An integer is a whole number with no decimal point. In Rust, integers are primitive types: basic building blocks for values such as 8, −8, and 0. Add a decimal point and Rust treats the number as a different kind of type, a float. section 1.3 section 1.4
Choose an integer type by asking two questions:
- Could the value be negative? Signed types can hold negative values, zero, and positive values; unsigned types hold only nonnegative values. section 1.3
- How large might it be? The number in a type name gives its bit width: u8 has 8 bits and can hold values up to 255; u16 can hold values up to 65,535. Bits store binary digits, each either 0 or 1. section 1.3
Let Rust infer a type—or choose one
In `let my_number = 8;`, Rust chooses `i32`. To choose a different type, write `let my_number: u8 = 8;` or put the type on the number: `let my_number = 8u8;`. Underscores can make a large number easier to read, as in `100_000_000_i32`; they do not change its value. section 1.4
error[E0604]: only u8 can be cast as char, not i32
--> src\main.rs:3:20
|
3 | println!("{}", my_number as char);
| ^^^^^^^^^^^^^^^^^
A compiler error shows why the precise integer type can matter: an inferred `i32` cannot be cast directly to `char` this way; the example requires `u8`. section 1.3
