[gdc] add support for kaltura embeds and update tests(closes #20575)

1 files changed, 57 insertions, 43 deletions

diff --git a/youtube_dl/extractor/gdcvault.py b/youtube_dl/extractor/gdcvault.py
index 8806dc48a..2f555c1d4 100644
--- a/youtube_dl/extractor/gdcvault.py
+++ b/youtube_dl/extractor/gdcvault.py
@@ -3,22 +3,24 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
+from .kaltura import KalturaIE
 from ..utils import (
     HEADRequest,
     sanitized_Request,
+    smuggle_url,
     urlencode_postdata,
 )
 
 
 class GDCVaultIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?gdcvault\.com/play/(?P<id>\d+)/(?P<name>(\w|-)+)?'
+    _VALID_URL = r'https?://(?:www\.)?gdcvault\.com/play/(?P<id>\d+)(?:/(?P<name>[\w-]+))?'
     _NETRC_MACHINE = 'gdcvault'
     _TESTS = [
         {
             'url': 'http://www.gdcvault.com/play/1019721/Doki-Doki-Universe-Sweet-Simple',
             'md5': '7ce8388f544c88b7ac11c7ab1b593704',
             'info_dict': {
-                'id': '1019721',
+                'id': '201311826596_AWNY',
                 'display_id': 'Doki-Doki-Universe-Sweet-Simple',
                 'ext': 'mp4',
                 'title': 'Doki-Doki Universe: Sweet, Simple and Genuine (GDC Next 10)'
@@ -27,7 +29,7 @@ class GDCVaultIE(InfoExtractor):
         {
             'url': 'http://www.gdcvault.com/play/1015683/Embracing-the-Dark-Art-of',
             'info_dict': {
-                'id': '1015683',
+                'id': '201203272_1330951438328RSXR',
                 'display_id': 'Embracing-the-Dark-Art-of',
                 'ext': 'flv',
                 'title': 'Embracing the Dark Art of Mathematical Modeling in AI'
@@ -56,7 +58,7 @@ class GDCVaultIE(InfoExtractor):
             'url': 'http://gdcvault.com/play/1023460/Tenacious-Design-and-The-Interface',
             'md5': 'a8efb6c31ed06ca8739294960b2dbabd',
             'info_dict': {
-                'id': '1023460',
+                'id': '840376_BQRC',
                 'ext': 'mp4',
                 'display_id': 'Tenacious-Design-and-The-Interface',
                 'title': 'Tenacious Design and The Interface of \'Destiny\'',
@@ -66,26 +68,38 @@ class GDCVaultIE(InfoExtractor):
             # Multiple audios
             'url': 'http://www.gdcvault.com/play/1014631/Classic-Game-Postmortem-PAC',
             'info_dict': {
-                'id': '1014631',
-                'ext': 'flv',
+                'id': '12396_1299111843500GMPX',
+                'ext': 'mp4',
                 'title': 'How to Create a Good Game - From My Experience of Designing Pac-Man',
             },
-            'params': {
-                'skip_download': True,  # Requires rtmpdump
-                'format': 'jp',  # The japanese audio
-            }
+            # 'params': {
+            #     'skip_download': True,  # Requires rtmpdump
+            #     'format': 'jp',  # The japanese audio
+            # }
         },
         {
             # gdc-player.html
             'url': 'http://www.gdcvault.com/play/1435/An-American-engine-in-Tokyo',
             'info_dict': {
-                'id': '1435',
+                'id': '9350_1238021887562UHXB',
                 'display_id': 'An-American-engine-in-Tokyo',
-                'ext': 'flv',
+                'ext': 'mp4',
                 'title': 'An American Engine in Tokyo:/nThe collaboration of Epic Games and Square Enix/nFor THE LAST REMINANT',
             },
+        },
+        {
+            # Kaltura Embed
+            'url': 'https://www.gdcvault.com/play/1026180/Mastering-the-Apex-of-Scaling',
+            'info_dict': {
+                'id': '0_h1fg8j3p',
+                'ext': 'mp4',
+                'title': 'Mastering the Apex of Scaling Game Servers (Presented by Multiplay)',
+                'timestamp': 1554401811,
+                'upload_date': '20190404',
+                'uploader_id': 'joe@blazestreaming.com',
+            },
             'params': {
-                'skip_download': True,  # Requires rtmpdump
+                'format': 'mp4-408',
             },
         },
     ]
@@ -114,10 +128,8 @@ class GDCVaultIE(InfoExtractor):
         return start_page
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-
-        video_id = mobj.group('id')
-        display_id = mobj.group('name') or video_id
+        video_id, name = re.match(self._VALID_URL, url).groups()
+        display_id = name or video_id
 
         webpage_url = 'http://www.gdcvault.com/play/' + video_id
         start_page = self._download_webpage(webpage_url, display_id)
@@ -127,12 +139,12 @@ class GDCVaultIE(InfoExtractor):
             start_page, 'url', default=None)
         if direct_url:
             title = self._html_search_regex(
-                r'<td><strong>Session Name</strong></td>\s*<td>(.*?)</td>',
+                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
-            head = self._request_webpage(HEADRequest(video_url), video_id)
-            video_url = head.geturl()
+            video_url = self._request_webpage(
+                HEADRequest(video_url), video_id).geturl()
 
             return {
                 'id': video_id,
@@ -141,34 +153,36 @@ class GDCVaultIE(InfoExtractor):
                 'title': title,
             }
 
-        PLAYER_REGEX = r'<iframe src="(?P<xml_root>.+?)/(?:gdc-)?player.*?\.html.*?".*?</iframe>'
-
-        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=(.+?\.xml).*?".*?</iframe>',
-            start_page, 'xml filename', default=None)
-        if xml_name is None:
-            # Fallback to the older format
+        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>'
+
+            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=".*?\?xmlURL=xml/(?P<xml_file>.+?\.xml).*?".*?</iframe>',
+                r'<iframe src=".*?\?xml(?:=|URL=xml/)(.+?\.xml).*?".*?</iframe>',
                 start_page, 'xml filename')
+            embed_url = '%s/xml/%s' % (xml_root, xml_name)
+            ie_key = 'DigitallySpeaking'
 
         return {
             '_type': 'url_transparent',
             'id': video_id,
             'display_id': display_id,
-            'url': '%s/xml/%s' % (xml_root, xml_name),
-            'ie_key': 'DigitallySpeaking',
+            'url': embed_url,
+            'ie_key': ie_key,
         }