这是适用于linux和windows以及标准库

paths = ['a/b/c/', 'a/b/c', '\\a\\b\\c', '\\a\\b\\c\\', 'a\\b\\c',
         'a/b/../../a/b/c/', 'a/b/../../a/b/c']

def path_leaf(path):
    return path.strip('/').strip('\\').split('/')[-1].split('\\')[-1]

[path_leaf(path) for path in paths]

结果:

['c', 'c', 'c', 'c', 'c', 'c', 'c']

这是适用于linux和windows以及标准库

paths = ['a/b/c/', 'a/b/c', '\\a\\b\\c', '\\a\\b\\c\\', 'a\\b\\c',
         'a/b/../../a/b/c/', 'a/b/../../a/b/c']

def path_leaf(path):
    return path.strip('/').strip('\\').split('/')[-1].split('\\')[-1]

[path_leaf(path) for path in paths]

结果:

['c', 'c', 'c', 'c', 'c', 'c', 'c']

如果您的文件路径不是以“/”结尾，且目录以“/”分隔，则使用以下代码。众所周知，path通常不以“/”结尾。

import os
path_str = "/var/www/index.html"
print(os.path.basename(path_str))

但在某些情况下，像url以“/”结尾，然后使用以下代码

import os
path_str = "/home/some_str/last_str/"
split_path = path_str.rsplit("/",1)
print(os.path.basename(split_path[0]))

但是当你的路径被“\”分开时，你通常在Windows路径中找到，然后你可以使用以下代码

import os
path_str = "c:\\var\www\index.html"
print(os.path.basename(path_str))

import os
path_str = "c:\\home\some_str\last_str\\"
split_path = path_str.rsplit("\\",1)
print(os.path.basename(split_path[0]))

您可以通过检查操作系统类型将两者组合成一个函数并返回结果。

带扩展名的文件名

filepath = './dir/subdir/filename.ext'
basename = os.path.basename(filepath)
print(basename)
# filename.ext

print(type(basename))
# <class 'str'>

不带扩展名的文件名

basename_without_ext = os.path.splitext(os.path.basename(filepath))[0]
print(basename_without_ext)
# filename

这是工作!

os.path.basename(name)

但是你不能在Linux中通过Windows文件路径获取文件名。Windows。 操作系统。不同操作系统上不同模块的路径加载:

Linux - posixpath Windows - npath

所以你可以用os。路径总是得到正确的结果

像其他人建议的那样使用os.path.split或os.path.basename并不能在所有情况下工作:如果您在Linux上运行脚本并试图处理经典的windows样式的路径，它将失败。

Windows路径可以使用反斜杠或正斜杠作为路径分隔符。因此，ntpath模块(相当于os. path)在windows上运行时的路径)将适用于所有平台上的所有(1)路径。

import ntpath
ntpath.basename("a/b/c")

当然，如果文件以斜杠结束，basename将为空，所以创建自己的函数来处理它:

def path_leaf(path):
    head, tail = ntpath.split(path)
    return tail or ntpath.basename(head)

验证:

>>> paths = ['a/b/c/', 'a/b/c', '\\a\\b\\c', '\\a\\b\\c\\', 'a\\b\\c', 
...     'a/b/../../a/b/c/', 'a/b/../../a/b/c']
>>> [path_leaf(path) for path in paths]
['c', 'c', 'c', 'c', 'c', 'c', 'c']

(1) There's one caveat: Linux filenames may contain backslashes. So on linux, r'a/b\c' always refers to the file b\c in the a folder, while on Windows, it always refers to the c file in the b subfolder of the a folder. So when both forward and backward slashes are used in a path, you need to know the associated platform to be able to interpret it correctly. In practice it's usually safe to assume it's a windows path since backslashes are seldom used in Linux filenames, but keep this in mind when you code so you don't create accidental security holes.