = self._download_json(
 165             'https://app.pluralsight.com/player/user/api/v1/player/payload',
 166             display_id, data=urlencode_postdata({'courseId': course_name}),
 167             headers={'Referer': url})
 168
 169         collection = course['modules']
 170
 171         module, clip = None, None
 172
 173         for module_ in collection:
 174             if name in (module_.get('moduleName'), module_.get('name')):
 175                 module = module_
 176                 for clip_ in module_.get('clips', []):
 177                     clip_index = clip_.get('clipIndex')
 178                     if clip_index is None:
 179                         clip_index = clip_.get('index')
 180                     if clip_index is None:
 181                         continue
 182                     if compat_str(clip_index) == clip_id:
 183                         clip = clip_
 184                         break
 185
 186         if not clip:
 187             raise ExtractorError('Unable to resolve clip')
 188
 189         title = '%s - %s' % (module['title'], clip['title'])
 190
 191         QUALITIES = {
 192             'low': {'width': 640, 'height': 480},
 193             'medium': {'width': 848, 'height': 640},
 194             'high': {'width': 1024, 'height': 768},
 195             'high-widescreen': {'width': 1280, 'height': 720},
 196         }
 197
 198         QUALITIES_PREFERENCE = ('low', 'medium', 'high', 'high-widescreen',)
 199         quality_key = qualities(QUALITIES_PREFERENCE)
 200
 201         AllowedQuality = collections.namedtuple('AllowedQuality', ['ext', 'qualities'])
 202
 203         ALLOWED_QUALITIES = (
 204             AllowedQuality('webm', ['high', ]),
 205             AllowedQuality('mp4', ['low', 'medium', 'high', ]),
 206         )
 207
 208         # Some courses also offer widescreen resolution for high quality (see
 209         # https://github.com/rg3/youtube-dl/issues/7766)
 210         widescreen = course.get('supportsWideScreenVideoFormats') is True
 211         best_quality = 'high-widescreen' if widescreen else 'high'
 212         if widescreen:
 213             for allowed_quality in ALLOWED_QUALITIES:
 214                 allowed_quality.qualities.append(best_quality)
 215
 216         # In order to minimize the number of calls to ViewClip API and reduce
 217         # the probability of being throttled or banned by Pluralsight we will request
 218         # only single format until formats listing was explicitly requested.
 219         if self._downloader.params.get('listformats', False):
 220             allowed_qualities = ALLOWED_QUALITIES
 221         else:
 222             def guess_allowed_qualities():
 223                 req_format = self._downloader.params.get('format') or 'best'
 224                 req_format_split = req_format.split('-', 1)
 225                 if len(req_format_split) > 1:
 226                     req_ext, req_quality = req_format_split
 227                     for allowed_quality in ALLOWED_QUALITIES:
 228                         if req_ext == allowed_quality.ext and req_quality in allowed_quality.qualities:
 229                             return (AllowedQuality(req_ext, (req_quality, )), )
 230                 req_ext = 'webm' if self._downloader.params.get('prefer_free_formats') else 'mp4'
 231                 return (AllowedQuality(req_ext, (best_quality, )), )
 232             allowed_qualities = guess_allowed_qualities()
 233
 234         formats = []
 235         for ext, qualities_ in allowed_qualities:
 236             for quality in qualities_:
 237                 f = QUALITIES[quality].copy()
 238                 clip_post = {
 239                     'author': author,
 240                     'includeCaptions': False,
 241                     'clipIndex': int(clip_id),
 242                     'courseName': course_name,
 243                     'locale': 'en',
 244                     'moduleName': name,
 245                     'mediaType': ext,
 246                     'quality': '%dx%d' % (f['width'], f['height']),
 247                 }
 248                 format_id = '%s-%s' % (ext, quality)
 249                 viewclip = self._download_json(
 250                     '%s/video/clips/viewclip' % self._API_BASE, display_id,
 251                     'Downloading %s viewclip JSON' % format_id, fatal=False,
 252                     data=json.dumps(clip_post).encode('utf-8'),
 253                     headers={'Content-Type': 'application/json;charset=utf-8'})
 254
 255                 # Pluralsight tracks multiple sequential calls to ViewClip API and start
 256                 # to return 429 HTTP errors after some time (see
 257                 # https://github.com/rg3/youtube-dl/pull/6989). Moreover it may even lead
 258                 # to account ban (see https://github.com/rg3/youtube-dl/issues/6842).
