Auto merge of rust-lang#137290 - matthiaskrgr:rollup-a7xdbi4, r=matthiaskrgr

Rollup of 8 pull requests

Successful merges:

 - rust-lang#120580 (Add `MAX_LEN_UTF8` and `MAX_LEN_UTF16` Constants)
 - rust-lang#132268 (Impl TryFrom<Vec<u8>> for String)
 - rust-lang#136093 (Match Ergonomics 2024: update old-edition behavior of feature gates)
 - rust-lang#136344 (Suggest replacing `.` with `::` in more error diagnostics.)
 - rust-lang#136690 (Use more explicit and reliable ptr select in sort impls)
 - rust-lang#136815 (CI: Stop /msys64/bin from being prepended to PATH in msys2 shell)
 - rust-lang#136923 (Lint `#[must_use]` attributes applied to methods in trait impls)
 - rust-lang#137155 (Organize `OsString`/`OsStr` shims)

r? `@ghost`
`@rustbot` modify labels: rollup

Author: bors

alloc/src/lib.rs

@@ -105,6 +105,7 @@
 #![feature(box_uninit_write)]
 #![feature(bstr)]
 #![feature(bstr_internals)]
+#![feature(char_max_len)]
 #![feature(clone_to_uninit)]
 #![feature(coerce_unsized)]
 #![feature(const_eval_select)]

alloc/src/string.rs

@@ -1419,7 +1419,9 @@ impl String {
     pub fn push(&mut self, ch: char) {
         match ch.len_utf8() {
             1 => self.vec.push(ch as u8),
-            _ => self.vec.extend_from_slice(ch.encode_utf8(&mut [0; 4]).as_bytes()),
+            _ => {
+                self.vec.extend_from_slice(ch.encode_utf8(&mut [0; char::MAX_LEN_UTF8]).as_bytes())
+            }
         }
     }
 
@@ -1716,7 +1718,7 @@ impl String {
     #[rustc_confusables("set")]
     pub fn insert(&mut self, idx: usize, ch: char) {
         assert!(self.is_char_boundary(idx));
-        let mut bits = [0; 4];
+        let mut bits = [0; char::MAX_LEN_UTF8];
         let bits = ch.encode_utf8(&mut bits).as_bytes();
 
         unsafe {
@@ -2797,7 +2799,7 @@ impl SpecToString for core::ascii::Char {
 impl SpecToString for char {
     #[inline]
     fn spec_to_string(&self) -> String {
-        String::from(self.encode_utf8(&mut [0; 4]))
+        String::from(self.encode_utf8(&mut [0; char::MAX_LEN_UTF8]))
     }
 }
 
@@ -3155,6 +3157,24 @@ impl From<String> for Vec<u8> {
     }
 }
 
+#[stable(feature = "try_from_vec_u8_for_string", since = "CURRENT_RUSTC_VERSION")]
+impl TryFrom<Vec<u8>> for String {
+    type Error = FromUtf8Error;
+    /// Converts the given [`Vec<u8>`] into a  [`String`] if it contains valid UTF-8 data.
+    ///
+    /// # Examples
+    ///
+    /// ```
+    /// let s1 = b"hello world".to_vec();
+    /// let v1 = String::try_from(s1).unwrap();
+    /// assert_eq!(v1, "hello world");
+    ///
+    /// ```
+    fn try_from(bytes: Vec<u8>) -> Result<Self, Self::Error> {
+        Self::from_utf8(bytes)
+    }
+}
+
 #[cfg(not(no_global_oom_handling))]
 #[stable(feature = "rust1", since = "1.0.0")]
 impl fmt::Write for String {

alloc/tests/lib.rs

@@ -3,6 +3,7 @@
 #![feature(iter_array_chunks)]
 #![feature(assert_matches)]
 #![feature(btree_extract_if)]
+#![feature(char_max_len)]
 #![feature(cow_is_borrowed)]
 #![feature(core_intrinsics)]
 #![feature(downcast_unchecked)]

alloc/tests/str.rs

@@ -2,6 +2,7 @@
 
 use std::assert_matches::assert_matches;
 use std::borrow::Cow;
+use std::char::MAX_LEN_UTF8;
 use std::cmp::Ordering::{Equal, Greater, Less};
 use std::str::{from_utf8, from_utf8_unchecked};
 
