Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/pluralsight.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4 import json
   5 import random
   6 import collections
   7
   8 from .common import InfoExtractor
   9 from ..compat import (
  10     compat_str,
  11     compat_urllib_parse,
  12     compat_urlparse,
  13 )
  14 from ..utils import (
  15     ExtractorError,
  16     int_or_none,
  17     parse_duration,
  18     qualities,
  19     sanitized_Request,
  20 )
  21
  22
  23 class PluralsightBaseIE(InfoExtractor):
  24     _API_BASE = 'http://app.pluralsight.com'
  25
  26
  27 class PluralsightIE(PluralsightBaseIE):
  28     IE_NAME = 'pluralsight'
  29     _VALID_URL = r'https?://(?:(?:www|app)\.)?pluralsight\.com/training/player\?'
  30     _LOGIN_URL = 'https://app.pluralsight.com/id/'
  31
  32     _NETRC_MACHINE = 'pluralsight'
  33
  34     _TESTS = [{
  35         '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',
  36         'md5': '4d458cf5cf4c593788672419a8dd4cf8',
  37         'info_dict': {
  38             'id': 'hosting-sql-server-windows-azure-iaas-m7-mgmt-04',
  39             'ext': 'mp4',
  40             'title': 'Management of SQL Server - Demo Monitoring',
  41             'duration': 338,
  42         },
  43         'skip': 'Requires pluralsight account credentials',
  44     }, {
  45         'url': 'https://app.pluralsight.com/training/player?course=angularjs-get-started&author=scott-allen&name=angularjs-get-started-m1-introduction&clip=0&mode=live',
  46         'only_matching': True,
  47     }, {
  48         # available without pluralsight account
  49         'url': 'http://app.pluralsight.com/training/player?author=scott-allen&name=angularjs-get-started-m1-introduction&mode=live&clip=0&course=angularjs-get-started',
  50         'only_matching': True,
  51     }]
  52
  53     def _real_initialize(self):
  54         self._login()
  55
  56     def _login(self):
  57         (username, password) = self._get_login_info()
  58         if username is None:
  59             return
  60
  61         login_page = self._download_webpage(
  62             self._LOGIN_URL, None, 'Downloading login page')
  63
  64         login_form = self._hidden_inputs(login_page)
  65
  66         login_form.update({
  67             'Username': username.encode('utf-8'),
  68             'Password': password.encode('utf-8'),
  69         })
  70
  71         post_url = self._search_regex(
  72             r'<form[^>]+action=(["\'])(?P<url>.+?)\1', login_page,
  73             'post url', default=self._LOGIN_URL, group='url')
  74
  75         if not post_url.startswith('http'):
  76             post_url = compat_urlparse.urljoin(self._LOGIN_URL, post_url)
  77
  78         request = sanitized_Request(
  79             post_url, compat_urllib_parse.urlencode(login_form).encode('utf-8'))
  80         request.add_header('Content-Type', 'application/x-www-form-urlencoded')
  81
  82         response = self._download_webpage(
  83             request, None, 'Logging in as %s' % username)
  84
  85         error = self._search_regex(
  86             r'<span[^>]+class="field-validation-error"[^>]*>([^<]+)</span>',
  87             response, 'error message', default=None)
  88         if error:
  89             raise ExtractorError('Unable to login: %s' % error, expected=True)
  90
  91         if all(p not in response for p in ('__INITIAL_STATE__', '"currentUser"')):
  92             raise ExtractorError('Unable to log in')
  93
  94     def _real_extract(self, url):
  95         qs = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
  96
  97         author = qs.get('author', [None])[0]
  98         name = qs.get('name', [None])[0]
  99         clip_id = qs.get('clip', [None])[0]
 100         course = qs.get('course', [None])[0]
 101
 102         if any(not f for f in (author, name, clip_id, course,)):
 103             raise ExtractorError('Invalid URL', expected=True)
 104
 105         display_id = '%s-%s' % (name, clip_id)
 106
 107         webpage = self._download_webpage(url, display_id)
 108
 109         modules = self._search_regex(
 110             r'moduleCollection\s*:\s*new\s+ModuleCollection\((\[.+?\])\s*,\s*\$rootScope\)',
 111             webpage, 'modules', default=None)
 112
 113         if modules:
 114             collection = self._parse_json(modules, display_id)
 115         else:
 116             # Webpage may be served in different layout (see
 117             # https://github.com/rg3/youtube-dl/issues/7607)
 118             collection = self._parse_json(
 119                 self._search_regex(
 120                     r'var\s+initialState\s*=\s*({.+?});\n', webpage, 'initial state'),
