string.Concat overload is internal instead of public

[LeviGNMBS]

Description

Hi, i wanted to report the following issue:

This string.Concat overload:

internal static string Concat(ReadOnlySpan<char> str0, ReadOnlySpan<char> str1, ReadOnlySpan<char> str2, ReadOnlySpan<char> str3, ReadOnlySpan<char> str4)

https://github.com/dotnet/dotnet/blob/95335fdf41878a6dbec1ba88eb69dda4a90ec655/src/runtime/src/libraries/System.Private.CoreLib/src/System/String.Manipulation.cs#L351
is internal, its counterparts however are all public:

public static string Concat(ReadOnlySpan<char> str0, ReadOnlySpan<char> str1)
public static string Concat(ReadOnlySpan<char> str0, ReadOnlySpan<char> str1, ReadOnlySpan<char> str2)
public static string Concat(ReadOnlySpan<char> str0, ReadOnlySpan<char> str1, ReadOnlySpan<char> str2, ReadOnlySpan<char> str3)

This causes user code cannot reach the method, even though its just the same logic as the others but with one more parameter? Could have been an oversight?

Thank you for checking if this is intended. Sorry for the short form report, i think this is very self-explanatory.

Reproduction Steps

Try to run:
string.Concat("a".AsSpan(), "2", "t", "4", "e");
gives an error instead of compiling

Expected behavior

Code compiles

Actual behavior

Code gives error

Regression?

No response

Known Workarounds

No response

Configuration

No response

Other information

No response

[dotnet-policy-service]

Tagging subscribers to this area: @dotnet/area-system-memory
See info in area-owners.md if you want to be subscribed.

[MihuBot]

I'm a bot. Here is a possible related and/or duplicate issue (I may be wrong):

• string.Concat(5 x ReadOnlySpan<char>) #126365

[LeviGNMBS]

That is indeed a duplicate i did not find. The same question remains though even after reading the comments. I get it requires effort to change it, but for such a small bug I cant imagine it to be that much effort? Right now trying to debug some core code i copied it into an empty test to debug it but it didn't compile due to the missing overload. I get i can change it, but this kinda sticks out like a sore thumb in the api. 😄

[huoyaoyuan]

I get it requires effort to change it, but for such a small bug I cant imagine it to be that much effort?

It's not a bug, neither "small". It's intentional to decide whether being public for every single method.

Right now trying to debug some core code i copied it into an empty test to debug it but it didn't compile due to the missing overload.

It's is a common situation for BCL. Many non well-designed helpers are kept internal, and you need to copy them together for manual testing.

[tannergooding]

We don't tend to expose such APIs without explicit need and in the case of 4+ you're typically going outside the ABI conventions for what can be efficiently passed in register and better off using params anyways. Hence most types only expose specific variants for 1-3 parameters, with a few exceptions for scenarios that had explicit need.