Skip to content

INTEGERS (NAMBA) AND FLOATS (DESIMALI) ​

Integers and floats are the basic numeric data types in Nuru, used for representing whole numbers and decimal numbers, respectively. This page covers the syntax and usage of integers and floats in Nuru, including precedence, unary increments, shorthand assignments, and negative numbers.

PRECEDENCE ​

Integers and floats behave as expected in mathematical operations, following the BODMAS rule:

go
2 + 3 * 5 // 17

fanya a = 2.5
fanya b = 3/5

a + b // 2.8
2 + 3 * 5 // 17

fanya a = 2.5
fanya b = 3/5

a + b // 2.8

UNARY INCREMENTS ​

You can perform unary increments (++ and --) on both floats and integers. These will add or subtract 1 from the current value. Note that the float or int have to be assigned to a variable for this operation to work. Here's an example:

go
fanya i = 2.4

i++ // 3.4
fanya i = 2.4

i++ // 3.4

SHORTHAND ASSIGNMENT ​

Nuru supports shorthand assignments with +=, -=, /=, *=, and %=: You

go
fanya i = 2

i *= 3 // 6
i /= 2 // 3
i += 100 // 103
i -= 10 // 93
i %= 90 // 3
fanya i = 2

i *= 3 // 6
i /= 2 // 3
i += 100 // 103
i -= 10 // 93
i %= 90 // 3

NEGATIVE NUMBERS ​

Negative numbers also behave as expected:

go
fanya i = -10

wakati (i < 0) {
    andika(i)
    i++
}
fanya i = -10

wakati (i < 0) {
    andika(i)
    i++
}

Output:

go
-10
-9
-8
-7
-6
-5
-4
-3
-2
-1
0
1
2
3
4
5
6
7
8
9
-10
-9
-8
-7
-6
-5
-4
-3
-2
-1
0
1
2
3
4
5
6
7
8
9

All code is open source if you can read Assembly