Numeric Types and Casting

Here we have some code that assigns the integer 12 to a variable named whole_number, and the fractional number 12.34 to a variable named fractional_number. That decimal point within the numbers is important, because it decides which of C#'s two most common numeric types the number will be treated as.

var wholeNumber = 12;
var fractionalNumber = 12.34;
Console.WriteLine(wholeNumber.GetType());
Console.WriteLine(fractionalNumber.GetType());

• The whole number automatically gets assigned a type of Int32, and the number with a decimal point gets a type of Double.
• As we saw in the previous video, even a number that's technically an integer will be treated as a Double if you include a decimal point.
  • Let's change the assignment to wholeNumber from 12 to 12.0.
  • It's still technically the same number, but the decimal point changes its default type.
  • You can see that the type for wholeNumber is now also Double.
• Remember that var implicitly sets the type of a variable based on the value you initially set it to.
  • We can also explicitly set the type of our variables.

Let me set wholeNumber's type to int... And fractionalNumber's type to double:

int wholeNumber = 12.0;
double fractionalNumber = 12.34;
Console.WriteLine(wholeNumber.GetType());
Console.WriteLine(fractionalNumber.GetType());

• But this gives us an error when assigning to wholeNumber: "Cannot implicitly convert type 'double' to 'int'." What does that mean?
• Well, this really doesn't seem to matter much when assigning 12.0 to an integer variable. But what if we were trying to assign 12.5 or 3.14 or some other number with a fractional part? int wholeNumber = 12.75
• For an integer variable to hold the double value, we have to convert it to an integer.
• But what do we do with the fractional part, the .75?
• Does it just not matter? Can we drop it? That seems unlikely. You wouldn't want the fractional dollars dropped from your bank account, for example.
• To protect you from issues like this, C# won't let you assign a double value to an int variable without converting it first.
• A conversion explicitly converts a numeric value of one type to another type.
• A conversion shows the compiler that you know this is a double value, but you definitely want to convert it to an int, and are certain that there either is no fractional portion to the number, or that losing the fractional portion won't cause any problems.
• You do a conversion by placing parentheses before the value you want to convert, containing the type you want to convert the value to: int wholeNumber = (int)12.75;

Conversions only work with numeric types. You can't convert from a string to an int or an int to a string, for example.

wholeNumber = (int)"12";
string myString = (string)12;

Cannot convert type 'string' to 'int'
Cannot convert type 'int' to 'string'

• There are often other ways to convert from one non-numeric type to another, but those are provided by libraries.
• Conversions, on the other hand, are provided by the language itself.
• We'll see how to convert from a string to a numeric type in an upcoming video.

Casting is an explicit conversion. It's required when making a conversion from a more-precise numeric type to a less-precise one, because data might be lost. Converting from a floating-point number to an integer loses all the decimal places, for example, so it requires an explicit cast.

But when you're converting from a less-precise numeric type to a more-precise one, there's no risk of data loss, because the more-precise numeric type will always be able to hold the full value from the less-precise one. For example, when converting from an integer to a floating-point number, you can just add ".0` onto the end. It consumes more computer memory to store it now, but it's still the same value.

When converting from a less-precise numeric type to a more-precise one, C# will do an implicit cast. An implicit cast is a conversion between numeric types that happens automatically, without the need to specify that it should.

• For example, suppose I wanted to change the initial value of fractionalNumber from 12.34 to 12.
• There's no need to write 12.0.
• I can just get rid of the decimal point and all the decimal places: double fractionalNumber = 12;
• It still works, and the resulting value still has a type of double because I declared fractionalNumber with a double type.
  • C# does an implicit conversion from the less-precise int type to the more-precise double type.

int wholeNumber = (int)12.5;
double fractionalNumber = 12;
Console.WriteLine(wholeNumber.GetType());
Console.WriteLine(fractionalNumber.GetType());