From e24a2001c5a5a3f3cb9b6f776bfeafb8ed062e20 Mon Sep 17 00:00:00 2001
From: CaptainB <bin@fit2cloud.com>
Date: Thu, 26 Jun 2025 16:23:32 +0800
Subject: [PATCH] feat: refine regex patterns in text_split_handle for improved
 comment detection
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

--bug=1057526 --user=刘瑞斌 【知识库】markdown文件导入知识库，分段详情中代码块展示异常 https://www.tapd.cn/62980211/s/1719131
---
 .../handle/impl/text/text_split_handle.py     | 21 +++++++++----------
 1 file changed, 10 insertions(+), 11 deletions(-)

diff --git a/apps/common/handle/impl/text/text_split_handle.py b/apps/common/handle/impl/text/text_split_handle.py
index b5e6ded3c..d412e1ce0 100644
--- a/apps/common/handle/impl/text/text_split_handle.py
+++ b/apps/common/handle/impl/text/text_split_handle.py
@@ -15,12 +15,14 @@ from charset_normalizer import detect
 from common.handle.base_split_handle import BaseSplitHandle
 from common.utils.split_model import SplitModel
 
-default_pattern_list = [re.compile('(?<=^)# .*|(?<=\\n)# .*'),
-                        re.compile('(?<=\\n)(?<!#)## (?!#).*|(?<=^)(?<!#)## (?!#).*'),
-                        re.compile("(?<=\\n)(?<!#)### (?!#).*|(?<=^)(?<!#)### (?!#).*"),
-                        re.compile("(?<=\\n)(?<!#)#### (?!#).*|(?<=^)(?<!#)#### (?!#).*"),
-                        re.compile("(?<=\\n)(?<!#)##### (?!#).*|(?<=^)(?<!#)##### (?!#).*"),
-                        re.compile("(?<=\\n)(?<!#)###### (?!#).*|(?<=^)(?<!#)###### (?!#).*")]
+default_pattern_list = [
+    re.compile('(?<=^)# (?!-\\*- coding:).*|(?<=\\n)# (?!-\\*- coding:).*'),
+    re.compile('(?<=\\n)(?<!#)## (?!#).*|(?<=^)(?<!#)## (?!#).*'),
+    re.compile("(?<=\\n)(?<!#)### (?!#).*|(?<=^)(?<!#)### (?!#).*"),
+    re.compile("(?<=\\n)(?<!#)#### (?!#).*|(?<=^)(?<!#)#### (?!#).*"),
+    re.compile("(?<=\\n)(?<!#)##### (?!#).*|(?<=^)(?<!#)##### (?!#).*"),
+    re.compile("(?<=\\n)(?<!#)###### (?!#).*|(?<=^)(?<!#)###### (?!#).*")
+]
 
 
 class TextSplitHandle(BaseSplitHandle):
@@ -45,11 +47,8 @@ class TextSplitHandle(BaseSplitHandle):
         try:
             content = buffer.decode(detect(buffer)['encoding'])
         except BaseException as e:
-            return {'name': file.name,
-                    'content': []}
-        return {'name': file.name,
-                'content': split_model.parse(content)
-                }
+            return {'name': file.name, 'content': []}
+        return {'name': file.name, 'content': split_model.parse(content)}
 
     def get_content(self, file, save_image):
         buffer = file.read()