Raphaël G. Git Repositories - youtubedl/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     update_url_query,
  22     urlencode_postdata,
  23 )
  24
  25
  26 class PluralsightBaseIE(InfoExtractor):
  27     _API_BASE = 'https://app.pluralsight.com'
  28
  29
  30 class PluralsightIE(PluralsightBaseIE):
  31     IE_NAME = 'pluralsight'
  32     _VALID_URL = r'https?://(?:(?:www|app)\.)?pluralsight\.com/(?:training/)?player\?'
  33     _LOGIN_URL = 'https://app.pluralsight.com/id/'
  34
  35     _NETRC_MACHINE = 'pluralsight'
  36
  37     _TESTS = [{
  38         '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',
  39         'md5': '4d458cf5cf4c593788672419a8dd4cf8',
  40         'info_dict': {
  41             'id': 'hosting-sql-server-windows-azure-iaas-m7-mgmt-04',
  42             'ext': 'mp4',
  43             'title': 'Management of SQL Server - Demo Monitoring',
  44             'duration': 338,
  45         },
  46         'skip': 'Requires pluralsight account credentials',
  47     }, {
  48         'url': 'https://app.pluralsight.com/training/player?course=angularjs-get-started&author=scott-allen&name=angularjs-get-started-m1-introduction&clip=0&mode=live',
  49         'only_matching': True,
  50     }, {
  51         # available without pluralsight account
  52         'url': 'http://app.pluralsight.com/training/player?author=scott-allen&name=angularjs-get-started-m1-introduction&mode=live&clip=0&course=angularjs-get-started',
  53         'only_matching': True,
  54     }, {
  55         'url': 'https://app.pluralsight.com/player?course=ccna-intro-networking&author=ross-bagurdes&name=ccna-intro-networking-m06&clip=0',
  56         'only_matching': True,
  57     }]
  58
  59     def _real_initialize(self):
  60         self._login()
  61
  62     def _login(self):
  63         (username, password) = self._get_login_info()
  64         if username is None:
  65             return
  66
  67         login_page = self._download_webpage(
  68             self._LOGIN_URL, None, 'Downloading login page')
  69
  70         login_form = self._hidden_inputs(login_page)
  71
  72         login_form.update({
  73             'Username': username,
  74             'Password': password,
  75         })
  76
  77         post_url = self._search_regex(
  78             r'<form[^>]+action=(["\'])(?P<url>.+?)\1', login_page,
  79             'post url', default=self._LOGIN_URL, group='url')
  80
  81         if not post_url.startswith('http'):
  82             post_url = compat_urlparse.urljoin(self._LOGIN_URL, post_url)
  83
  84         response = self._download_webpage(
  85             post_url, None, 'Logging in as %s' % username,
  86             data=urlencode_postdata(login_form),
  87             headers={'Content-Type': 'application/x-www-form-urlencoded'})
  88
  89         error = self._search_regex(
  90             r'<span[^>]+class="field-validation-error"[^>]*>([^<]+)</span>',
  91             response, 'error message', default=None)
  92         if error:
  93             raise ExtractorError('Unable to login: %s' % error, expected=True)
  94
  95         if all(p not in response for p in ('__INITIAL_STATE__', '"currentUser"')):
  96             BLOCKED = 'Your account has been blocked due to suspicious activity'
  97             if BLOCKED in response:
  98                 raise ExtractorError(
  99                     'Unable to login: %s' % BLOCKED, expected=True)
 100             raise ExtractorError('Unable to log in')
 101
 102     def _get_subtitles(self, author, clip_id, lang, name, duration, video_id):
 103         captions_post = {
 104             'a': author,
 105             'cn': clip_id,
 106             'lc': lang,
 107             'm': name,
 108         }
 109         captions = self._download_json(
 110             '%s/player/retrieve-captions' % self._API_BASE, video_id,
 111             'Downloading captions JSON', 'Unable to download captions JSON',
 112             fatal=False, data=json.dumps(captions_post).encode('utf-8'),
 113             headers={'Content-Type': 'application/json;charset=utf-8'})
 114         if captions:
 115             return {
 116                 lang: [{
 117                     'ext': 'json',
 118                     'data': json.dumps(captions),
 119                 }, {
 120                     'ext': 'srt',
 121                     'data': self._convert_subtitles(duration, captions),
 122                 }]
 123             }
 124
 125     @staticmethod
 126     def _convert_subtitles(duration, subs):
 127         srt = ''
 128         TIME_OFFSET_KEYS = ('displayTimeOffset', 'DisplayTimeOffset')
 129         TEXT_KEYS = ('text', 'Text')
 130         for num, current in enumerate(subs):
 131             current = subs[num]
 132             start, text = (
 133                 float_or_none(dict_get(current, TIME_OFFSET_KEYS)),
 134                 dict_get(current, TEXT_KEYS))
 135             if start is None or text is None:
 136                 continue
 137             end = duration if num == len(subs) - 1 else float_or_none(
 138                 dict_get(subs[num + 1], TIME_OFFSET_KEYS))
 139             if end is None:
 140                 continue
 141             srt += os.linesep.join(
 142                 (
 143                     '%d' % num,
 144                     '%s --> %s' % (
 145                         srt_subtitles_timecode(start),
 146                         srt_subtitles_timecode(end)),
 147                     text,
