python批量删除注释_删除c,c++,java源文件中全部注释的Python脚本

#!D:\Python32

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

# 过滤JAVA程序中的注释

# 如果字符串中有注释符号的话会有问题。

import os

import re

import io

# 改这个目录！！！

top_dir = "E:\\work2\\";

# 状态

S_INIT = 0;

S_SLASH = 1;

S_BLOCK_COMMENT = 2;

S_BLOCK_COMMENT_DOT = 3;

S_LINE_COMMENT = 4;

S_STR = 5;

S_STR_ESCAPE = 6;

def trim_dir(path):

print("dir：" + path);

for root, dirs, files in os.walk(path):

for name in files:

trim_file(os.path.join(root, name))

#for name in dirs:

#trim_dir(os.path.join(root, name))

def trim_file(path):

print("file:" + path);

if re.match(r".*?\.(java|c|cpp|h)$", path):

print("process");

else:

print("ignore");

return;

bak_file = path + ".bak";

try:

os.rename(path, bak_file);

except:

print "bak except",bak_file;

fp_src = open(bak_file);

fp_dst = open(path, 'w');

state = S_INIT;

for line in fp_src.readlines():

for c in line:

if state == S_INIT:

if c == '/':

state = S_SLASH;

elif c == '"':

state = S_STR;

fp_dst.write(c);

else:

fp_dst.write(c);

elif state == S_SLASH:

if c == '*':

state = S_BLOCK_COMMENT;

elif c == '/':

state = S_LINE_COMMENT;

else:

fp_dst.write('/');

fp_dst.write(c);

state=S_INIT;

elif state == S_BLOCK_COMMENT:

if c == '*':

state = S_BLOCK_COMMENT_DOT;

elif state == S_BLOCK_COMMENT_DOT:

if c == '/':

state = S_INIT;

elif c=='*':

state=S_BLOCK_COMMENT_DOT; #再次碰到*号还是要继续状态，否则会出错

else:

state = S_BLOCK_COMMENT;

elif state == S_LINE_COMMENT:

if c == '\n':

state = S_INIT;

fp_dst.write(c);

elif state == S_STR:

if c == '\\':

state = S_STR_ESCAPE;

elif c == '"':

state = S_INIT;

fp_dst.write(c);

elif state == S_STR_ESCAPE:

# 这里未完全实现全部序列，如\oNNN \xHH \u1234 \U12345678，但没影响

state = S_STR;

fp_dst.write(c);

fp_src.close();

fp_dst.close();

#os.remove(bak_file);

trim_dir(top_dir);

但是里面有两个问题，会引起错误，现在经验证可以在我的项目中使用，但是不保证完全没有错误，有错误希望和我联系，也可以和原作者联系