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