【转】如何编写Python脚本替换文件中的多行字符

 在大概3个月之前,Python对我来说一直是个迷。然而,就在3个月前我经理给我一个任务——删除(替换)所有项目源码文件中包含特定几行内容的所有注释。整个项目源码的大小有1G,在Linux服务器(中高档)上编译需要半个多小时,可见代码量之大,不可能手动去一个一个改。肯定得用脚本去处理,于是我想到了Python。在这之前没有接触过Python,花了2个星期一顿恶补之后,总算顺利交差了。

一直很想和大家分享一下碰到的问题及我如何解决的(可能我的方案并不好,但是他能够解决我的问题),但一直拖到现在是因为我感觉我还对Python的了解还不够。因为要在短时间内完成上面交下来的任务,在学习Python的时候,都是走马观花,对解决自己的问题不相关的直接跳过,看资料也静不下心,脑海里都是问题。前几天我静下心把Python的书从头到尾浏览了一遍,感觉现在是时候要进行总结了。

本文的主要内容如下:

问题描述 
解题思路 
代码实现 
Python的特点 
1、问题描述
项目源码很大,属于C/C++混合的那种,编程风格也很多样,有'.c'、'.cc'、'cpp'、'.h'、'.hh'等文件。我要完成的任务是:把包含特定几行内容的注释删掉,如(声明:下面的内容只是我随便举的一个例子,项目源码中不涉及下面的内容。)

/* 
* Copyright 2002 Sun Microsystems, Inc. All rights reserved.

*

* Redistribution and use in source and binary forms, with or without

* modification, are permitted provided that the following conditions

* are met:

*

* - Redistributions of source code must retain the above copyright

* notice, this list of conditions and the following disclaimer.

*

* - Redistribution in binary form must reproduce the above copyright

* notice, this list of conditions and the following disclaimer in

* the documentation and/or other materials provided with the

* distribution.

*

* Neither the name of Sun Microsystems, Inc. or the names of

* contributors may be used to endorse or promote products derived

* from this software without specific prior written permission.

*/

但是格式有很多种,如有的在“ Copyright 2002 Sun Microsystems, Inc. All rights reserved.”前面有一段关于本源码文件的描述、有的在“from this software without specific prior written permission.”后面有一段关于本源码文件的描述、有的是C++风格的注释用"//",而不是“/**/”、还有的没有

“ * - Redistribution in binary form must reproduce the above copyright

* notice, this list of conditions and the following disclaimer in

* the documentation and/or other materials provided with the

* distribution.”等等还有其他一些。总之一句话,我要删除的包含特定几行内容的注释有很多中格式!

于是我决定要用Python来编写脚本处理。要匹配特定的内容,我想到了用正则表达式,但苦于不知道如何去构建正则来匹配上面描述的内容(您知道的话,希望能够告诉我)!我只有另辟路径了。

2、解题思路
我的思路——要删除所有项目源码中包含特定几行内容的注释,脚本要满足以下几点功能:

脚本要能够遍历所有的源码文件('.c'、'.cc'、'cpp'、'.h'、'.hh'),并只处理上面的几种类型的文件 
找出包含特定几行内容的注释,并删除之 
能够处理一些特殊情况,如软连接文件 
上面的几点的处理步骤可以表示如下:

Step 1:输入要处理源码文件夹名,或者源码文件名;

Step 2:如果是文件名,检查文件的类型是否为'.c'、'.cc'、'cpp'、'.h'、'.hh',否则不处理;

Step 3:检查文件是否是软连接,如果是软连接则不处理;

Step 4:查找文件中是否存在匹配的注释,存在则删掉,否则不处理;

Step 5:如果是文件夹,则对文件夹中的每个文件、文件夹进行处理,转Step2.

思路很明确,关键是如何查找文件中是否包含匹配的内容,并删除!还有就是,对于一个没用过Python等脚本语言的人来说,如何编码实现也是一个问题!

