[Feature Request] nameof(X)u8 : Generating UTF-8 byte sequences using nameof()

[xin9le]

Background and Motivation

UTF-8 literals were introduced in C# 11, which has been extremely beneficial for developers performing optimizations using UTF-8 byte sequences. However, at present, it is not possible to apply u8 to nameof(). This proposal aims to simplify the generation of UTF-8 byte sequences in a manner that follows name changes. Additionally, it will bridge the asymmetry gap between the generation of regular strings and UTF-8 byte sequences.

// string
string t1 = "CSharp";
string t2 = nameof(Language.CSharp);

// UTF-8 string
ReadOnlySpan<byte> = "CSharp"u8;
ReadOnlySpan<byte> = nameof(Language.CSharp)u8;  // currently, doesn't support

Proposed API

ReadOnlySpan<byte> text = nameof(X)u8;

Usage Examples

MyEnumExtensions
    .IsDefined<HttpStatusCode>(nameof(HttpStatusCode.OK)u8)
    .Should()
    .BeTrue();

Alternative Designs

If $"nameof(X)"u8 were to be implemented, it would enable equivalent functionality. However, this would not resolve the asymmetry issue mentioned earlier.

• [Proposal]: u8 string interpolation csharplang#7072

Risks

I guess it depends on being able to parse nameof(X)u8, but no further concerns come to mind.

[huoyaoyuan]

This is a C# language feature and should go to https://github.com/dotnet/csharplang/discussions. The API request category is for API provided by the compiler, to support analyzers and IDEs etc.

[xin9le]

@huoyaoyuan
Thank you for your observation. I had the feeling that I might have posted in the wrong place, so I appreciate your guidance in directing me to the appropriate location.