GH-48095: [Python][Docs] Add missing {pyarrow,compute} functions to API docs

[tadeja]

Issue #48095 and #45322

Rationale for this change

Missing API doc entries.

What changes are included in this PR?

I've added strings for functions and options to the API doc page.

Are these changes tested?

I've built docs locally.

Are there any user-facing changes?

Docs only.

• GitHub Issue: [Python][Docs] Add missing {pyarrow,compute} functions to API docs #48095
• Also covers: [Docs][Python] Add Hyperbolic Trig functions to API docs #45322

[github-actions]

⚠️ GitHub issue #48095 has been automatically assigned in GitHub to PR creator.

[tadeja]

I haven’t included arange as docs were not generated for it and I cannot say why yet.

[dangotbanned]

Thanks @tadeja!

I haven’t included arange as docs were not generated for it and I cannot say why yet.

I think it should have been here:

arrow/docs/source/python/api/arrays.rst

Lines 24 to 33 in 0c20874

	Factory Functions
	-----------------
	These functions create new Arrow arrays:
	.. autosummary::
	:toctree: ../generated/
	array
	nulls

Which also made me realize that some others in that module are missing too:

• asarray
  • https://github.com/rok/arrow/blob/04916ca84c1a6a9d3ecc0bde7d809e9718acef2b/python/pyarrow-stubs/pyarrow/array.pyi#L136-L140
• repeat
  • https://github.com/rok/arrow/blob/04916ca84c1a6a9d3ecc0bde7d809e9718acef2b/python/pyarrow-stubs/pyarrow/array.pyi#L150-L154

[AlenkaF]

@dangotbanned that should be correct. I would also add infer_type as it seems to be missing too (though it would need a separate section or be in a subsection of Array Types).

[AlenkaF]

Thank you for updating our API docs @tadeja!
I have added minor suggestions otherwise the changes look good. I also checked the rendered pages and found no issues.

Note, the changes in this PR would close two issues:

• #48095
• #45322

Maybe the second issue could be linked from the PR description also.

[tadeja]

Thanks @AlenkaF, @dangotbanned.
Just note that asarray does not seem to get picked up on the Arrays doc page. I am checking locally if adding it to pyarrow/init.py could do the trick.

[github-actions]

⚠️ GitHub issue #48095 has been automatically assigned in GitHub to PR creator.

[github-actions]

Revision: 56043a1

Submitted crossbow builds: ursacomputing/crossbow @ actions-0ec9937bf0

Task	Status
preview-docs

[rok]

@github-actions crossbow submit preview-docs

[github-actions]

Revision: bd372aa

Submitted crossbow builds: ursacomputing/crossbow @ actions-ff8cf53890

Task	Status
preview-docs

[AlenkaF]

LGTM, thanks!!

[tadeja]

@AlenkaF
Note: This PR is now adding asarray to __ init __.py, and making it publicly available where it wasn't so far.

A side note regarding added
index_in_meta_binary and is_in_meta_binary - see older items here (first #29556 then #29839)

[AlenkaF]

Note: This PR is now adding asarray to __ init __.py, and making it publicly available where it wasn't so far.

Oh, yeah. I seem to remember we have talked about this with Joris in the past. It is a bit confusing for the user to have array and asarray available specially because asarray is a thin wrapper around array(). We might want to keep it as is for now and not add asarray to the API docs.

A side note regarding added
index_in_meta_binary and is_in_meta_binary - see older items here (first #29556 then #29839)

I think #29556 can be closed and we might want to wait for #29839 to be solved before adding index_in_meta_binary and is_in_meta_binary to the API docs?

[AlenkaF]

Thanks @tadeja!
@raulcd would you mind having a look before I merge?

[tadeja]

@AlenkaF, thank you. So I've removed asarray, index_in_meta_binary and is_in_meta_binary from this PR for now as advised.

[AlenkaF]

Not sure why the linter job is failing on an unrelated code.

[AlenkaF]

There is already an open PR for the linter error here: #48173.
It is unrelated to this PR so can be merged after another pair of 👀

[raulcd]

I'll try to look at this later today

[raulcd]

I've gone over the rendered docs and this looks great to me! Thank you very much @tadeja for the work here!

[github-actions]

⚠️ GitHub issue #48095 has been automatically assigned in GitHub to PR creator.

[conbench-apache-arrow]

After merging your PR, Conbench analyzed the 4 benchmarking runs that have been run so far on merge-commit e4b5b11.

There were no benchmark performance regressions. 🎉

The full Conbench report has more details. It also includes information about 9 possible false positives for unstable benchmarks that are known to sometimes produce them.