This is less repetitive, and IMO still readable.

That's not the same result, you're missing ^1 on the end.

Personally I find it confusing that this uses the string length of a literal to work out how many characters to remove, particularly as the literals don't have the right quotation marks in them, nor would the text in question have double quotes. I also don't see why this can't be in the language agnostic layer, since isVerbatim will always be false for VB anyway. I would simply write this as:

private static string GetTextWithoutQuotes(string text, bool isVerbatim)
{
    // Trim off an extra character for the @ symbol for verbatim strings
	var startIndex = isVerbatim ? 2 : 1;
	return text[startIndex..^1];
}

also, please onl yhave this refactoring if there are no diagnotics on the expr. and probably a good time to ensure that the string is terminated. and pleas have test for thie case "foo{ (i.e. an unterminated literal).

That's not the same result, you're missing ^1 on the end.

Personally I find it confusing that this uses the string length of a literal to work out how many characters to remove, particularly as the literals don't have the right quotation marks in them, nor would the text in question have double quotes. I also don't see why this can't be in the language agnostic layer, since isVerbatim will always be false for VB anyway. I would simply write this as:

private static string GetTextWithoutQuotes(string text, bool isVerbatim)
{
    // Trim off an extra character for the @ symbol for verbatim strings
	var startIndex = isVerbatim ? 2 : 1;
	return text[startIndex..^1];
}

I agree, this is a better way of doing things.

(The code I used was copied from https://github.com/dotnet/roslyn/blob/master/src/Features/CSharp/Portable/ConvertToInterpolatedString/CSharpConvertConcatenationToInterpolatedStringRefactoringProvider.cs#L25)

feel free to extract all of this into a helper utilities class they can both call into.

That's not the same result, you're missing ^1 on the end.

Personally I find it confusing that this uses the string length of a literal to work out how many characters to remove, particularly as the literals don't have the right quotation marks in them, nor would the text in question have double quotes. I also don't see why this can't be in the language agnostic layer, since isVerbatim will always be false for VB anyway. I would simply write this as:

private static string GetTextWithoutQuotes(string text, bool isVerbatim)
{
    // Trim off an extra character for the @ symbol for verbatim strings
	var startIndex = isVerbatim ? 2 : 1;
	return text[startIndex..^1];
}