From 422f42ca7faa1d0834b1b503e87d605ad55f1ef8 Mon Sep 17 00:00:00 2001
From: oobabooga <112222186+oobabooga@users.noreply.github.com>
Date: Sat, 4 Apr 2026 23:51:15 -0700
Subject: [PATCH] Pre-compile LaTeX regex in html_generator.py

---
 modules/html_generator.py | 11 ++++++++---
 1 file changed, 8 insertions(+), 3 deletions(-)

diff --git a/modules/html_generator.py b/modules/html_generator.py
index 8dd46850..e3ebea8d 100644
--- a/modules/html_generator.py
+++ b/modules/html_generator.py
@@ -14,6 +14,13 @@ from modules.reasoning import extract_reasoning
 from modules.sane_markdown_lists import SaneListExtension
 from modules.utils import get_available_chat_styles
 
+# Pre-compiled regex for protecting markdown-sensitive characters inside LaTeX.
+# Covers $$...$$, \[...\], \(...\), and inline $...$ (when content contains \\).
+_LATEX_PATTERN = re.compile(
+    r'((?:^|[\r\n\s])\$\$[^`]*?\$\$)|\\\[(.*?)\\\]|\\\((.*?)\\\)|(?<!\$)\$(?!\$)([^\$\n]*\\\\[^\$\n]*?)\$(?!\$)',
+    re.DOTALL
+)
+
 # This is to store the paths to the thumbnails of the profile pictures
 image_cache = {}
 
@@ -241,9 +248,7 @@ def process_markdown_content(string):
     string = re.sub(r"(.)```", r"\1\n```", string)
 
     # Protect asterisks and underscores within all LaTeX blocks before markdown conversion
-    latex_pattern = re.compile(r'((?:^|[\r\n\s])\$\$[^`]*?\$\$)|\\\[(.*?)\\\]|\\\((.*?)\\\)|(?<!\$)\$(?!\$)([^\$\n]*\\\\[^\$\n]*?)\$(?!\$)',
-                               re.DOTALL)
-    string = latex_pattern.sub(protect_asterisks_underscores_in_latex, string)
+    string = _LATEX_PATTERN.sub(protect_asterisks_underscores_in_latex, string)
 
     result = ''
     is_code = False