From 4ef52ec6950fb331900ba95f2bbf02fd7290fda1 Mon Sep 17 00:00:00 2001
From: yangdx <gzdaniel@me.com>
Date: Sun, 28 Dec 2025 15:16:59 +0800
Subject: [PATCH] Change DOCX extraction to use HTML tags for whitespace

- Replace tabs with HTML em spaces
- Convert all newlines to break tags
---
 lightrag/api/routers/document_routes.py | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/lightrag/api/routers/document_routes.py b/lightrag/api/routers/document_routes.py
index d906aa5cf9..a6360aaa30 100644
--- a/lightrag/api/routers/document_routes.py
+++ b/lightrag/api/routers/document_routes.py
@@ -1015,10 +1015,10 @@ def escape_cell(cell_value: str | None) -> str:
         # CRITICAL: Escape backslash first to avoid double-escaping
         return (
             text.replace("\\", "\\\\")  # Must be first: \ -> \\
-            .replace("\t", "\\t")  # Tab -> \t (visible)
-            .replace("\r\n", "\\n")  # Windows newline -> \n
-            .replace("\r", "\\n")  # Mac newline -> \n
-            .replace("\n", "\\n")  # Unix newline -> \n
+            .replace("\t", "&emsp;&emsp;")  # Tab -> \t (visible)
+            .replace("\r\n", "<br>")  # Windows newline -> \n
+            .replace("\r", "<br>")  # Mac newline -> \n
+            .replace("\n", "<br>")  # Unix newline -> \n
         )
 
     content_parts = []