Runes in Go

Go's rune type lets you work with Unicode code points that can take more than one byte in a UTF-8 string.

All the content from our Boot.dev courses are available for free here on the blog. This one is the "Constants and Formatting" chapter of Learn Go. If you want to try the far more immersive version of the course, do check it out!

Characters and Bytes

In many programming languages (cough, C, cough), a "character" is a single byte. Using ASCII encoding, the standard for the C programming language, we can represent 128 characters with 7 bits. This is enough for the English alphabet, numbers, and some special characters.

Runes

In Go, strings are just sequences of bytes: they can hold arbitrary data. However, Go also has a special type, rune, which is an alias for int32. This means that a rune is a 32-bit integer, which is large enough to hold any Unicode code point.

UTF-8 String Encoding

When you're working with strings, you need to be aware of the encoding (bytes -> representation). Go uses UTF-8 encoding, which is a variable-length encoding.

What Does This Mean?

There are 2 main takeaways:

1. When you need to work with individual Unicode code points in a string, you should use the rune type. It breaks strings up into code points, which can be more than one byte long.
2. We can include a wide variety of Unicode characters in our strings, such as emojis and Chinese characters, and Go will handle them just fine.

A Bear Emoji

The simple string boots has 5 bytes and 5 runes. The bear emoji is a different story:

package main

import (
	"fmt"
	"unicode/utf8"
)

func main() {
	fmt.Println(len("boots"))                    // 5
	fmt.Println(utf8.RuneCountInString("boots")) // 5

	fmt.Println(len("🐻"))                    // 4
	fmt.Println(utf8.RuneCountInString("🐻")) // 1
}

The emoji is only one rune, but it takes up 4 bytes.