In Python, how to find all the files under a directory, including the files in subdirectories?

Is there any built in functions to find all the files under a particular directory including files under subdirectories ? I have tried this code, but not working...may be t开发者_如何学JAVAhe logic itself is wrong...

def fun(mydir):
    lis=glob.glob(mydir)
    length=len(lis)
    l,i=0,0
    if len(lis):
        while(l+i<length):
            if os.path.isfile(lis[i]):
                final.append(lis[i])
                lis.pop(i)
                l=l+1
                i=i+1
            else:
                i=i+1
            print final
        fun(lis)
    else:
        print final

There is no built-in function, but using os.walk it's trivial to construct it:

import os
def recursive_file_gen(mydir):
    for root, dirs, files in os.walk(mydir):
        for file in files:
            yield os.path.join(root, file)

ETA: the os.walk function walks directory tree recursively; the recursive_file_gen function is a generator (uses yield keyword to produce next file). To get the resulting list do:

list(recursive_file_gen(mydir))

I highly recommend this path module, written by Jason Orendorff:

http://pypi.python.org/pypi/path.py/2.2

Unfortunately, his website is down now, but you can still download from the above link (or through easy_install, if you prefer).

Using this path module, you can do various actions on paths, including the walking files you requested. Here's an example:

from path import path

my_path = path('.')

for file in my_path.walkfiles():
    print file

for file in my_path.walkfiles('*.pdf'):
    print file

There are also convenience functions for many other things to do with paths:

In [1]: from path import path

In [2]: my_dir = path('my_dir')

In [3]: my_file = path('readme.txt')

In [5]: print my_dir / my_file
my_dir/readme.txt

In [6]: joined_path = my_dir / my_file

In [7]: print joined_path
my_dir/readme.txt

In [8]: print joined_path.parent
my_dir

In [9]: print joined_path.name
readme.txt

In [10]: print joined_path.namebase
readme

In [11]: print joined_path.ext
.txt

In [12]: joined_path.copy('some_output_path.txt')

In [13]: print path('some_output_path.txt').isfile()
True

In [14]: print path('some_output_path.txt').isdir()
False

There are more operations that can be done too, but these are some of the ones that I use most often. Notice that the path class inherits from string, so it can be used wherever a string is used. Also, notice that two or more path objects can easily be joined together by using the overridden / operator.

Hope this helps!

os.walk() is what you need.

But for added performance, try the package scandir. It also part of the standard library in Python 3.5 and is described in PEP 471