yt_dlp/extractor/gdcvault.py

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