Address issue with links for Google Sites connector

onyx-dot-app · Nov 3, 2023 · 329824a · 329824a · vercel · Nov 3, 2023
1 parent b0f76b9
commit 329824a
Showing 1 changed file with 4 additions and 1 deletion.
diff --git 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