fix: use correct argument value in error message and propagate JSDoc fixes to sibling packages

[Planeshifter]

Description

Propagating fixes merged to develop between 2026-05-25 13:11 UTC and 2026-05-26 13:11 UTC to sibling packages.

fix: use correct argument in format() error message

blas/base/dgemm had a RangeError for N < 0 whose format() value interpolated M instead of N. Three *syr2 siblings carry the analogous defect: the zero-stride check on strideY reports strideX. The non-native ssyr2.js (and both ndarray.js files) already report strideY correctly, confirming the helper-file copy-paste.

• Source: 6c22721
• Targets:
  • blas/base/ssyr2/lib/ssyr2.native.js
  • blas/base/dsyr2/lib/dsyr2.js
  • blas/base/dsyr2/lib/dsyr2.native.js

docs: align scalar parameter descriptions to package style

blas/ext/base/gxsa aligned * @param {number} alpha - scalar to ... - scalar constant to match the package's declared types. The same wording mismatch lives in eighteen BLAS helper files where alpha or beta is documented as just scalar while the package's own index.d.ts (and, for the GEMM helpers, the sibling JSDoc later in the same file) uses scalar constant.

• Source: 8f8c656
• Targets:
  • blas/base/sspr (ndarray.js, sspr.js, ndarray.native.js, base.js, sspr.native.js)
  • blas/base/dspr (ndarray.js, dspr.js, base.js)
  • blas/base/ssyr2 (ndarray.js, ssyr2.js, base.js)
  • blas/base/gscal (ndarray.js, accessors.js)
  • blas/ext/base/gapx (ndarray.js)
  • blas/base/dgemm/lib/base.js, blas/base/sgemm/lib/base.js
  • blas/base/ggemm (base.js, accessors.js)

docs: convert imperative to noun phrase in constant-module JSDoc

os/num-cpus changed its @module description from Return the number of CPUs. to Number of CPUs. because the module exports a value, not a function. assert/is-little-endian is the lone outlier among the is-* endian/platform constants — is-big-endian, is-darwin, and is-windows all use the noun phrase.

• Source: 4531b43
• Target: assert/is-little-endian/lib/index.js

Related Issues

None.

Questions

No.

Other

Validation

Search scope was lib/node_modules/@stdlib/blas/** (Patterns A and B) and lib/node_modules/@stdlib/**/lib/index.js exporting constants (Pattern C). Each site was independently confirmed by two opus validation agents, verified for self-containment by an adaptation pass, and checked for style consistency against the package's own d.ts / siblings by a final pass.

Deliberately excluded:

• ~33 BLAS sites whose alpha - scalar JSDoc lives in packages whose canonical wording is just "constant" (e.g., *scal, *axpy, *fill, gaxpy, wasm/*scal, wasm/*axpy) — these require human judgment about which canonical wording the package should adopt.
• The tempory → temporary typo pattern from 4531b43 — zero additional occurrences in the repo.
• Other throw new ...( format(...) ) mismatches in BLAS that, on inspection, turned out to be intentional (e.g., transA reported when ta = resolveStr(transA) is null; LDA bound variables computed from order).

No refactor: commits propagated this run — b3c6c81b / 04fe53bd (isMatrixTranspose → resolveStr in sgemm/ggemm) change signatures and warrant human review before being applied to dgemm and the complex GEMM variants.

Checklist

• Read, understood, and followed the contributing guidelines.

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

This PR was prepared by Claude Code as part of an automated fix-propagation routine. Each propagation site was independently confirmed by two opus validation agents and a style-consistency pass before patches were applied; the routine refuses to propagate any site that fails consensus.

---

@stdlib-js/reviewers

---

Generated by Claude Code

[stdlib-bot]

Coverage Report

Package	Statements	Branches	Functions	Lines
assert/is-little-endian	$\color{green}148/148$ $\color{green}+100.00%$	$\color{green}4/4$ $\color{green}+100.00%$	$\color{green}1/1$ $\color{green}+100.00%$	$\color{green}148/148$ $\color{green}+100.00%$
blas/base/dgemm	$\color{green}843/843$ $\color{green}+100.00%$	$\color{green}119/119$ $\color{green}+100.00%$	$\color{green}8/8$ $\color{green}+100.00%$	$\color{green}843/843$ $\color{green}+100.00%$
blas/base/dspr	$\color{green}381/381$ $\color{green}+100.00%$	$\color{green}44/44$ $\color{green}+100.00%$	$\color{green}3/3$ $\color{green}+100.00%$	$\color{green}381/381$ $\color{green}+100.00%$
blas/base/dsyr2	$\color{green}667/667$ $\color{green}+100.00%$	$\color{green}92/92$ $\color{green}+100.00%$	$\color{green}5/5$ $\color{green}+100.00%$	$\color{green}667/667$ $\color{green}+100.00%$
blas/base/ggemm	$\color{red}1335/1344$ $\color{green}+99.33%$	$\color{red}173/176$ $\color{green}+98.30%$	$\color{green}14/14$ $\color{green}+100.00%$	$\color{red}1335/1344$ $\color{green}+99.33%$
blas/base/gscal	$\color{green}314/314$ $\color{green}+100.00%$	$\color{green}34/34$ $\color{green}+100.00%$	$\color{green}3/3$ $\color{green}+100.00%$	$\color{green}314/314$ $\color{green}+100.00%$
blas/base/sgemm	$\color{green}844/844$ $\color{green}+100.00%$	$\color{green}119/119$ $\color{green}+100.00%$	$\color{green}8/8$ $\color{green}+100.00%$	$\color{green}844/844$ $\color{green}+100.00%$
blas/base/sspr	$\color{green}535/535$ $\color{green}+100.00%$	$\color{green}49/49$ $\color{green}+100.00%$	$\color{green}5/5$ $\color{green}+100.00%$	$\color{green}535/535$ $\color{green}+100.00%$
blas/base/ssyr2	$\color{green}633/633$ $\color{green}+100.00%$	$\color{green}91/91$ $\color{green}+100.00%$	$\color{green}5/5$ $\color{green}+100.00%$	$\color{green}633/633$ $\color{green}+100.00%$
blas/ext/base/gapx	$\color{green}277/277$ $\color{green}+100.00%$	$\color{green}23/23$ $\color{green}+100.00%$	$\color{green}3/3$ $\color{green}+100.00%$	$\color{green}277/277$ $\color{green}+100.00%$

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