delete UnicodeDecodeError

Author: joyguoguo

fuzz/collect_fuzz_python.py

@@ -431,12 +431,11 @@ def substitute_one_repo(
         with open(input_path, "rb") as f_input:
             lines = f_input.readlines()
 
-        # 文件已关闭，现在处理数据
-        for line in lines:
-            try:
-                # 尝试解码行以检查内容
+            # 文件已关闭，现在处理数据
+            for line in lines:
+                # 使用 errors='replace' 确保解码不会失败
                 decoded = line.decode("utf-8", errors="replace")
-
+                
                 # 只处理以 b' 或 b" 开头的行
                 if decoded.startswith(("b'", 'b"')):
                     if decoded.startswith("b'") and decoded.endswith("'\n"):
@@ -445,11 +444,11 @@ def substitute_one_repo(
                         byte_data = line[2:-2]
                     else:
                         continue
-
+                    
                     if 0 < len(byte_data) <= max_len:
                         valid_inputs.append(byte_data)
-            except UnicodeDecodeError:
-                if 0 < len(line) <= max_len:
+                # 对于其他行，如果长度在范围内且不是以 b' 或 b" 开头，也考虑加入
+                elif 0 < len(line) <= max_len:
                     valid_inputs.append(line)
 
         if not valid_inputs:
@@ -492,12 +491,19 @@ def substitute_one_repo(
                 with open(out_path, "w") as f_out:
                     f_out.write(new_code)
 
-                # 格式化代码
-                try:
-                    subprocess.run(["black", out_path], check=False)
-                except FileNotFoundError:
-                    logging.warning("Black formatter not found, skipping formatting")
-
+            # 格式化代码
+            formatter_installed = True
+            try:
+                subprocess.run(["black", out_path], 
+                            check=False, 
+                            stdout=subprocess.DEVNULL,  # 隐藏输出
+                            stderr=subprocess.DEVNULL)  # 隐藏错误
+            except FileNotFoundError:
+                if formatter_installed:  # 避免多次记录
+                    logging.warning("Black code formatter not found. For better formatting, install with:")
+                    logging.warning("pip install black")
+                    formatter_installed = False
+                
             except SyntaxError as e:
                 logging.error(f"Syntax error when processing {source_file}: {e}")
             except Exception as e: