Trait PartialRangeExt 

pub trait PartialRangeExt<Idx = usize>: RangeStart<Idx>
where
    Idx: Clone + Ord,
{
    // Provided methods
    fn end_bound(&self) -> Option<RangeEnd<Idx>> { ... }
    fn is_empty(&self) -> bool { ... }
    fn clamp_exclusive(
        &self,
        start: impl Into<Idx>,
        exclusive_end: impl Into<Idx>,
    ) -> Range<Idx>
       where Idx: Add<Output = Idx> + One { ... }
    fn clamp_inclusive(
        &self,
        start: impl Into<Idx>,
        inclusive_end: impl Into<Idx>,
    ) -> Range<Idx> { ... }
    fn clamp_right_inclusive(&self, inclusive_end: impl Into<Idx>) -> Range<Idx> { ... }
    fn clamp_right_exclusive(&self, exclusive_end: impl Into<Idx>) -> Range<Idx>
       where Idx: Add<Output = Idx> + One { ... }
    fn clamp_left(&self, start: impl Into<Idx>) -> PartialRange<Idx>  { ... }
    fn intersection<R>(&self, other: &R) -> PartialRange<Idx> 
       where R: PartialRangeExt<Idx> { ... }
}

A range that may be unbounded above or empty.

Implementors provide start via the RangeStart supertrait and override end_bound when the range is bounded; the default returns None, describing a range unbounded above. An implementor whose emptiness is not derivable from those two accessors, such as a type with hidden exhaustion state like std::ops::RangeInclusive, must also override is_empty; every derived operation consults is_empty before the bounds. The standard library leaves the endpoint values of a drained std::ops::RangeInclusive unspecified, so the position of an empty result derived from one is likewise unspecified; only emptiness is guaranteed.

The operations interpret Idx as a discrete domain in which x + one is the successor of x, and convert a bound between its inclusive and exclusive forms only where the converted bound is provably at most the window’s end. Empty results from the PartialRange-returning operations anchor at the raised or joint start; the std::ops::Range- and Range-returning operations instead lower the start to the computed end, so their results are always slice-safe.

Examples

use ps_range::{PartialRange, PartialRangeExt, Range};

assert_eq!((5usize..=10).clamp_exclusive(0usize, 7usize), 5..7);
assert_eq!((5usize..).clamp_inclusive(0usize, 7usize), Range::inclusive(5, 7));
assert!((5usize..=10).clamp_exclusive(20usize, 25usize).is_empty());

assert_eq!(
    (5usize..=10).intersection(&(8usize..)),
    PartialRange::Inclusive { start: 8, end: 10 }
);

Provided Methods

fn end_bound(&self) -> Option<RangeEnd<Idx>>

Returns the upper bound, or None if the range is unbounded above.

fn is_empty(&self) -> bool

Returns true if the range contains no indices.

fn clamp_exclusive( &self, start: impl Into<Idx>, exclusive_end: impl Into<Idx>, ) -> Range<Idx>

where Idx: Add<Output = Idx> + One,

Restricts the range to the window start..exclusive_end.

The result is slice-safe: its start does not exceed its end, and its end does not exceed the window’s. An empty input yields an empty range at its start clamped into the window; a disjoint input yields one at the computed end, which may fall below the window’s start.

fn clamp_inclusive( &self, start: impl Into<Idx>, inclusive_end: impl Into<Idx>, ) -> Range<Idx>

Restricts the range to the window start..=inclusive_end.

The result is always bounded, so it is expressed as a Range. An empty input yields an empty exclusive Range at its start clamped into the window; a disjoint input yields one at the computed end, which may fall below the window’s start.

fn clamp_right_inclusive(&self, inclusive_end: impl Into<Idx>) -> Range<Idx>

Caps the upper bound at inclusive_end.

fn clamp_right_exclusive(&self, exclusive_end: impl Into<Idx>) -> Range<Idx>

where Idx: Add<Output = Idx> + One,

Caps the upper bound at exclusive_end.

fn clamp_left(&self, start: impl Into<Idx>) -> PartialRange<Idx>

Raises the lower bound to at least start, preserving boundedness above.

If the range is empty, or start is disjoint from it, the result is the Empty variant anchored at the raised start.

fn intersection<R>(&self, other: &R) -> PartialRange<Idx>

where R: PartialRangeExt<Idx>,

Returns the overlap between this range and other.

The result is unbounded above only when both ranges are. If either range is empty, or the ranges are disjoint, the result is the Empty variant anchored at the joint start.

Dyn Compatibility

This trait is not dyn compatible.

In older versions of Rust, dyn compatibility was called "object safety".

Implementations on Foreign Types

impl<Idx, T> PartialRangeExt<Idx> for &T

where Idx: Clone + Ord, T: PartialRangeExt<Idx>,

fn end_bound(&self) -> Option<RangeEnd<Idx>>

fn is_empty(&self) -> bool

impl<Idx, T> PartialRangeExt<Idx> for &mut T

where Idx: Clone + Ord, T: PartialRangeExt<Idx>,

fn end_bound(&self) -> Option<RangeEnd<Idx>>

fn is_empty(&self) -> bool

impl<Idx, T> PartialRangeExt<Idx> for Option<T>

where Idx: Clone + Ord + Zero, T: PartialRangeExt<Idx>,

None is the empty range, anchored at Idx::zero(), so a possibly-absent range remains usable as a range.

fn end_bound(&self) -> Option<RangeEnd<Idx>>

fn is_empty(&self) -> bool

impl<Idx: Clone + Ord + Zero> PartialRangeExt<Idx> for RangeFull

impl<Idx: Clone + Ord + Zero> PartialRangeExt<Idx> for RangeTo<Idx>

fn end_bound(&self) -> Option<RangeEnd<Idx>>

impl<Idx: Clone + Ord + Zero> PartialRangeExt<Idx> for RangeToInclusive<Idx>

fn end_bound(&self) -> Option<RangeEnd<Idx>>

impl<Idx: Clone + Ord> PartialRangeExt<Idx> for Range<Idx>

fn end_bound(&self) -> Option<RangeEnd<Idx>>

impl<Idx: Clone + Ord> PartialRangeExt<Idx> for RangeFrom<Idx>

impl<Idx: Clone + Ord> PartialRangeExt<Idx> for RangeInclusive<Idx>

fn end_bound(&self) -> Option<RangeEnd<Idx>>

fn is_empty(&self) -> bool

Implementors

impl<Idx: Clone + Ord + Zero> PartialRangeExt<Idx> for PartialRange<Idx>

impl<Idx: Clone + Ord> PartialRangeExt<Idx> for ps_range::Range<Idx>