Rollup merge of #54058 - Kerollmops:slice-dedup, r=shepmaster

Introduce the partition_dedup/by/by_key methods for slices

This PR propose to add three methods to the slice type, the `partition_dedup`, `partition_dedup_by` and `partition_dedup_by_key`. The two other methods are based on `slice::partition_dedup_by`.

These methods take a mutable slice, deduplicates it and moves all duplicates to the end of it, returning two mutable slices, the first containing the deduplicated elements and the second all the duplicates unordered.

```rust
let mut slice = [1, 2, 2, 3, 3, 2];

let (dedup, duplicates) = slice.partition_dedup();

assert_eq!(dedup, [1, 2, 3, 2]);
assert_eq!(duplicates, [3, 2]);
```

The benefits of adding these methods is that it is now possible to:
  - deduplicate a slice without having to allocate and possibly clone elements on the heap, really useful for embedded stuff that can't allocate for example.
  - not loose duplicate elements, because, when using `Vec::dedup`, duplicates elements are dropped. These methods add more flexibillity to the user.

Note that this is near a copy/paste of the `Vec::dedup_by` function, once this method is stable the goal is to replace the algorithm in `Vec` by the following.

```rust
pub fn Vec::dedup_by<F>(&mut self, same_bucket: F)
    where F: FnMut(&mut T, &mut T) -> bool
{
    let (dedup, _) = self.as_mut_slice().partition_dedup_by(same_bucket);
    let len = dedup.len();
    self.truncate(len);
}
```

Author: pietroalbini

src/liballoc/lib.rs

@@ -119,6 +119,7 @@
 #![feature(exact_chunks)]
 #![feature(rustc_const_unstable)]
 #![feature(const_vec_new)]
+#![feature(slice_partition_dedup)]
 #![feature(maybe_uninit)]
 
 // Allow testing this library

src/liballoc/vec.rs

@@ -947,10 +947,9 @@ impl<T> Vec<T> {
     /// Removes all but the first of consecutive elements in the vector satisfying a given equality
     /// relation.
     ///
-    /// The `same_bucket` function is passed references to two elements from the vector, and
-    /// returns `true` if the elements compare equal, or `false` if they do not. The elements are
-    /// passed in opposite order from their order in the vector, so if `same_bucket(a, b)` returns
-    /// `true`, `a` is removed.
+    /// The `same_bucket` function is passed references to two elements from the vector and
+    /// must determine if the elements compare equal. The elements are passed in opposite order
+    /// from their order in the slice, so if `same_bucket(a, b)` returns `true`, `a` is removed.
     ///
     /// If the vector is sorted, this removes all duplicates.
     ///
@@ -964,90 +963,12 @@ impl<T> Vec<T> {
     /// assert_eq!(vec, ["foo", "bar", "baz", "bar"]);
     /// ```
     #[stable(feature = "dedup_by", since = "1.16.0")]
-    pub fn dedup_by<F>(&mut self, mut same_bucket: F) where F: FnMut(&mut T, &mut T) -> bool {
-        unsafe {
-            // Although we have a mutable reference to `self`, we cannot make
-            // *arbitrary* changes. The `same_bucket` calls could panic, so we
-            // must ensure that the vector is in a valid state at all time.
-            //
-            // The way that we handle this is by using swaps; we iterate
-            // over all the elements, swapping as we go so that at the end
-            // the elements we wish to keep are in the front, and those we
-            // wish to reject are at the back. We can then truncate the
-            // vector. This operation is still O(n).
-            //
-            // Example: We start in this state, where `r` represents "next
-            // read" and `w` represents "next_write`.
-            //
-            //           r
-            //     +---+---+---+---+---+---+
-            //     | 0 | 1 | 1 | 2 | 3 | 3 |
-            //     +---+---+---+---+---+---+
-            //           w
-            //
-            // Comparing self[r] against self[w-1], this is not a duplicate, so
-            // we swap self[r] and self[w] (no effect as r==w) and then increment both
-            // r and w, leaving us with:
-            //
-            //               r
-            //     +---+---+---+---+---+---+
-            //     | 0 | 1 | 1 | 2 | 3 | 3 |
-            //     +---+---+---+---+---+---+
-            //               w
-            //
-            // Comparing self[r] against self[w-1], this value is a duplicate,
-            // so we increment `r` but leave everything else unchanged:
-            //
-            //                   r
-            //     +---+---+---+---+---+---+
-            //     | 0 | 1 | 1 | 2 | 3 | 3 |
-            //     +---+---+---+---+---+---+
-            //               w
-            //
-            // Comparing self[r] against self[w-1], this is not a duplicate,
-            // so swap self[r] and self[w] and advance r and w:
-            //
-            //                       r
-            //     +---+---+---+---+---+---+
-            //     | 0 | 1 | 2 | 1 | 3 | 3 |
-            //     +---+---+---+---+---+---+
-            //                   w
-            //
-            // Not a duplicate, repeat:
-            //
-            //                           r
-            //     +---+---+---+---+---+---+
-            //     | 0 | 1 | 2 | 3 | 1 | 3 |
-            //     +---+---+---+---+---+---+
-            //                       w
-            //
-            // Duplicate, advance r. End of vec. Truncate to w.
-
-            let ln = self.len();
-            if ln <= 1 {
-                return;
-            }
-
-            // Avoid bounds checks by using raw pointers.
-            let p = self.as_mut_ptr();
-            let mut r: usize = 1;
-            let mut w: usize = 1;
-
-            while r < ln {
-                let p_r = p.add(r);
-                let p_wm1 = p.add(w - 1);
-                if !same_bucket(&mut *p_r, &mut *p_wm1) {
-                    if r != w {
-                        let p_w = p_wm1.offset(1);
-                        mem::swap(&mut *p_r, &mut *p_w);
-                    }
-                    w += 1;
-                }
-                r += 1;
-            }
-
-            self.truncate(w);
-        }
+    pub fn dedup_by<F>(&mut self, same_bucket: F) where F: FnMut(&mut T, &mut T) -> bool {
+        let len = {
+            let (dedup, _) = self.as_mut_slice().partition_dedup_by(same_bucket);
+            dedup.len()
+        };
+        self.truncate(len);
     }
 
