MINOR: [Python] Array.from_buffers accepts None buffers

[robsdedude]

Rationale for this change

Fix the docs. It seems this is a feature and not just happens to work as the code a little below even has an explicit comment:

arrow/python/pyarrow/array.pxi

Line 1347 in 0dfae70

# None will produce a null buffer pointer

What changes are included in this PR?

Document that Array.from_buffers takes a list[Buffer | None] instead of strictly list[Buffer].

Are these changes tested?

not applicable

Are there any user-facing changes?

Yes, but not in an API sense :)

[github-actions]

Thanks for opening a pull request!

If this is not a minor PR. Could you open an issue for this pull request on GitHub? https://github.com/apache/arrow/issues/new/choose

Opening GitHub issues ahead of time contributes to the Openness of the Apache Arrow project.

Then could you also rename the pull request title in the following format?

GH-${GITHUB_ISSUE_ID}: [${COMPONENT}] ${SUMMARY}

or

MINOR: [${COMPONENT}] ${SUMMARY}

See also:

• Other pull requests
• Contribution Guidelines - Contributing Overview

[robsdedude]

Then could you also rename the pull request title in the following format?

I can in fact not... I think this classifies as a minor change, but I seem to lack permissions to rename my own PR... The button for is isn't showing. Must be a GH thing.

[AlenkaF]

I think this change is OK and would also update the DictionaryArray method docstrings. cc @raulcd what do you think?

[raulcd]

This makes sense to me, even though this will probably be reworked as part of the adding type hints initiative that @rok is leading.
@rok does this look good to you?

[robsdedude]

I think this change is OK and would also update the DictionaryArray method docstrings.

@AlenkaF would you prefer a separate PR or should I extend this one?

[AlenkaF]

Please, extend this one.

[robsdedude]

Done. While at it, I also saw that RunEndEncodedArray might need updating?

[robsdedude]

@AlenkaF opinions on RunEndEncodedArray?

[AlenkaF]

@robsdedude I think RunEndEncodedArray is good as is, the buffers there should always be empty, see:

arrow/python/pyarrow/array.pxi

Lines 4403 to 4404 in f2b4e34

	RunEndEncodedArrays do not have buffers, only children arrays, but this
	implementation is needed to satisfy the Array interface.

and the docstrings are correct there:

arrow/python/pyarrow/array.pxi

Line 4414 in f2b4e34

Empty List or [None].

[AlenkaF]

Thanks!

[conbench-apache-arrow]

After merging your PR, Conbench analyzed the 3 benchmarking runs that have been run so far on merge-commit b095098.

There were no benchmark performance regressions. 🎉

The full Conbench report has more details. It also includes information about 8 possible false positives for unstable benchmarks that are known to sometimes produce them.