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         parsed_url = compat_urlparse.urlparse(url)
 161
 162         payload_url = compat_urlparse.urlunparse(parsed_url._replace(
 163             netloc='app.pluralsight.com', path='player/api/v1/payload'))
 164
 165         course = self._download_json(
 166             payload_url, display_id, headers={'Referer': url})['payload']['course']
 167
 168         collection = course['modules']
 169
 170         module, clip = None, None
 171
 172         for module_ in collection:
 173             if name in (module_.get('moduleName'), module_.get('name')):
 174                 module = module_
 175                 for clip_ in module_.get('clips', []):
 176                     clip_index = clip_.get('clipIndex')
 177                     if clip_index is None:
 178                         clip_index = clip_.get('index')
 179                     if clip_index is None:
 180                         continue
 181                     if compat_str(clip_index) == clip_id:
 182                         clip = clip_
 183                         break
 184
 185         if not clip:
 186             raise ExtractorError('Unable to resolve clip')
 187
 188         title = '%s - %s' % (module['title'], clip['title'])
 189
 190         QUALITIES = {
 191             'low': {'width': 640, 'height': 480},
 192             'medium': {'width': 848, 'height': 640},
 193             'high': {'width': 1024, 'height': 768},
 194             'high-widescreen': {'width': 1280, 'height': 720},
 195         }
 196
 197         QUALITIES_PREFERENCE = ('low', 'medium', 'high', 'high-widescreen',)
 198         quality_key = qualities(QUALITIES_PREFERENCE)
 199
 200         AllowedQuality = collections.namedtuple('AllowedQuality', ['ext', 'qualities'])
 201
 202         ALLOWED_QUALITIES = (
 203             AllowedQuality('webm', ['high', ]),
 204             AllowedQuality('mp4', ['low', 'medium', 'high', ]),
 205         )
 206
 207         # Some courses also offer widescreen resolution for high quality (see
 208         # https://github.com/rg3/youtube-dl/issues/7766)
 209         widescreen = course.get('supportsWideScreenVideoFormats') is True
 210         best_quality = 'high-widescreen' if widescreen else 'high'
 211         if widescreen:
 212             for allowed_quality in ALLOWED_QUALITIES:
 213                 allowed_quality.qualities.append(best_quality)
 214
 215         # In order to minimize the number of calls to ViewClip API and reduce
 216         # the probability of being throttled or banned by Pluralsight we will request
 217         # only single format until formats listing was explicitly requested.
 218         if self._downloader.params.get('listformats', False):
 219             allowed_qualities = ALLOWED_QUALITIES
 220         else:
 221             def guess_allowed_qualities():
 222                 req_format = self._downloader.params.get('format') or 'best'
 223                 req_format_split = req_format.split('-', 1)
 224                 if len(req_format_split) > 1:
 225                     req_ext, req_quality = req_format_split
 226                     for allowed_quality in ALLOWED_QUALITIES:
 227                         if req_ext == allowed_quality.ext and req_quality in allowed_quality.qualities:
 228                             return (AllowedQuality(req_ext, (req_quality, )), )
 229                 req_ext = 'webm' if self._downloader.params.get('prefer_free_formats') else 'mp4'
 230                 return (AllowedQuality(req_ext, (best_quality, )), )
 231             allowed_qualities = guess_allowed_qualities()
 232
 233         formats = []
 234         for ext, qualities_ in allowed_qualities:
 235             for quality in qualities_:
 236                 f = QUALITIES[quality].copy()
 237                 clip_post = {
 238                     'author': author,
 239                     'includeCaptions': False,
 240                     'clipIndex': int(clip_id),
 241                     'courseName': course_name,
 242                     'locale': 'en',
 243                     'moduleName': name,
 244                     'mediaType': ext,
 245                     'quality': '%dx%d' % (f['width'], f['height']),
 246                 }
 247                 format_id = '%s-%s' % (ext, quality)
 248                 viewclip = self._download_json(
 249                     '%s/video/clips/viewclip' % self._API_BASE, display_id,
 250                     'Downloading %s viewclip JSON' % format_id, fatal=False,
 251                     data=json.dumps(clip_post).encode('utf-8'),
 252                     headers={'Content-Type': 'application/json;charset=utf-8'})
 253
 254                 # Pluralsight tracks multiple sequential calls to ViewClip API and start
 255                 # to return 429 HTTP errors after some time (see
 256                 # https://github.com/rg3/youtube-dl/pull/6989). Moreover it may even lead
 257                 # to account ban (see https://github.com/rg3/youtube-dl/issues/6842).
