yt_dlp/extractor/gdcvault.py

   1 import re
   2
   3 from .common import InfoExtractor
   4 from .kaltura import KalturaIE
   5 from ..utils import (
   6     HEADRequest,
   7     remove_start,
   8     sanitized_Request,
   9     smuggle_url,
  10     urlencode_postdata,
  11 )
  12
  13
  14 class GDCVaultIE(InfoExtractor):
  15     _VALID_URL = r'https?://(?:www\.)?gdcvault\.com/play/(?P<id>\d+)(?:/(?P<name>[\w-]+))?'
  16     _NETRC_MACHINE = 'gdcvault'
  17     _TESTS = [
  18         {
  19             'url': 'http://www.gdcvault.com/play/1019721/Doki-Doki-Universe-Sweet-Simple',
  20             'md5': '7ce8388f544c88b7ac11c7ab1b593704',
  21             'info_dict': {
  22                 'id': '201311826596_AWNY',
  23                 'display_id': 'Doki-Doki-Universe-Sweet-Simple',
  24                 'ext': 'mp4',
  25                 'title': 'Doki-Doki Universe: Sweet, Simple and Genuine (GDC Next 10)'
  26             }
  27         },
  28         {
  29             'url': 'http://www.gdcvault.com/play/1015683/Embracing-the-Dark-Art-of',
  30             'info_dict': {
  31                 'id': '201203272_1330951438328RSXR',
  32                 'display_id': 'Embracing-the-Dark-Art-of',
  33                 'ext': 'flv',
  34                 'title': 'Embracing the Dark Art of Mathematical Modeling in AI'
  35             },
  36             'params': {
  37                 'skip_download': True,  # Requires rtmpdump
  38             }
  39         },
  40         {
  41             'url': 'http://www.gdcvault.com/play/1015301/Thexder-Meets-Windows-95-or',
  42             'md5': 'a5eb77996ef82118afbbe8e48731b98e',
  43             'info_dict': {
  44                 'id': '1015301',
  45                 'display_id': 'Thexder-Meets-Windows-95-or',
  46                 'ext': 'flv',
  47                 'title': 'Thexder Meets Windows 95, or Writing Great Games in the Windows 95 Environment',
  48             },
  49             'skip': 'Requires login',
  50         },
  51         {
  52             'url': 'http://gdcvault.com/play/1020791/',
  53             'only_matching': True,
  54         },
  55         {
  56             # Hard-coded hostname
  57             'url': 'http://gdcvault.com/play/1023460/Tenacious-Design-and-The-Interface',
  58             'md5': 'a8efb6c31ed06ca8739294960b2dbabd',
  59             'info_dict': {
  60                 'id': '840376_BQRC',
  61                 'ext': 'mp4',
  62                 'display_id': 'Tenacious-Design-and-The-Interface',
  63                 'title': 'Tenacious Design and The Interface of \'Destiny\'',
  64             },
  65         },
  66         {
  67             # Multiple audios
  68             'url': 'http://www.gdcvault.com/play/1014631/Classic-Game-Postmortem-PAC',
  69             'info_dict': {
  70                 'id': '12396_1299111843500GMPX',
  71                 'ext': 'mp4',
  72                 'title': 'How to Create a Good Game - From My Experience of Designing Pac-Man',
  73             },
  74             # 'params': {
  75             #     'skip_download': True,  # Requires rtmpdump
  76             #     'format': 'jp',  # The japanese audio
  77             # }
  78         },
  79         {
  80             # gdc-player.html
  81             'url': 'http://www.gdcvault.com/play/1435/An-American-engine-in-Tokyo',
  82             'info_dict': {
  83                 'id': '9350_1238021887562UHXB',
  84                 'display_id': 'An-American-engine-in-Tokyo',
  85                 'ext': 'mp4',
  86                 'title': 'An American Engine in Tokyo:/nThe collaboration of Epic Games and Square Enix/nFor THE LAST REMINANT',
  87             },
  88         },
  89         {
  90             # Kaltura Embed
  91             'url': 'https://www.gdcvault.com/play/1026180/Mastering-the-Apex-of-Scaling',
  92             'info_dict': {
  93                 'id': '0_h1fg8j3p',
  94                 'ext': 'mp4',
  95                 'title': 'Mastering the Apex of Scaling Game Servers (Presented by Multiplay)',
  96                 'timestamp': 1554401811,
  97                 'upload_date': '20190404',
  98                 'uploader_id': 'joe@blazestreaming.com',
  99             },
 100             'params': {
 101                 'format': 'mp4-408',
 102             },
 103         },
 104         {
 105             # Kaltura embed, whitespace between quote and embedded URL in iframe's src
 106             'url': 'https://www.gdcvault.com/play/1025699',
 107             'info_dict': {
 108                 'id': '0_zagynv0a',
 109                 'ext': 'mp4',
 110                 'title': 'Tech Toolbox',
 111                 'upload_date': '20190408',
 112                 'uploader_id': 'joe@blazestreaming.com',
 113                 'timestamp': 1554764629,
 114             },
 115             'params': {
 116                 'skip_download': True,
 117             },
 118         },
 119         {
 120             # HTML5 video
 121             'url': 'http://www.gdcvault.com/play/1014846/Conference-Keynote-Shigeru',
 122             'only_matching': True,
 123         },
 124     ]
 125
 126     def _login(self, webpage_url, display_id):
 127         username, password = self._get_login_info()
 128         if username is None or password is None:
 129             self.report_warning('It looks like ' + webpage_url + ' requires a login. Try specifying a username and password and try again.')
