[Python-ideas] pathlib.Path.walk

[Michel Desmoulin]

Like os.walk, but from a Path instance.

We have Path.iterdir, but it's not recursive. Which you use either
os.scandir, or os.walk. In any case, you end up doing:

import os
import pathlib

directory = pathlib.Path(get_dir())

# do things with directory

for root, dirs, files os.walk(directory):
root = Path(root)
for f in files:
f = root / f
# do something with file

Which is not that bad, but you waste a lot of time discovering how to do
that since you look first for something like Path.walk.

_______________________________________________
Python-ideas mailing list
Python...@python.org
https://mail.python.org/mailman/listinfo/python-ideas
Code of Conduct: http://python.org/psf/codeofconduct/

[INADA Naoki]

Why not Path.glob?
https://docs.python.org/3.6/library/pathlib.html#pathlib.Path.glob

[INADA Naoki]

On Mon, Apr 3, 2017 at 5:33 PM, Michel Desmoulin
<desmoul...@gmail.com> wrote:

> Like os.walk, but from a Path instance.
>
> We have Path.iterdir, but it's not recursive. Which you use either
> os.scandir, or os.walk. In any case, you end up doing:
>
> import os
> import pathlib
>
> directory = pathlib.Path(get_dir())
>
> # do things with directory
>
> for root, dirs, files os.walk(directory):
> root = Path(root)
> for f in files:
> f = root / f
> # do something with file
>
> Which is not that bad, but you waste a lot of time discovering how to do
> that since you look first for something like Path.walk.
>

OK, but what Path.walk should be return?
If all of dirs and files are Path object, it will have significant
performance overhead.

for root, dirs, files in directory.walk():
# all dirs are path object, but not used.
for f in files:
f = root / f # And Path overhead can be avoided for files too.