From aa351e27e20f1724fcb9d8fc25f1fc3776127be5 Mon Sep 17 00:00:00 2001
From: Alain St-Denis <alain@zenfolie.org>
Date: Sun, 29 Oct 2023 10:03:09 -0400
Subject: [PATCH] fix url regex

---
 federation/utils/text.py | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/federation/utils/text.py b/federation/utils/text.py
index ab4b8ec..bf1f235 100644
--- a/federation/utils/text.py
+++ b/federation/utils/text.py
@@ -11,7 +11,8 @@ TAG_PATTERN = re.compile(r'(#[\w\-]+)([)\]_!?*%/.,;\s]+\s*|\Z)', re.UNICODE)
 # This will match non matching braces. I don't think it's an issue.
 MENTION_PATTERN = re.compile(r'(@\{?(?:[\w\-. \u263a-\U0001f645]*; *)?[\w]+@[\w\-.]+\.[\w]+}?)', re.UNICODE)
 # based on https://stackoverflow.com/a/6041965
-URL_PATTERN = re.compile(r'((?:(?:https?|ftp)://|^|(?<=[("<\s]))+(?:[\w\-]+(?:(?:\.[\w\-]+)+))(?:[\w.,@?^=%&:\/~+#-]*[\w@?^=%&\/~+#-]))',
+URL_PATTERN = re.compile(r'((?:(?:https?|ftp)://|^|(?<=[("<\s]))+(?:[\w\-]+(?:(?:\.[\w\-]+)+))'
+                         r'[\w.,;:@?!$()*^=%&/~+\-#]*(?<![:;,).>"]))',
                          re.UNICODE)
 
 def decode_if_bytes(text):