From f137e4c27ccaecae3fd0acb58a01e8e5faccaa88 Mon Sep 17 00:00:00 2001
From: pukkandan <pukkandan.ytdlp@gmail.com>
Date: Sun, 19 Sep 2021 17:45:41 +0530
Subject: [PATCH] [utils] Improve `extract_timezone` Code taken from:
 https://github.com/ytdl-org/youtube-dl/pull/29845 Fixes:
 https://github.com/ytdl-org/youtube-dl/issues/29948 Authored by: dirkf

---
 yt_dlp/utils.py | 12 ++++++++++--
 1 file changed, 10 insertions(+), 2 deletions(-)

diff --git a/yt_dlp/utils.py b/yt_dlp/utils.py
index 983ca6ced..4c0ac5a25 100644
--- a/yt_dlp/utils.py
+++ b/yt_dlp/utils.py
@@ -3034,8 +3034,16 @@ def redirect_request(self, req, fp, code, msg, headers, newurl):
 
 def extract_timezone(date_str):
     m = re.search(
-        r'^.{8,}?(?P<tz>Z$| ?(?P<sign>\+|-)(?P<hours>[0-9]{2}):?(?P<minutes>[0-9]{2})$)',
-        date_str)
+        r'''(?x)
+            ^.{8,}?                                              # >=8 char non-TZ prefix, if present
+            (?P<tz>Z|                                            # just the UTC Z, or
+                (?:(?<=.\b\d{4}|\b\d{2}:\d\d)|                   # preceded by 4 digits or hh:mm or
+                   (?<!.\b[a-zA-Z]{3}|[a-zA-Z]{4}|..\b\d\d))     # not preceded by 3 alpha word or >= 4 alpha or 2 digits
+                   [ ]?                                          # optional space
+                (?P<sign>\+|-)                                   # +/-
+                (?P<hours>[0-9]{2}):?(?P<minutes>[0-9]{2})       # hh[:]mm
+            $)
+        ''', date_str)
     if not m:
         timezone = datetime.timedelta()
     else: