更新文本编码错误

2025-08-08 19:41:16 +08:00 · 2025-08-08 19:41:16 +08:00 · 54751fe89e
commit 54751fe89e
parent 249bfbcb9f
1 changed files with 9 additions and 2 deletions
--- a/AITrain/utils.py
+++ b/AITrain/utils.py
@ -86,5 +86,12 @@ class ReadFiles:
    @classmethod
    def read_text(cls, file_path: str):
        # 读取文本文件
-        with open(file_path, 'r', encoding='utf-8') as file:
-            return file.read()
+        encodings = ['gb18030', 'gbk', 'gb2312', 'big5', 'iso-8859-5', 'cp1252', 'latin1']
+        for enc in encodings:
+            try:
+                with open(file_path, 'r', encoding=enc) as f:
+                    return  f.read()
+            except :
+                continue
+        # with open(file_path, 'r', encoding='utf-8') as file:
+        #     return file.read()