Path.with_name can construct invalid paths

[anntzer]

BPO	21714
Nosy	@pitrou, @anntzer
Files	pathlib-with_name-with_suffix.patch

^{Note: these values reflect the state of the issue at the time it was migrated and might not reflect the current state.}

Show more details

GitHub fields:

assignee = None
closed_at = <Date 2014-07-07.01:40:30.122>
created_at = <Date 2014-06-11.05:39:26.783>
labels = ['type-bug', 'library']
title = 'Path.with_name can construct invalid paths'
updated_at = <Date 2014-07-07.01:40:30.120>
user = 'https://github.com/anntzer'

bugs.python.org fields:

activity = <Date 2014-07-07.01:40:30.120>
actor = 'pitrou'
assignee = 'none'
closed = True
closed_date = <Date 2014-07-07.01:40:30.122>
closer = 'pitrou'
components = ['Library (Lib)']
creation = <Date 2014-06-11.05:39:26.783>
creator = 'Antony.Lee'
dependencies = []
files = ['35735']
hgrepos = []
issue_num = 21714
keywords = ['patch']
message_count = 7.0
messages = ['220235', '220269', '220327', '221263', '221329', '222429', '222431']
nosy_count = 3.0
nosy_names = ['pitrou', 'python-dev', 'Antony.Lee']
pr_nums = []
priority = 'normal'
resolution = 'fixed'
stage = 'resolved'
status = 'closed'
superseder = None
type = 'behavior'
url = 'https://bugs.python.org/issue21714'
versions = ['Python 3.4', 'Python 3.5']

[anntzer]

Path.with_name can be used to construct paths containing slashes as name contents (rather than as separators), as demonstrated below.

$ python -c 'from pathlib import Path; print(Path("foo").with_name("bar/baz").name)'
bar/baz

This should be changed to either raise a ValueError, or behave like path.parent / new_name. Given the amount of checking in the related with_suffix method (and the fact that the second option can be readily implemented by hand), the first option seems preferable.

[pitrou]

I think raising ValueError is the way to go. Would you like to try writing a patch for it?

[anntzer]

I have the patch almost ready, but ran into another issue: should "path.with_name('foo/')" be allowed? It may make sense to treat it like "path.with_name('foo')", just like 'Path("foo/") == Path("foo")'.

The implementation is also simpler with it, as it can reuse the "parse_parts" approach used by "with_suffix". But this also raises a separate issue, which is that with the current implementation, we have "Path('foo').with_suffix('.bar') == Path('foo').with_suffix('.bar/')", and that is less reasonable. Should I open a separate issue for that?

[pitrou]

should "path.with_name('foo/')" be allowed?

For sanity, I think path separators should be disallowed.

[anntzer]

The attached patch fixes all the issues mentioned, and also integrates the fixes of bpo-20639 (issues with with_suffix) as they are quite similar.

[python-dev]

New changeset c2636b5816a3 by Antoine Pitrou in branch '3.4':
Issue bpo-21714: Disallow the construction of invalid paths using Path.with_name(). Original patch by Antony Lee.
http://hg.python.org/cpython/rev/c2636b5816a3

[pitrou]

Ok, patch now applied (as well as the original patch for with_suffix()). Thank you very much!