user0/rust
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Auto merge of #145916 - matthiaskrgr:rollup-cnvhq2z, r=matthiaskrgr

Browse source 
Rollup of 6 pull requests

Successful merges:

 - rust-lang/rust#142215 (Use -Zmir-opt-level=0 in tests for MIR building)
 - rust-lang/rust#143341 (Mention that casting to *const () is a way to roundtrip with from_raw_parts)
 - rust-lang/rust#145078 (Fix wrong cache line size of riscv64)
 - rust-lang/rust#145290 (Improve std::fs::read_dir docs)
 - rust-lang/rust#145335 (Move WTF-8 code from std into core and alloc)
 - rust-lang/rust#145904 (Move `riscv64-gc-unknown-linux-musl` from Tier 2 with Host tools to Tier 2)

r? `@ghost`
`@rustbot` modify labels: rollup
This commit is contained in:
bors 2025-08-27 06:15:42 +00:00
commit 4f808ba6bf
46 changed files with 1406 additions and 1261 deletions
Download patch file Download diff file Expand all files Collapse all files

3
library/alloc/src/lib.rs
View file

@ -159,6 +159,7 @@
#![feature(unicode_internals)]
#![feature(unsize)]
#![feature(unwrap_infallible)]
#![feature(wtf8_internals)]
// tidy-alphabetical-end
//
// Language features:
@ -232,6 +233,8 @@ pub mod sync;
#[cfg(all(not(no_global_oom_handling), not(no_rc), not(no_sync)))]
pub mod task;
pub mod vec;
#[cfg(all(not(no_rc), not(no_sync), not(no_global_oom_handling)))]
pub mod wtf8;
#[doc(hidden)]
#[unstable(feature = "liballoc_internals", issue = "none", reason = "implementation detail")]

562
library/alloc/src/wtf8/mod.rs Normal file
View file

@ -0,0 +1,562 @@
//! Heap-allocated counterpart to core `wtf8` module.
#![unstable(
feature = "wtf8_internals",
issue = "none",
reason = "this is internal code for representing OsStr on some platforms and not a public API"
)]
// rustdoc bug: doc(hidden) on the module won't stop types in the module from showing up in trait
// implementations, so, we'll have to add more doc(hidden)s anyway
#![doc(hidden)]
// Note: This module is also included in the alloctests crate using #[path] to
// run the tests. See the comment there for an explanation why this is the case.
#[cfg(test)]
mod tests;
use core::char::{MAX_LEN_UTF8, encode_utf8_raw};
use core::hash::{Hash, Hasher};
pub use core::wtf8::{CodePoint, Wtf8};
#[cfg(not(test))]
pub use core::wtf8::{EncodeWide, Wtf8CodePoints};
use core::{fmt, mem, ops, str};
use crate::borrow::{Cow, ToOwned};
use crate::boxed::Box;
use crate::collections::TryReserveError;
#[cfg(not(test))]
use crate::rc::Rc;
use crate::string::String;
#[cfg(all(not(test), target_has_atomic = "ptr"))]
use crate::sync::Arc;
use crate::vec::Vec;
/// An owned, growable string of well-formed WTF-8 data.
///
/// Similar to `String`, but can additionally contain surrogate code points
/// if they’re not in a surrogate pair.
#[derive(Eq, PartialEq, Ord, PartialOrd, Clone)]
#[doc(hidden)]
pub struct Wtf8Buf {
bytes: Vec<u8>,
/// Do we know that `bytes` holds a valid UTF-8 encoding? We can easily
/// know this if we're constructed from a `String` or `&str`.
///
/// It is possible for `bytes` to have valid UTF-8 without this being
/// set, such as when we're concatenating `&Wtf8`'s and surrogates become
/// paired, as we don't bother to rescan the entire string.
is_known_utf8: bool,
}
impl ops::Deref for Wtf8Buf {
type Target = Wtf8;
fn deref(&self) -> &Wtf8 {
self.as_slice()
}
}
impl ops::DerefMut for Wtf8Buf {
fn deref_mut(&mut self) -> &mut Wtf8 {
self.as_mut_slice()
}
}
/// Formats the string in double quotes, with characters escaped according to
/// [`char::escape_debug`] and unpaired surrogates represented as `\u{xxxx}`,
/// where each `x` is a hexadecimal digit.
