fix underscores inside links

files/helpers/regex.py

@@ -1,5 +1,6 @@
 import random
 import re
+import regex
 from random import choice, choices
 from typing import List, Optional, Union
 
@@ -193,3 +194,5 @@ def command_regex_matcher(match, upper=False):
 reason_regex = re.compile('(/(post|comment)/[0-9]+)', flags=re.A)
 
 discord_username_regex = re.compile("(\s|^|>).{2,32}#[0-9]{4}(?=[^0-9]|$)", flags=re.A)
+
+underscore_in_link_regex = regex.compile("(?<=https:\/\/[^\s)]+)_", flags=regex.A)

files/helpers/sanitize.py

@@ -291,7 +291,10 @@ def sanitize(sanitized, golden=True, limit_pings=0, showmore=True, count_marseys
 	if FEATURES['MARKUP_COMMANDS']:
 		sanitized = command_regex.sub(command_regex_matcher, sanitized)
 
+	sanitized = underscore_in_link_regex.sub(r'\_', sanitized)
+
 	sanitized = markdown(sanitized)
+
 	sanitized = strikethrough_regex.sub(r'\1<del>\2</del>', sanitized)
 
 	# replacing zero width characters, overlines, fake colons

migrations/20230127-fix-underscores-in-links.sh

@@ -0,0 +1 @@
+pip3 install regex

requirements.txt

@@ -23,6 +23,7 @@ pyotp
 pywebpush
 qrcode
 redis
+regex
 requests
 SQLAlchemy
 tinycss2