Proposal: Add a method to check if a string is a OneByteString

[theweipeng]

What is the problem this feature will solve?

In the Buffer module, we have a series of methods for handling string writing and reading, such as latin1WriteStatic and utf8WriteStatic. However, in some situations, we don't know the actual encoding of the string without checking every character. Checking the encoding will introduce overhead, especially when the string is large since SIMD is not accessible on the JavaScript side. In some string processing programs like the serialize framework (https://fury.apache.org/), high performance in string processing is highly beneficial for such programs.

What is the feature you are proposing to solve the problem?

Add isOneByteString function on the javascript side.

function isOneByteString(str) {
    if (typeof str  !== "string") {
        return null;
    }
    return getIsOneByte(str);
}

Add the getIsOneByte function on the C++ side. There is GetIsOneByteSlow for the slow mode, which is used when the place where it is being used cannot be compiled by TurboFan. And there is GetIsOneByteFast for the fast mode. This function is only applicable when the input string is a FastOneByteString, and in such a case, it will return true directly.

void GetIsOneByteSlow(
    const v8::FunctionCallbackInfo<v8::Value>& info) {
  DCHECK(ValidateCallbackInfo(info));
  if (info.Length() != 1 || !info[0]->IsString()) {
    info.GetIsolate()->ThrowError(
        "isOneByteString() requires a single string argument.");
    return;
  }
  bool is_one_byte = Utils::OpenDirectHandle(*info[0].As<v8::String>())
                         ->IsOneByteRepresentation();
  info.GetReturnValue().Set(is_one_byte);
}

bool GetIsOneByteFast(v8::Local<v8::Value> receiver,
                  const v8::FastOneByteString &source) {
  return true;
}

What alternatives have you considered?

No response