///
/// For example, the code units [U+0061, U+D800, U+000A] are formatted as
/// `"a\u{D800}\n"`.
impl fmt::Debug for Wtf8Buf {
#[inline]
fn fmt(&self, formatter: &mut fmt::Formatter<'_>) -> fmt::Result {
fmt::Debug::fmt(&**self, formatter)
}
}
/// Formats the string with unpaired surrogates substituted with the replacement
/// character, U+FFFD.
impl fmt::Display for Wtf8Buf {
fn fmt(&self, formatter: &mut fmt::Formatter<'_>) -> fmt::Result {
if let Some(s) = self.as_known_utf8() {
fmt::Display::fmt(s, formatter)
} else {
fmt::Display::fmt(&**self, formatter)
}
}
}
#[cfg_attr(test, allow(dead_code))]
impl Wtf8Buf {
/// Creates a new, empty WTF-8 string.
#[inline]
pub fn new() -> Wtf8Buf {
Wtf8Buf { bytes: Vec::new(), is_known_utf8: true }
}
/// Creates a new, empty WTF-8 string with pre-allocated capacity for `capacity` bytes.
#[inline]
pub fn with_capacity(capacity: usize) -> Wtf8Buf {
Wtf8Buf { bytes: Vec::with_capacity(capacity), is_known_utf8: true }
}
/// Creates a WTF-8 string from a WTF-8 byte vec.
///
/// Since the byte vec is not checked for valid WTF-8, this function is
/// marked unsafe.
#[inline]
pub unsafe fn from_bytes_unchecked(value: Vec<u8>) -> Wtf8Buf {
Wtf8Buf { bytes: value, is_known_utf8: false }
}
/// Creates a WTF-8 string from a UTF-8 `String`.
///
/// This takes ownership of the `String` and does not copy.
///
/// Since WTF-8 is a superset of UTF-8, this always succeeds.
#[inline]
pub const fn from_string(string: String) -> Wtf8Buf {
Wtf8Buf { bytes: string.into_bytes(), is_known_utf8: true }
}
/// Creates a WTF-8 string from a UTF-8 `&str` slice.
///
/// This copies the content of the slice.
///
/// Since WTF-8 is a superset of UTF-8, this always succeeds.
#[inline]
pub fn from_str(s: &str) -> Wtf8Buf {
Wtf8Buf { bytes: s.as_bytes().to_vec(), is_known_utf8: true }
}
pub fn clear(&mut self) {
self.bytes.clear();
self.is_known_utf8 = true;
}
/// Creates a WTF-8 string from a potentially ill-formed UTF-16 slice of 16-bit code units.
///
/// This is lossless: calling `.encode_wide()` on the resulting string
/// will always return the original code units.
pub fn from_wide(v: &[u16]) -> Wtf8Buf {
let mut string = Wtf8Buf::with_capacity(v.len());
for item in char::decode_utf16(v.iter().cloned()) {
match item {
Ok(ch) => string.push_char(ch),
Err(surrogate) => {
let surrogate = surrogate.unpaired_surrogate();
// Surrogates are known to be in the code point range.
let code_point = unsafe { CodePoint::from_u32_unchecked(surrogate as u32) };
// The string will now contain an unpaired surrogate.
string.is_known_utf8 = false;
// Skip the WTF-8 concatenation check,
// surrogate pairs are already decoded by decode_utf16
unsafe {
string.push_code_point_unchecked(code_point);
}
}
}
}
string
}
/// Appends the given `char` to the end of this string.
/// This does **not** include the WTF-8 concatenation check or `is_known_utf8` check.
/// Copied from String::push.
unsafe fn push_code_point_unchecked(&mut self, code_point: CodePoint) {
let mut bytes = [0; MAX_LEN_UTF8];
let bytes = encode_utf8_raw(code_point.to_u32(), &mut bytes);
self.bytes.extend_from_slice(bytes)
}
#[inline]
pub fn as_slice(&self) -> &Wtf8 {
unsafe { Wtf8::from_bytes_unchecked(&self.bytes) }
}
#[inline]
pub fn as_mut_slice(&mut self) -> &mut Wtf8 {
// Safety: `Wtf8` doesn't expose any way to mutate the bytes that would
// cause them to change from well-formed UTF-8 to ill-formed UTF-8,
// which would break the assumptions of the `is_known_utf8` field.
unsafe { Wtf8::from_mut_bytes_unchecked(&mut self.bytes) }
}
/// Converts the string to UTF-8 without validation, if it was created from
/// valid UTF-8.
