From fd61f317bf71b91abbc5cd256027877b70f6dbce Mon Sep 17 00:00:00 2001
From: Tim Mann <tim@tim-mann.org>
Date: Sat, 13 Feb 2021 16:26:33 -0800
Subject: [PATCH] Step 1 of a rewrite to find and parse embedded JSON instead
of just running a regexp over the whole page. This version passes the tests,
but more work is needed.
---
youtube_dl/extractor/pac12.py | 14 +++++++++++---
1 file changed, 11 insertions(+), 3 deletions(-)
diff --git a/youtube_dl/extractor/pac12.py b/youtube_dl/extractor/pac12.py
index 534261e15..07da749d3 100644
--- a/youtube_dl/extractor/pac12.py
+++ b/youtube_dl/extractor/pac12.py
@@ -1,5 +1,7 @@
# coding: utf-8
from __future__ import unicode_literals
+from __future__ import print_function #XXX
+import pprint #XXX
import re
@@ -32,9 +34,15 @@ class Pac12IE(InfoExtractor):
def _real_extract(self, url):
video_id = self._match_id(url)
webpage = self._download_webpage(url, video_id)
- video_url = \
- self._search_regex(r'"manifest_url":"(?P<url>https:[^"]+)"',
- webpage, 'url', group='url', default=None)
+
+ drupal_settings = self._parse_json(
+ self._search_regex(
+ r'<script[^>]+type="application/json"[^>]*data-drupal-selector="drupal-settings-json">([^<]+)</script>',
+ webpage, 'drupal settings'), video_id)
+ pprint.pprint(drupal_settings.get('currentVideo'))
+
+ video_url = drupal_settings.get('currentVideo', {}).get('manifest_url')
+
vod_url = None
if (video_url is None) or ('vod-' not in url):
vod_url = self._search_regex(r'(https?://(?:embed\.)?pac-12\.com/(?:embed/)?vod-[0-9a-zA-Z]+)',