# u-num-it
A simple procedural macro for matching `typenum::consts` in a given range or array.
It helps you write concise numeric typenum matches like:
```rust
use typenum::{Bit, ToInt, Unsigned};
let x: isize = 3;
let y: isize = u_num_it::u_num_it!(-5..5, match x {
N => {
// negative integer types implement ToInt
NumType::to_int()
},
False => {
// False (B0) is a Bit, not an Integer. Use to_u8() and cast.
NumType::to_u8() as isize
},
P => {
// positive integer types implement ToInt
NumType::to_int()
}
});
assert_eq!(y, 3);
```
Or with an array of arbitrary numbers:
```rust
use typenum::{Bit, ToInt, Unsigned};
let x: isize = 8;
let y: isize = u_num_it::u_num_it!([1, 2, 8, 22], match x {
P => {
NumType::to_int()
},
_ => panic!("unexpected")
});
assert_eq!(y, 8);
```
Instead of writing the explicit (and repetitive) match yourself:
```rust
use typenum::{Bit, ToInt, Unsigned};
let x: isize = 3;
let y: isize = match x {
-5 => typenum::consts::N5::to_int(),
-4 => typenum::consts::N4::to_int(),
-3 => typenum::consts::N3::to_int(),
-2 => typenum::consts::N2::to_int(),
-1 => typenum::consts::N1::to_int(),
0 => typenum::consts::False::to_u8() as isize, // False is a Bit (B0)
1 => typenum::consts::P1::to_int(),
2 => typenum::consts::P2::to_int(),
3 => typenum::consts::P3::to_int(),
4 => typenum::consts::P4::to_int(),
i => panic!("out of range: {i}")
};
assert_eq!(y, 3);
```
`NumType` is available inside each match arm, bound to the concrete typenum type for the matched value:
```rust
use typenum::{Bit, ToInt};
let x: isize = 3;
let y: isize = u_num_it::u_num_it!(-5..5, match x {
N => {
// NumType is a negative integer type (e.g. typenum::consts::N3 when x = -3)
NumType::to_int()
},
False => {
// NumType is typenum::consts::False (B0). Use to_u8() and cast if you need an isize.
let zero: u8 = NumType::to_u8();
zero as isize
},
P => {
// NumType is typenum::consts::P3 when x = 3
let val: isize = NumType::to_int();
val
}
});
assert_eq!(y, 3);
```
## Notes
- Use `ToInt` for signed/unsigned integer typenum constants (`P*`, `N*`).
- Use `to_u8()` for bit types (`False` = `B0`, `True` = `B1` if you ever introduce it), casting as needed.
- Avoid mixing `P` and `U` in the same macro call (the macro enforces this).
- Literal `0` and `False` are treated as the same value; don't use both in one invocation.