@@ -1231,7 +1232,7 @@ fn test_to_uppercase_rev_iterator() {
 #[test]
 #[cfg_attr(miri, ignore)] // Miri is too slow
 fn test_chars_decoding() {
-    let mut bytes = [0; 4];
+    let mut bytes = [0; MAX_LEN_UTF8];
     for c in (0..0x110000).filter_map(std::char::from_u32) {
         let s = c.encode_utf8(&mut bytes);
         if Some(c) != s.chars().next() {
@@ -1243,7 +1244,7 @@ fn test_chars_decoding() {
 #[test]
 #[cfg_attr(miri, ignore)] // Miri is too slow
 fn test_chars_rev_decoding() {
-    let mut bytes = [0; 4];
+    let mut bytes = [0; MAX_LEN_UTF8];
     for c in (0..0x110000).filter_map(std::char::from_u32) {
         let s = c.encode_utf8(&mut bytes);
         if Some(c) != s.chars().rev().next() {

core/src/char/methods.rs

@@ -71,6 +71,16 @@ impl char {
     #[stable(feature = "assoc_char_consts", since = "1.52.0")]
     pub const MAX: char = '\u{10FFFF}';
 
+    /// The maximum number of bytes required to [encode](char::encode_utf8) a `char` to
+    /// UTF-8 encoding.
+    #[unstable(feature = "char_max_len", issue = "121714")]
+    pub const MAX_LEN_UTF8: usize = 4;
+
+    /// The maximum number of two-byte units required to [encode](char::encode_utf16) a `char`
+    /// to UTF-16 encoding.
+    #[unstable(feature = "char_max_len", issue = "121714")]
+    pub const MAX_LEN_UTF16: usize = 2;
+
     /// `U+FFFD REPLACEMENT CHARACTER` (�) is used in Unicode to represent a
     /// decoding error.
     ///

core/src/char/mod.rs

@@ -95,6 +95,16 @@ const MAX_THREE_B: u32 = 0x10000;
 #[stable(feature = "rust1", since = "1.0.0")]
 pub const MAX: char = char::MAX;
 
+/// The maximum number of bytes required to [encode](char::encode_utf8) a `char` to
+/// UTF-8 encoding.
+#[unstable(feature = "char_max_len", issue = "121714")]
+pub const MAX_LEN_UTF8: usize = char::MAX_LEN_UTF8;
+
+/// The maximum number of two-byte units required to [encode](char::encode_utf16) a `char`
+/// to UTF-16 encoding.
+#[unstable(feature = "char_max_len", issue = "121714")]
+pub const MAX_LEN_UTF16: usize = char::MAX_LEN_UTF16;
+
 /// `U+FFFD REPLACEMENT CHARACTER` (�) is used in Unicode to represent a
 /// decoding error. Use [`char::REPLACEMENT_CHARACTER`] instead.
 #[stable(feature = "decode_utf16", since = "1.9.0")]

core/src/fmt/mod.rs

@@ -3,7 +3,7 @@
 #![stable(feature = "rust1", since = "1.0.0")]
 
 use crate::cell::{Cell, Ref, RefCell, RefMut, SyncUnsafeCell, UnsafeCell};
-use crate::char::EscapeDebugExtArgs;
+use crate::char::{EscapeDebugExtArgs, MAX_LEN_UTF8};
 use crate::marker::PhantomData;
 use crate::num::fmt as numfmt;
 use crate::ops::Deref;
@@ -187,7 +187,7 @@ pub trait Write {
     /// ```
     #[stable(feature = "fmt_write_char", since = "1.1.0")]
     fn write_char(&mut self, c: char) -> Result {
-        self.write_str(c.encode_utf8(&mut [0; 4]))
+        self.write_str(c.encode_utf8(&mut [0; MAX_LEN_UTF8]))
     }
 
     /// Glue for usage of the [`write!`] macro with implementors of this trait.
@@ -2768,7 +2768,7 @@ impl Display for char {
         if f.options.width.is_none() && f.options.precision.is_none() {
             f.write_char(*self)
         } else {
-            f.pad(self.encode_utf8(&mut [0; 4]))
+            f.pad(self.encode_utf8(&mut [0; MAX_LEN_UTF8]))
         }
     }
 }

core/src/slice/sort/shared/smallsort.rs

@@ -387,7 +387,7 @@ unsafe fn swap_if_less<T, F>(v_base: *mut T, a_pos: usize, b_pos: usize, is_less
 where
     F: FnMut(&T, &T) -> bool,
 {
-    // SAFETY: the caller must guarantee that `a` and `b` each added to `v_base` yield valid
+    // SAFETY: the caller must guarantee that `a_pos` and `b_pos` each added to `v_base` yield valid
     // pointers into `v_base`, and are properly aligned, and part of the same allocation.
     unsafe {
         let v_a = v_base.add(a_pos);
@@ -404,16 +404,16 @@ where
         // The equivalent code with a branch would be:
         //
         // if should_swap {
-        //     ptr::swap(left, right, 1);
+        //     ptr::swap(v_a, v_b, 1);
         // }
 