#[inline]
fn as_known_utf8(&self) -> Option<&str> {
if self.is_known_utf8 {
// SAFETY: The buffer is known to be valid UTF-8.
Some(unsafe { str::from_utf8_unchecked(self.as_bytes()) })
} else {
None
}
}
/// Reserves capacity for at least `additional` more bytes to be inserted
/// in the given `Wtf8Buf`.
/// The collection may reserve more space to avoid frequent reallocations.
///
/// # Panics
///
/// Panics if the new capacity exceeds `isize::MAX` bytes.
#[inline]
pub fn reserve(&mut self, additional: usize) {
self.bytes.reserve(additional)
}
/// Tries to reserve capacity for at least `additional` more bytes to be
/// inserted in the given `Wtf8Buf`. The `Wtf8Buf` may reserve more space to
/// avoid frequent reallocations. After calling `try_reserve`, capacity will
/// be greater than or equal to `self.len() + additional`. Does nothing if
/// capacity is already sufficient. This method preserves the contents even
/// if an error occurs.
///
/// # Errors
///
/// If the capacity overflows, or the allocator reports a failure, then an error
/// is returned.
#[inline]
pub fn try_reserve(&mut self, additional: usize) -> Result<(), TryReserveError> {
self.bytes.try_reserve(additional)
}
#[inline]
pub fn reserve_exact(&mut self, additional: usize) {
self.bytes.reserve_exact(additional)
}
/// Tries to reserve the minimum capacity for exactly `additional` more
/// bytes to be inserted in the given `Wtf8Buf`. After calling
/// `try_reserve_exact`, capacity will be greater than or equal to
/// `self.len() + additional` if it returns `Ok(())`.
/// Does nothing if the capacity is already sufficient.
///
/// Note that the allocator may give the `Wtf8Buf` more space than it
/// requests. Therefore, capacity can not be relied upon to be precisely
/// minimal. Prefer [`try_reserve`] if future insertions are expected.
///
/// [`try_reserve`]: Wtf8Buf::try_reserve
///
/// # Errors
///
/// If the capacity overflows, or the allocator reports a failure, then an error
/// is returned.
#[inline]
pub fn try_reserve_exact(&mut self, additional: usize) -> Result<(), TryReserveError> {
self.bytes.try_reserve_exact(additional)
}
#[inline]
pub fn shrink_to_fit(&mut self) {
self.bytes.shrink_to_fit()
}
#[inline]
pub fn shrink_to(&mut self, min_capacity: usize) {
self.bytes.shrink_to(min_capacity)
}
#[inline]
pub fn leak<'a>(self) -> &'a mut Wtf8 {
unsafe { Wtf8::from_mut_bytes_unchecked(self.bytes.leak()) }
}
/// Returns the number of bytes that this string buffer can hold without reallocating.
#[inline]
pub fn capacity(&self) -> usize {
self.bytes.capacity()
}
/// Append a UTF-8 slice at the end of the string.
#[inline]
pub fn push_str(&mut self, other: &str) {
self.bytes.extend_from_slice(other.as_bytes())
}
/// Append a WTF-8 slice at the end of the string.
///
/// This replaces newly paired surrogates at the boundary
/// with a supplementary code point,
/// like concatenating ill-formed UTF-16 strings effectively would.
#[inline]
pub fn push_wtf8(&mut self, other: &Wtf8) {
match ((&*self).final_lead_surrogate(), other.initial_trail_surrogate()) {
// Replace newly paired surrogates by a supplementary code point.
(Some(lead), Some(trail)) => {
let len_without_lead_surrogate = self.len() - 3;
self.bytes.truncate(len_without_lead_surrogate);
let other_without_trail_surrogate = &other.as_bytes()[3..];
// 4 bytes for the supplementary code point
self.bytes.reserve(4 + other_without_trail_surrogate.len());
self.push_char(decode_surrogate_pair(lead, trail));
self.bytes.extend_from_slice(other_without_trail_surrogate);
}
_ => {
// If we'll be pushing a string containing a surrogate, we may
// no longer have UTF-8.
if self.is_known_utf8 && other.next_surrogate(0).is_some() {
self.is_known_utf8 = false;
}
self.bytes.extend_from_slice(other.as_bytes());
}
}
}
/// Append a Unicode scalar value at the end of the string.
