feat: add C implementation of math/base/special/kernel-betainc

[nirmaljb]

---

type: pre_commit_static_analysis_report
description: Results of running static analysis checks when committing changes. report:

• task: lint_filenames status: passed
• task: lint_editorconfig status: passed
• task: lint_markdown status: passed
• task: lint_package_json status: passed
• task: lint_repl_help status: na
• task: lint_javascript_src status: passed
• task: lint_javascript_cli status: na
• task: lint_javascript_examples status: na
• task: lint_javascript_tests status: na
• task: lint_javascript_benchmarks status: passed
• task: lint_python status: na
• task: lint_r status: na
• task: lint_c_src status: missing_dependencies
• task: lint_c_examples status: missing_dependencies
• task: lint_c_benchmarks status: missing_dependencies
• task: lint_c_tests_fixtures status: na
• task: lint_shell status: na
• task: lint_typescript_declarations status: passed
• task: lint_typescript_tests status: passed
• task: lint_license_headers status: passed ---

Progresses #649 .

Description

What is the purpose of this pull request?

This pull request:

• Progresses a part of : [RFC]: Add C implementations to base special math functions (tracking issue) #649

Related Issues

Does this pull request have any related issues?

This pull request has the following related issues:

• [RFC]: Add C implementations to base special math functions (tracking issue) #649

Questions

Any questions for reviewers of this pull request?

No.

Other

Any other information relevant to this pull request? This may include screenshots, references, and/or implementation notes.

No.

Checklist

Please ensure the following tasks are completed before submitting this pull request.

• Read, understood, and followed the contributing guidelines.

AI Assistance

When authoring the changes proposed in this PR, did you use any kind of AI assistance?

• Yes
• No

If you answered "yes" above, how did you use AI assistance?

• Code generation (e.g., when writing an implementation or fixing a bug)
• Test/benchmark generation
• Documentation (including examples)
• Research and understanding

Disclosure

If you answered "yes" to using AI assistance, please provide a short disclosure indicating how you used AI assistance. This helps reviewers determine how much scrutiny to apply when reviewing your contribution. Example disclosures: "This PR was written primarily by Claude Code." or "I consulted ChatGPT to understand the codebase, but the proposed changes were fully authored manually by myself.".

I used codex to find references, but the proposed changes were fully authored manually by myself.

---

@stdlib-js/reviewers

[stdlib-bot]

Coverage Report

Package	Statements	Branches	Functions	Lines
math/base/special/kernel-betainc	$\color{red}1645/1788$ $\color{green}+0.00%$	$\color{red}195/226$ $\color{green}+0.00%$	$\color{red}14/16$ $\color{green}+0.00%$	$\color{red}1645/1788$ $\color{green}+0.00%$

The above coverage report was generated for the changes in this PR.

[nirmaljb]

/stdlib update-copyright-years

[nirmaljb]

Thank you for the suggestions! @gunjjoshi

[nirmaljb]

Fixed the issues
Thank you for pointing out! @gunjjoshi

[nirmaljb]

Note: The following three Rocco Romeo cases are not added out because they require long double (80-bit extended) precision to match the Wolfram reference values. Boost passes these by internally promoting double to long double (its default promote_double policy). JavaScript and C double-only computation loses too much precision in the intermediate x^a * y^b / Beta(a,b) power term for these extreme x values.

1. beta( 2, 4, ldexp(1+1/1024, -351) ): rel error ~9.3e-10 (~4.2M eps)
2. beta( 2, 4, ldexp(1+1/2048, -351) ): rel error ~2.4e-7 (~1B eps)
3. ibeta( 3, 5, ldexp(1+15/16, -268) ): rel error ~6.5e-3 (~29B eps)

[nirmaljb]

Thank you, @Neerajpathak07, for taking the time to review this function. It's a huge function and would require line-by-line review, and I really appreciate you doing that.
I would be working on reviewing it myself, and I look forward to more review changes. Hope to get this merged as soon as possible.

[anandkaranubc]

They seem high ULP diff numbers here. Is there a reason why that is the case?

[rautelaKamal]

Hey @nirmaljb, pulled your branch locally since kernel-betainc is a dependency for some of my downstream work, wanted to make sure things look good end to end.

Overall looks solid. Two small things I noticed:

In examples/c/example.c around line 39, the printf statement seems to be missing a \n at the end, output lines will run together when executed.

The ULP tolerances in test.native.js get fairly wide in some spots (e.g., 24k+ ULP for the large_range upper regularized tests around line 574). Just wanted to check, is that expected behavior matching the JS tests, or is there room to tighten those?

Great work on this overall, its a complex port.

[Neerajpathak07]

syncing these ranges up with the corresponding JS benchmark. Used random_uniform something we also did in a few other special functions.
cc:- @kgryte

[nirmaljb]

@Neerajpathak07 Thank you for taking the time and effort to make these changes!

[Neerajpathak07]

@kgryte Here if expected values are very large and the kernelBetainc function is making large number of internal calls then the high ULP Diff could be viable if I am not wrong. Plus we can add a comment mentioning the reason behind it as well.

[Neerajpathak07]

Suggested change

	if ( stdlib_base_floor( a ) == a && stdlib_base_floor( b ) == b && a < STDLIB_CONSTANT_INT32_MAX - 100 ) {
	if ( stdlib_base_floor( a ) == a && stdlib_base_floor( b ) == b && a < STDLIB_CONSTANT_INT32_MAX - 100 && y != 1 ) {

Would be good to add a safe-gaurd for y!=1 as per used in the boost implementation.
Re:-

if((floor(a) == a) && (floor(b) == b) && (a < (std::numeric_limits<int>::max)() - 100) && (y != 1))

[nirmaljb]

Thanks for noticing, I realised it as well, but I decided not to put this condition because it was not present in the javascript implementation. (Line: 349-354).

if ( b < 40.0 ) {
			if (
				floor(a) === a &&
				floor(b) === b &&
				a < MAX_INT32 - 100
			) {

What do you think we should here?

[Neerajpathak07]

Naturally we are interested in aligning the C implementation here to the reference Boost one. But for this we need to investigate what value this adds to the function.

[Planeshifter]

Thanks for tackling this rather involved PR!