 121                 display_id)['course']['modules']
 122
 123         module, clip = None, None
 124
 125         for module_ in collection:
 126             if name in (module_.get('moduleName'), module_.get('name')):
 127                 module = module_
 128                 for clip_ in module_.get('clips', []):
 129                     clip_index = clip_.get('clipIndex')
 130                     if clip_index is None:
 131                         clip_index = clip_.get('index')
 132                     if clip_index is None:
 133                         continue
 134                     if compat_str(clip_index) == clip_id:
 135                         clip = clip_
 136                         break
 137
 138         if not clip:
 139             raise ExtractorError('Unable to resolve clip')
 140
 141         QUALITIES = {
 142             'low': {'width': 640, 'height': 480},
 143             'medium': {'width': 848, 'height': 640},
 144             'high': {'width': 1024, 'height': 768},
 145             'high-widescreen': {'width': 1280, 'height': 720},
 146         }
 147
 148         QUALITIES_PREFERENCE = ('low', 'medium', 'high', 'high-widescreen',)
 149         quality_key = qualities(QUALITIES_PREFERENCE)
 150
 151         AllowedQuality = collections.namedtuple('AllowedQuality', ['ext', 'qualities'])
 152
 153         ALLOWED_QUALITIES = (
 154             AllowedQuality('webm', ['high', ]),
 155             AllowedQuality('mp4', ['low', 'medium', 'high', ]),
 156         )
 157
 158         # Some courses also offer widescreen resolution for high quality (see
 159         # https://github.com/rg3/youtube-dl/issues/7766)
 160         widescreen = True if re.search(
 161             r'courseSupportsWidescreenVideoFormats\s*:\s*true', webpage) else False
 162         best_quality = 'high-widescreen' if widescreen else 'high'
 163         if widescreen:
 164             for allowed_quality in ALLOWED_QUALITIES:
 165                 allowed_quality.qualities.append(best_quality)
 166
 167         # In order to minimize the number of calls to ViewClip API and reduce
 168         # the probability of being throttled or banned by Pluralsight we will request
 169         # only single format until formats listing was explicitly requested.
 170         if self._downloader.params.get('listformats', False):
 171             allowed_qualities = ALLOWED_QUALITIES
 172         else:
 173             def guess_allowed_qualities():
 174                 req_format = self._downloader.params.get('format') or 'best'
 175                 req_format_split = req_format.split('-', 1)
 176                 if len(req_format_split) > 1:
 177                     req_ext, req_quality = req_format_split
 178                     for allowed_quality in ALLOWED_QUALITIES:
 179                         if req_ext == allowed_quality.ext and req_quality in allowed_quality.qualities:
 180                             return (AllowedQuality(req_ext, (req_quality, )), )
 181                 req_ext = 'webm' if self._downloader.params.get('prefer_free_formats') else 'mp4'
 182                 return (AllowedQuality(req_ext, (best_quality, )), )
 183             allowed_qualities = guess_allowed_qualities()
 184
 185         formats = []
 186         for ext, qualities_ in allowed_qualities:
 187             for quality in qualities_:
 188                 f = QUALITIES[quality].copy()
 189                 clip_post = {
 190                     'a': author,
 191                     'cap': 'false',
 192                     'cn': clip_id,
 193                     'course': course,
 194                     'lc': 'en',
 195                     'm': name,
 196                     'mt': ext,
 197                     'q': '%dx%d' % (f['width'], f['height']),
 198                 }
 199                 request = sanitized_Request(
 200                     '%s/training/Player/ViewClip' % self._API_BASE,
 201                     json.dumps(clip_post).encode('utf-8'))
 202                 request.add_header('Content-Type', 'application/json;charset=utf-8')
 203                 format_id = '%s-%s' % (ext, quality)
 204                 clip_url = self._download_webpage(
 205                     request, display_id, 'Downloading %s URL' % format_id, fatal=False)
 206
 207                 # Pluralsight tracks multiple sequential calls to ViewClip API and start
 208                 # to return 429 HTTP errors after some time (see
 209                 # https://github.com/rg3/youtube-dl/pull/6989). Moreover it may even lead
 210                 # to account ban (see https://github.com/rg3/youtube-dl/issues/6842).
