Understanding Go’s Constants: More Flexible Than They Seem

When I first got into Go, I thought constants were simple and limited—just fixed values, nothing fancy. But as I delve deeper, I find they're quite versatile. Yes, they're fixed values, but Go handles them in ways that are both flexible and efficient. It's pretty cool. Let's see what that means with some practical examples.

Constants as Type-Free Values (Until They're Used)

In Go, constants are often untyped until you actually use them. They have a default kind but can be assigned to variables of different types, as long as the value fits. This makes them adaptable in a way that's unusual for a statically typed language.

Here's how that looks:

const x = 10
var i int = x
var f float64 = x
var b byte = x

Bit analogous to Schrödinger’s paradox, x can be an int, a float64, or even a byte until you assign it. This temporary flexibility lets x work smoothly with different types in your code. No need for casting, which keeps things neat.

You can even mix constants of different types in expression and Go will figure out the best type for the result:

const a = 1.5
const b = 2
const result = a * b // result is float64

Since a is a floating-point number, Go promotes the whole expression to float64. So you don't have to worry about losing precision—Go handles it. But be careful: if you try to assign result to an int, you'll get an error. Go doesn't allow implicit conversions that might lose data.

Limitations

This flexibility only goes far. Once you assign a constant to a variable, that variable's type is set:

const y = 10
var z int = y       // z is an int
var k float64 = y   // y can still be used as float64

But if you try this:

const y = 10.5
var m int = y       // Error: constant 10.5 truncated to integer

Go will throw an error because it won't automatically convert a floating-point constant to an integer without an explicit cast. So while constants are flexible, they won't change type to fit incompatible variables.

Understanding Type Defaults

When you use untyped constants without specifying a type, they assume a default type:

  • Untyped Integer Constants default to int.

  • Untyped Floating-Point Constants default to float64.

  • Untyped Rune Constants default to rune (which is int32).

  • Untyped Complex Constants default to complex128.

  • Untyped String Constants default to string.

  • Untyped Boolean Constants default to bool.

Here's a quick table:

Constant KindCan Adapt To
Untyped Integerint, int8, int16, int32, int64, uint, uint8, uint16, uint32, uint64, uintptr, float32, float64, complex64, complex128
Untyped Floatfloat32, float64, complex64, complex128
Untyped Complexcomplex64, complex128
Untyped Runerune, int32, any integer type that can hold the value
Untyped Stringstring
Untyped Booleanbool

Compile-Time Evaluation and Performance

Go doesn't just evaluate constants at compile time—it also optimizes constant expressions. That means you can use constants in calculations, and Go will compute the result during compilation:

const a = 100
const b = 5
const c = a * b + 20 // c is computed at compile time

So c isn't recalculated at runtime; Go has already figured out it's 520 at compile time. This can boost performance, especially in code where speed matters. By using constants, Go handles the calculations once, instead of doing them every time your program runs.

Constants in Conditional Compilation

Go doesn't have a preprocessor like some other languages, but you can use constants in if statements to include or exclude code at compile time.

const debug = false

func main() {
    if debug {
        fmt.Println("Debugging enabled")
    }
    // The above block might be removed by the compiler if debug is false
}

When debug is false, the compiler knows the if condition will never be true and might leave out the code inside the block. This can make your final binary smaller. Pretty handy, right?

Working with Big Numbers

One powerful feature of Go's constants is that they support very large numbers. Untyped numeric constants in Go have "infinite" precision, limited only by memory and the compiler.

const bigNum = 1e1000 // This is a valid constant

Even though bigNum is way bigger than any built-in numeric type like float64 or int, Go lets you define it as a constant. You can do calculations with these large numbers at compile time:

const (
    a = 1e20
    b = 1e30
    c = a * b // c is 1e50
)

Typed Constants with iota

If you've been using Go, you've probably seen iota for creating enumerated constants. It's useful because it automatically assigns incremental values.

You can also use expressions in constant declarations with iota to create related constants.

const (
    _ = iota
    KB = 1 << (10 * iota)
    MB
    GB
    TB
)

This code defines constants for kilobyte, megabyte, gigabyte, and terabyte using bit shifting. It's calculated at compile time. It's a neat way to generate a series of related constants.

I find iota really helpful for this kind of stuff. As Go doesn't have a built-in enum type, you can effectively simulate enums using the iota identifier and custom types.

Constants with Bitwise Operations and Shifts

Constants can use bitwise operations and shifts, even resulting in values that are bigger than any built-in type.

const (
    shiftAmount = 100
    shiftedValue = 1 << shiftAmount // shiftedValue is a huge number (1267650600228229401496703205376)
)

Here, shiftedValue becomes a very large number because of the big shift amount. This value is too big for standard integer types but is valid as a constant until you try to assign it:

var n int = shiftedValue // Error: constant overflows int

This shows that constants can represent values you can't store in variables, allowing for compile-time calculations with very large numbers.


Limitations with Constants

While Go’s constants are flexible, there are some things they can't do.

Constants Cannot Be Referenced by Pointers

Constants don't have a memory address at runtime. So you can't take the address of a constant or use a pointer to it.

const x = 10
var p = &x // Error: cannot take the address of x

Constants with Typed nil Pointers

While nil can be assigned to variables of pointer, slice, map, channel, and function types, you cannot create a constant that holds a typed nil pointer.

const nilPtr = (*int)(nil) // Error: const initializer (*int)(nil) is not a constant

This adds to the immutability and compile-time nature of constants in Go.

Function Calls in Constant Declarations

Only certain built-in functions can be used in constant expressions, like len, cap, real, imag, and complex.

const str = "hello"
const length = len(str) // This works

const pow = math.Pow(2, 3) // Error: math.Pow cannot be used in constant expressions

This is because, those built-in functions can be used

Composite Types and Constants

Constants can't directly represent composite types like slices, maps, or structs. But you can use constants to initialize them.

const mySlice = []int{1, 2, 3} // Error: []int{…} is not constant

The code above doesn't work because you can't declare a slice as a constant. However, you can use constants inside a variable slice:

const a = 1
const b = 2
const c = 3

var mySlice = []int{a, b, c} // This is fine

Just remember, the types like slice itself isn't a constant—you can't declare it as one. The elements inside can be constants, though.

Explicit Conversion When Needed

If an untyped constant can't be directly assigned due to a type mismatch or possible loss of precision, you need to use an explicit type conversion.

const y = 1.9999

var i int = int(y) // This works, but you lose the decimal part

I hope this gives you a better idea about constants in Go. They're not just simple fixed values; they're a powerful feature that can make your code more expressive and efficient.

If you found this helpful, I'd really appreciate your feedback. I'm eager to learn and improve!

0
Subscribe to my newsletter

Read articles from Ashwin Gopalsamy directly inside your inbox. Subscribe to the newsletter, and don't miss out.

Written by

Ashwin Gopalsamy
Ashwin Gopalsamy

Product-first engineer, blogger and open-source contributor with around 4 years of experience in software development, cloud-native architecture and distributed systems. I build fintech products that process millions of transactions daily and drive substantial revenue. My expertise spans designing, architecting and deploying scalable software, focusing on the business under the code. I collaborate closely with engineers, product owners, and guilds, known for my clear communication and team-centric approach in dynamic environments. Colleagues appreciate my adaptability, openness and focus on diverse, meaningful contributions. Beyond coding, I’m recognized for my documentation, ownership and presentation skills, which drive clarity and engagement across teams. Bilingual in English and Deustch, I bridge cross-functional teams across geographies, ensuring smooth, efficient communication. I’m always open to new opportunities for connection and collaboration. Let’s connect and explore ways to create together.