Struct Duration

pub struct Duration { pub(in ::time) secs: u64, pub(in ::time) nanos: Nanoseconds }

A Duration type to represent a span of time, typically used for system timeouts.

Each Duration is composed of a whole number of seconds and a fractional part represented in nanoseconds. If the underlying system does not support nanosecond-level precision, APIs binding a system timeout will typically round up the number of nanoseconds.

Durations implement many common traits, including Add, Sub, and other ops traits. It implements Default by returning a zero-length Duration.

Examples

use std::time::Duration;

let five_seconds = Duration::new(5, 0);
let five_seconds_and_five_nanos = five_seconds + Duration::new(0, 5);

assert_eq!(five_seconds_and_five_nanos.as_secs(), 5);
assert_eq!(five_seconds_and_five_nanos.subsec_nanos(), 5);

let ten_millis = Duration::from_millis(10);

Formatting Duration values

Duration intentionally does not have a Display impl, as there are a variety of ways to format spans of time for human readability. Duration provides a Debug impl that shows the full precision of the value.

The Debug output uses the non-ASCII "µs" suffix for microseconds. If your program output may appear in contexts that cannot rely on full Unicode compatibility, you may wish to format Duration objects yourself or use a crate to do so.

Fields

secs: u64
nanos: Nanoseconds

Implementations

impl Duration

const SECOND: Duration = _;

The duration of one second.

Examples

#![feature(duration_constants)]
use std::time::Duration;

assert_eq!(Duration::SECOND, Duration::from_secs(1));

const MILLISECOND: Duration = _;

The duration of one millisecond.

Examples

#![feature(duration_constants)]
use std::time::Duration;

assert_eq!(Duration::MILLISECOND, Duration::from_millis(1));

const MICROSECOND: Duration = _;

The duration of one microsecond.

Examples

#![feature(duration_constants)]
use std::time::Duration;

assert_eq!(Duration::MICROSECOND, Duration::from_micros(1));

const NANOSECOND: Duration = _;

The duration of one nanosecond.

Examples

#![feature(duration_constants)]
use std::time::Duration;

assert_eq!(Duration::NANOSECOND, Duration::from_nanos(1));

const ZERO: Duration = _;

A duration of zero time.

Examples

use std::time::Duration;

let duration = Duration::ZERO;
assert!(duration.is_zero());
assert_eq!(duration.as_nanos(), 0);

const MAX: Duration = _;

The maximum duration.

May vary by platform as necessary. Must be able to contain the difference between two instances of Instant or two instances of SystemTime. This constraint gives it a value of about 584,942,417,355 years in practice, which is currently used on all platforms.

Examples

use std::time::Duration;

assert_eq!(Duration::MAX, Duration::new(u64::MAX, 1_000_000_000 - 1));

const fn new(secs: u64, nanos: u32) -> Duration

Creates a new Duration from the specified number of whole seconds and additional nanoseconds.

If the number of nanoseconds is greater than 1 billion (the number of nanoseconds in a second), then it will carry over into the seconds provided.

Panics

This constructor will panic if the carry from the nanoseconds overflows the seconds counter.

Examples

use std::time::Duration;

let five_seconds = Duration::new(5, 0);

const fn from_secs(secs: u64) -> Duration

Creates a new Duration from the specified number of whole seconds.

Examples

use std::time::Duration;

let duration = Duration::from_secs(5);

assert_eq!(5, duration.as_secs());
assert_eq!(0, duration.subsec_nanos());

const fn from_millis(millis: u64) -> Duration

Creates a new Duration from the specified number of milliseconds.

Examples

use std::time::Duration;

let duration = Duration::from_millis(2_569);

assert_eq!(2, duration.as_secs());
assert_eq!(569_000_000, duration.subsec_nanos());

const fn from_micros(micros: u64) -> Duration

Creates a new Duration from the specified number of microseconds.

