一个类似于ghostdog的复制粘贴解决方案:

def get_all_filepaths(root_path, ext):
    """
    Search all files which have a given extension within root_path.

    This ignores the case of the extension and searches subdirectories, too.

    Parameters
    ----------
    root_path : str
    ext : str

    Returns
    -------
    list of str

    Examples
    --------
    >>> get_all_filepaths('/run', '.lock')
    ['/run/unattended-upgrades.lock',
     '/run/mlocate.daily.lock',
     '/run/xtables.lock',
     '/run/mysqld/mysqld.sock.lock',
     '/run/postgresql/.s.PGSQL.5432.lock',
     '/run/network/.ifstate.lock',
     '/run/lock/asound.state.lock']
    """
    import os
    all_files = []
    for root, dirs, files in os.walk(root_path):
        for filename in files:
            if filename.lower().endswith(ext):
                all_files.append(os.path.join(root, filename))
    return all_files

你也可以使用yield来创建一个生成器，从而避免组装完整的列表:

def get_all_filepaths(root_path, ext):
    import os
    for root, dirs, files in os.walk(root_path):
        for filename in files:
            if filename.lower().endswith(ext):
                yield os.path.join(root, filename)

我喜欢os.walk():

import os

for root, dirs, files in os.walk(dir):
    for f in files:
        if os.path.splitext(f)[1] == '.txt':
            fullpath = os.path.join(root, f)
            print(fullpath)

或者使用生成器:

import os

fileiter = (os.path.join(root, f)
    for root, _, files in os.walk(dir)
    for f in files)
txtfileiter = (f for f in fileiter if os.path.splitext(f)[1] == '.txt')
for txt in txtfileiter:
    print(txt)

你可以简单地使用pathlibs glob 1:

import pathlib

list(pathlib.Path('your_directory').glob('*.txt'))

或在循环中:

for txt_file in pathlib.Path('your_directory').glob('*.txt'):
    # do something with "txt_file"

如果你想递归你可以使用。glob('**/*.txt')

1 pathlib模块被包含在python 3.4的标准库中。但是你甚至可以在旧的Python版本(即使用conda或pip)上安装该模块的反向端口:pathlib和pathlib2。

子目录的功能性解决方案:

from fnmatch import filter
from functools import partial
from itertools import chain
from os import path, walk

print(*chain(*(map(partial(path.join, root), filter(filenames, "*.txt")) for root, _, filenames in walk("mydir"))))

如果文件夹包含大量文件或内存受限，可以考虑使用生成器:

def yield_files_with_extensions(folder_path, file_extension):
   for _, _, files in os.walk(folder_path):
       for file in files:
           if file.endswith(file_extension):
               yield file

选项A:迭代

for f in yield_files_with_extensions('.', '.txt'): 
    print(f)

选项B:全部获取

files = [f for f in yield_files_with_extensions('.', '.txt')]

我做了一个测试(Python 3.6.4, W7x64)，看看对于一个文件夹，没有子目录，哪种解决方案是最快的，以获得具有特定扩展名的文件的完整文件路径列表。

简而言之，对于这个任务，os.listdir()是最快的，比次优os.walk()快1.7倍，比pathlib快2.7倍，比os.scandir()快3.2倍，比glob快3.3倍。 请记住，当您需要递归结果时，这些结果将会改变。如果您复制/粘贴下面的一个方法，请添加.lower()，否则在搜索.ext时将找不到.ext。

import os
import pathlib
import timeit
import glob

def a():
    path = pathlib.Path().cwd()
    list_sqlite_files = [str(f) for f in path.glob("*.sqlite")]

def b(): 
    path = os.getcwd()
    list_sqlite_files = [f.path for f in os.scandir(path) if os.path.splitext(f)[1] == ".sqlite"]

def c():
    path = os.getcwd()
    list_sqlite_files = [os.path.join(path, f) for f in os.listdir(path) if f.endswith(".sqlite")]

def d():
    path = os.getcwd()
    os.chdir(path)
    list_sqlite_files = [os.path.join(path, f) for f in glob.glob("*.sqlite")]

def e():
    path = os.getcwd()
    list_sqlite_files = [os.path.join(path, f) for f in glob.glob1(str(path), "*.sqlite")]

def f():
    path = os.getcwd()
    list_sqlite_files = []
    for root, dirs, files in os.walk(path):
        for file in files:
            if file.endswith(".sqlite"):
                list_sqlite_files.append( os.path.join(root, file) )
        break



print(timeit.timeit(a, number=1000))
print(timeit.timeit(b, number=1000))
print(timeit.timeit(c, number=1000))
print(timeit.timeit(d, number=1000))
print(timeit.timeit(e, number=1000))
print(timeit.timeit(f, number=1000))

结果:

# Python 3.6.4
0.431
0.515
0.161
0.548
0.537
0.274