_ Git - youtube-dl/blob - youtube_dl/extractor/pluralsight.py

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