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