 211                 # To somewhat reduce the probability of these consequences
 212                 # we will sleep random amount of time before each call to ViewClip.
 213                 self._sleep(
 214                     random.randint(2, 5), display_id,
 215                     '%(video_id)s: Waiting for %(timeout)s seconds to avoid throttling')
 216
 217                 if not clip_url:
 218                     continue
 219                 f.update({
 220                     'url': clip_url,
 221                     'ext': ext,
 222                     'format_id': format_id,
 223                     'quality': quality_key(quality),
 224                 })
 225                 formats.append(f)
 226         self._sort_formats(formats)
 227
 228         # TODO: captions
 229         # http://www.pluralsight.com/training/Player/ViewClip + cap = true
 230         # or
 231         # http://www.pluralsight.com/training/Player/Captions
 232         # { a = author, cn = clip_id, lc = end, m = name }
 233
 234         return {
 235             'id': clip.get('clipName') or clip['name'],
 236             'title': '%s - %s' % (module['title'], clip['title']),
 237             'duration': int_or_none(clip.get('duration')) or parse_duration(clip.get('formattedDuration')),
 238             'creator': author,
 239             'formats': formats
 240         }
 241
 242
 243 class PluralsightCourseIE(PluralsightBaseIE):
 244     IE_NAME = 'pluralsight:course'
 245     _VALID_URL = r'https?://(?:(?:www|app)\.)?pluralsight\.com/(?:library/)?courses/(?P<id>[^/]+)'
 246     _TESTS = [{
 247         # Free course from Pluralsight Starter Subscription for Microsoft TechNet
 248         # https://offers.pluralsight.com/technet?loc=zTS3z&prod=zOTprodz&tech=zOttechz&prog=zOTprogz&type=zSOz&media=zOTmediaz&country=zUSz
 249         'url': 'http://www.pluralsight.com/courses/hosting-sql-server-windows-azure-iaas',
 250         'info_dict': {
 251             'id': 'hosting-sql-server-windows-azure-iaas',
 252             'title': 'Hosting SQL Server in Microsoft Azure IaaS Fundamentals',
 253             'description': 'md5:61b37e60f21c4b2f91dc621a977d0986',
 254         },
 255         'playlist_count': 31,
 256     }, {
 257         # available without pluralsight account
 258         'url': 'https://www.pluralsight.com/courses/angularjs-get-started',
 259         'only_matching': True,
 260     }, {
 261         'url': 'https://app.pluralsight.com/library/courses/understanding-microsoft-azure-amazon-aws/table-of-contents',
 262         'only_matching': True,
 263     }]
 264
 265     def _real_extract(self, url):
 266         course_id = self._match_id(url)
 267
 268         # TODO: PSM cookie
 269
 270         course = self._download_json(
 271             '%s/data/course/%s' % (self._API_BASE, course_id),
 272             course_id, 'Downloading course JSON')
 273
 274         title = course['title']
 275         description = course.get('description') or course.get('shortDescription')
 276
 277         course_data = self._download_json(
 278             '%s/data/course/content/%s' % (self._API_BASE, course_id),
 279             course_id, 'Downloading course data JSON')
 280
 281         entries = []
 282         for module in course_data:
 283             for clip in module.get('clips', []):
 284                 player_parameters = clip.get('playerParameters')
 285                 if not player_parameters:
 286                     continue
 287                 entries.append(self.url_result(
 288                     '%s/training/player?%s' % (self._API_BASE, player_parameters),
 289                     'Pluralsight'))
 290
 291         return self.playlist_result(entries, course_id, title, description)