Examples

use std::time::Duration;

let duration = Duration::from_micros(1_000_002);

assert_eq!(1, duration.as_secs());
assert_eq!(2_000, duration.subsec_nanos());

const fn from_nanos(nanos: u64) -> Duration

Creates a new Duration from the specified number of nanoseconds.

Note: Using this on the return value of as_nanos() might cause unexpected behavior: as_nanos() returns a u128, and can return values that do not fit in u64, e.g. 585 years. Instead, consider using the pattern Duration::new(d.as_secs(), d.subsec_nanos()) if you cannot copy/clone the Duration directly.

Examples

use std::time::Duration;

let duration = Duration::from_nanos(1_000_000_123);

assert_eq!(1, duration.as_secs());
assert_eq!(123, duration.subsec_nanos());

const fn from_nanos_u128(nanos: u128) -> Duration

Creates a new Duration from the specified number of nanoseconds.

Panics

Panics if the given number of nanoseconds is greater than Duration::MAX.

Examples

use std::time::Duration;

let nanos = 10_u128.pow(24) + 321;
let duration = Duration::from_nanos_u128(nanos);

assert_eq!(10_u64.pow(15), duration.as_secs());
assert_eq!(321, duration.subsec_nanos());

const fn from_weeks(weeks: u64) -> Duration

Creates a new Duration from the specified number of weeks.

Panics

Panics if the given number of weeks overflows the Duration size.

Examples

#![feature(duration_constructors)]
use std::time::Duration;

let duration = Duration::from_weeks(4);

assert_eq!(4 * 7 * 24 * 60 * 60, duration.as_secs());
assert_eq!(0, duration.subsec_nanos());

const fn from_days(days: u64) -> Duration

Creates a new Duration from the specified number of days.

Panics

Panics if the given number of days overflows the Duration size.

Examples

#![feature(duration_constructors)]
use std::time::Duration;

let duration = Duration::from_days(7);

assert_eq!(7 * 24 * 60 * 60, duration.as_secs());
assert_eq!(0, duration.subsec_nanos());

const fn from_hours(hours: u64) -> Duration

Creates a new Duration from the specified number of hours.

Panics

Panics if the given number of hours overflows the Duration size.

Examples

use std::time::Duration;

let duration = Duration::from_hours(6);

assert_eq!(6 * 60 * 60, duration.as_secs());
assert_eq!(0, duration.subsec_nanos());

const fn from_mins(mins: u64) -> Duration

Creates a new Duration from the specified number of minutes.

Panics

Panics if the given number of minutes overflows the Duration size.

Examples

use std::time::Duration;

let duration = Duration::from_mins(10);

assert_eq!(10 * 60, duration.as_secs());
assert_eq!(0, duration.subsec_nanos());

const fn is_zero(&self) -> bool

Returns true if this Duration spans no time.

Examples

use std::time::Duration;

assert!(Duration::ZERO.is_zero());
assert!(Duration::new(0, 0).is_zero());
assert!(Duration::from_nanos(0).is_zero());
assert!(Duration::from_secs(0).is_zero());

assert!(!Duration::new(1, 1).is_zero());
assert!(!Duration::from_nanos(1).is_zero());
assert!(!Duration::from_secs(1).is_zero());

const fn as_secs(&self) -> u64

Returns the number of whole seconds contained by this Duration.

The returned value does not include the fractional (nanosecond) part of the duration, which can be obtained using subsec_nanos.

Examples

use std::time::Duration;

let duration = Duration::new(5, 730_023_852);
assert_eq!(duration.as_secs(), 5);

To determine the total number of seconds represented by the Duration including the fractional part, use as_secs_f64 or as_secs_f32

const fn subsec_millis(&self) -> u32

Returns the fractional part of this Duration, in whole milliseconds.

This method does not return the length of the duration when represented by milliseconds. The returned number always represents a fractional portion of a second (i.e., it is less than one thousand).

Examples

use std::time::Duration;

let duration = Duration::from_millis(5_432);
assert_eq!(duration.as_secs(), 5);
assert_eq!(duration.subsec_millis(), 432);

const fn subsec_micros(&self) -> u32

Returns the fractional part of this Duration, in whole microseconds.