         // The goal is to generate cmov instructions here.
-        let left_swap = if should_swap { v_b } else { v_a };
-        let right_swap = if should_swap { v_a } else { v_b };
+        let v_a_swap = should_swap.select_unpredictable(v_b, v_a);
+        let v_b_swap = should_swap.select_unpredictable(v_a, v_b);
 
-        let right_swap_tmp = ManuallyDrop::new(ptr::read(right_swap));
-        ptr::copy(left_swap, v_a, 1);
-        ptr::copy_nonoverlapping(&*right_swap_tmp, v_b, 1);
+        let v_b_swap_tmp = ManuallyDrop::new(ptr::read(v_b_swap));
+        ptr::copy(v_a_swap, v_a, 1);
+        ptr::copy_nonoverlapping(&*v_b_swap_tmp, v_b, 1);
     }
 }
 
@@ -640,26 +640,21 @@ pub unsafe fn sort4_stable<T, F: FnMut(&T, &T) -> bool>(
         //  1,  1 |  c   b    a         d
         let c3 = is_less(&*c, &*a);
         let c4 = is_less(&*d, &*b);
-        let min = select(c3, c, a);
-        let max = select(c4, b, d);
-        let unknown_left = select(c3, a, select(c4, c, b));
-        let unknown_right = select(c4, d, select(c3, b, c));
+        let min = c3.select_unpredictable(c, a);
+        let max = c4.select_unpredictable(b, d);
+        let unknown_left = c3.select_unpredictable(a, c4.select_unpredictable(c, b));
+        let unknown_right = c4.select_unpredictable(d, c3.select_unpredictable(b, c));
 
         // Sort the last two unknown elements.
         let c5 = is_less(&*unknown_right, &*unknown_left);
-        let lo = select(c5, unknown_right, unknown_left);
-        let hi = select(c5, unknown_left, unknown_right);
+        let lo = c5.select_unpredictable(unknown_right, unknown_left);
+        let hi = c5.select_unpredictable(unknown_left, unknown_right);
 
         ptr::copy_nonoverlapping(min, dst, 1);
         ptr::copy_nonoverlapping(lo, dst.add(1), 1);
         ptr::copy_nonoverlapping(hi, dst.add(2), 1);
         ptr::copy_nonoverlapping(max, dst.add(3), 1);
     }
-
-    #[inline(always)]
-    fn select<T>(cond: bool, if_true: *const T, if_false: *const T) -> *const T {
-        if cond { if_true } else { if_false }
-    }
 }
 
 /// SAFETY: The caller MUST guarantee that `v_base` is valid for 8 reads and

core/src/str/pattern.rs

@@ -38,6 +38,7 @@
     issue = "27721"
 )]
 
+use crate::char::MAX_LEN_UTF8;
 use crate::cmp::Ordering;
 use crate::convert::TryInto as _;
 use crate::slice::memchr;
@@ -561,8 +562,8 @@ impl Pattern for char {
     type Searcher<'a> = CharSearcher<'a>;
 
     #[inline]
-    fn into_searcher(self, haystack: &str) -> Self::Searcher<'_> {
-        let mut utf8_encoded = [0; 4];
+    fn into_searcher<'a>(self, haystack: &'a str) -> Self::Searcher<'a> {
+        let mut utf8_encoded = [0; MAX_LEN_UTF8];
         let utf8_size = self
             .encode_utf8(&mut utf8_encoded)
             .len()

coretests/tests/char.rs

@@ -1,3 +1,4 @@
+use std::char::MAX_LEN_UTF8;
 use std::str::FromStr;
 use std::{char, str};
 
@@ -259,7 +260,7 @@ fn test_escape_unicode() {
 #[test]
 fn test_encode_utf8() {
     fn check(input: char, expect: &[u8]) {
-        let mut buf = [0; 4];
+        let mut buf = [0; MAX_LEN_UTF8];
         let ptr = buf.as_ptr();
         let s = input.encode_utf8(&mut buf);
         assert_eq!(s.as_ptr() as usize, ptr as usize);