python fnmatch & glob
1,转载:Python模块学习 - fnmatch & glob - Dahlhin - 博客园 (cnblogs.com)
介绍
fnmatch 和 glob 模块都是用来做字符串匹配文件名的标准库。
fnmatch模块
大部分情况下使用字符串匹配查找特定的文件就能满足需求,如果需要更加灵活的字符串匹配,就没有办法了,这里介绍标准库fnmatch。这个库专门用来做文件名匹配
fnmatch支持的通配符
fnmatch支持的通配如下:
| 通配符 | 含义 |
| * | 匹配任何数量的字符 |
| ? | 匹配单个字符 |
| [seq] | 匹配seq中的字符 |
| [!seq] | 匹配除seq以外的任何字符 |
fnmatch的基本使用
fnmatch这个库相对比较简单,只有4个函数,分别是fnmatch、fnmatchcase、filter和translate,其中最常用的是fnmatch。主要功能如下:
-
- fnmatch:判断文件名是否符合特定的模式。
- fnmatchcase:判断文件名是否符合特定的模式,区分大小写。
- filter:返回输入列表中,符合特定模式的文件名列表。
- translate:将通配符模式转换成正则表达式。
例子
fnmatch和fnmatchcase用法相同,判断名称是否符合表达式,返回True or False
|
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
|
>>> os.listdir(os.curdir)['A1.jpg', 'a1.txt', 'a2.txt', 'aA.txt', 'b3.jpg', 'b2.jpg', 'b1.jpg']>>> [ name for name in os.listdir(os.curdir) if fnmatch.fnmatch(name,'*.jpg') ]['A1.jpg', 'b3.jpg', 'b2.jpg', 'b1.jpg']>>> [ name for name in os.listdir(os.curdir) if fnmatch.fnmatch(name,"[ab]*") ]['a1.txt', 'a2.txt', 'aA.txt', 'b3.jpg', 'b2.jpg', 'b1.jpg']>>> [ name for name in os.listdir(os.curdir) if fnmatch.fnmatch(name,"[!a]*") ]['A1.jpg', 'b3.jpg', 'b2.jpg', 'b1.jpg']>>> [ name for name in os.listdir(os.curdir) if fnmatch.fnmatch(name,"b?.jpg") ]['b3.jpg', 'b2.jpg', 'b1.jpg']>>> [ name for name in os.listdir(os.curdir) if fnmatch.fnmatchcase(name,"A?.jpg") ]['A1.jpg'] |
filter和fnmatch类似,只不过filter接受的第一个参数是一个文件名列表,返回符合表达式的列表(即:筛选)
|
1
2
3
4
5
6
|
>>> name = os.listdir(os.curdir)>>> name['A1.jpg', 'a1.txt', 'a2.txt', 'aA.txt', 'b3.jpg', 'b2.jpg', 'b1.jpg']>>> fnmatch.filter(name,'*.txt')['a1.txt', 'a2.txt', 'aA.txt']>>> |
glob模块
我们前面的fnmatch模块,都是利用os.listdir获取文件列表,然后通过字符串fnmatch模块进行文件名匹配的,而在Python中还有更加简单的方式,即使用glob库。
glob的作用就相当于os.listdir 加上 fnmatch。使用glob以后就不用使用os.listdir获取文件列表了。
glob比fnmatch更简单,因为他只有 glob,iglob,escape三个函数。
glob基本使用
glob和iglob的区别在于glob返回的是一个列表,iglob返回的是一个生成器对象
|
1
2
3
4
5
6
7
8
9
10
11
|
>>> import glob>>> glob.glob('*.txt')['a1.txt', 'a2.txt', 'aA.txt']>>> g = glob.iglob('*.txt') # 使用iglob返回的是一个生成器>>> g<generator object _iglob at 0x1050bbba0>>>> list(g)['a1.txt', 'a2.txt', 'aA.txt']>>> |
PS:glob同样支持通配符和fnmatch相同,这里不在列举,并且在通配符表达式中支持路径
|
1
2
3
|
>>> glob.glob('/Users/DahlHin/github/test/*.txt')['/Users/DahlHin/github/test/a1.txt','/Users/DahlHin/github/test/a2.txt','/Users/DahlHin/github/test/aA.txt']>>> |
所有巧合的是要么是上天注定要么是一个人偷偷的在努力。
2,参考:
浙公网安备 33010602011771号