This method does not return the length of the duration when represented by microseconds. The returned number always represents a fractional portion of a second (i.e., it is less than one million).

Examples

use std::time::Duration;

let duration = Duration::from_micros(1_234_567);
assert_eq!(duration.as_secs(), 1);
assert_eq!(duration.subsec_micros(), 234_567);

const fn subsec_nanos(&self) -> u32

Returns the fractional part of this Duration, in nanoseconds.

This method does not return the length of the duration when represented by nanoseconds. The returned number always represents a fractional portion of a second (i.e., it is less than one billion).

Examples

use std::time::Duration;

let duration = Duration::from_millis(5_010);
assert_eq!(duration.as_secs(), 5);
assert_eq!(duration.subsec_nanos(), 10_000_000);

const fn as_millis(&self) -> u128

Returns the total number of whole milliseconds contained by this Duration.

Examples

use std::time::Duration;

let duration = Duration::new(5, 730_023_852);
assert_eq!(duration.as_millis(), 5_730);

const fn as_micros(&self) -> u128

Returns the total number of whole microseconds contained by this Duration.

Examples

use std::time::Duration;

let duration = Duration::new(5, 730_023_852);
assert_eq!(duration.as_micros(), 5_730_023);

const fn as_nanos(&self) -> u128

Returns the total number of nanoseconds contained by this Duration.

Examples

use std::time::Duration;

let duration = Duration::new(5, 730_023_852);
assert_eq!(duration.as_nanos(), 5_730_023_852);

const fn abs_diff(self, other: Duration) -> Duration

Computes the absolute difference between self and other.

Examples

use std::time::Duration;

assert_eq!(Duration::new(100, 0).abs_diff(Duration::new(80, 0)), Duration::new(20, 0));
assert_eq!(Duration::new(100, 400_000_000).abs_diff(Duration::new(110, 0)), Duration::new(9, 600_000_000));

const fn checked_add(self, rhs: Duration) -> Option<Duration>

Checked Duration addition. Computes self + other, returning None if overflow occurred.

Examples

use std::time::Duration;

assert_eq!(Duration::new(0, 0).checked_add(Duration::new(0, 1)), Some(Duration::new(0, 1)));
assert_eq!(Duration::new(1, 0).checked_add(Duration::new(u64::MAX, 0)), None);

const fn saturating_add(self, rhs: Duration) -> Duration

Saturating Duration addition. Computes self + other, returning Duration::MAX if overflow occurred.

Examples

use std::time::Duration;

assert_eq!(Duration::new(0, 0).saturating_add(Duration::new(0, 1)), Duration::new(0, 1));
assert_eq!(Duration::new(1, 0).saturating_add(Duration::new(u64::MAX, 0)), Duration::MAX);

const fn checked_sub(self, rhs: Duration) -> Option<Duration>

Checked Duration subtraction. Computes self - other, returning None if the result would be negative or if overflow occurred.

Examples

use std::time::Duration;

assert_eq!(Duration::new(0, 1).checked_sub(Duration::new(0, 0)), Some(Duration::new(0, 1)));
assert_eq!(Duration::new(0, 0).checked_sub(Duration::new(0, 1)), None);

const fn saturating_sub(self, rhs: Duration) -> Duration

Saturating Duration subtraction. Computes self - other, returning Duration::ZERO if the result would be negative or if overflow occurred.

Examples

use std::time::Duration;

assert_eq!(Duration::new(0, 1).saturating_sub(Duration::new(0, 0)), Duration::new(0, 1));
assert_eq!(Duration::new(0, 0).saturating_sub(Duration::new(0, 1)), Duration::ZERO);

const fn checked_mul(self, rhs: u32) -> Option<Duration>

Checked Duration multiplication. Computes self * other, returning None if overflow occurred.

