What is the minimum decimal which can be converted to float and then back to decimal?-CodePudding

var t = (decimal)((float)decimal.MinValue)

fails with error CS0031: Constant value '-7.922816E 28' cannot be converted to a 'decimal', but why?

decimal.MinValue is -79228162514264337593543950335m. It successfully converts to float as -79228160000000000000000000000f wichh is larger than decimal.MinValue:

(float)decimal.MinValue > decimal.MinValue

Why this value can't be converted? I understand it's something about the significand and the exponent but what exactly happen? And which is minimum decimal which can be converted to float and then back to decimal in C#?

CodePudding user response：

You are not printing the full precision of the float, so you cannot see the value that is causing it to fail.

If you change the code as follows:

var decMin = (float)decimal.MinValue;
Console.WriteLine(decMin.ToString("R")); // "-7.9228163E 28"

// Next line throws System.OverflowException: Value was either too large or too small for a Decimal.
var dec = (decimal)decMin;

The value of decMin is displayed as -7.9228163E 28. To make this easier to compare, let's rewrite that without the exponent and compare it to decimal.MinValue:

decMin           = -79_228_163_000_000_000_000_000_000_000
decimal.MinValue = -79_228_162_514_264_337_593_543_950_335

Now you can see that decMin is MORE NEGATIVE than decimal.MinValue - which explains why it fails to convert.