# Golang rand, crypto: Random Number Generators

Use the rand package and generate random numbers. Call rand.Int and Seed.**Rand.** Everything, even the color of the grass, may be deterministic in this world. But many events seem random. We use "math/rand" in Go to simulate this effect.

**With rand.Int** we get random integers. And with rand.Seed we initialize the pseudo-random number generator that the Go language uses.

**An intro program.** This program generates 5 pseudo-random numbers. It uses a default seed value (which initializes the generator) of 1.

**So:** The program generates the same numbers each time. Try running it multiple times—it always has the same result.

**Int:** This method returns a positive (non-negative) int. No fractional or negative values are possible.

**Tip:** To change the initial series of values returned by rand.Int, use the Seed method.

**Golang program that uses rand.Int**
package main
import (
"fmt"
"math/rand"
)
func main() {*
// Loop five times.
*for i := 0; i < 5; i++ {*
// Get random positive integer.
*value := __rand.Int__()
fmt.Println(value)
}
}
**Output**
5577006791947779410
8674665223082153551
6129484611666145821
4037200794235010051
3916589616287113937

**Seed.** This method initializes the random source. When we use a seed based on the current time, each program execution will start with a different random sequence.

**Here:** We use the current Nanoseconds from the time. We cast this to an int64 to seed the number generator.

Time**Note:** If you run this program many times, each execution will print a different random number.

**Golang program that uses Seed**
package main
import (
"fmt"
"math/rand"
"time"
)
func main() {*
// Call Seed, using current nanoseconds.
*rand.__Seed__(int64(time.Now().Nanosecond()))*
// Random int will be different each program execution.
*value := rand.Int()
fmt.Println(value)
}
**Output**
6912970929678809162

**Rand.Perm.** This function is useful. It generates slices of Ints with values from 0 to the max index of the slice. Each value appears only once. So it creates permutations of Ints.

**Here:** We call rand.Perm with an argument of 5. So the numbers 0, 1, 2, 3, and 4 are randomly arranged in slices of 5 elements.

**Important:** Perm uses the default random source. So we would need to use Seed to avoid the same output on each program run.

**Golang program that uses rand.Perm**
package main
import (
"fmt"
"math/rand"
)
func main() {
for i := 0; i < 5; i++ {*
// Use rand.Perm to generate a random array of numbers.
*numbers := rand.__Perm__(*5*)
fmt.Println(numbers)
}
}
**Output**
[0 4 2 3 1]
[4 1 2 0 3]
[2 3 0 4 1]
[0 4 2 1 3]
[1 0 4 2 3]

**Crypto rand.** Random numbers come in varying levels of randomness. For the best random numbers built into Go, we should use crypto rand. We must use the "math/big" package.

**Program:** We call the rand.Int method in "crypto/rand" 20 times. We get values that are from 0 to 99 inclusive (we never get 100).

**Golang program that uses crypto rand**
package main
import (
"crypto/rand"
"math/big"
"fmt"
)
func main() {*
// Generate 20 random numbers with exclusive max of 100.
// ... So max value returned is 99.
// All values returned are 0 or greater as well.
*for i := 0; i < 20; i++ {
result, _ := __rand.Int__(rand.Reader, big.NewInt(100))
fmt.Println(result)
}
}
**Output**
54
77
56
21
56
81
55
16
57
88
95
58
12
74
23
35
72
29
25
68

**Some research.** Here is some research on Seed. If we do not use Seed each program execution will yield the same numbers. So we should seed with time or another source of randomness.

**Note:** The programs in this example set should all use rand.Seed for optimally random behavior.

**Quote:** Use the Seed function to initialize the default Source if different behavior is required for each run.

Rand: golang.org

**Global versus local.** With top-level rand methods, we use a global source. But we can create Rand instances that are local. This helps with many randomness requirements.

© 2007-2019 Sam Allen. Every person is special and unique. Send bug reports to info@dotnetperls.com.