Examples

use std::time::Duration;

assert_eq!(Duration::new(0, 500_000_001).checked_mul(2), Some(Duration::new(1, 2)));
assert_eq!(Duration::new(u64::MAX - 1, 0).checked_mul(2), None);

const fn saturating_mul(self, rhs: u32) -> Duration

Saturating Duration multiplication. Computes self * other, returning Duration::MAX if overflow occurred.

Examples

use std::time::Duration;

assert_eq!(Duration::new(0, 500_000_001).saturating_mul(2), Duration::new(1, 2));
assert_eq!(Duration::new(u64::MAX - 1, 0).saturating_mul(2), Duration::MAX);

const fn checked_div(self, rhs: u32) -> Option<Duration>

Checked Duration division. Computes self / other, returning None if other == 0.

Examples

use std::time::Duration;

assert_eq!(Duration::new(2, 0).checked_div(2), Some(Duration::new(1, 0)));
assert_eq!(Duration::new(1, 0).checked_div(2), Some(Duration::new(0, 500_000_000)));
assert_eq!(Duration::new(2, 0).checked_div(0), None);

const fn as_secs_f64(&self) -> f64

Returns the number of seconds contained by this Duration as f64.

The returned value includes the fractional (nanosecond) part of the duration.

Examples

use std::time::Duration;

let dur = Duration::new(2, 700_000_000);
assert_eq!(dur.as_secs_f64(), 2.7);

const fn as_secs_f32(&self) -> f32

Returns the number of seconds contained by this Duration as f32.

The returned value includes the fractional (nanosecond) part of the duration.

Examples

use std::time::Duration;

let dur = Duration::new(2, 700_000_000);
assert_eq!(dur.as_secs_f32(), 2.7);

const fn as_millis_f64(&self) -> f64

Returns the number of milliseconds contained by this Duration as f64.

The returned value includes the fractional (nanosecond) part of the duration.

Examples

#![feature(duration_millis_float)]
use std::time::Duration;

let dur = Duration::new(2, 345_678_000);
assert_eq!(dur.as_millis_f64(), 2_345.678);

const fn as_millis_f32(&self) -> f32

Returns the number of milliseconds contained by this Duration as f32.

The returned value includes the fractional (nanosecond) part of the duration.

Examples

#![feature(duration_millis_float)]
use std::time::Duration;

let dur = Duration::new(2, 345_678_000);
assert_eq!(dur.as_millis_f32(), 2_345.678);

fn from_secs_f64(secs: f64) -> Duration

Creates a new Duration from the specified number of seconds represented as f64.

Panics

This constructor will panic if secs is negative, overflows Duration or not finite.

Examples

use std::time::Duration;

let res = Duration::from_secs_f64(0.0);
assert_eq!(res, Duration::new(0, 0));
let res = Duration::from_secs_f64(1e-20);
assert_eq!(res, Duration::new(0, 0));
let res = Duration::from_secs_f64(4.2e-7);
assert_eq!(res, Duration::new(0, 420));
let res = Duration::from_secs_f64(2.7);
assert_eq!(res, Duration::new(2, 700_000_000));
let res = Duration::from_secs_f64(3e10);
assert_eq!(res, Duration::new(30_000_000_000, 0));
// subnormal float
let res = Duration::from_secs_f64(f64::from_bits(1));
assert_eq!(res, Duration::new(0, 0));
// conversion uses rounding
let res = Duration::from_secs_f64(0.999e-9);
assert_eq!(res, Duration::new(0, 1));

fn from_secs_f32(secs: f32) -> Duration

Creates a new Duration from the specified number of seconds represented as f32.

Panics

This constructor will panic if secs is negative, overflows Duration or not finite.

