Crate count_digits
source ·Expand description
A no_std trait to count the digits of integer types in various number bases.
Compatible with all primitive integer types and all non-zero integer types.
pub trait CountDigits: Copy + Sized {
type Radix;
/// Returns the count of bits in an integer.
fn count_bits(self) -> u32;
/// Returns the count of octal digits in an integer.
fn count_octal_digits(self) -> u32;
/// Returns the count of hexadecimal digits in an integer.
fn count_hex_digits(self) -> u32;
/// Returns the count of decimal digits in an integer.
fn count_digits(self) -> usize;
/// Returns the count of digits in an integer for a given radix.
fn count_digits_radix(self, radix: Self::Radix) -> usize;
}
§Examples
use count_digits::CountDigits;
assert_eq!(16___u32, 0b1111000000001101.count_bits());
assert_eq!(16_usize, 0b1111000000001101.count_digits_radix(2_u32));
assert_eq!(06___u32, 0o170015.count_octal_digits());
assert_eq!(06_usize, 0o170015.count_digits_radix(8_u32));
assert_eq!(04___u32, 0xF00D.count_hex_digits());
assert_eq!(04_usize, 0xF00D.count_digits_radix(16_u32));
assert_eq!(05_usize, 61453.count_digits());
assert_eq!(05_usize, 61453.count_digits_radix(10_u32));
The named functions for which the radix is a power of two ( count_bits(), count_octal_digits(), and count_hex_digits() ) return a u32 for compatibility with Rust’s bit-shifting functions and constants, which all use u32 for arguments and return types.
assert_eq!(0b1011___u8.count_bits(), u8::BITS - 0b1011___u8.leading_zeros());
assert_eq!(0b1011__i32.count_bits(), i32::BITS - 0b1011__i32.leading_zeros());
assert_eq!(0b1011_u128.count_bits(), u128::BITS - 0b1011_u128.leading_zeros());
The base-10 count_digits() function returns usize for compatibility with Rust’s formatting macros.
let max_digits = [1, 2, 15, 105]
.iter()
.map(CountDigits::count_digits)
.max()
.unwrap();
for n in [1, 2, 15, 105] {
assert_eq!(3, format!("{n:0>pad$}", pad = max_digits).len());
}
In the case of formatting binary, octal, or hex numbers, the count_digits_radix(2 | 8 | 16) function can be used to retrieve the desired count directly as a usize.
let max_bits = [0b1, 0b10, 0b101, 0b1011]
.iter()
.map(|n| n.count_digits_radix(2u32))
.max()
.unwrap();
for n in [0b1, 0b10, 0b101, 0b1011] {
assert_eq!(4, format!("{n:0>pad$}", pad = max_bits).len());
}
assert_eq!(5, 12345_i32.wrapping_neg().count_digits());
assert_eq!(5, 12345_i32.wrapping_neg().count_digits_radix(10));
Since negative numbers represented in base-10 are displayed with a negative sign, the base-10 digit count of a positive number will be equal to the base-10 digit count of the number’s negated value, assuming no wrapping occurred.
assert_eq!(
867_5309_i32.count_digits(),
867_5309_i32.wrapping_neg().count_digits(),
);
However, the digit counts of negative numbers represented in other bases reflect the twos-complement representation, and the digit count of a positive number will not be the same as the count of its negated value.
for radix in 2..=16 {
match radix {
10 => assert_eq!(
0xF00D_i32.count_digits_radix(radix),
0xF00D_i32.wrapping_neg().count_digits_radix(radix),
),
_ => assert_ne!(
0xBAD_i32.count_digits_radix(radix),
0xBAD_i32.wrapping_neg().count_digits_radix(radix),
),
}
}
This behavior is consistent with Rust’s display format.
assert_eq!(1, format!("{:b}", 1_i8).chars().count());
assert_eq!(1, format!("{:o}", 1_i8).chars().count());
assert_eq!(1, format!("{ }", 1_i8).chars().count());
assert_eq!(1, format!("{:x}", 1_i8).chars().count());
assert_eq!(8, format!("{:b}", -1_i8).chars().count());
assert_eq!(3, format!("{:o}", -1_i8).chars().count());
assert_eq!(1, format!("{ }", -1_i8).strip_prefix('-').unwrap().chars().count());
assert_eq!(2, format!("{:x}", -1_i8).chars().count());
§Benchmarks
Traits§
- A no_std trait to count the digits of integer types in various number bases.