 130             return None
 131
 132         mobj = re.match(r'(?P<root_url>https?://.*?/).*', webpage_url)
 133         login_url = mobj.group('root_url') + 'api/login.php'
 134         logout_url = mobj.group('root_url') + 'logout'
 135
 136         login_form = {
 137             'email': username,
 138             'password': password,
 139         }
 140
 141         request = sanitized_Request(login_url, urlencode_postdata(login_form))
 142         request.add_header('Content-Type', 'application/x-www-form-urlencoded')
 143         self._download_webpage(request, display_id, 'Logging in')
 144         start_page = self._download_webpage(webpage_url, display_id, 'Getting authenticated video page')
 145         self._download_webpage(logout_url, display_id, 'Logging out')
 146
 147         return start_page
 148
 149     def _real_extract(self, url):
 150         video_id, name = self._match_valid_url(url).groups()
 151         display_id = name or video_id
 152
 153         webpage_url = 'http://www.gdcvault.com/play/' + video_id
 154         start_page = self._download_webpage(webpage_url, display_id)
 155
 156         direct_url = self._search_regex(
 157             r's1\.addVariable\("file",\s*encodeURIComponent\("(/[^"]+)"\)\);',
 158             start_page, 'url', default=None)
 159         if direct_url:
 160             title = self._html_search_regex(
 161                 r'<td><strong>Session Name:?</strong></td>\s*<td>(.*?)</td>',
 162                 start_page, 'title')
 163             video_url = 'http://www.gdcvault.com' + direct_url
 164             # resolve the url so that we can detect the correct extension
 165             video_url = self._request_webpage(
 166                 HEADRequest(video_url), video_id).geturl()
 167
 168             return {
 169                 'id': video_id,
 170                 'display_id': display_id,
 171                 'url': video_url,
 172                 'title': title,
 173             }
 174
 175         embed_url = KalturaIE._extract_url(start_page)
 176         if embed_url:
 177             embed_url = smuggle_url(embed_url, {'source_url': url})
 178             ie_key = 'Kaltura'
 179         else:
 180             PLAYER_REGEX = r'<iframe src="(?P<xml_root>.+?)/(?:gdc-)?player.*?\.html.*?".*?</iframe>'
 181
 182             xml_root = self._html_search_regex(
 183                 PLAYER_REGEX, start_page, 'xml root', default=None)
 184             if xml_root is None:
 185                 # Probably need to authenticate
 186                 login_res = self._login(webpage_url, display_id)
 187                 if login_res is None:
 188                     self.report_warning('Could not login.')
 189                 else:
 190                     start_page = login_res
 191                     # Grab the url from the authenticated page
 192                     xml_root = self._html_search_regex(
 193                         PLAYER_REGEX, start_page, 'xml root')
 194
 195             xml_name = self._html_search_regex(
 196                 r'<iframe src=".*?\?xml(?:=|URL=xml/)(.+?\.xml).*?".*?</iframe>',
 197                 start_page, 'xml filename', default=None)
 198             if not xml_name:
 199                 info = self._parse_html5_media_entries(url, start_page, video_id)[0]
 200                 info.update({
 201                     'title': remove_start(self._search_regex(
 202                         r'>Session Name:\s*<.*?>\s*<td>(.+?)</td>', start_page,
 203                         'title', default=None) or self._og_search_title(
 204                         start_page, default=None), 'GDC Vault - '),
 205                     'id': video_id,
 206                     'display_id': display_id,
 207                 })
 208                 return info
 209             embed_url = '%s/xml/%s' % (xml_root, xml_name)
 210             ie_key = 'DigitallySpeaking'
 211
 212         return {
 213             '_type': 'url_transparent',
 214             'id': video_id,
 215             'display_id': display_id,
 216             'url': embed_url,
 217             'ie_key': ie_key,
 218         }