From c74352046b09b7992955d0423dc138de6c5417b3 Mon Sep 17 00:00:00 2001
From: shaohuzhang1 <80892890+shaohuzhang1@users.noreply.github.com>
Date: Wed, 20 Aug 2025 11:36:38 +0800
Subject: [PATCH] fix: The website knowledge base lacks comprehensive methods
 for determining files and folders, resulting in ineffective access to links
 within the final document #2935 (#3893)

---
 apps/common/util/fork.py | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/apps/common/util/fork.py b/apps/common/util/fork.py
index 90527db9c..a8df9f7b2 100644
--- a/apps/common/util/fork.py
+++ b/apps/common/util/fork.py
@@ -3,6 +3,7 @@ import logging
 import re
 import traceback
 from functools import reduce
+from pathlib import Path
 from typing import List, Set
 from urllib.parse import urljoin, urlparse, ParseResult, urlsplit, urlunparse
 
@@ -70,6 +71,8 @@ class Fork:
 
     def __init__(self, base_fork_url: str, selector_list: List[str]):
         base_fork_url = remove_fragment(base_fork_url)
+        if any([True for end_str in ['index.html', '.htm', '.html'] if base_fork_url.endswith(end_str)]):
+            self.base_fork_url = str(Path(base_fork_url).parent)
         self.base_fork_url = urljoin(base_fork_url if base_fork_url.endswith("/") else base_fork_url + '/', '.')
         parsed = urlsplit(base_fork_url)
         query = parsed.query