Examples

use std::time::Duration;

let res = Duration::from_secs_f32(0.0);
assert_eq!(res, Duration::new(0, 0));
let res = Duration::from_secs_f32(1e-20);
assert_eq!(res, Duration::new(0, 0));
let res = Duration::from_secs_f32(4.2e-7);
assert_eq!(res, Duration::new(0, 420));
let res = Duration::from_secs_f32(2.7);
assert_eq!(res, Duration::new(2, 700_000_048));
let res = Duration::from_secs_f32(3e10);
assert_eq!(res, Duration::new(30_000_001_024, 0));
// subnormal float
let res = Duration::from_secs_f32(f32::from_bits(1));
assert_eq!(res, Duration::new(0, 0));
// conversion uses rounding
let res = Duration::from_secs_f32(0.999e-9);
assert_eq!(res, Duration::new(0, 1));

fn mul_f64(self, rhs: f64) -> Duration

Multiplies Duration by f64.

Panics

This method will panic if result is negative, overflows Duration or not finite.

Examples

use std::time::Duration;

let dur = Duration::new(2, 700_000_000);
assert_eq!(dur.mul_f64(3.14), Duration::new(8, 478_000_000));
assert_eq!(dur.mul_f64(3.14e5), Duration::new(847_800, 0));

Note that f64 does not have enough bits (f64::MANTISSA_DIGITS) to represent the full range of possible Duration with nanosecond precision, so rounding may occur even for trivial operations like multiplying by 1.

# #![feature(float_exact_integer_constants)]
use std::time::Duration;

// This is about 14.9 weeks, remaining precise to the nanosecond:
let weeks = Duration::from_nanos(f64::MAX_EXACT_INTEGER as u64);
assert_eq!(weeks, weeks.mul_f64(1.0));

// A larger value incurs rounding in the floating-point operation:
let weeks = Duration::from_nanos(u64::MAX);
assert_ne!(weeks, weeks.mul_f64(1.0));

// This is over 285 million years, remaining precise to the second:
let years = Duration::from_secs(f64::MAX_EXACT_INTEGER as u64);
assert_eq!(years, years.mul_f64(1.0));

// And again larger values incur rounding:
let years = Duration::from_secs(u64::MAX / 2);
assert_ne!(years, years.mul_f64(1.0));

# use std::time::Duration;
// In the extreme, rounding can even overflow `Duration`, which panics.
let _ = Duration::from_secs(u64::MAX).mul_f64(1.0);

fn mul_f32(self, rhs: f32) -> Duration

Multiplies Duration by f32.

Since the significand of f32 is quite limited compared to the range of Duration -- only about 16.8ms of exact nanosecond precision -- this method currently forwards to [mul_f64][Self::mul_f64] for greater accuracy.

Panics

This method will panic if result is negative, overflows Duration or not finite.

Examples

use std::time::Duration;

let dur = Duration::new(2, 700_000_000);
// Note that this `3.14_f32` argument already has more floating-point
// representation error than a direct `3.14_f64` would, so the result
// is slightly different from the ideal 8.478s.
assert_eq!(dur.mul_f32(3.14), Duration::new(8, 478_000_283));
assert_eq!(dur.mul_f32(3.14e5), Duration::new(847_800, 0));

fn div_f64(self, rhs: f64) -> Duration

Divides Duration by f64.

Panics

This method will panic if result is negative, overflows Duration or not finite.

Examples

use std::time::Duration;

let dur = Duration::new(2, 700_000_000);
assert_eq!(dur.div_f64(3.14), Duration::new(0, 859_872_611));
assert_eq!(dur.div_f64(3.14e5), Duration::new(0, 8_599));

Note that f64 does not have enough bits (f64::MANTISSA_DIGITS) to represent the full range of possible Duration with nanosecond precision, so rounding may occur even for trivial operations like dividing by 1.

