Python3.9使用fileinput模块遇编码及参数错误求助

错误原因分析 #

1. TypeError：Python 3.9的fileinput.FileInput构造函数不支持直接传入encoding参数，该参数是Python 3.10版本才新增的特性。
2. UnicodeDecodeError：你的XML文件编码并非标准UTF-8，大概率是带BOM的UTF-8（0xef是UTF-8 BOM的起始字节）或GBK等中文编码，导致用UTF-8解码失败。

解决方案 #

步骤1：替换encoding参数为openhook（解决TypeError） #

在Python 3.9中，需通过fileinput.hook_encoded()作为openhook参数来指定文件编码，而非直接传encoding。

步骤2：匹配文件实际编码（解决UnicodeDecodeError） #

优先尝试以下常见编码：

• utf-8-sig：适配带BOM的UTF-8文件（针对你遇到的0xef字节问题最可能有效）
• gbk：中文Windows环境下的常见编码
• 若不确定编码，可使用chardet库自动检测（需先执行pip install chardet安装）

修改后的代码示例 #

示例1：针对带BOM的UTF-8文件 #

import fileinput
import os

FILENAME = []
# 修复：原代码只收集文件名，子目录文件会找不到，需拼接完整路径
for root, dirs, files in os.walk(r'.', topdown=False):
    for filename in files:
        if filename.endswith('.xml'):
            full_path = os.path.join(root, filename)
            FILENAME.append(full_path)
            print(full_path)

codeskin = input('Enter Code:')
Prefab = 'prefab_skill_effects/hero_skill_effects/'
codeskinmod = 'prefab_skill_effects/hero_skill_effects/'
Word = Prefab + codeskin
CarNew = input('Enter the Character To Replace:')
CarNews = codeskinmod + CarNew

# 用hook_encoded指定编码，替代直接传encoding参数
with fileinput.FileInput(FILENAME, inplace=True, backup='.bak', openhook=fileinput.hook_encoded('utf-8-sig')) as file:
    for line in file:
        print(line.replace(Word, CarNews), end='')
# with语句会自动关闭文件，无需手动调用file.close()

示例2：自动检测文件编码 #

import fileinput
import os
import chardet

# 检测单个文件的编码
def detect_encoding(file_path):
    with open(file_path, 'rb') as f:
        result = chardet.detect(f.read())
    return result['encoding']

FILENAME = []
for root, dirs, files in os.walk(r'.', topdown=False):
    for filename in files:
        if filename.endswith('.xml'):
            full_path = os.path.join(root, filename)
            FILENAME.append(full_path)
            print(full_path)

codeskin = input('Enter Code:')
Prefab = 'prefab_skill_effects/hero_skill_effects/'
codeskinmod = 'prefab_skill_effects/hero_skill_effects/'
Word = Prefab + codeskin
CarNew = input('Enter the Character To Replace:')
CarNews = codeskinmod + CarNew

# 自定义openhook，根据每个文件的编码打开
def custom_openhook(filename, mode):
    encoding = detect_encoding(filename)
    return open(filename, mode, encoding=encoding, errors='ignore')

with fileinput.FileInput(FILENAME, inplace=True, backup='.bak', openhook=custom_openhook) as file:
    for line in file:
        print(line.replace(Word, CarNews), end='')

内容的提问来源于stack exchange，提问作者Khánh Đàm