Struct RingBuf 

pub struct RingBuf<T: Copy, const N: usize> { /* private fields */ }

A ring buffer of N elements stored entirely on the stack.

Once full, new pushes overwrite the oldest entry. Iteration with iter() yields elements from oldest to newest.

Safety note

Slots are stored as MaybeUninit<T> so that T: Default is not required. Exactly one invariant makes every read sound: the len entries ending at head have all been written by push. Every read goes through the private index helper, which addresses only that range, and every public accessor checks len before calling it. A change that lets len outrun the number of writes is undefined behaviour, not a logic bug.

Implementations

impl<T: Copy, const N: usize> RingBuf<T, N>

pub const fn new() -> Self

Create a new, empty ring buffer.

This is a const fn, so the buffer can be built in a const or static initialiser. Note that push and clear take &mut self, so a bare static RingBuf is read-only and of little use, and static mut is a hard error to reference under edition 2024. The pattern this actually enables is const-initialising the buffer inside an interior-mutability wrapper, which is how a single-owner buffer is reached from an interrupt context:

// with critical-section, cortex-m, or similar:
static LOG: Mutex<RefCell<RingBuf<u32, 64>>> =
    Mutex::new(RefCell::new(RingBuf::new()));

Without a const new that initialiser is impossible and you need a StaticCell or a OnceCell and a runtime init step.

Capacity 0 is a build failure

The N > 0 check is a const assertion, so a zero-capacity ring cannot be constructed at all – there is no runtime panic left to catch, and therefore no way to write the negative case as a #[test] (zero_capacity_panics was deleted for exactly this reason). This compile_fail doctest is that coverage, and pinning the error code keeps it honest: a bare compile_fail would also pass on a typo.

let _ = ph_eventing::RingBuf::<u32, 0>::new();

Panics

Does not panic.

pub fn push(&mut self, val: T)

Append a value, overwriting the oldest entry once the ring is full.

This never fails and never blocks; if losing the oldest entry is not acceptable, use crate::EventBuf, whose push reports when full.

pub fn len(&self) -> usize

Number of elements currently stored, always in 0..=N.

pub fn is_empty(&self) -> bool

Returns true if the ring holds no elements.

pub fn is_full(&self) -> bool

Returns true if the ring is at capacity, so the next push will overwrite the oldest entry.

pub const fn capacity(&self) -> usize

Number of elements the ring can hold.

pub fn clear(&mut self)

Drop every element, resetting the ring to empty.

The backing array is left as-is; only the cursors are reset, so this is O(1) and does not touch the stored values.

pub fn get(&self, i: usize) -> Option<T>

Read the i-th element (0 = oldest).

pub fn latest(&self) -> Option<T>

Most recently pushed element.

pub fn iter(&self) -> RingIter<'_, T, N>

Iterate over elements oldest→newest.

Trait Implementations

impl<T: Copy, const N: usize> Debug for RingBuf<T, N>

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

Formats the value using the given formatter.

impl<T: Copy, const N: usize> Default for RingBuf<T, N>

fn default() -> Self

Returns the “default value” for a type.

impl<'a, T: Copy, const N: usize> IntoIterator for &'a RingBuf<T, N>

type Item = T

The type of the elements being iterated over.

type IntoIter = RingIter<'a, T, N>

Which kind of iterator are we turning this into?

fn into_iter(self) -> RingIter<'a, T, N>

Creates an iterator from a value.

impl<T: Copy, const N: usize> Sink<T> for RingBuf<T, N>

type Error = Infallible

The error returned when the sink cannot accept the value.

fn try_push(&mut self, val: T) -> Result<(), Infallible>

Push a value into the sink.