# #![feature(float_exact_integer_constants)]
use std::time::Duration;

// This is about 14.9 weeks, remaining precise to the nanosecond:
let weeks = Duration::from_nanos(f64::MAX_EXACT_INTEGER as u64);
assert_eq!(weeks, weeks.div_f64(1.0));

// A larger value incurs rounding in the floating-point operation:
let weeks = Duration::from_nanos(u64::MAX);
assert_ne!(weeks, weeks.div_f64(1.0));

// This is over 285 million years, remaining precise to the second:
let years = Duration::from_secs(f64::MAX_EXACT_INTEGER as u64);
assert_eq!(years, years.div_f64(1.0));

// And again larger values incur rounding:
let years = Duration::from_secs(u64::MAX / 2);
assert_ne!(years, years.div_f64(1.0));

# use std::time::Duration;
// In the extreme, rounding can even overflow `Duration`, which panics.
let _ = Duration::from_secs(u64::MAX).div_f64(1.0);

fn div_f32(self, rhs: f32) -> Duration

Divides Duration by f32.

Since the significand of f32 is quite limited compared to the range of Duration -- only about 16.8ms of exact nanosecond precision -- this method currently forwards to [div_f64][Self::div_f64] for greater accuracy.

Panics

This method will panic if result is negative, overflows Duration or not finite.

Examples

use std::time::Duration;

let dur = Duration::new(2, 700_000_000);
// Note that this `3.14_f32` argument already has more floating-point
// representation error than a direct `3.14_f64` would, so the result
// is slightly different from the ideally rounded 0.859_872_611.
assert_eq!(dur.div_f32(3.14), Duration::new(0, 859_872_583));
assert_eq!(dur.div_f32(3.14e5), Duration::new(0, 8_599));

const fn div_duration_f64(self, rhs: Duration) -> f64

Divides Duration by Duration and returns f64.

Examples

use std::time::Duration;

let dur1 = Duration::new(2, 700_000_000);
let dur2 = Duration::new(5, 400_000_000);
assert_eq!(dur1.div_duration_f64(dur2), 0.5);

const fn div_duration_f32(self, rhs: Duration) -> f32

Divides Duration by Duration and returns f32.

Examples

use std::time::Duration;

let dur1 = Duration::new(2, 700_000_000);
let dur2 = Duration::new(5, 400_000_000);
assert_eq!(dur1.div_duration_f32(dur2), 0.5);

const fn div_duration_floor(self, rhs: Duration) -> u128

Divides Duration by Duration and returns u128, rounding the result towards zero.

Examples

#![feature(duration_integer_division)]
use std::time::Duration;

let dur = Duration::new(2, 0);
assert_eq!(dur.div_duration_floor(Duration::new(1, 000_000_001)), 1);
assert_eq!(dur.div_duration_floor(Duration::new(1, 000_000_000)), 2);
assert_eq!(dur.div_duration_floor(Duration::new(0, 999_999_999)), 2);

const fn div_duration_ceil(self, rhs: Duration) -> u128

Divides Duration by Duration and returns u128, rounding the result towards positive infinity.

Examples

#![feature(duration_integer_division)]
use std::time::Duration;

let dur = Duration::new(2, 0);
assert_eq!(dur.div_duration_ceil(Duration::new(1, 000_000_001)), 2);
assert_eq!(dur.div_duration_ceil(Duration::new(1, 000_000_000)), 2);
assert_eq!(dur.div_duration_ceil(Duration::new(0, 999_999_999)), 3);

impl Duration

fn try_from_secs_f32(secs: f32) -> Result<Duration, TryFromFloatSecsError>

The checked version of from_secs_f32.

This constructor will return an Err if secs is negative, overflows Duration or not finite.

Examples

use std::time::Duration;

let res = Duration::try_from_secs_f32(0.0);
assert_eq!(res, Ok(Duration::new(0, 0)));
let res = Duration::try_from_secs_f32(1e-20);
assert_eq!(res, Ok(Duration::new(0, 0)));
let res = Duration::try_from_secs_f32(4.2e-7);
assert_eq!(res, Ok(Duration::new(0, 420)));
let res = Duration::try_from_secs_f32(2.7);
assert_eq!(res, Ok(Duration::new(2, 700_000_048)));
let res = Duration::try_from_secs_f32(3e10);
assert_eq!(res, Ok(Duration::new(30_000_001_024, 0)));
// subnormal float:
let res = Duration::try_from_secs_f32(f32::from_bits(1));
assert_eq!(res, Ok(Duration::new(0, 0)));

let res = Duration::try_from_secs_f32(-5.0);
assert!(res.is_err());
let res = Duration::try_from_secs_f32(f32::NAN);
assert!(res.is_err());
let res = Duration::try_from_secs_f32(2e19);
assert!(res.is_err());