     /// Appends an element to the back of a collection.
@@ -1533,7 +1454,8 @@ impl<'a> Drop for SetLenOnDrop<'a> {
 }
 
 impl<T: PartialEq> Vec<T> {
-    /// Removes consecutive repeated elements in the vector.
+    /// Removes consecutive repeated elements in the vector according to the
+    /// [`PartialEq`] trait implementation.
     ///
     /// If the vector is sorted, this removes all duplicates.
     ///

src/libcore/slice/mod.rs

@@ -1402,6 +1402,178 @@ impl<T> [T] {
         sort::quicksort(self, |a, b| f(a).lt(&f(b)));
     }
 
+    /// Moves all consecutive repeated elements to the end of the slice according to the
+    /// [`PartialEq`] trait implementation.
+    ///
+    /// Returns two slices. The first contains no consecutive repeated elements.
+    /// The second contains all the duplicates in no specified order.
+    ///
+    /// If the slice is sorted, the first returned slice contains no duplicates.
+    ///
+    /// # Examples
+    ///
+    /// ```
+    /// #![feature(slice_partition_dedup)]
+    ///
+    /// let mut slice = [1, 2, 2, 3, 3, 2, 1, 1];
+    ///
+    /// let (dedup, duplicates) = slice.partition_dedup();
+    ///
+    /// assert_eq!(dedup, [1, 2, 3, 2, 1]);
+    /// assert_eq!(duplicates, [2, 3, 1]);
+    /// ```
+    #[unstable(feature = "slice_partition_dedup", issue = "54279")]
+    #[inline]
+    pub fn partition_dedup(&mut self) -> (&mut [T], &mut [T])
+        where T: PartialEq
+    {
+        self.partition_dedup_by(|a, b| a == b)
+    }
+
+    /// Moves all but the first of consecutive elements to the end of the slice satisfying
+    /// a given equality relation.
+    ///
+    /// Returns two slices. The first contains no consecutive repeated elements.
+    /// The second contains all the duplicates in no specified order.
+    ///
+    /// The `same_bucket` function is passed references to two elements from the slice and
+    /// must determine if the elements compare equal. The elements are passed in opposite order
+    /// from their order in the slice, so if `same_bucket(a, b)` returns `true`, `a` is moved
+    /// at the end of the slice.
+    ///
+    /// If the slice is sorted, the first returned slice contains no duplicates.
+    ///
+    /// # Examples
+    ///
+    /// ```
+    /// #![feature(slice_partition_dedup)]
+    ///
+    /// let mut slice = ["foo", "Foo", "BAZ", "Bar", "bar", "baz", "BAZ"];
+    ///
+    /// let (dedup, duplicates) = slice.partition_dedup_by(|a, b| a.eq_ignore_ascii_case(b));
+    ///
+    /// assert_eq!(dedup, ["foo", "BAZ", "Bar", "baz"]);
+    /// assert_eq!(duplicates, ["bar", "Foo", "BAZ"]);
+    /// ```
+    #[unstable(feature = "slice_partition_dedup", issue = "54279")]
+    #[inline]
+    pub fn partition_dedup_by<F>(&mut self, mut same_bucket: F) -> (&mut [T], &mut [T])
+        where F: FnMut(&mut T, &mut T) -> bool
+    {
+        // Although we have a mutable reference to `self`, we cannot make
+        // *arbitrary* changes. The `same_bucket` calls could panic, so we
+        // must ensure that the slice is in a valid state at all times.
+        //
+        // The way that we handle this is by using swaps; we iterate
+        // over all the elements, swapping as we go so that at the end
+        // the elements we wish to keep are in the front, and those we
+        // wish to reject are at the back. We can then split the slice.
+        // This operation is still O(n).
+        //
+        // Example: We start in this state, where `r` represents "next
+        // read" and `w` represents "next_write`.
+        //
+        //           r
+        //     +---+---+---+---+---+---+
+        //     | 0 | 1 | 1 | 2 | 3 | 3 |
