Utf8Parser discrepancy in double parsing between .NET 2.2 and 3.0

[ppekrol]

Hi,

I have following code:

            string bigNumString = string.Join(string.Empty, "17976931348623158" + string.Join("", Enumerable.Repeat(0,293)));
            var bytes = Encoding.UTF8.GetBytes(bigNumString);

            if (Utf8Parser.TryParse(bytes.AsSpan(), out double value, out var consumed) && bytes.Length == consumed)
            {
                Console.WriteLine(value); // .NET Core 3.0
                return;
            }

            throw new InvalidOperationException("Not parsed!"); // .NET Core 2.2

In .NET Core 2.2 it throws an exception, but in .NET Core 3.0 (preview8) it returns Infinity. Is this by design?

[stephentoub]

in .NET Core 3.0 (preview8) it returns Infinity. Is this by design?

Yes.
https://github.com/dotnet/corefx/issues/39296#issuecomment-509382614
cc: @tannergooding

[ppekrol]

Thanks @stephentoub
That explains a lot!