feat: add C implementation for math/base/special/roundb

[gunjjoshi]

Description

What is the purpose of this pull request?

This pull request:

• adds C implementation for math/base/special/roundb.

double stdlib_base_roundb( const double x, const double n, const double b );

Related Issues

Does this pull request have any related issues?

This pull request:

• resolves a part of [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.

---

@stdlib-js/reviewers

[kgryte]

@gunjjoshi Would it make more sense to make n and b 32-bit integers, similar to n in roundn?

[gunjjoshi]

@gunjjoshi Would it make more sense to make n and b 32-bit integers, similar to n in roundn?

Sure @kgryte, I will change that.

[gunjjoshi]

@kgryte, Do we have STDLIB_MATH_BASE_NAPI_MODULE_DII_D( stdlib_base_roundb )? Or should we use something else here?

Edit: checked it, we don't have this.

[gunjjoshi]

Should I add napi_value stdlib_math_base_napi_dii_d( napi_env env, napi_callback_info info, double (*fcn)( double, int32_t, int32_t ) ) in math/base/napi/ternary/src/main.c in a seperate PR?
And then we can use it here.

[kgryte]

@gunjjoshi Yes, that would be great!

[kgryte]

I believe you'll also need to add a corresponding macro to that same package.

[gunjjoshi]

Yes. I'll add the macro in the .h file as well.

[gunjjoshi]

@kgryte Our math/base/assert/is-nan and math/base/assert/is-infinite take double as input. But we have 2 integer values, n and b to handle here. We have some test cases which fail due to this. I checked roundn, but over there, we do not have test cases which give NaN or +-infinite as the arguments for the integer n.

[kgryte]

As discussed offline, the plan here is to remove non-applicable tests for the native bindings for n and b with regard to the C implementation and its restriction to only accept integer arguments.

[gunjjoshi]

I have also removed the checks in main.c where we check if b or n is NaN or infinite.

[Planeshifter]

Thanks, @gunjjoshi! Will merge once CI passes.