fix: parse_file避免内存泄漏

cp_app.py

@@ -411,55 +411,62 @@ def ask(input:str, p_name:str, stream=False):
 def parse_file(file_content, file_type):
     try:
         if file_type == "pdf":
-            # 将文件内容转换为字节流
             pdf_bytes = file_content.read()
             pdf_stream = io.BytesIO(pdf_bytes)
+            doc = None
+            try:
                 doc = fitz.open(stream=pdf_stream, filetype="pdf")
                 text_content = ""
                 
-            # 首先尝试直接提取文本
                 for page_num in range(len(doc)):
                     page = doc[page_num]
                     text_content += page.get_text() + "\n"
                 
                 t_plain = text_content.strip()
-            doc.close()
                 if t_plain:
                     return t_plain, None
                 else:
                     return None, "无法直接提取文本，请使用 OCR 处理"
+            finally:
+                if doc:
+                    doc.close()
+                pdf_stream.close()  # 显式关闭流
 
         elif file_type == "docx":
-            # 将文件内容转换为字节流
             doc_stream = io.BytesIO(file_content.read())
+            try:
                 doc = Document(doc_stream)
-
-            # 提取所有段落的文本
                 text_content = ""
+                
                 for paragraph in doc.paragraphs:
                     text_content += paragraph.text + "\n"
                 
-            # 提取表格中的文本
                 for table in doc.tables:
                     for row in table.rows:
                         for cell in row.cells:
                             text_content += cell.text + " "
                         text_content += "\n"
                 return text_content, None
+            finally:
+                doc_stream.close()  # 确保流被关闭
 
         elif file_type == "doc":
             file_name = f'{uuid.uuid4()}.doc'
             file_path = os.path.join(CURRENT_DIR, file_name)
+            try:
                 file_content.save(file_path)
                 completed = subprocess.run(['catdoc', file_path], capture_output=True, text=True)
-            os.remove(file_path)
                 if completed.returncode != 0:
                     return None, completed.stderr
                 return completed.stdout, None
+            finally:
+                if os.path.exists(file_path):
+                    os.remove(file_path)  # 确保临时文件被删除
 
         elif file_type == "txt":
             text_content = file_content.read().decode("utf-8")
             return text_content, None
+            
         return None, "不支持的文件类型"
     except Exception as e:
         return None, f"文件解析错误: {str(e)}"