Number Validators for Go

Valgo provides a generic validator for all supported numeric types and family-specific constructors that preserve the exact input type.

Number

Number() accepts any value whose underlying type is a signed integer, unsigned integer, float32, or float64.

v.Is(v.Number(10).EqualTo(10))
v.Is(v.Number(11).GreaterThan(10))
v.Is(v.Number(11).Between(10, 12)) // inclusive
v.Is(v.Number(0).Zero())
v.Is(v.Number(20).InSlice([]int{10, 20, 30}))

Number() provides equality, ordering, inclusive Between(), Zero(), InSlice(), and Passing(). Use NumberP() for pointers; it additionally provides Nil() and ZeroOrNil().

var amount *int
v.Is(v.NumberP(amount, "amount").Nil())

Signed integers

Int() accepts every signed integer type: int, int8, int16, int32, int64, and custom types based on them. Use IntP() for pointers. Rune() and RuneP() remain available when the int32 value is semantically a rune.

v.Is(v.Int(10).Positive())
v.Is(v.Int(int16(10)).GreaterThan(int16(5)))
v.Is(v.Int(int64(-1)).Negative())

age := int16(21)
v.Is(v.IntP(&age).GreaterOrEqualTo(int16(18)))

Signed integer validators add Positive() and Negative() to the common numeric rules. The width-specific constructors Int8(), Int16(), Int32(), Int64() and their pointer forms are deprecated aliases; use Int() or IntP() in new code.

Unsigned integers

Uint() accepts every unsigned integer type: uint, uint8, uint16, uint32, uint64, and custom types based on them. Use UintP() for pointers. Byte() and ByteP() remain available when the uint8 value is semantically a byte.

v.Is(v.Uint(uint64(10)).LessOrEqualTo(uint64(10)))
v.Is(v.Byte(byte(1)).GreaterThan(byte(0)))

The width-specific constructors Uint8(), Uint16(), Uint32(), Uint64() and their pointer forms are deprecated aliases; use Uint() or UintP() in new code.

Floating-point numbers

Float() accepts float32, float64, and custom types based on them. Use FloatP() for pointers.

v.Is(v.Float(float32(1.5)).GreaterThan(float32(1.0)))
v.Is(v.Float(3.14).Finite())

Float validators add Positive(), Negative(), NaN(), Infinite(), and Finite() to the common numeric rules. Float32(), Float64(), Float32P(), and Float64P() are deprecated aliases; use Float() or FloatP() in new code.

Use the is predicate package when you need the same numeric checks as standalone boolean functions without a validation session or error messages.