#[inline]
pub fn push_char(&mut self, c: char) {
// SAFETY: It's always safe to push a char.
unsafe { self.push_code_point_unchecked(CodePoint::from_char(c)) }
}
/// Append a code point at the end of the string.
///
/// This replaces newly paired surrogates at the boundary
/// with a supplementary code point,
/// like concatenating ill-formed UTF-16 strings effectively would.
#[inline]
pub fn push(&mut self, code_point: CodePoint) {
if let Some(trail) = code_point.to_trail_surrogate() {
if let Some(lead) = (&*self).final_lead_surrogate() {
let len_without_lead_surrogate = self.len() - 3;
self.bytes.truncate(len_without_lead_surrogate);
self.push_char(decode_surrogate_pair(lead, trail));
return;
}
// We're pushing a trailing surrogate.
self.is_known_utf8 = false;
} else if code_point.to_lead_surrogate().is_some() {
// We're pushing a leading surrogate.
self.is_known_utf8 = false;
}
// No newly paired surrogates at the boundary.
unsafe { self.push_code_point_unchecked(code_point) }
}
/// Shortens a string to the specified length.
///
/// # Panics
///
/// Panics if `new_len` > current length,
/// or if `new_len` is not a code point boundary.
#[inline]
pub fn truncate(&mut self, new_len: usize) {
assert!(self.is_code_point_boundary(new_len));
self.bytes.truncate(new_len)
}
/// Consumes the WTF-8 string and tries to convert it to a vec of bytes.
#[inline]
pub fn into_bytes(self) -> Vec<u8> {
self.bytes
}
/// Consumes the WTF-8 string and tries to convert it to UTF-8.
///
/// This does not copy the data.
///
/// If the contents are not well-formed UTF-8
/// (that is, if the string contains surrogates),
/// the original WTF-8 string is returned instead.
pub fn into_string(self) -> Result<String, Wtf8Buf> {
if self.is_known_utf8 || self.next_surrogate(0).is_none() {
Ok(unsafe { String::from_utf8_unchecked(self.bytes) })
} else {
Err(self)
}
}
/// Consumes the WTF-8 string and converts it lossily to UTF-8.
///
/// This does not copy the data (but may overwrite parts of it in place).
///
/// Surrogates are replaced with `"\u{FFFD}"` (the replacement character “<>”)
pub fn into_string_lossy(mut self) -> String {
if !self.is_known_utf8 {
let mut pos = 0;
while let Some((surrogate_pos, _)) = self.next_surrogate(pos) {
pos = surrogate_pos + 3;
// Surrogates and the replacement character are all 3 bytes, so
// they can substituted in-place.
self.bytes[surrogate_pos..pos].copy_from_slice("\u{FFFD}".as_bytes());
}
}
unsafe { String::from_utf8_unchecked(self.bytes) }
}
/// Converts this `Wtf8Buf` into a boxed `Wtf8`.
#[inline]
pub fn into_box(self) -> Box<Wtf8> {
// SAFETY: relies on `Wtf8` being `repr(transparent)`.
unsafe { mem::transmute(self.bytes.into_boxed_slice()) }
}
/// Converts a `Box<Wtf8>` into a `Wtf8Buf`.
pub fn from_box(boxed: Box<Wtf8>) -> Wtf8Buf {
let bytes: Box<[u8]> = unsafe { mem::transmute(boxed) };
Wtf8Buf { bytes: bytes.into_vec(), is_known_utf8: false }
}
/// Provides plumbing to core `Vec::extend_from_slice`.
/// More well behaving alternative to allowing outer types
/// full mutable access to the core `Vec`.
#[inline]
pub unsafe fn extend_from_slice_unchecked(&mut self, other: &[u8]) {
self.bytes.extend_from_slice(other);
self.is_known_utf8 = false;
}
}
/// Creates a new WTF-8 string from an iterator of code points.
///
/// This replaces surrogate code point pairs with supplementary code points,
/// like concatenating ill-formed UTF-16 strings effectively would.
impl FromIterator<CodePoint> for Wtf8Buf {
fn from_iter<T: IntoIterator<Item = CodePoint>>(iter: T) -> Wtf8Buf {
let mut string = Wtf8Buf::new();
string.extend(iter);
string
}
}
/// Append code points from an iterator to the string.