+        //     +---+---+---+---+---+---+
+        //           w
+        //
+        // Comparing self[r] against self[w-1], this is not a duplicate, so
+        // we swap self[r] and self[w] (no effect as r==w) and then increment both
+        // r and w, leaving us with:
+        //
+        //               r
+        //     +---+---+---+---+---+---+
+        //     | 0 | 1 | 1 | 2 | 3 | 3 |
+        //     +---+---+---+---+---+---+
+        //               w
+        //
+        // Comparing self[r] against self[w-1], this value is a duplicate,
+        // so we increment `r` but leave everything else unchanged:
+        //
+        //                   r
+        //     +---+---+---+---+---+---+
+        //     | 0 | 1 | 1 | 2 | 3 | 3 |
+        //     +---+---+---+---+---+---+
+        //               w
+        //
+        // Comparing self[r] against self[w-1], this is not a duplicate,
+        // so swap self[r] and self[w] and advance r and w:
+        //
+        //                       r
+        //     +---+---+---+---+---+---+
+        //     | 0 | 1 | 2 | 1 | 3 | 3 |
+        //     +---+---+---+---+---+---+
+        //                   w
+        //
+        // Not a duplicate, repeat:
+        //
+        //                           r
+        //     +---+---+---+---+---+---+
+        //     | 0 | 1 | 2 | 3 | 1 | 3 |
+        //     +---+---+---+---+---+---+
+        //                       w
+        //
+        // Duplicate, advance r. End of slice. Split at w.
+
+        let len = self.len();
+        if len <= 1 {
+            return (self, &mut [])
+        }
+
+        let ptr = self.as_mut_ptr();
+        let mut next_read: usize = 1;
+        let mut next_write: usize = 1;
+
+        unsafe {
+            // Avoid bounds checks by using raw pointers.
+            while next_read < len {
+                let ptr_read = ptr.add(next_read);
+                let prev_ptr_write = ptr.add(next_write - 1);
+                if !same_bucket(&mut *ptr_read, &mut *prev_ptr_write) {
+                    if next_read != next_write {
+                        let ptr_write = prev_ptr_write.offset(1);
+                        mem::swap(&mut *ptr_read, &mut *ptr_write);
+                    }
+                    next_write += 1;
+                }
+                next_read += 1;
+            }
+        }
+
+        self.split_at_mut(next_write)
+    }
+
+    /// Moves all but the first of consecutive elements to the end of the slice that resolve
+    /// to the same key.
+    ///
+    /// Returns two slices. The first contains no consecutive repeated elements.
+    /// The second contains all the duplicates in no specified order.
+    ///
+    /// If the slice is sorted, the first returned slice contains no duplicates.
+    ///
+    /// # Examples
+    ///
+    /// ```
+    /// #![feature(slice_partition_dedup)]
+    ///
+    /// let mut slice = [10, 20, 21, 30, 30, 20, 11, 13];
+    ///
+    /// let (dedup, duplicates) = slice.partition_dedup_by_key(|i| *i / 10);
+    ///
+    /// assert_eq!(dedup, [10, 20, 30, 20, 11]);
+    /// assert_eq!(duplicates, [21, 30, 13]);
+    /// ```
+    #[unstable(feature = "slice_partition_dedup", issue = "54279")]
+    #[inline]
+    pub fn partition_dedup_by_key<K, F>(&mut self, mut key: F) -> (&mut [T], &mut [T])
+        where F: FnMut(&mut T) -> K,
+              K: PartialEq,
+    {
+        self.partition_dedup_by(|a, b| key(a) == key(b))
+    }
+
     /// Rotates the slice in-place such that the first `mid` elements of the
     /// slice move to the end while the last `self.len() - mid` elements move to
     /// the front. After calling `rotate_left`, the element previously at index

