How to remove a path prefix in python?

Question

I wanted to know what is the pythonic function for this :

I want to remove everything before the wa path.

p = path.split('/')
counter = 0
while True:
    if p[counter] == 'wa':
        break
    counter += 1
path = '/'+'/'.join(p[counter:])

For instance, I want '/book/html/wa/foo/bar/' to become '/wa/foo/bar/'.

Fyi, when dealing with paths better use the split/join functions from the `os.path` module — ThiefMaster, Jan 01 '12 at 12:34

score 230 · Accepted Answer · edited May 24 '21 at 21:34

230

A better answer would be to use os.path.relpath:

http://docs.python.org/3/library/os.path.html#os.path.relpath

>>> import os
>>> full_path = '/book/html/wa/foo/bar/'
>>> relative_path = '/book/html'
>>> print(os.path.relpath(full_path, relative_path))
'wa/foo/bar'

edited May 24 '21 at 21:34

Community

1
1

answered Nov 08 '13 at 10:35

Mitch ミッチ

2,382
2
13
5

12

This is a much better answer because it avoids any issues with different path separators. – intrepidhero Sep 18 '15 at 22:59
1

Totally agree with @intrepidhero's comment, plus this works whether or not `full_path` contains the trailing `/` character or not—so it's even more general than that. – martineau Dec 10 '18 at 10:50

pjgranahan · Answer 2 · 2017-11-10T00:16:03.370

36

For Python 3.4+, you should use pathlib.PurePath.relative_to. From the documentation:

>>> p = PurePosixPath('/etc/passwd')
>>> p.relative_to('/')
PurePosixPath('etc/passwd')

>>> p.relative_to('/etc')
PurePosixPath('passwd')

>>> p.relative_to('/usr')
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "pathlib.py", line 694, in relative_to
    .format(str(self), str(formatted)))
ValueError: '/etc/passwd' does not start with '/usr'

Also see this StackOverflow question for more answers to your question.

edited Nov 10 '17 at 00:16

answered Apr 24 '17 at 23:06

pjgranahan

575
6
15

1

While the `pathlib` module is very "user friendly", it wasn't created until very late in the game. Personally I still prefer using `os.path.relpath()` as shown in the [accepted answer](https://stackoverflow.com/a/19856910/355230) because it will work in most versions of Python (including Python 2). – martineau Dec 10 '18 at 10:45

score 27 · Answer 3 · answered Jan 01 '12 at 12:20

27

>>> path = '/book/html/wa/foo/bar/'
>>> path[path.find('/wa'):]
'/wa/foo/bar/'

answered Jan 01 '12 at 12:20

Felix Loether

6,010
2
32
23

+1: compared to using a regular expression, this is simpler, and probably about as fast. – Eric O. Lebigot Jan 01 '12 at 12:23
2

This returns the last character if the string doesn't contain `/wa` (`path[-1:]`), so if that might happen you'd want to check `if "/wa" in path` first – dbr Jan 01 '12 at 12:50
4

alternately, you can use `str.index` instead of `str.find` to raise an exception when the needle is not in the haystack. – SingleNegationElimination Jan 01 '12 at 14:27
2

This doesn't work with multiple folders of same names. – nn0p Jun 15 '17 at 15:09

score 0 · Answer 4 · answered Jan 01 '12 at 12:29

0

import re

path = '/book/html/wa/foo/bar/'
m = re.match(r'.*(/wa/[a-z/]+)',path)
print m.group(1)

answered Jan 01 '12 at 12:29

vivek

4,951
4
25
33

This helps for my second question which was how to remove the last path if it is a integer. Nice :) – Natim Jan 01 '12 at 12:32

How to remove a path prefix in python?

4 Answers4

Linked