///
/// This replaces surrogate code point pairs with supplementary code points,
/// like concatenating ill-formed UTF-16 strings effectively would.
impl Extend<CodePoint> for Wtf8Buf {
fn extend<T: IntoIterator<Item = CodePoint>>(&mut self, iter: T) {
let iterator = iter.into_iter();
let (low, _high) = iterator.size_hint();
// Lower bound of one byte per code point (ASCII only)
self.bytes.reserve(low);
iterator.for_each(move |code_point| self.push(code_point));
}
#[inline]
fn extend_one(&mut self, code_point: CodePoint) {
self.push(code_point);
}
#[inline]
fn extend_reserve(&mut self, additional: usize) {
// Lower bound of one byte per code point (ASCII only)
self.bytes.reserve(additional);
}
}
/// Creates an owned `Wtf8Buf` from a borrowed `Wtf8`.
pub(super) fn to_owned(slice: &Wtf8) -> Wtf8Buf {
Wtf8Buf { bytes: slice.as_bytes().to_vec(), is_known_utf8: false }
}
/// Lossily converts the string to UTF-8.
/// Returns a UTF-8 `&str` slice if the contents are well-formed in UTF-8.
///
/// Surrogates are replaced with `"\u{FFFD}"` (the replacement character “<>”).
///
/// This only copies the data if necessary (if it contains any surrogate).
pub(super) fn to_string_lossy(slice: &Wtf8) -> Cow<'_, str> {
let Some((surrogate_pos, _)) = slice.next_surrogate(0) else {
return Cow::Borrowed(unsafe { str::from_utf8_unchecked(slice.as_bytes()) });
};
let wtf8_bytes = slice.as_bytes();
let mut utf8_bytes = Vec::with_capacity(slice.len());
utf8_bytes.extend_from_slice(&wtf8_bytes[..surrogate_pos]);
utf8_bytes.extend_from_slice("\u{FFFD}".as_bytes());
let mut pos = surrogate_pos + 3;
loop {
match slice.next_surrogate(pos) {
Some((surrogate_pos, _)) => {
utf8_bytes.extend_from_slice(&wtf8_bytes[pos..surrogate_pos]);
utf8_bytes.extend_from_slice("\u{FFFD}".as_bytes());
pos = surrogate_pos + 3;
}
None => {
utf8_bytes.extend_from_slice(&wtf8_bytes[pos..]);
return Cow::Owned(unsafe { String::from_utf8_unchecked(utf8_bytes) });
}
}
}
}
#[inline]
pub(super) fn clone_into(slice: &Wtf8, buf: &mut Wtf8Buf) {
buf.is_known_utf8 = false;
slice.as_bytes().clone_into(&mut buf.bytes);
}
#[cfg(not(test))]
impl Wtf8 {
#[rustc_allow_incoherent_impl]
pub fn to_owned(&self) -> Wtf8Buf {
to_owned(self)
}
#[rustc_allow_incoherent_impl]
pub fn clone_into(&self, buf: &mut Wtf8Buf) {
clone_into(self, buf)
}
#[rustc_allow_incoherent_impl]
pub fn to_string_lossy(&self) -> Cow<'_, str> {
to_string_lossy(self)
}
#[rustc_allow_incoherent_impl]
pub fn into_box(&self) -> Box<Wtf8> {
let boxed: Box<[u8]> = self.as_bytes().into();
unsafe { mem::transmute(boxed) }
}
#[rustc_allow_incoherent_impl]
pub fn empty_box() -> Box<Wtf8> {
let boxed: Box<[u8]> = Default::default();
unsafe { mem::transmute(boxed) }
}
#[cfg(target_has_atomic = "ptr")]
#[rustc_allow_incoherent_impl]
pub fn into_arc(&self) -> Arc<Wtf8> {
let arc: Arc<[u8]> = Arc::from(self.as_bytes());
unsafe { Arc::from_raw(Arc::into_raw(arc) as *const Wtf8) }
}
#[rustc_allow_incoherent_impl]
pub fn into_rc(&self) -> Rc<Wtf8> {
let rc: Rc<[u8]> = Rc::from(self.as_bytes());
unsafe { Rc::from_raw(Rc::into_raw(rc) as *const Wtf8) }
}
#[inline]
#[rustc_allow_incoherent_impl]
pub fn to_ascii_lowercase(&self) -> Wtf8Buf {
