diff options
author | Barney Gale <barney.gale@gmail.com> | 2024-08-26 16:05:34 (GMT) |
---|---|---|
committer | GitHub <noreply@github.com> | 2024-08-26 16:05:34 (GMT) |
commit | 7bd6ebf696efcd5cf8e4e7946f9d8d8aee05664c (patch) | |
tree | 32e8e1088b2018438394b8a6409b95887f572b60 /Python/jit.c | |
parent | 033d537cd4b8c12f2441f1c23960c2153122140a (diff) | |
download | cpython-7bd6ebf696efcd5cf8e4e7946f9d8d8aee05664c.zip cpython-7bd6ebf696efcd5cf8e4e7946f9d8d8aee05664c.tar.gz cpython-7bd6ebf696efcd5cf8e4e7946f9d8d8aee05664c.tar.bz2 |
GH-73991: Prune `pathlib.Path.copy()` and `copy_into()` arguments (#123337)
Remove *ignore* and *on_error* arguments from `pathlib.Path.copy[_into]()`,
because these arguments are under-designed. Specifically:
- *ignore* is appropriated from `shutil.copytree()`, but it's not clear
how it should apply when the user copies a non-directory. We've changed
the callback signature from the `shutil` version, but I'm not confident
the new signature is as good as it can be.
- *on_error* is a generalisation of `shutil.copytree()`'s error handling,
which is to accumulate exceptions and raise a single `shutil.Error` at
the end. It's not obvious which solution is better.
Additionally, this arguments may be challenging to implement in future user
subclasses of `PathBase`, which might utilise a native recursive copying
method.
Diffstat (limited to 'Python/jit.c')
0 files changed, 0 insertions, 0 deletions