// the conversion uses rounding with tie resolution to even
let res = Duration::try_from_secs_f32(0.999e-9);
assert_eq!(res, Ok(Duration::new(0, 1)));

// this float represents exactly 976562.5e-9
let val = f32::from_bits(0x3A80_0000);
let res = Duration::try_from_secs_f32(val);
assert_eq!(res, Ok(Duration::new(0, 976_562)));

// this float represents exactly 2929687.5e-9
let val = f32::from_bits(0x3B40_0000);
let res = Duration::try_from_secs_f32(val);
assert_eq!(res, Ok(Duration::new(0, 2_929_688)));

// this float represents exactly 1.000_976_562_5
let val = f32::from_bits(0x3F802000);
let res = Duration::try_from_secs_f32(val);
assert_eq!(res, Ok(Duration::new(1, 976_562)));

// this float represents exactly 1.002_929_687_5
let val = f32::from_bits(0x3F806000);
let res = Duration::try_from_secs_f32(val);
assert_eq!(res, Ok(Duration::new(1, 2_929_688)));

fn try_from_secs_f64(secs: f64) -> Result<Duration, TryFromFloatSecsError>

The checked version of from_secs_f64.

This constructor will return an Err if secs is negative, overflows Duration or not finite.

Examples

use std::time::Duration;

let res = Duration::try_from_secs_f64(0.0);
assert_eq!(res, Ok(Duration::new(0, 0)));
let res = Duration::try_from_secs_f64(1e-20);
assert_eq!(res, Ok(Duration::new(0, 0)));
let res = Duration::try_from_secs_f64(4.2e-7);
assert_eq!(res, Ok(Duration::new(0, 420)));
let res = Duration::try_from_secs_f64(2.7);
assert_eq!(res, Ok(Duration::new(2, 700_000_000)));
let res = Duration::try_from_secs_f64(3e10);
assert_eq!(res, Ok(Duration::new(30_000_000_000, 0)));
// subnormal float
let res = Duration::try_from_secs_f64(f64::from_bits(1));
assert_eq!(res, Ok(Duration::new(0, 0)));

let res = Duration::try_from_secs_f64(-5.0);
assert!(res.is_err());
let res = Duration::try_from_secs_f64(f64::NAN);
assert!(res.is_err());
let res = Duration::try_from_secs_f64(2e19);
assert!(res.is_err());

// the conversion uses rounding with tie resolution to even
let res = Duration::try_from_secs_f64(0.999e-9);
assert_eq!(res, Ok(Duration::new(0, 1)));
let res = Duration::try_from_secs_f64(0.999_999_999_499);
assert_eq!(res, Ok(Duration::new(0, 999_999_999)));
let res = Duration::try_from_secs_f64(0.999_999_999_501);
assert_eq!(res, Ok(Duration::new(1, 0)));
let res = Duration::try_from_secs_f64(42.999_999_999_499);
assert_eq!(res, Ok(Duration::new(42, 999_999_999)));
let res = Duration::try_from_secs_f64(42.999_999_999_501);
assert_eq!(res, Ok(Duration::new(43, 0)));

// this float represents exactly 976562.5e-9
let val = f64::from_bits(0x3F50_0000_0000_0000);
let res = Duration::try_from_secs_f64(val);
assert_eq!(res, Ok(Duration::new(0, 976_562)));

// this float represents exactly 2929687.5e-9
let val = f64::from_bits(0x3F68_0000_0000_0000);
let res = Duration::try_from_secs_f64(val);
assert_eq!(res, Ok(Duration::new(0, 2_929_688)));