Wtf8Buf { bytes: self.as_bytes().to_ascii_lowercase(), is_known_utf8: false }
}
#[inline]
#[rustc_allow_incoherent_impl]
pub fn to_ascii_uppercase(&self) -> Wtf8Buf {
Wtf8Buf { bytes: self.as_bytes().to_ascii_uppercase(), is_known_utf8: false }
}
}
#[inline]
fn decode_surrogate_pair(lead: u16, trail: u16) -> char {
let code_point = 0x10000 + ((((lead - 0xD800) as u32) << 10) | (trail - 0xDC00) as u32);
unsafe { char::from_u32_unchecked(code_point) }
}
impl Hash for Wtf8Buf {
#[inline]
fn hash<H: Hasher>(&self, state: &mut H) {
state.write(&self.bytes);
0xfeu8.hash(state)
}
}

209
library/std/src/sys_common/wtf8/tests.rs → library/alloc/src/wtf8/tests.rs
View file

@ -1,3 +1,5 @@
use realalloc::string::ToString;
use super::*;
#[test]
@ -82,82 +84,85 @@ fn code_point_to_char_lossy() {
#[test]
fn wtf8buf_new() {
assert_eq!(Wtf8Buf::new().bytes, b"");
assert_eq!(Wtf8Buf::new().as_bytes(), b"");
}
#[test]
fn wtf8buf_from_str() {
assert_eq!(Wtf8Buf::from_str("").bytes, b"");
assert_eq!(Wtf8Buf::from_str("aé 💩").bytes, b"a\xC3\xA9 \xF0\x9F\x92\xA9");
assert_eq!(Wtf8Buf::from_str("").as_bytes(), b"");
assert_eq!(Wtf8Buf::from_str("aé 💩").as_bytes(), b"a\xC3\xA9 \xF0\x9F\x92\xA9");
}
#[test]
fn wtf8buf_from_string() {
assert_eq!(Wtf8Buf::from_string(String::from("")).bytes, b"");
assert_eq!(Wtf8Buf::from_string(String::from("aé 💩")).bytes, b"a\xC3\xA9 \xF0\x9F\x92\xA9");
assert_eq!(Wtf8Buf::from_string(String::from("")).as_bytes(), b"");
assert_eq!(
Wtf8Buf::from_string(String::from("aé 💩")).as_bytes(),
b"a\xC3\xA9 \xF0\x9F\x92\xA9"
);
}
#[test]
fn wtf8buf_from_wide() {
let buf = Wtf8Buf::from_wide(&[]);
assert_eq!(buf.bytes, b"");
assert_eq!(buf.as_bytes(), b"");
assert!(buf.is_known_utf8);
let buf = Wtf8Buf::from_wide(&[0x61, 0xE9, 0x20, 0xD83D, 0xDCA9]);
assert_eq!(buf.bytes, b"a\xC3\xA9 \xF0\x9F\x92\xA9");
assert_eq!(buf.as_bytes(), b"a\xC3\xA9 \xF0\x9F\x92\xA9");
assert!(buf.is_known_utf8);
let buf = Wtf8Buf::from_wide(&[0x61, 0xE9, 0x20, 0xD83D, 0xD83D, 0xDCA9]);
assert_eq!(buf.bytes, b"a\xC3\xA9 \xED\xA0\xBD\xF0\x9F\x92\xA9");
assert_eq!(buf.as_bytes(), b"a\xC3\xA9 \xED\xA0\xBD\xF0\x9F\x92\xA9");
assert!(!buf.is_known_utf8);
let buf = Wtf8Buf::from_wide(&[0xD800]);
assert_eq!(buf.bytes, b"\xED\xA0\x80");
assert_eq!(buf.as_bytes(), b"\xED\xA0\x80");
assert!(!buf.is_known_utf8);
let buf = Wtf8Buf::from_wide(&[0xDBFF]);
assert_eq!(buf.bytes, b"\xED\xAF\xBF");
assert_eq!(buf.as_bytes(), b"\xED\xAF\xBF");
assert!(!buf.is_known_utf8);
let buf = Wtf8Buf::from_wide(&[0xDC00]);
assert_eq!(buf.bytes, b"\xED\xB0\x80");
assert_eq!(buf.as_bytes(), b"\xED\xB0\x80");
assert!(!buf.is_known_utf8);
let buf = Wtf8Buf::from_wide(&[0xDFFF]);
assert_eq!(buf.bytes, b"\xED\xBF\xBF");
assert_eq!(buf.as_bytes(), b"\xED\xBF\xBF");
assert!(!buf.is_known_utf8);
}
#[test]
fn wtf8buf_push_str() {
let mut string = Wtf8Buf::new();
assert_eq!(string.bytes, b"");
assert_eq!(string.as_bytes(), b"");
assert!(string.is_known_utf8);
string.push_str("aé 💩");
assert_eq!(string.bytes, b"a\xC3\xA9 \xF0\x9F\x92\xA9");
assert_eq!(string.as_bytes(), b"a\xC3\xA9 \xF0\x9F\x92\xA9");
assert!(string.is_known_utf8);
}
#[test]