src/libcore/tests/lib.rs

@@ -39,6 +39,7 @@
 #![feature(inner_deref)]
 #![feature(slice_internals)]
 #![feature(option_replace)]
+#![feature(slice_partition_dedup)]
 #![feature(copy_within)]
 
 extern crate core;

src/libcore/tests/slice.rs

@@ -1001,6 +1001,65 @@ fn test_align_to_empty_mid() {
     }
 }
 
+#[test]
+fn test_slice_partition_dedup_by() {
+    let mut slice: [i32; 9] = [1, -1, 2, 3, 1, -5, 5, -2, 2];
+
+    let (dedup, duplicates) = slice.partition_dedup_by(|a, b| a.abs() == b.abs());
+
+    assert_eq!(dedup, [1, 2, 3, 1, -5, -2]);
+    assert_eq!(duplicates, [5, -1, 2]);
+}
+
+#[test]
+fn test_slice_partition_dedup_empty() {
+    let mut slice: [i32; 0] = [];
+
+    let (dedup, duplicates) = slice.partition_dedup();
+
+    assert_eq!(dedup, []);
+    assert_eq!(duplicates, []);
+}
+
+#[test]
+fn test_slice_partition_dedup_one() {
+    let mut slice = [12];
+
+    let (dedup, duplicates) = slice.partition_dedup();
+
+    assert_eq!(dedup, [12]);
+    assert_eq!(duplicates, []);
+}
+
+#[test]
+fn test_slice_partition_dedup_multiple_ident() {
+    let mut slice = [12, 12, 12, 12, 12, 11, 11, 11, 11, 11, 11];
+
+    let (dedup, duplicates) = slice.partition_dedup();
+
+    assert_eq!(dedup, [12, 11]);
+    assert_eq!(duplicates, [12, 12, 12, 12, 11, 11, 11, 11, 11]);
+}
+
+#[test]
+fn test_slice_partition_dedup_partialeq() {
+    #[derive(Debug)]
+    struct Foo(i32, i32);
+
+    impl PartialEq for Foo {
+        fn eq(&self, other: &Foo) -> bool {
+            self.0 == other.0
+        }
+    }
+
+    let mut slice = [Foo(0, 1), Foo(0, 5), Foo(1, 7), Foo(1, 9)];
+
+    let (dedup, duplicates) = slice.partition_dedup();
+
+    assert_eq!(dedup, [Foo(0, 1), Foo(1, 7)]);
+    assert_eq!(duplicates, [Foo(0, 5), Foo(1, 9)]);
+}
+
 #[test]
 fn test_copy_within() {
     // Start to end, with a RangeTo.