Explain what "normal" is in Double.IsNormal

[sfiruch]

Summary

The documentation for Double.IsNormal is wholly unhelpful. It says it checks if a number is "Normal". Only in the remarks, "normal" is explained. This changes it to make it clear what it actually checks.

[dotnet-policy-service]

Tagging subscribers to this area: @dotnet/area-system-runtime

[Copilot]

Pull request overview

Updates the API documentation for Double.IsNormal to clarify what “normal” means directly in the summary text.

Changes:

• Replaces the IsNormal <summary> text with an explicit definition (not NaN, not infinite, not subnormal, and not zero).

[Copilot]

The updated

explains what “normal” means, but the text still relies on the term “normal” without the definition. Consider updating to use the same explicit criteria (or add the definition there) so readers don’t have to infer it from the summary/remarks.

Also, the criteria list is now repeated in both

and , which creates duplication that can drift over time. Consider removing the redundant list from or using for additional context instead of repeating the same definition.

Suggested change

	<see langword="true" /> if the value is normal; <see langword="false" /> otherwise.</returns>
	<remarks>This effectively checks the value is not NaN, not infinite, not subnormal, and not zero.</remarks>
	<see langword="true" /> if the value is not NaN, not infinite, not subnormal, and not zero; <see langword="false" /> otherwise.</returns>
	<remarks>Use this method to determine whether a double-precision floating-point number is a normal value before performing calculations that assume finite, nonzero magnitudes.</remarks>

[tannergooding]

If you're wanting to not be repetitive, then repeating subnormal doesn't really help either.. The technical definition is:

Suggested change

	<summary>Determines whether the specified value is not NaN, not infinite, not subnormal, and not zero.</summary>
	<summary>Determines whether the specified value is a finite non-zero value with magnitude greater than or equal to <c>2<sup>-1022</sup></c>.</summary>

[tannergooding]

I would expect Single, Half, NFloat, and INumberBase<TSelf> to all be updated in the same PR.

I would also expect other definitions like IsInfinite, IsSubnormal, IsFinite, etc to be updated if required.

[tannergooding]

And as per the Copilot guidance below, the summary and remarks are effectively just mirroring eachother now, so the remarks need to either be removed or updated.