python文本编码转换_python转换文件编码

每次在更新主题或者是wordpress的时候，经常会出现乱码，发现是文件的编码格式有问题，都不是utf-8格式，但是因为文件太多，懒得手动去修改，固写一个脚本程序来修改整个目录。

#-*- coding: utf-8 -*-

__author__ = 'L'

import sys

import os

import stat

import codecs

import chardet

import shutil

utf8_encoder = codecs.lookup('utf-8')

def change_file_code_to_uft8(path):

for item in os.listdir(path):

sub_path = os.path.join(path, item)

mode = os.stat(sub_path)[stat.ST_MODE]

if stat.S_ISDIR(mode):

continue

#change_file_code_to_uft8(sub_path)

else:

if (False == sub_path.endswith('.bak')):

print(sub_path)

with open(sub_path, "rb") as file_handle:

f = chardet.detect(file_handle.read())

print(f)

#print(f['encoding'])

if ('utf-8' != f['encoding'] and False == sub_path.endswith('.bak')):

new_file = sub_path + '.bak'

if (not os.path.exists(new_file)):

shutil.copyfile(sub_path, new_file)

#file_handle = codecs.open(sub_path, 'r', 'gbk')

file_handle = codecs.open(sub_path, 'r', f['encoding'])

try:

context = file_handle.read()

utf8_context = utf8_encoder.encode(context, 'ignore')

with open(sub_path, "wb") as file_handle:

file_handle.write(utf8_context[0])

except UnicodeDecodeError as err:

s = sys.exc_info()

print("Error '%s' happended on line %d" % (s[1], s[2].tb_lineno))

pass

if (False == sub_path.endswith('.bak')):

with open(sub_path, "rb") as file_handle:

f = chardet.detect(file_handle.read())

print(f['encoding'])

if len(sys.argv) >= 2:

current_path = sys.argv[1]

else:

print('please input path.')

sys.exit(0)

change_file_code_to_uft8(current_path)

上面的代码中没有使用递归修改，因为我只想修改目录下面的文件，如果你想递归修改，则可以把continue修改成下面注释的语句。而且对于二进制文件修改起来有问题，最好只是修改文本文件的编码。当然也可以指定后缀名来修改。怕修改文件可能出现错误，故在修改之前，先做了一下备份，如果修改之后没有问题，可以再写个脚本删除之，或者也可以直接在转化成功之后删除。