From 329824ab22ce3a68321d7244fe2fca83262c9633 Mon Sep 17 00:00:00 2001
From: Weves <chrisweaver101@gmail.com>
Date: Thu, 2 Nov 2023 21:48:55 -0700
Subject: [PATCH] Address issue with links for Google Sites connector

---
 backend/danswer/connectors/google_site/connector.py | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/backend/danswer/connectors/google_site/connector.py b/backend/danswer/connectors/google_site/connector.py
index c92040cbb95..6a11bee16f8 100644
--- a/backend/danswer/connectors/google_site/connector.py
+++ b/backend/danswer/connectors/google_site/connector.py
@@ -1,4 +1,5 @@
 import os
+import re
 import urllib.parse
 from typing import Any
 from typing import cast
@@ -29,7 +30,9 @@ def process_link(element: BeautifulSoup | Tag) -> str:
     href = urllib.parse.unquote(href)
     href = href.rstrip(".html").lower()
     href = href.replace("_", "")
-    href = href.replace(" ", "-")
+    href = re.sub(
+        r"([\s-]+)", "-", href
+    )  # replace all whitespace/'-' groups with a single '-'
 
     return href