 258                 # To somewhat reduce the probability of these consequences
 259                 # we will sleep random amount of time before each call to ViewClip.
 260                 self._sleep(
 261                     random.randint(2, 5), display_id,
 262                     '%(video_id)s: Waiting for %(timeout)s seconds to avoid throttling')
 263
 264                 if not viewclip:
 265                     continue
 266
 267                 clip_urls = viewclip.get('urls')
 268                 if not isinstance(clip_urls, list):
 269                     continue
 270
 271                 for clip_url_data in clip_urls:
 272                     clip_url = clip_url_data.get('url')
 273                     if not clip_url:
 274                         continue
 275                     cdn = clip_url_data.get('cdn')
 276                     clip_f = f.copy()
 277                     clip_f.update({
 278                         'url': clip_url,
 279                         'ext': ext,
 280                         'format_id': '%s-%s' % (format_id, cdn) if cdn else format_id,
 281                         'quality': quality_key(quality),
 282                         'source_preference': int_or_none(clip_url_data.get('rank')),
 283                     })
 284                     formats.append(clip_f)
 285
 286         self._sort_formats(formats)
 287
 288         duration = int_or_none(
 289             clip.get('duration')) or parse_duration(clip.get('formattedDuration'))
 290
 291         # TODO: other languages?
 292         subtitles = self.extract_subtitles(
 293             author, clip_id, 'en', name, duration, display_id)
 294
 295         return {
 296             'id': clip.get('clipName') or clip['name'],
 297             'title': title,
 298             'duration': duration,
 299             'creator': author,
 300             'formats': formats,
 301             'subtitles': subtitles,
 302         }
 303
 304
 305 class PluralsightCourseIE(PluralsightBaseIE):
 306     IE_NAME = 'pluralsight:course'
 307     _VALID_URL = r'https?://(?:(?:www|app)\.)?pluralsight\.com/(?:library/)?courses/(?P<id>[^/]+)'
 308     _TESTS = [{
 309         # Free course from Pluralsight Starter Subscription for Microsoft TechNet
 310         # https://offers.pluralsight.com/technet?loc=zTS3z&prod=zOTprodz&tech=zOttechz&prog=zOTprogz&type=zSOz&media=zOTmediaz&country=zUSz
 311         'url': 'http://www.pluralsight.com/courses/hosting-sql-server-windows-azure-iaas',
 312         'info_dict': {
 313             'id': 'hosting-sql-server-windows-azure-iaas',
 314             'title': 'Hosting SQL Server in Microsoft Azure IaaS Fundamentals',
 315             'description': 'md5:61b37e60f21c4b2f91dc621a977d0986',
 316         },
 317         'playlist_count': 31,
 318     }, {
 319         # available without pluralsight account
 320         'url': 'https://www.pluralsight.com/courses/angularjs-get-started',
 321         'only_matching': True,
 322     }, {
 323         'url': 'https://app.pluralsight.com/library/courses/understanding-microsoft-azure-amazon-aws/table-of-contents',
 324         'only_matching': True,
 325     }]
 326
 327     def _real_extract(self, url):
 328         course_id = self._match_id(url)
 329
 330         # TODO: PSM cookie
 331
 332         course = self._download_json(
 333             '%s/data/course/%s' % (self._API_BASE, course_id),
 334             course_id, 'Downloading course JSON')
 335
 336         title = course['title']
 337         description = course.get('description') or course.get('shortDescription')
 338
 339         course_data = self._download_json(
 340             '%s/data/course/content/%s' % (self._API_BASE, course_id),
 341             course_id, 'Downloading course data JSON')
 342
 343         entries = []
 344         for num, module in enumerate(course_data, 1):
 345             for clip in module.get('clips', []):
 346                 player_parameters = clip.get('playerParameters')
 347                 if not player_parameters:
 348                     continue
 349                 entries.append({
 350                     '_type': 'url_transparent',
 351                     'url': '%s/training/player?%s' % (self._API_BASE, player_parameters),
 352                     'ie_key': PluralsightIE.ie_key(),
 353                     'chapter': module.get('title'),
 354                     'chapter_number': num,
 355                     'chapter_id': module.get('moduleRef'),
 356                 })
 357
 358         return self.playlist_result(entries, course_id, title, description)