 148                     os.linesep,
 149                 ))
 150         return srt
 151
 152     def _real_extract(self, url):
 153         qs = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
 154
 155         author = qs.get('author', [None])[0]
 156         name = qs.get('name', [None])[0]
 157         clip_id = qs.get('clip', [None])[0]
 158         course_name = qs.get('course', [None])[0]
 159
 160         if any(not f for f in (author, name, clip_id, course_name,)):
 161             raise ExtractorError('Invalid URL', expected=True)
 162
 163         display_id = '%s-%s' % (name, clip_id)
 164
 165         course = self._download_json(
 166             'https://app.pluralsight.com/player/user/api/v1/player/payload',
 167             display_id, data=urlencode_postdata({'courseId': course_name}),
 168             headers={'Referer': url})
 169
 170         collection = course['modules']
 171
 172         module, clip = None, None
 173
 174         for module_ in collection:
 175             if name in (module_.get('moduleName'), module_.get('name')):
 176                 module = module_
 177                 for clip_ in module_.get('clips', []):
 178                     clip_index = clip_.get('clipIndex')
 179                     if clip_index is None:
 180                         clip_index = clip_.get('index')
 181                     if clip_index is None:
 182                         continue
 183                     if compat_str(clip_index) == clip_id:
 184                         clip = clip_
 185                         break
 186
 187         if not clip:
 188             raise ExtractorError('Unable to resolve clip')
 189
 190         title = '%s - %s' % (module['title'], clip['title'])
 191
 192         QUALITIES = {
 193             'low': {'width': 640, 'height': 480},
 194             'medium': {'width': 848, 'height': 640},
 195             'high': {'width': 1024, 'height': 768},
 196             'high-widescreen': {'width': 1280, 'height': 720},
 197         }
 198
 199         QUALITIES_PREFERENCE = ('low', 'medium', 'high', 'high-widescreen',)
 200         quality_key = qualities(QUALITIES_PREFERENCE)
 201
 202         AllowedQuality = collections.namedtuple('AllowedQuality', ['ext', 'qualities'])
 203
 204         ALLOWED_QUALITIES = (
 205             AllowedQuality('webm', ['high', ]),
 206             AllowedQuality('mp4', ['low', 'medium', 'high', ]),
 207         )
 208
 209         # Some courses also offer widescreen resolution for high quality (see
 210         # https://github.com/rg3/youtube-dl/issues/7766)
 211         widescreen = course.get('supportsWideScreenVideoFormats') is True
 212         best_quality = 'high-widescreen' if widescreen else 'high'
 213         if widescreen:
 214             for allowed_quality in ALLOWED_QUALITIES:
 215                 allowed_quality.qualities.append(best_quality)
 216
 217         # In order to minimize the number of calls to ViewClip API and reduce
 218         # the probability of being throttled or banned by Pluralsight we will request
 219         # only single format until formats listing was explicitly requested.
 220         if self._downloader.params.get('listformats', False):
 221             allowed_qualities = ALLOWED_QUALITIES
 222         else:
 223             def guess_allowed_qualities():
 224                 req_format = self._downloader.params.get('format') or 'best'
 225                 req_format_split = req_format.split('-', 1)
 226                 if len(req_format_split) > 1:
 227                     req_ext, req_quality = req_format_split
 228                     for allowed_quality in ALLOWED_QUALITIES:
 229                         if req_ext == allowed_quality.ext and req_quality in allowed_quality.qualities:
 230                             return (AllowedQuality(req_ext, (req_quality, )), )
 231                 req_ext = 'webm' if self._downloader.params.get('prefer_free_formats') else 'mp4'
 232                 return (AllowedQuality(req_ext, (best_quality, )), )
 233             allowed_qualities = guess_allowed_qualities()
 234
 235         formats = []
 236         for ext, qualities_ in allowed_qualities:
 237             for quality in qualities_:
 238                 f = QUALITIES[quality].copy()
 239                 clip_post = {
 240                     'author': author,
 241                     'includeCaptions': False,
 242                     'clipIndex': int(clip_id),
 243                     'courseName': course_name,
 244                     'locale': 'en',
 245                     'moduleName': name,
 246                     'mediaType': ext,
 247                     'quality': '%dx%d' % (f['width'], f['height']),
 248                 }
 249                 format_id = '%s-%s' % (ext, quality)
 250                 viewclip = self._download_json(
 251                     '%s/video/clips/viewclip' % self._API_BASE, display_id,
 252                     'Downloading %s viewclip JSON' % format_id, fatal=False,
 253                     data=json.dumps(clip_post).encode('utf-8'),
 254                     headers={'Content-Type': 'application/json;charset=utf-8'})
 255
 256                 # Pluralsight tracks multiple sequential calls to ViewClip API and start
 257                 # to return 429 HTTP errors after some time (see
 258                 # https://github.com/rg3/youtube-dl/pull/6989). Moreover it may even lead
 259                 # to account ban (see https://github.com/rg3/youtube-dl/issues/6842).
