introduce benchmarks of BTreeSet.intersection
16 tests combining 4 kinds of contents with different sizes exposing edge cases.
The ones with asymmetric sizes are addressed by https://github.com/rust-lang/rust/pull/58577.
The pos_vs_neg cases seems (are were meant to be) the same as the neg_vs_pos case (same thing, reverse order) but reality shows a surprsing 25% difference.
override `VecDeque::try_rfold`, also update iterator
This keeps the slice based iteration and updates the iterator state after each slice. It also uses a loop to reduce the amount of code.
This uses unsafe code, so some thorough review would be appreciated. Cc @RalfJung
This keeps the slice based iteration and updates the iterator
state after each slice. It also uses a loop to reduce the amount
of code.
This uses unsafe code, so some thorough review would be
appreciated.
Non-naive implementation of `VecDeque.append`
Replaces the old, simple implementation with a more manual (and **unsafe** 😱) one. I've added 1 more test and verified that it covers all 6 code paths in the function.
This new implementation was about 60% faster than the old naive one when I tried benchmarking it.
Add slice::sort_by_cached_key as a memoised sort_by_key
At present, `slice::sort_by_key` calls its key function twice for each comparison that is made. When the key function is expensive (which can often be the case when `sort_by_key` is chosen over `sort_by`), this can lead to very suboptimal behaviour.
To address this, I've introduced a new slice method, `sort_by_cached_key`, which has identical semantic behaviour to `sort_by_key`, except that it guarantees the key function will only be called once per element.
Where there are `n` elements and the key function is `O(m)`:
- `slice::sort_by_cached_key` time complexity is `O(m n log m n)`, compared to `slice::sort_by_key`'s `O(m n + n log n)`.
- `slice::sort_by_cached_key` space complexity remains at `O(n + m)`. (Technically, it now reserves a slice of size `n`, whereas before it reserved a slice of size `n/2`.)
`slice::sort_unstable_by_key` has not been given an analogue, as it is important that unstable sorts are in-place, which is not a property that is guaranteed here. However, this also means that `slice::sort_unstable_by_key` is likely to be slower than `slice::sort_by_cached_key` when the key function does not have negligible complexity. We might want to explore this trade-off further in the future.
Benchmarks (for a vector of 100 `i32`s):
```
# Lexicographic: `|x| x.to_string()`
test bench_sort_by_key ... bench: 112,638 ns/iter (+/- 19,563)
test bench_sort_by_cached_key ... bench: 15,038 ns/iter (+/- 4,814)
# Identity: `|x| *x`
test bench_sort_by_key ... bench: 1,346 ns/iter (+/- 238)
test bench_sort_by_cached_key ... bench: 1,839 ns/iter (+/- 765)
# Power: `|x| x.pow(31)`
test bench_sort_by_key ... bench: 3,624 ns/iter (+/- 738)
test bench_sort_by_cached_key ... bench: 1,997 ns/iter (+/- 311)
# Abs: `|x| x.abs()`
test bench_sort_by_key ... bench: 1,546 ns/iter (+/- 174)
test bench_sort_by_cached_key ... bench: 1,668 ns/iter (+/- 790)
```
(So it seems functions that are single operations do perform slightly worse with this method, but for pretty much any more complex key, you're better off with this optimisation.)
I've definitely found myself using expensive keys in the past and wishing this optimisation was made (e.g. for https://github.com/rust-lang/rust/pull/47415). This feels like both desirable and expected behaviour, at the small cost of slightly more stack allocation and minute degradation in performance for extremely trivial keys.
Resolves#34447.
Background
==========
Slices currently have an unstable [`rotate`] method which rotates
elements in the slice to the _left_ N positions. [Here][tracking] is the
tracking issue for this unstable feature.
```rust
let mut a = ['a', 'b' ,'c', 'd', 'e', 'f'];
a.rotate(2);
assert_eq!(a, ['c', 'd', 'e', 'f', 'a', 'b']);
```
Proposal
========
Deprecate the [`rotate`] method and introduce `rotate_left` and
`rotate_right` methods.
```rust
let mut a = ['a', 'b' ,'c', 'd', 'e', 'f'];
a.rotate_left(2);
assert_eq!(a, ['c', 'd', 'e', 'f', 'a', 'b']);
```
```rust
let mut a = ['a', 'b' ,'c', 'd', 'e', 'f'];
a.rotate_right(2);
assert_eq!(a, ['e', 'f', 'a', 'b', 'c', 'd']);
```
Justification
=============
I used this method today for my first time and (probably because I’m a
naive westerner who reads LTR) was surprised when the docs mentioned that
elements get rotated in a left-ward direction. I was in a situation
where I needed to shift elements in a right-ward direction and had to
context switch from the main problem I was working on and think how much
to rotate left in order to accomplish the right-ward rotation I needed.
Ruby’s `Array.rotate` shifts left-ward, Python’s `deque.rotate` shifts
right-ward. Both of their implementations allow passing negative numbers
to shift in the opposite direction respectively.
Introducing `rotate_left` and `rotate_right` would:
- remove ambiguity about direction (alleviating need to read docs 😉)
- make it easier for people who need to rotate right
[`rotate`]: https://doc.rust-lang.org/std/primitive.slice.html#method.rotate
[tracking]: https://github.com/rust-lang/rust/issues/41891
Many AsciiExt imports have become useless thanks to the inherent ascii
methods added in the last commits. These were removed. In some places, I
fully specified the ascii method being called to enforce usage of the
AsciiExt trait. Note that some imports are not removed but tagged with
a `#[cfg(stage0)]` attribute. This is necessary, because certain ascii
methods are not yet available in stage0. All those imports will be
removed later.
Additionally, failing tests were fixed. The test suite should exit
successfully now.
Stein Somers