Skip to content

Latest commit

 

History

History
56 lines (47 loc) · 1.77 KB

console.md

File metadata and controls

56 lines (47 loc) · 1.77 KB

Console

Writing UTF-8 strings to the Console

Console.Write methods use System.String which is UTF-16. In some scenarios we already have UTF-8 data that we want to write to the console (for example, JSON), and on some systems the default Console.OutputStream is UTF-8. It would be wasteful to allocate a string just to convert it back to UTF-8 bytes, so instead we can write the bytes directly to the output stream.

Important

It's up to the caller to ensure that the value is valid UTF-8. We can validate it using Utf8.IsValid (or Utf8.IndexOfInvalidSubsequence to find the error), but it might be wasteful if the API used to produce the bytes already does that.

public static class Utf8Console
{
    private static readonly Stream s_outputStream = Console.OpenStandardOutput();
    private static readonly ReadOnlyMemory<byte> s_newLine = Encoding.UTF8.GetBytes(Environment.NewLine);

    private static bool IsUtf8 => Console.OutputEncoding is UTF8Encoding;

    public static void Write(ReadOnlySpan<byte> value)
    {
        if (IsUtf8)
        {
            lock (Console.Out)
            {
                s_outputStream.Write(value);
            }
        }
        else
        {
            Console.Write(Encoding.UTF8.GetString(value));
        }
    }

    public static void WriteLine(ReadOnlySpan<byte> value)
    {
        if (IsUtf8)
        {
            lock (Console.Out)
            {
                s_outputStream.Write(value);
                s_outputStream.Write(s_newLine.Span);
            }
        }
        else
        {
            Console.WriteLine(Encoding.UTF8.GetString(value));
        }
    }
}

Examples:

Utf8Console.WriteLine("Hello, World"u8);
Utf8Console.WriteLine(JsonSerializer.SerializeToUtf8Bytes(new { a = 1, b = "2" }));