Macro diem_infallible::checked

macro_rules! checked {
    ($a:tt + $b:tt) => { ... };
    ($a:tt - $b:tt) => { ... };
    ($a:tt * $b:tt) => { ... };
    ($a:tt / $b:tt) => { ... };
    ($a:tt + $($tokens:tt)*) => { ... };
    ($a:tt - $($tokens:tt)*) => { ... };
    ($a:tt * $($tokens:tt)*) => { ... };
    ($a:tt / $($tokens:tt)*) => { ... };
}

Utility macro for writing secure arithmetic operations in order to avoid integer overflows.

Examples

let a: i64 = 1;
let b: i64 = 2;
let c: i64 = 3;

assert_eq!(checked!(a + b).unwrap(), 3);
assert_eq!(checked!(a + b + c).unwrap(), 6);

// When doing multiple different operations, it's important to use parentheses in order
// to guarantee the order of operation!
assert_eq!(checked!(a + ((b - c) * c)).unwrap(), -2);

// When using numeric literals, the compiler might not be able to infer the type properly,
// so if it complains, just add the type to the number.
assert_eq!(checked!(10_u32 / 2_u32).unwrap(), 5);
assert_eq!(checked!(10_u32 * 2_u32).unwrap(), 20);
assert_eq!(checked!(10_u32 - 2_u32).unwrap(), 8);
assert_eq!(checked!(2_i32 - 10_i32).unwrap(), -8);
assert_eq!(checked!(10_u32 + 2_u32).unwrap(), 12);

// Casts using `as` operator must appear within parenthesis
assert_eq!(checked!(10_u32 + (2_u16 as u32)).unwrap(), 12);

assert_eq!(checked!(10_u32 / (1_u32 + 1_u32)).unwrap(), 5);
assert_eq!(checked!(10_u32 * (1_u32 + 1_u32)).unwrap(), 20);
assert_eq!(checked!(10_u32 - (1_u32 + 1_u32)).unwrap(), 8);
assert_eq!(checked!(10_u32 + (1_u32 + 1_u32)).unwrap(), 12);

let max = u32::max_value();
assert!(checked!(max + 1_u32).is_err());
assert!(checked!(0_u32 - 1_u32).is_err());

// When one of the operands is an associated function or member, due to limitations with the
// macro syntax which disallows an `expr` to precede a `+` sign, make sure to wrap the expression
// in parenthesis
assert_eq!(checked!((foo.bar) + 1_i32).unwrap(), 2);
assert_eq!(checked!(1_i32 + (Foo::one())).unwrap(), 2);