Overview

‘mat/rand’ package of golang contains a Intn function that can be used to generate a pseudo-random number between [0,n). Bracket at the end means that n is exclusive. This function can be utilized to generate a random string from a character set.

To know more about what pseudo-random number means, checkout this post – https://golangbyexample.com/generate-random-number-golang

Below is the signature of this method. It takes input a number n and will return a number x in range 0<=x<n.

func Intn(n int) int

Above function can be used to generate a random string. Basically we first select a charSet. Then we use the above function to generate a random number and then use that random number to get a random character from the charSet. This random character is added to a string until we have a random string of desired length.

Code

package main

import (
    "fmt"
    "math/rand"
    "strings"
    "time"
)

func main() {
    rand.Seed(time.Now().Unix())

    //Only lowercase
    charSet := "abcdedfghijklmnopqrst"
    var output strings.Builder
    length := 10
    for i := 0; i < length; i++ {
        random := rand.Intn(len(charSet))
        randomChar := charSet[random]
        output.WriteString(string(randomChar))
    }
    fmt.Println(output.String())
    output.Reset()

    //Lowercase and Uppercase Both
    charSet = "abcdedfghijklmnopqrstABCDEFGHIJKLMNOP"
    length = 20
    for i := 0; i < length; i++ {
        random := rand.Intn(len(charSet))
        randomChar := charSet[random]
        output.WriteString(string(randomChar))
    }
    fmt.Println(output.String())
}

Output:

Below is the output on my machine. On your's it might give a different output

himsemkpkd
nHaiEpccEdBfCFPtaBbi

In above program we are using character set as

abcdedfghijklmnopqrst and abcdedfghijklmnopqrstABCDEFGHIJKLMNOP

All the above characters in charSet were ASCII characters hence we were able to index a character in charSet string. But this could be problem if the charSet contains non-ASCII character.

In Golang string is a sequence of bytes. A string literal actually represents a UTF-8 sequence of bytes. In UTF-8, ASCII characters are single-byte corresponding to the first 128 Unicode characters. All other characters are between 1 to 4 bytes. Due to this it is not possible to index a character in a string.  In GO, rune data type represents a Unicode point.  Once a string is converted to an array of rune then it is possible to index a character in that array of rune.

So in case, the character set contains some characters that are not ASCII they might occupy more than 1 bytes. In that case, we cannot use the above code to generate a random string as we cannot index into the charSet. For this case, we have to first convert a string into a rune array so that we can index into the rune array to the character and then incrementally form the random string.

As in the below example, our charSet contains a non-ASCII character '£'. This character occupies two bytes

package main

import (
    "fmt"
    "math/rand"
    "strings"
    "time"
)

func main() {
    rand.Seed(time.Now().Unix())
    //Only lowercase and £
    charSet := []rune("abcdedfghijklmnopqrst£")
    var output strings.Builder
    length := 10
    for i := 0; i < length; i++ {
        random := rand.Intn(len(charSet))
        randomChar := charSet[random]
        output.WriteRune(randomChar)
    }
    fmt.Println(output.String())
    output.Reset()
 
   //Lowercase and Uppercase Both and £
    charSet = []rune("abcdedfghijklmnopqrstABCDEFGHIJKLMNOP£")
    length = 20
    for i := 0; i < length; i++ {
        random := rand.Intn(len(charSet))
        randomChar := charSet[random]
        output.WriteRune(randomChar)
    }
    fmt.Println(output.String())
}

Output:

Below is the output on my machine. On yours it might give a different output

aidqpbse£j
rebhjblsePsLpGBPOhfB