Numbers
By default all of the Lua interpreters come with double as their default numeric type; While it is possible to compile Lua for embedded environments without support for floating point numbers by configuring Lua to use int or long as its underlying type, We assume it being always double in the Fuse programming language and any number literal is being treated as if it is reflecting the IEEE 754 standard.
NOTE: From Lua 5.3 and onward it has support for integer types(either 64-bit or 32-bit depending on the configuration in the build step), But it will break compatibility with alternative runtimes and also isn’t backward compatible with older versions of Lua itself.
A number literal contains an integer part and an optional fractal part which are separated by a dot(.), We can also use decimal exponents in our numbers.
Here are some different ways to define number literals.
let a = 1
let b = 0.5
let b = 9.8
let c = 5.43e-21
let d = 0.123e45
let e = 3e+14
let f = 100_000_000_000
let g = 0.000_000_000_001
NOTE: Underlines can be used freely within any number literal to enhance its readability!
We can also define numbers using Hexadecimal or Binary literals.
let hex = 0xFF
let bin = 0b11111111
Operations
All Fuse numbers come with a default implementation for all the standard arithmetic operators supported in Lua 5.4, These operators are addition(+), subtraction(-), multiplication(*), division(/), floor division(//), modulo(%), exponentiation(^) and unary minus(-).
assert_eq(1 + 2, 3)
assert_eq(100_000 - 1, 99_999)
assert_eq(25 * 4, 100)
assert_eq(12 / 3, 4)
assert_eq(10 // 3, 3)
assert_eq(101 % 10, 1)
assert_eq(2 ^ 10, 1024)
assert_eq(-42, 0 - 42)
Bitwise
In addition to these arithmetic operators mentioned above, Numbers also implement bitwise operators. Bitwise operators will implicitly cast the number to an integer(either a native or simulated one in case of targeting runtimes with no support for integers).
assert_eq(0xf0 & 0xff, 0xf0) -- bitwise AND
assert_eq(0x0f | 0xf0, 0xff) -- bitwise OR
assert_eq(0xff ! 0x0f == 0xf0) -- bitwise exclusive NOT
assert_eq(0xffff >> 8, 0xff) -- bitwise right shift
assert_eq(0xff << 8, 0xff00) -- bitwise left shift
assert_eq(!0xffffffffffffff0f | 0xf0, 0xff) -- unary bitwise NOT
Comparison
Numbers can be compared to each other using the following operations.
assert_eq(5 == 5, true) -- equality operator
assert_eq(1 != 5, true) -- inequality operator
assert_eq(1 < 5, true) -- less than operator
assert_eq(1 > 5, false) -- greater than operator
assert_eq(5 <= 5, true) -- less than or equal operator
assert_eq(5 >= 5, true) -- greater than or equal operator
NOTE: You can alter how these operators behave by reimplementing their respective traits. Read more on that on the Operator Traits page.