youtube_dl/extractor/gdcvault.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..utils import (
   7     HEADRequest,
   8     sanitized_Request,
   9     urlencode_postdata,
  10 )
  11
  12
  13 class GDCVaultIE(InfoExtractor):
  14     _VALID_URL = r'https?://(?:www\.)?gdcvault\.com/play/(?P<id>\d+)/(?P<name>(\w|-)+)?'
  15     _NETRC_MACHINE = 'gdcvault'
  16     _TESTS = [
  17         {
  18             'url': 'http://www.gdcvault.com/play/1019721/Doki-Doki-Universe-Sweet-Simple',
  19             'md5': '7ce8388f544c88b7ac11c7ab1b593704',
  20             'info_dict': {
  21                 'id': '1019721',
  22                 'display_id': 'Doki-Doki-Universe-Sweet-Simple',
  23                 'ext': 'mp4',
  24                 'title': 'Doki-Doki Universe: Sweet, Simple and Genuine (GDC Next 10)'
  25             }
  26         },
  27         {
  28             'url': 'http://www.gdcvault.com/play/1015683/Embracing-the-Dark-Art-of',
  29             'info_dict': {
  30                 'id': '1015683',
  31                 'display_id': 'Embracing-the-Dark-Art-of',
  32                 'ext': 'flv',
  33                 'title': 'Embracing the Dark Art of Mathematical Modeling in AI'
  34             },
  35             'params': {
  36                 'skip_download': True,  # Requires rtmpdump
  37             }
  38         },
  39         {
  40             'url': 'http://www.gdcvault.com/play/1015301/Thexder-Meets-Windows-95-or',
  41             'md5': 'a5eb77996ef82118afbbe8e48731b98e',
  42             'info_dict': {
  43                 'id': '1015301',
  44                 'display_id': 'Thexder-Meets-Windows-95-or',
  45                 'ext': 'flv',
  46                 'title': 'Thexder Meets Windows 95, or Writing Great Games in the Windows 95 Environment',
  47             },
  48             'skip': 'Requires login',
  49         },
  50         {
  51             'url': 'http://gdcvault.com/play/1020791/',
  52             'only_matching': True,
  53         },
  54         {
  55             # Hard-coded hostname
  56             'url': 'http://gdcvault.com/play/1023460/Tenacious-Design-and-The-Interface',
  57             'md5': 'a8efb6c31ed06ca8739294960b2dbabd',
  58             'info_dict': {
  59                 'id': '1023460',
  60                 'ext': 'mp4',
  61                 'display_id': 'Tenacious-Design-and-The-Interface',
  62                 'title': 'Tenacious Design and The Interface of \'Destiny\'',
  63             },
  64         },
  65         {
  66             # Multiple audios
  67             'url': 'http://www.gdcvault.com/play/1014631/Classic-Game-Postmortem-PAC',
  68             'info_dict': {
  69                 'id': '1014631',
  70                 'ext': 'flv',
  71                 'title': 'How to Create a Good Game - From My Experience of Designing Pac-Man',
  72             },
  73             'params': {
  74                 'skip_download': True,  # Requires rtmpdump
  75                 'format': 'jp',  # The japanese audio
  76             }
  77         },
  78     ]
  79
  80     def _login(self, webpage_url, display_id):
  81         (username, password) = self._get_login_info()
  82         if username is None or password is None:
  83             self.report_warning('It looks like ' + webpage_url + ' requires a login. Try specifying a username and password and try again.')
  84             return None
  85
  86         mobj = re.match(r'(?P<root_url>https?://.*?/).*', webpage_url)
  87         login_url = mobj.group('root_url') + 'api/login.php'
  88         logout_url = mobj.group('root_url') + 'logout'
  89
  90         login_form = {
  91             'email': username,
  92             'password': password,
  93         }
  94
  95         request = sanitized_Request(login_url, urlencode_postdata(login_form))
  96         request.add_header('Content-Type', 'application/x-www-form-urlencoded')
  97         self._download_webpage(request, display_id, 'Logging in')
  98         start_page = self._download_webpage(webpage_url, display_id, 'Getting authenticated video page')
  99         self._download_webpage(logout_url, display_id, 'Logging out')
 100
 101         return start_page
 102
 103     def _real_extract(self, url):
 104         mobj = re.match(self._VALID_URL, url)
 105
 106         video_id = mobj.group('id')
 107         display_id = mobj.group('name') or video_id
 108
 109         webpage_url = 'http://www.gdcvault.com/play/' + video_id
 110         start_page = self._download_webpage(webpage_url, display_id)
 111
 112         direct_url = self._search_regex(
 113             r's1\.addVariable\("file",\s*encodeURIComponent\("(/[^"]+)"\)\);',
 114             start_page, 'url', default=None)
 115         if direct_url:
 116             title = self._html_search_regex(
 117                 r'<td><strong>Session Name</strong></td>\s*<td>(.*?)</td>',
 118                 start_page, 'title')
 119             video_url = 'http://www.gdcvault.com' + direct_url
 120             # resolve the url so that we can detect the correct extension
 121             head = self._request_webpage(HEADRequest(video_url), video_id)
 122             video_url = head.geturl()
 123
 124             return {
 125                 'id': video_id,
 126                 'display_id': display_id,
 127                 'url': video_url,
 128                 'title': title,
 129             }
 130
 131         PLAYER_REGEX = r'<iframe src="(?P<xml_root>.+?)/player.*?\.html.*?".*?</iframe>'
 132
 133         xml_root = self._html_search_regex(
 134             PLAYER_REGEX, start_page, 'xml root', default=None)
 135         if xml_root is None:
 136             # Probably need to authenticate
 137             login_res = self._login(webpage_url, display_id)
 138             if login_res is None:
 139                 self.report_warning('Could not login.')
 140             else:
 141                 start_page = login_res
 142                 # Grab the url from the authenticated page
 143                 xml_root = self._html_search_regex(
 144                     PLAYER_REGEX, start_page, 'xml root')
 145
 146         xml_name = self._html_search_regex(
 147             r'<iframe src=".*?\?xml=(.+?\.xml).*?".*?</iframe>',
 148             start_page, 'xml filename', default=None)
 149         if xml_name is None:
 150             # Fallback to the older format
 151             xml_name = self._html_search_regex(
 152                 r'<iframe src=".*?\?xmlURL=xml/(?P<xml_file>.+?\.xml).*?".*?</iframe>',
 153                 start_page, 'xml filename')
 154
 155         return {
 156             '_type': 'url_transparent',
 157             'id': video_id,
 158             'display_id': display_id,
 159             'url': '%s/xml/%s' % (xml_root, xml_name),
 160             'ie_key': 'DigitallySpeaking',
 161         }