[gunjjoshi]

---

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

Resolves stdlib-js/metr-issue-tracker#1215.

Description

What is the purpose of this pull request?

This pull request:

• adds fft/base/fftpack/float32/rffti, which will be the single-precision variant of fft/base/fftpack/generic/rffti.

Related Issues

Does this pull request have any related issues?

This pull request has the following related issues:

• [RFC]: add fft/base/fftpack/float32/rffti metr-issue-tracker#1215

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
• Code review

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.".

Used Claude Code to review my implementation and to do some R&D.

---

@stdlib-js/reviewers

[stdlib-bot]

Coverage Report

Package	Statements	Branches	Functions	Lines
fft/base/fftpack/float32/rffti	$\\color{green}414/414$ $\\color{green}+0.00\\%$	$\\color{green}12/12$ $\\color{green}+0.00\\%$	$\\color{green}2/2$ $\\color{green}+0.00\\%$	$\\color{green}414/414$ $\\color{green}+0.00\\%$

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

[gunjjoshi]

Not having the DFFTPACK_DOUBLE_PRECISION flag here, which is present in https://github.com/stdlib-js/stdlib/blob/develop/lib/node_modules/%40stdlib/fft/base/fftpack/float64/rffti/test/fixtures/c/fftpack/Makefile#L104, since this is for single-precision. Ref: https://github.com/marton78/pffft/blob/0b4ee12c4ba45a4a8e567550c16d96d1679f50ce/src/fftpack.c#L60

[kgryte]

Thanks for flagging.

[kgryte]

LGTM. Removed some of the float32 emulation after looking at the reference implementation. In particular, FFTPACK reinterprets real memory as int and then performs integer ops, not float ops. So, in order to avoid accuracy issues once exceeding the maximum safe float32 integer, I removed various f32 wrapper calls and already updated float32/decompose, as well.

We could, in principle, exactly mimic their logic. Nothing is stopping us from creating a new, say, Int32Array view over the underlying buffer, but I am fine keeping as is, for now.

It is possible that users will want to compute an FFT having more than 16.7M elements

In [42]: FLOAT32_MAX_SAFE_INTEGER
Out[42]: 16777215

but we will cross that bridge when we get there, as things get a bit messy if you need to support int64 arrays (which I don't think we do (!), as surely even the largest practical sequence length will have less than 2^31-1 factors). And given that factors would be written as int32 and not float, it would make fixture and example logic more complicated. But this is also doable, as would require creating an int32 array view to reinterpret raw memory.

We should discuss more during our next sync.