 259                 # To somewhat reduce the probability of these consequences
 260                 # we will sleep random amount of time before each call to ViewClip.
 261                 self._sleep(
 262                     random.randint(2, 5), display_id,
 263                     '%(video_id)s: Waiting for %(timeout)s seconds to avoid throttling')
 264
 265                 if not viewclip:
 266                     continue
 267
 268                 clip_urls = viewclip.get('urls')
 269                 if not isinstance(clip_urls, list):
 270                     continue
 271
 272                 for clip_url_data in clip_urls:
 273                     clip_url = clip_url_data.get('url')
 274                     if not clip_url:
 275                         continue
 276                     cdn = clip_url_data.get('cdn')
 277                     clip_f = f.copy()
 278                     clip_f.update({
 279                         'url': clip_url,
 280                         'ext': ext,
 281                         'format_id': '%s-%s' % (format_id, cdn) if cdn else format_id,
 282                         'quality': quality_key(quality),
 283                         'source_preference': int_or_none(clip_url_data.get('rank')),
 284                     })
 285                     formats.append(clip_f)
 286
 287         self._sort_formats(formats)
 288
 289         duration = int_or_none(
 290             clip.get('duration')) or parse_duration(clip.get('formattedDuration'))
 291
 292         # TODO: other languages?
 293         subtitles = self.extract_subtitles(
 294             author, clip_id, 'en', name, duration, display_id)
 295
 296         return {
 297             'id': clip.get('clipName') or clip['name'],
 298             'title': title,
 299             'duration': duration,
 300             'creator': author,
 301             'formats': formats,
 302             'subtitles': subtitles,
 303         }
 304
 305
 306 class PluralsightCourseIE(PluralsightBaseIE):
 307     IE_NAME = 'pluralsight:course'
 308     _VALID_URL = r'https?://(?:(?:www|app)\.)?pluralsight\.com/(?:library/)?courses/(?P<id>[^/]+)'
 309     _TESTS = [{
 310         # Free course from Pluralsight Starter Subscription for Microsoft TechNet
 311         # https://offers.pluralsight.com/technet?loc=zTS3z&prod=zOTprodz&tech=zOttechz&prog=zOTprogz&type=zSOz&media=zOTmediaz&country=zUSz
 312         'url': 'http://www.pluralsight.com/courses/hosting-sql-server-windows-azure-iaas',
 313         'info_dict': {
 314             'id': 'hosting-sql-server-windows-azure-iaas',
 315             'title': 'Hosting SQL Server in Microsoft Azure IaaS Fundamentals',
 316             'description': 'md5:61b37e60f21c4b2f91dc621a977d0986',
 317         },
 318         'playlist_count': 31,
 319     }, {
 320         # available without pluralsight account
 321         'url': 'https://www.pluralsight.com/courses/angularjs-get-started',
 322         'only_matching': True,
 323     }, {
 324         'url': 'https://app.pluralsight.com/library/courses/understanding-microsoft-azure-amazon-aws/table-of-contents',
 325         'only_matching': True,
 326     }]
 327
 328     def _real_extract(self, url):
 329         course_id = self._match_id(url)
 330
 331         # TODO: PSM cookie
 332
 333         course = self._download_json(
 334             '%s/data/course/%s' % (self._API_BASE, course_id),
 335             course_id, 'Downloading course JSON')
 336
 337         title = course['title']
 338         description = course.get('description') or course.get('shortDescription')
 339
 340         course_data = self._download_json(
 341             '%s/data/course/content/%s' % (self._API_BASE, course_id),
 342             course_id, 'Downloading course data JSON')
 343
 344         entries = []
 345         for num, module in enumerate(course_data, 1):
 346             for clip in module.get('clips', []):
 347                 player_parameters = clip.get('playerParameters')
 348                 if not player_parameters:
 349                     continue
 350                 entries.append({
 351                     '_type': 'url_transparent',
 352                     'url': '%s/training/player?%s' % (self._API_BASE, player_parameters),
 353                     'ie_key': PluralsightIE.ie_key(),
 354                     'chapter': module.get('title'),
 355                     'chapter_number': num,
 356                     'chapter_id': module.get('moduleRef'),
 357                 })
 358
 359         return self.playlist_result(entries, course_id, title, description)