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