fn wtf8buf_push_char() {
let mut string = Wtf8Buf::from_str("aé ");
assert_eq!(string.bytes, b"a\xC3\xA9 ");
assert_eq!(string.as_bytes(), b"a\xC3\xA9 ");
assert!(string.is_known_utf8);
string.push_char('💩');
assert_eq!(string.bytes, b"a\xC3\xA9 \xF0\x9F\x92\xA9");
assert_eq!(string.as_bytes(), b"a\xC3\xA9 \xF0\x9F\x92\xA9");
assert!(string.is_known_utf8);
}
#[test]
fn wtf8buf_push() {
let mut string = Wtf8Buf::from_str("aé ");
assert_eq!(string.bytes, b"a\xC3\xA9 ");
assert_eq!(string.as_bytes(), b"a\xC3\xA9 ");
assert!(string.is_known_utf8);
string.push(CodePoint::from_char('💩'));
assert_eq!(string.bytes, b"a\xC3\xA9 \xF0\x9F\x92\xA9");
assert_eq!(string.as_bytes(), b"a\xC3\xA9 \xF0\x9F\x92\xA9");
assert!(string.is_known_utf8);
fn c(value: u32) -> CodePoint {
@ -168,53 +173,53 @@ fn wtf8buf_push() {
string.push(c(0xD83D)); // lead
assert!(!string.is_known_utf8);
string.push(c(0xDCA9)); // trail
assert_eq!(string.bytes, b"\xF0\x9F\x92\xA9"); // Magic!
assert_eq!(string.as_bytes(), b"\xF0\x9F\x92\xA9"); // Magic!
let mut string = Wtf8Buf::new();
string.push(c(0xD83D)); // lead
assert!(!string.is_known_utf8);
string.push(c(0x20)); // not surrogate
string.push(c(0xDCA9)); // trail
assert_eq!(string.bytes, b"\xED\xA0\xBD \xED\xB2\xA9");
assert_eq!(string.as_bytes(), b"\xED\xA0\xBD \xED\xB2\xA9");
let mut string = Wtf8Buf::new();
string.push(c(0xD800)); // lead
assert!(!string.is_known_utf8);
string.push(c(0xDBFF)); // lead
assert_eq!(string.bytes, b"\xED\xA0\x80\xED\xAF\xBF");
assert_eq!(string.as_bytes(), b"\xED\xA0\x80\xED\xAF\xBF");
let mut string = Wtf8Buf::new();
string.push(c(0xD800)); // lead
assert!(!string.is_known_utf8);
string.push(c(0xE000)); // not surrogate
assert_eq!(string.bytes, b"\xED\xA0\x80\xEE\x80\x80");
assert_eq!(string.as_bytes(), b"\xED\xA0\x80\xEE\x80\x80");
let mut string = Wtf8Buf::new();
string.push(c(0xD7FF)); // not surrogate
assert!(string.is_known_utf8);
string.push(c(0xDC00)); // trail
assert!(!string.is_known_utf8);
assert_eq!(string.bytes, b"\xED\x9F\xBF\xED\xB0\x80");
assert_eq!(string.as_bytes(), b"\xED\x9F\xBF\xED\xB0\x80");
let mut string = Wtf8Buf::new();
string.push(c(0x61)); // not surrogate, < 3 bytes
assert!(string.is_known_utf8);
string.push(c(0xDC00)); // trail
assert!(!string.is_known_utf8);
assert_eq!(string.bytes, b"\x61\xED\xB0\x80");
assert_eq!(string.as_bytes(), b"\x61\xED\xB0\x80");
let mut string = Wtf8Buf::new();
string.push(c(0xDC00)); // trail
assert!(!string.is_known_utf8);
assert_eq!(string.bytes, b"\xED\xB0\x80");
assert_eq!(string.as_bytes(), b"\xED\xB0\x80");
}
#[test]
fn wtf8buf_push_wtf8() {
let mut string = Wtf8Buf::from_str("aé");
assert_eq!(string.bytes, b"a\xC3\xA9");
assert_eq!(string.as_bytes(), b"a\xC3\xA9");
string.push_wtf8(Wtf8::from_str(" 💩"));
assert_eq!(string.bytes, b"a\xC3\xA9 \xF0\x9F\x92\xA9");
assert_eq!(string.as_bytes(), b"a\xC3\xA9 \xF0\x9F\x92\xA9");
assert!(string.is_known_utf8);
fn w(v: &[u8]) -> &Wtf8 {
@ -224,42 +229,42 @@ fn wtf8buf_push_wtf8() {
let mut string = Wtf8Buf::new();
string.push_wtf8(w(b"\xED\xA0\xBD")); // lead
string.push_wtf8(w(b"\xED\xB2\xA9")); // trail
assert_eq!(string.bytes, b"\xF0\x9F\x92\xA9"); // Magic!