如何确定注释是否为包含特定几行内容的注释?我的思路如下:(因为正则表达式学的不好,只有通过下面的方法了)

如果是/*、//则记录下当前的文件行数,即行号startLine 
以行为单位查找是否存在特定的几行,如“ Copyright 2002 Sun Microsystems, Inc. All rights reserved.”等等 
直到遇到*/,或注释结束了(对于//)。如果存在,则记录下注释结束的行号endLine 
最后,删掉这从startLine ~ endLine的内容。 
3、代码实现
废话我不多说了,直接按照上面的实例实现代码,如果你对Python不熟,请参阅相关资料。
#!/usr/bin/env python
#
Filename: comment.py

import os, sys, fileinput

#-------------------------------------------------------------
def usage():
    
print u'''
    help: comment.py <filename | dirname>

    [dirname]: Option, select a directory to operate
    [filename]: Option, select a file to operate

    Example: python comment.py /home/saylor/test
    
'''
#--------------------------------------------------------------
def commentFile(src, fileList):
    
'''
    description: comment files
    param src: Operate file name
    
'''
    
#if file exist?
    ifnot os.path.exists(src):
        
print'Error: file - %s doesn\'t exist.'% src
        return False
    
if os.path.islink(src):
        
print'Error: file - %s is just a link, will not handle it.'
        
return False
    filetype 
= (os.path.splitext(src))[1]
    
ifnot filetype in ['.c','.h']:
        
return False
    
try:
        
ifnot os.access(src, os.W_OK):
            os.chmod(src, 
0664)
    
except:
        
print'Error: you can not chang %s\'s mode.'% src
    try:
        inputf 
= open(src, 'r')
        outputfilename 
= src +'.tmp'
        outputf 
= open(outputfilename, 'w')
    beginLine 
= 0
    endLine 
=100000000
    isMatched 
= False

    
#-----find the beginLine and endLine -------------------
        for eachline in fileinput.input(src):
        
if eachline.find('/*'>= 0:
        beginLine 
= fileinput.lineno()
        
if eachline.find('Copyright 2002 Sun Microsystems, Inc. All rights reserved.'>= 0:
            isMatched 
= True
        
if eachline.find('*/'>= 0 and isMatched:
        endLine 
= fileinput.lineno()
        
break
    
    
#-----delete the content between beginLine and endLine-----
    print beginLine, endLine
    lineNo 
=1
    
for eachline in inputf:
        
if lineNo < beginLine:
        
print eachline
        outputf.write(eachline)
        
elif lineNo > endLine:
        
print eachline
        outputf.write(eachline)
        lineNo 
= lineNo +1
        
        inputf.close()
        outputf.close()
        os.rename(outputfilename, src)
        fileList.append(src)
    
except:
        
print'Error: unexcept error.'
        inputf.close()
        outputf.close()
    
return True

#--------------------------------------------------------------
def commentDir(src, fileList):
    
'''
    description:
         comment files in src(dir)
    param src:
         operate files in src(dir)
    
'''
    
#if dir exist?
    ifnot os.path.exists(src):
        
print'Error: dir - %s is not exist.'%s (src)
        
return False
    filelists 
= os.listdir(src)
    
for eachfile in filelists:
        eachfile 
= src +'/'+eachfile
        
if os.path.isdir(eachfile):
            commentDir(eachfile, fileList)
        
elif os.path.isfile(eachfile):
            commentFile(eachfile, fileList)
    
return True

#--------------------------------------------------------------
def main():
    
if len(sys.argv) <2:
        usage()
        sys.exit(
1)
    src 
= sys.argv[1]
    
if os.path.isdir(src):
        dire 
= os.path.abspath(src)
        dirFlag 
= True
    
elif os.path.isfile(src):
        fl 
= os.path.abspath(src)
        dirFlag 
= False
    
else:
        
print'Error'
    fileList 
= []
    
if dirFlag:
        commentDir(dire, fileList)
    
else:
        commentFile(fl, fileList)
    
if fileList:
        
print'Successful handle file: ...'
        
for eachfile in fileList:
            
print eachfile
    
print'Done'
    
return True

#--------------------------------------------------------------
if__name__=='__main__':
    main()

posted @ 2012-10-19 23:02  swordzj  阅读(1177)  评论(0编辑  收藏  举报