Update to ytdl-commit-a726009

[blinkx] Remove extractor
a726009987

yt_dlp/extractor/gdcvault.py

@@ -5,7 +5,10 @@ import re
 from .common import InfoExtractor
 from .kaltura import KalturaIE
 from ..utils import (
+    HEADRequest,
+    remove_start,
     sanitized_Request,
+    smuggle_url,
     urlencode_postdata,
 )
 
@@ -100,6 +103,26 @@ class GDCVaultIE(InfoExtractor):
                 'format': 'mp4-408',
             },
         },
+        {
+            # Kaltura embed, whitespace between quote and embedded URL in iframe's src
+            'url': 'https://www.gdcvault.com/play/1025699',
+            'info_dict': {
+                'id': '0_zagynv0a',
+                'ext': 'mp4',
+                'title': 'Tech Toolbox',
+                'upload_date': '20190408',
+                'uploader_id': 'joe@blazestreaming.com',
+                'timestamp': 1554764629,
+            },
+            'params': {
+                'skip_download': True,
+            },
+        },
+        {
+            # HTML5 video
+            'url': 'http://www.gdcvault.com/play/1014846/Conference-Keynote-Shigeru',
+            'only_matching': True,
+        },
     ]
 
     def _login(self, webpage_url, display_id):
@@ -120,38 +143,78 @@ class GDCVaultIE(InfoExtractor):
         request = sanitized_Request(login_url, urlencode_postdata(login_form))
         request.add_header('Content-Type', 'application/x-www-form-urlencoded')
         self._download_webpage(request, display_id, 'Logging in')
-        webpage = self._download_webpage(webpage_url, display_id, 'Getting authenticated video page')
+        start_page = self._download_webpage(webpage_url, display_id, 'Getting authenticated video page')
         self._download_webpage(logout_url, display_id, 'Logging out')
 
-        return webpage
+        return start_page
 
     def _real_extract(self, url):
         video_id, name = re.match(self._VALID_URL, url).groups()
         display_id = name or video_id
 
-        webpage = self._download_webpage(url, display_id)
+        webpage_url = 'http://www.gdcvault.com/play/' + video_id
+        start_page = self._download_webpage(webpage_url, display_id)
 
-        title = self._html_search_regex(
-            r'<td><strong>Session Name:?</strong></td>\s*<td>(.*?)</td>',
-            webpage, 'title')
+        direct_url = self._search_regex(
+            r's1\.addVariable\("file",\s*encodeURIComponent\("(/[^"]+)"\)\);',
+            start_page, 'url', default=None)
+        if direct_url:
+            title = self._html_search_regex(
+                r'<td><strong>Session Name:?</strong></td>\s*<td>(.*?)</td>',
+                start_page, 'title')
+            video_url = 'http://www.gdcvault.com' + direct_url
+            # resolve the url so that we can detect the correct extension
+            video_url = self._request_webpage(
+                HEADRequest(video_url), video_id).geturl()
 
-        PLAYER_REGEX = r'<iframe src=\"(?P<manifest_url>.*?)\".*?</iframe>'
-        manifest_url = self._html_search_regex(
-            PLAYER_REGEX, webpage, 'manifest_url')
+            return {
+                'id': video_id,
+                'display_id': display_id,
+                'url': video_url,
+                'title': title,
+            }
 
-        partner_id = self._search_regex(
-            r'/p(?:artner_id)?/(\d+)', manifest_url, 'partner id',
-            default='1670711')
+        embed_url = KalturaIE._extract_url(start_page)
+        if embed_url:
+            embed_url = smuggle_url(embed_url, {'source_url': url})
+            ie_key = 'Kaltura'
+        else:
+            PLAYER_REGEX = r'<iframe src="(?P<xml_root>.+?)/(?:gdc-)?player.*?\.html.*?".*?</iframe>'
 
-        kaltura_id = self._search_regex(
-            r'entry_id=(?P<id>(?:[^&])+)', manifest_url,
-            'kaltura id', group='id')
+            xml_root = self._html_search_regex(
+                PLAYER_REGEX, start_page, 'xml root', default=None)
+            if xml_root is None:
+                # Probably need to authenticate
+                login_res = self._login(webpage_url, display_id)
+                if login_res is None:
+                    self.report_warning('Could not login.')
+                else:
+                    start_page = login_res
+                    # Grab the url from the authenticated page
+                    xml_root = self._html_search_regex(
+                        PLAYER_REGEX, start_page, 'xml root')
+
+            xml_name = self._html_search_regex(
+                r'<iframe src=".*?\?xml(?:=|URL=xml/)(.+?\.xml).*?".*?</iframe>',
+                start_page, 'xml filename', default=None)
+            if not xml_name:
+                info = self._parse_html5_media_entries(url, start_page, video_id)[0]
+                info.update({
+                    'title': remove_start(self._search_regex(
+                        r'>Session Name:\s*<.*?>\s*<td>(.+?)</td>', start_page,
+                        'title', default=None) or self._og_search_title(
+                        start_page, default=None), 'GDC Vault - '),
+                    'id': video_id,
+                    'display_id': display_id,
+                })
+                return info
+            embed_url = '%s/xml/%s' % (xml_root, xml_name)
+            ie_key = 'DigitallySpeaking'
 
         return {
             '_type': 'url_transparent',
-            'url': 'kaltura:%s:%s' % (partner_id, kaltura_id),
-            'ie_key': KalturaIE.ie_key(),
             'id': video_id,
             'display_id': display_id,
-            'title': title,
+            'url': embed_url,
+            'ie_key': ie_key,
         }