Let’s say I have this directory structure.
├── root1 │ └── root2 │ ├── bar │ │ └── file1 │ ├── foo │ │ ├── file2 │ │ └── file3 │ └── zoom │ └── z1 │ └── file41
I want to isolate path components relative to root1/root2, i.e. strip out the leading root part, giving relative directories:
bar/file1 foo/file3 zoom/z1/file41
The root depth can be arbitrary and the files, the node of this tree, can also reside at different levels.
This code does it, but I am looking for Pathlib’s pythonic way to do it.
from pathlib import Path
import os
#these would come from os.walk or some glob...
file1 = Path("root1/root2/bar/file1")
file2 = Path("root1/root2/foo/file3")
file41 = Path("root1/root2/zoom/z1/file41")
root = Path("root1/root2")
#take out the root prefix by string replacement.
for file_ in [file1, file2, file41]:
    #is there a PathLib way to do this?🤔
    file_relative = Path(str(file_).replace(str(root),"").lstrip(os.path.sep))
    print("  %s" % (file_relative))
Advertisement
Answer
TLDR: use Path.relative_to:
Path("a/b/c").relative_to("a/b")  # returns PosixPath('c')
Full example:
from pathlib import Path
import os
# these would come from os.walk or some glob...
file1 = Path("root1/root2/bar/file1")
file2 = Path("root1/root2/foo/file3")
file41 = Path("root1/root2/zoom/z1/file41")
root = Path("root1/root2")
# take out the root prefix by string replacement.
for file_ in [file1, file2, file41]:
    # is there a PathLib way to do this?🤔
    file_relative = file_.relative_to(root)
    print("  %s" % (file_relative))
Prints
barfile1 foofile3 zoomz1file41