// this float represents exactly 1.000_976_562_5
let val = f64::from_bits(0x3FF0_0400_0000_0000);
let res = Duration::try_from_secs_f64(val);
assert_eq!(res, Ok(Duration::new(1, 976_562)));

// this float represents exactly 1.002_929_687_5
let val = f64::from_bits(0x3_FF00_C000_0000_000);
let res = Duration::try_from_secs_f64(val);
assert_eq!(res, Ok(Duration::new(1, 2_929_688)));

Trait Implementations

impl Add for Duration

type Output = Duration;
fn add(self, rhs: Duration) -> Duration

impl AddAssign for Duration

fn add_assign(&mut self, rhs: Duration)

impl Clone for Duration

fn clone(&self) -> Duration

impl Copy for Duration

impl Debug for Duration

fn fmt(&self, f: &mut Formatter<'_>) -> Result

impl Default for Duration

fn default() -> Duration

impl Div<u32> for Duration

type Output = Duration;
fn div(self, rhs: u32) -> Duration

impl DivAssign<u32> for Duration

fn div_assign(&mut self, rhs: u32)

impl Eq for Duration

fn assert_fields_are_eq(&self)

impl Hash for Duration

fn hash<__H: Hasher>(&self, state: &mut __H)

impl Mul<u32> for Duration

type Output = Duration;
fn mul(self, rhs: u32) -> Duration

impl MulAssign<u32> for Duration

fn mul_assign(&mut self, rhs: u32)

impl Ord for Duration

fn cmp(&self, other: &Duration) -> Ordering

impl PartialEq for Duration

fn eq(&self, other: &Duration) -> bool

impl PartialOrd for Duration

fn partial_cmp(&self, other: &Duration) -> Option<Ordering>

impl StructuralPartialEq for Duration

impl Sub for Duration

type Output = Duration;
fn sub(self, rhs: Duration) -> Duration

impl SubAssign for Duration

fn sub_assign(&mut self, rhs: Duration)

impl Sum for Duration

fn sum<I: Iterator<Item = Duration>>(iter: I) -> Duration

impl TrivialClone for Duration

impl<'a> Sum<&'a Duration> for Duration

fn sum<I: Iterator<Item = &'a Duration>>(iter: I) -> Duration

Auto Trait Implementations

impl Freeze for Duration

impl RefUnwindSafe for Duration

impl Send for Duration

impl Sync for Duration

impl Unpin for Duration

impl UnsafeUnpin for Duration

impl UnwindSafe for Duration

Blanket Implementations

impl<T> Any for Duration where T: 'static + ?Sized,

fn type_id(&self) -> TypeId

impl<T> Borrow<T> for Duration where T: ?Sized,

fn borrow(&self) -> &T

impl<T> BorrowMut<T> for Duration where T: ?Sized,

fn borrow_mut(&mut self) -> &mut T

impl<T> CloneToUninit for Duration where T: Clone,

unsafe fn clone_to_uninit(&self, dest: *mut u8)

impl<T> From<T> for Duration

fn from(t: T) -> T

Returns the argument unchanged.

impl<T> Printable for Duration where T: Copy + Debug,

impl<T> SizeHint for Duration where T: ?Sized,

fn lower_bound(&self) -> usize
fn upper_bound(&self) -> Option<usize>

impl<T> SizedTypeProperties for Duration

impl<T, U> Into<U> for Duration where U: From<T>,

fn into(self) -> U

Calls U::from(self).

That is, this conversion is whatever the implementation of [From]<T> for U chooses to do.

impl<T, U> TryFrom<U> for Duration where U: Into<T>,

type Error = Infallible;
fn try_from(value: U) -> Result<T, <T as TryFrom<U>>::Error>

impl<T, U> TryInto<U> for Duration where U: TryFrom<T>,

type Error = <U as TryFrom<T>>::Error;
fn try_into(self) -> Result<U, <U as TryFrom<T>>::Error>