[RFC]: tracking issue for missing float16 constants

[opbot-xd]

Description

What is the feature you are proposing?

A tracking issue to coordinate the implementation of missing constants/float16/* packages. Currently, multiple contributors are submitting PRs for float16 constants without visibility into what's already being worked on, leading to duplicate efforts.

Rationale

Why is this feature important?

1. Prevent duplicate PRs - Contributors can see what's already being worked on
2. Organize the work - Clear checklist of what needs to be done
3. Reduce maintainer burden - Less time spent reviewing duplicate PRs

There are currently multiple open PRs for the same constants (e.g., ln-two has #8972 and #9028, log2-e has #8991 and #9029).

Proposed Solution

How do you propose to implement this feature?

Create a checklist of all float16 constants that exist in float32 but not in float16. Contributors can submit PRs referencing this issue.

Constants to Implement

The following constants exist in float32 but are missing from float16:

Mathematical Constants

• glaisher-kinkelin - Glaisher-Kinkelin constant
• sqrt-half - Square root of 1/2
• sqrt-half-pi - Square root of π/2
• sqrt-pi - Square root of π
• sqrt-three - Square root of 3
• sqrt-two-pi - Square root of 2π

Logarithmic Constants

• ln-half - Natural logarithm of 1/2
• ln-pi - Natural logarithm of π
• ln-ten - Natural logarithm of 10
• ln-two - Natural logarithm of 2
• ln-two-pi - Natural logarithm of 2π
• ln-sqrt-two-pi - Natural logarithm of √(2π)
• half-ln-two - Half of natural logarithm of 2
• log10-e - Base 10 logarithm of e
• log2-e - Base 2 logarithm of e

Technical/Format Constants

• abs-mask - Absolute value mask
• nan - NaN value
• fourth-root-eps - Fourth root of machine epsilon
• [ ] gamma-lanczos-g - Lanczos approximation g value
• min-base10-exponent - Minimum base 10 exponent
• min-base10-exponent-subnormal - Minimum base 10 exponent (subnormal)
• max-base10-exponent-subnormal - Maximum base 10 exponent (subnormal)
• min-base2-exponent-subnormal - Minimum base 2 exponent (subnormal)

Factorial/Sequence Constants

• max-nth-factorial - Maximum nth factorial
• max-nth-double-factorial - Maximum nth double factorial
• max-safe-fibonacci - Maximum safe Fibonacci number
• max-safe-lucas - Maximum safe Lucas number
• max-safe-nth-fibonacci - Maximum safe nth Fibonacci
• max-safe-nth-lucas - Maximum safe nth Lucas
• max-safe-nth-double-factorial - Maximum safe nth double factorial
• max-safe-nth-factorial - Maximum safe nth factorial
• max-safe-nth-tribonacci - Maximum safe nth Tribonacci

---

Related Issues

No existing tracking issue found for float16 constants.

Questions

No.

Other

No.

Checklist

• I have read and understood the Code of Conduct.
• Searched for existing issues and pull requests.
• The issue name begins with RFC:.

[stdlib-bot]

👋 Hi there! 👋

And thank you for opening your first issue! We will get back to you shortly. 🏃 💨

If you have any further questions while waiting for a response, please join our Gitter channel to chat with project maintainers and other community members.