 260                 # To somewhat reduce the probability of these consequences
 261                 # we will sleep random amount of time before each call to ViewClip.
 262                 self._sleep(
 263                     random.randint(2, 5), display_id,
 264                     '%(video_id)s: Waiting for %(timeout)s seconds to avoid throttling')
 265
 266                 if not viewclip:
 267                     continue
 268
 269                 clip_urls = viewclip.get('urls')
 270                 if not isinstance(clip_urls, list):
 271                     continue
 272
 273                 for clip_url_data in clip_urls:
 274                     clip_url = clip_url_data.get('url')
 275                     if not clip_url:
 276                         continue
 277                     cdn = clip_url_data.get('cdn')
 278                     clip_f = f.copy()
 279                     clip_f.update({
 280                         'url': clip_url,
 281                         'ext': ext,
 282                         'format_id': '%s-%s' % (format_id, cdn) if cdn else format_id,
 283                         'quality': quality_key(quality),
 284                         'source_preference': int_or_none(clip_url_data.get('rank')),
 285                     })
 286                     formats.append(clip_f)
 287
 288         self._sort_formats(formats)
 289
 290         duration = int_or_none(
 291             clip.get('duration')) or parse_duration(clip.get('formattedDuration'))
 292
 293         # TODO: other languages?
 294         subtitles = self.extract_subtitles(
 295             author, clip_id, 'en', name, duration, display_id)
 296
 297         return {
 298             'id': clip.get('clipName') or clip['name'],
 299             'title': title,
 300             'duration': duration,
 301             'creator': author,
 302             'formats': formats,
 303             'subtitles': subtitles,
 304         }
 305
 306
 307 class PluralsightCourseIE(PluralsightBaseIE):
 308     IE_NAME = 'pluralsight:course'
 309     _VALID_URL = r'https?://(?:(?:www|app)\.)?pluralsight\.com/(?:library/)?courses/(?P<id>[^/]+)'
 310     _TESTS = [{
 311         # Free course from Pluralsight Starter Subscription for Microsoft TechNet
 312         # https://offers.pluralsight.com/technet?loc=zTS3z&prod=zOTprodz&tech=zOttechz&prog=zOTprogz&type=zSOz&media=zOTmediaz&country=zUSz
 313         'url': 'http://www.pluralsight.com/courses/hosting-sql-server-windows-azure-iaas',
 314         'info_dict': {
 315             'id': 'hosting-sql-server-windows-azure-iaas',
 316             'title': 'Hosting SQL Server in Microsoft Azure IaaS Fundamentals',
 317             'description': 'md5:61b37e60f21c4b2f91dc621a977d0986',
 318         },
 319         'playlist_count': 31,
 320     }, {
 321         # available without pluralsight account
 322         'url': 'https://www.pluralsight.com/courses/angularjs-get-started',
 323         'only_matching': True,
 324     }, {
 325         'url': 'https://app.pluralsight.com/library/courses/understanding-microsoft-azure-amazon-aws/table-of-contents',
 326         'only_matching': True,
 327     }]
 328
 329     def _real_extract(self, url):
 330         course_id = self._match_id(url)
 331
 332         # TODO: PSM cookie
 333
 334         course = self._download_json(
 335             '%s/player/functions/rpc' % self._API_BASE, course_id,
 336             'Downloading course JSON',
 337             data=json.dumps({
 338                 'fn': 'bootstrapPlayer',
 339                 'payload': {
 340                     'courseId': course_id,
 341                 }
 342             }).encode('utf-8'),
 343             headers={
 344                 'Content-Type': 'application/json;charset=utf-8'
 345             })['payload']['course']
 346
 347         title = course['title']
 348         course_name = course['name']
 349         course_data = course['modules']
 350         description = course.get('description') or course.get('shortDescription')
 351
 352         entries = []
 353         for num, module in enumerate(course_data, 1):
 354             author = module.get('author')
 355             module_name = module.get('name')
 356             if not author or not module_name:
 357                 continue
 358             for clip in module.get('clips', []):
 359                 clip_index = int_or_none(clip.get('index'))
 360                 if clip_index is None:
 361                     continue
 362                 clip_url = update_url_query(
 363                     '%s/player' % self._API_BASE, query={
 364                         'mode': 'live',
 365                         'course': course_name,
 366                         'author': author,
 367                         'name': module_name,
 368                         'clip': clip_index,
 369                     })
 370                 entries.append({
 371                     '_type': 'url_transparent',
 372                     'url': clip_url,
 373                     'ie_key': PluralsightIE.ie_key(),
 374                     'chapter': module.get('title'),
 375                     'chapter_number': num,
 376                     'chapter_id': module.get('moduleRef'),
 377                 })
 378
 379         return self.playlist_result(entries, course_id, title, description)