Integral numeric types (original) (raw)

The integral numeric types represent integer numbers. All integral numeric types are value types. They're also simple types and can be initialized with literals. All integral numeric types support arithmetic, bitwise logical, comparison, and equality operators.

Characteristics of the integral types

C# supports the following predefined integral types:

C# type/keyword	Range	Size	.NET type
sbyte	-128 to 127	Signed 8-bit integer	System.SByte
byte	0 to 255	Unsigned 8-bit integer	System.Byte
short	-32,768 to 32,767	Signed 16-bit integer	System.Int16
ushort	0 to 65,535	Unsigned 16-bit integer	System.UInt16
int	-2,147,483,648 to 2,147,483,647	Signed 32-bit integer	System.Int32
uint	0 to 4,294,967,295	Unsigned 32-bit integer	System.UInt32
long	-9,223,372,036,854,775,808 to 9,223,372,036,854,775,807	Signed 64-bit integer	System.Int64
ulong	0 to 18,446,744,073,709,551,615	Unsigned 64-bit integer	System.UInt64
nint	Depends on platform (computed at runtime)	Signed 32-bit or 64-bit integer	System.IntPtr
nuint	Depends on platform (computed at runtime)	Unsigned 32-bit or 64-bit integer	System.UIntPtr

In all of the table rows except the last two, each C# type keyword from the leftmost column is an alias for the corresponding .NET type. The keyword and .NET type name are interchangeable. For example, the following declarations declare variables of the same type:

int a = 123;
System.Int32 b = 123;

The nint and nuint types in the last two rows of the table are native-sized integers. You can use the nint and nuint contextual keywords to define native-sized integers. These are 32-bit integers when running in a 32-bit process, or 64-bit integers when running in a 64-bit process. They can be used for interop scenarios, low-level libraries, and to optimize performance in scenarios where integer math is used extensively.

The native-sized integer types are represented internally as the .NET types System.IntPtr and System.UIntPtr. Starting in C# 11, the nint and nuint types are aliases for the underlying types.

The default value of each integral type is zero, 0.

Each of the integral types has MinValue and MaxValue properties that provide the minimum and maximum value of that type. These properties are compile-time constants except for the case of the native-sized types (nint and nuint). The MinValue and MaxValue properties are calculated at runtime for native-sized types. The sizes of those types depend on the process settings.

Use the System.Numerics.BigInteger structure to represent a signed integer with no upper or lower bounds.

Integer literals

Integer literals can be

decimal: without any prefix
hexadecimal: with the 0x or 0X prefix
binary: with the 0b or 0B prefix

The following code demonstrates an example of each:

var decimalLiteral = 42;
var hexLiteral = 0x2A;
var binaryLiteral = 0b_0010_1010;

The preceding example also shows the use of _ as a digit separator. You can use the digit separator with all kinds of numeric literals.

The type of an integer literal is determined by its suffix as follows:

If the literal has no suffix, its type is the first of the following types in which its value can be represented: int, uint, long, ulong.
Note
Literals are interpreted as positive values. For example, the literal 0xFF_FF_FF_FF represents the number 4294967295 of the uint type, though it has the same bit representation as the number -1 of the int type. If you need a value of a certain type, cast a literal to that type. Use the unchecked operator, if a literal value cannot be represented in the target type. For example, unchecked((int)0xFF_FF_FF_FF) produces -1.
If the literal is suffixed by U or u, its type is the first of the following types in which its value can be represented: uint, ulong.
If the literal is suffixed by L or l, its type is the first of the following types in which its value can be represented: long, ulong.
Note
You can use the lowercase letter l as a suffix. However, this generates a compiler warning because the letter l can be confused with the digit 1. Use L for clarity.
If the literal is suffixed by UL, Ul, uL, ul, LU, Lu, lU, or lu, its type is ulong.

If the value represented by an integer literal exceeds UInt64.MaxValue, a compiler error CS1021 occurs.

If the determined type of an integer literal is int and the value represented by the literal is within the range of the destination type, the value can be implicitly converted to sbyte, byte, short, ushort, uint, ulong, nint or nuint:

byte a = 17;
byte b = 300;   // CS0031: Constant value '300' cannot be converted to a 'byte'

As the preceding example shows, if the literal's value isn't within the range of the destination type, a compiler error CS0031 occurs.

You can also use a cast to convert the value represented by an integer literal to the type other than the determined type of the literal:

var signedByte = (sbyte)42;
var longVariable = (long)42;

Conversions

You can convert any integral numeric type to any other integral numeric type. If the destination type can store all values of the source type, the conversion is implicit. Otherwise, you need to use a cast expression to perform an explicit conversion. For more information, see Built-in numeric conversions.

Native sized integers

Native sized integer types have special behavior because the storage is determined by the natural integer size on the target machine.

To get the size of a native-sized integer at run time, you can use sizeof(). However, the code must be compiled in an unsafe context. For example:

Console.WriteLine($"size of nint = {sizeof(nint)}");  
Console.WriteLine($"size of nuint = {sizeof(nuint)}");  
// output when run in a 64-bit process  
//size of nint = 8  
//size of nuint = 8  
// output when run in a 32-bit process  
//size of nint = 4  
//size of nuint = 4

You can also get the equivalent value from the static IntPtr.Size and UIntPtr.Size properties.

To get the minimum and maximum values of native-sized integers at run time, use MinValue and MaxValue as static properties with the nint and nuint keywords, as in the following example:

Console.WriteLine($"nint.MinValue = {nint.MinValue}");  
Console.WriteLine($"nint.MaxValue = {nint.MaxValue}");  
Console.WriteLine($"nuint.MinValue = {nuint.MinValue}");  
Console.WriteLine($"nuint.MaxValue = {nuint.MaxValue}");  
// output when run in a 64-bit process  
//nint.MinValue = -9223372036854775808  
//nint.MaxValue = 9223372036854775807  
//nuint.MinValue = 0  
//nuint.MaxValue = 18446744073709551615  
// output when run in a 32-bit process  
//nint.MinValue = -2147483648  
//nint.MaxValue = 2147483647  
//nuint.MinValue = 0  
//nuint.MaxValue = 4294967295

You can use constant values in the following ranges:
- For nint: Int32.MinValue to Int32.MaxValue.
- For nuint: UInt32.MinValue to UInt32.MaxValue.
The compiler provides implicit and explicit conversions to other numeric types. For more information, see Built-in numeric conversions.
There's no direct syntax for native-sized integer literals. There's no suffix to indicate that a literal is a native-sized integer, such as L to indicate a long. You can use implicit or explicit casts of other integer values instead. For example:

nint a = 42  
nint a = (nint)42;

C# language specification