Update common.py

magic_pdf/tools/common.py

 import copy
 import json as json_parse
 import os
+import re
 
 import click
 from loguru import logger
@@ -16,9 +17,13 @@ from magic_pdf.rw.AbsReaderWriter import AbsReaderWriter
 from magic_pdf.rw.DiskReaderWriter import DiskReaderWriter
 from magic_pdf.post_proc.remove_spaces_html import remove_extra_spaces_html_txt
 
+def sanitize_filename(filename: str) -> str:
+    return re.sub(r'[^a-zA-Z0-9_\u4e00-\u9fff-]', '_', filename)
 
 def prepare_env(output_dir, pdf_file_name, method):
-    local_parent_dir = os.path.join(output_dir, pdf_file_name, method)
+    # logger.info(f'pdf_file_name:{pdf_file_name}')
+    pdf_file_name = sanitize_filename(pdf_file_name)
+    local_parent_dir = os.path.join(output_dir, pdf_file_name)
 
     local_image_dir = os.path.join(str(local_parent_dir), 'images')
     local_md_dir = local_parent_dir
@@ -116,6 +121,7 @@ def do_parse(
                                        md_make_mode=f_make_md_mode)
 
     try:
+        pdf_file_name = sanitize_filename(pdf_file_name)
         txt_file = f'{pdf_file_name}.txt'
         md_writer.write(
             content=md_content,