今天第一次进行 文件遍历,自己递归写的时候还调试了好久,(主要因为分隔符号的问题),后来发现了os.walk方法,就忍不住和大家分享下.
先看下代码:
import os
for i in os.walk(‘c:‘+os.sep+‘ant‘):
print i[1]
下面是输出:
c:\ant
c:\ant\bin
c:\ant\docs
c:\ant\docs\ant2
c:\ant\docs\antlibs
c:\ant\docs\antlibs\antunit
c:\ant\docs\antlibs\compress
c:\ant\docs\antlibs\dotnet
c:\ant\docs\antlibs\props
c:\ant\docs\antlibs\svn
c:\ant\docs\images
c:\ant\docs\manual
c:\ant\docs\manual\api
c:\ant\docs\manual\api\org
c:\ant\docs\manual\api\org\apache
c:\ant\docs\manual\api\org\apache\tools
c:\ant\docs\manual\api\org\apache\tools\ant
c:\ant\docs\manual\api\org\apache\tools\ant\dispatch
c:\ant\docs\manual\api\org\apache\tools\ant\filters
后面还有很长.
如果不使用这个方法,遍历同样能达到效果.不过使用 os.walk 方便很多了.这个方法返回的是一个三元tupple(dirpath, dirnames, filenames),
其中第一个为起始路径,
第二个为起始路径下的文件夹,
第三个是起始路径下的文件.
dirpath是一个string,代表目录的路径,
dirnames是一个list,包含了dirpath下所有子目录的名字,
filenames是一个list,包含了非目录文件的名字.这些名字不包含路径信息,如果需要得到全路径,需要使用 os.path.join(dirpath, name).
下面是可以看到 os.walk 方法返回的内容.
代码:
import os
for i in os.walk(‘c:‘+os.sep+‘ant‘):
print i
输出:
(‘c:\\ant‘, [‘bin‘, ‘docs‘, ‘etc‘, ‘lib‘, ‘Project‘], [‘fetch.xml‘, ‘get-m2.xml‘, ‘INSTALL‘, ‘KEYS‘, ‘LICENSE‘, ‘NOTICE‘, ‘README‘, ‘WHATSNEW‘])
(‘c:\\ant\\bin‘, [], [‘ant‘, ‘ant.bat‘, ‘ant.cmd‘, ‘antenv.cmd‘, ‘antRun‘, ‘antRun.bat‘, ‘antRun.pl‘, ‘complete-ant-cmd.pl‘, ‘envset.cmd‘, ‘lcp.bat‘, ‘runant.pl‘, ‘runant.py‘, ‘runrc.cmd‘])
(‘c:\\ant\\docs‘, [‘ant2‘, ‘antlibs‘, ‘images‘, ‘manual‘, ‘projects‘, ‘slides‘, ‘webtest‘], [‘antnews.html‘, ‘ant_in_anger.html‘, ‘ant_task_guidelines.html‘, ‘appendix_e.pdf‘, ‘breadcrumbs.js‘, ‘bugs.html‘, ‘bylaws.html‘, ‘contributors.html‘, ‘external.html‘, ‘faq.html‘, ‘favicon.ico‘, ‘index.html‘, ‘legal.html‘, ‘LICENSE‘, ‘license.html‘, ‘mail.html‘, ‘mission.html‘, ‘nightlies.html‘, ‘page.css‘, ‘problems.html‘, ‘projects.html‘, ‘resources.html‘, ‘svn.html‘])
(‘c:\\ant\\docs\\ant2‘, [], [‘actionlist.html‘, ‘features.html‘, ‘FunctionalRequirements.html‘, ‘original-specification.html‘, ‘requested-features.html‘, ‘requested-features.txt‘, ‘VFS.txt‘])
(‘c:\\ant\\docs\\antlibs‘, [‘antunit‘, ‘compress‘, ‘dotnet‘, ‘props‘, ‘svn‘], [‘bindownload.cgi‘, ‘bindownload.html‘, ‘charter.html‘, ‘index.html‘, ‘proper.html‘, ‘sandbox.html‘, ‘srcdownload.cgi‘, ‘srcdownload.html‘])
(‘c:\\ant\\docs\\antlibs\\antunit‘, [], [‘index.html‘])
(‘c:\\ant\\docs\\antlibs\\compress‘, [], [‘index.html‘])
(‘c:\\ant\\docs\\antlibs\\dotnet‘, [], [‘index.html‘])
(‘c:\\ant\\docs\\antlibs\\props‘, [], [‘index.html‘])
...
当然后面还有很长了.
有了这个函数无论是遍历文件夹,还是遍历文件都很方便.
下面是我是自己用递归实现的遍历文件方法.
代码:
def listdir(leval,path):
for i in os.listdir(path):
print(‘| ‘*(leval + 1) + i)
if os.path.isdir(path+i):
listdir(leval+1, path+i)
path = ‘c:‘+os.sep+‘ant‘
#或者直接 path=‘C:/ant‘
print(path+os.sep)
listdir(0, path+os.sep)
下面是输出:
c:\ant\
| bin
| | ant
| | ant.bat
| | ant.cmd
| | antenv.cmd
| | antRun
| | antRun.bat
| | antRun.pl
| | complete-ant-cmd.pl
| | envset.cmd
| | lcp.bat
| | runant.pl
| | runant.py
| | runrc.cmd
| docs
| | ant2
| | antlibs
| | antnews.html
| | ant_in_anger.html
| | ant_task_guidelines.html
| | appendix_e.pdf
| | breadcrumbs.js
| | bugs.html
| | bylaws.html
| | contributors.html
| | external.html
| | faq.html
| | favicon.ico
| | images
| | index.html
| | legal.html
| | LICENSE
| | license.html
| | mail.html
| | manual
| | mission.html
| | nightlies.html
| | page.css
| | problems.html
| | projects
| | projects.html
| | resources.html
| | slides
| | svn.html
| | webtest
| etc
| | ant-bootstrap.jar
| | changelog.xsl
| | checkstyle
| | coverage-frames.xsl
| | jdepend-frames.xsl
| | jdepend.xsl
| | junit-frames-xalan1.xsl
| | junit-frames.xsl
| | junit-noframes.xsl
| | log.xsl
| | maudit-frames.xsl
| | mmetrics-frames.xsl
| | tagdiff.xsl
| fetch.xml
| get-m2.xml
| INSTALL
| KEYS
| lib
| | ant-1.8.0.pom
| | ant-1.8.0.pom.md5
| | ant-1.8.0.pom.sha1
| | ant-1.8.0.pom.sha512
..
如果只想得到文件夹,而不要文件,把要做的事情放到
if os.path.isdir(path+i):
里面就好了,比如: print()
O(∩_∩)O~