assert_eq!(string.as_bytes(), b"\xF0\x9F\x92\xA9"); // Magic!
let mut string = Wtf8Buf::new();
string.push_wtf8(w(b"\xED\xA0\xBD")); // lead
string.push_wtf8(w(b" ")); // not surrogate
string.push_wtf8(w(b"\xED\xB2\xA9")); // trail
assert_eq!(string.bytes, b"\xED\xA0\xBD \xED\xB2\xA9");
assert_eq!(string.as_bytes(), b"\xED\xA0\xBD \xED\xB2\xA9");
assert!(!string.is_known_utf8);
let mut string = Wtf8Buf::new();
string.push_wtf8(w(b"\xED\xA0\x80")); // lead
string.push_wtf8(w(b"\xED\xAF\xBF")); // lead
assert_eq!(string.bytes, b"\xED\xA0\x80\xED\xAF\xBF");
assert_eq!(string.as_bytes(), b"\xED\xA0\x80\xED\xAF\xBF");
assert!(!string.is_known_utf8);
let mut string = Wtf8Buf::new();
string.push_wtf8(w(b"\xED\xA0\x80")); // lead
string.push_wtf8(w(b"\xEE\x80\x80")); // not surrogate
assert_eq!(string.bytes, b"\xED\xA0\x80\xEE\x80\x80");
assert_eq!(string.as_bytes(), b"\xED\xA0\x80\xEE\x80\x80");
assert!(!string.is_known_utf8);
let mut string = Wtf8Buf::new();
string.push_wtf8(w(b"\xED\x9F\xBF")); // not surrogate
string.push_wtf8(w(b"\xED\xB0\x80")); // trail
assert_eq!(string.bytes, b"\xED\x9F\xBF\xED\xB0\x80");
assert_eq!(string.as_bytes(), b"\xED\x9F\xBF\xED\xB0\x80");
assert!(!string.is_known_utf8);
let mut string = Wtf8Buf::new();
string.push_wtf8(w(b"a")); // not surrogate, < 3 bytes
string.push_wtf8(w(b"\xED\xB0\x80")); // trail
assert_eq!(string.bytes, b"\x61\xED\xB0\x80");
assert_eq!(string.as_bytes(), b"\x61\xED\xB0\x80");
assert!(!string.is_known_utf8);
let mut string = Wtf8Buf::new();
string.push_wtf8(w(b"\xED\xB0\x80")); // trail
assert_eq!(string.bytes, b"\xED\xB0\x80");
assert_eq!(string.as_bytes(), b"\xED\xB0\x80");
assert!(!string.is_known_utf8);
}
@ -269,15 +274,15 @@ fn wtf8buf_truncate() {
assert!(string.is_known_utf8);
string.truncate(3);
assert_eq!(string.bytes, b"a\xC3\xA9");
assert_eq!(string.as_bytes(), b"a\xC3\xA9");
assert!(string.is_known_utf8);
string.truncate(1);
assert_eq!(string.bytes, b"a");
assert_eq!(string.as_bytes(), b"a");
assert!(string.is_known_utf8);
string.truncate(0);
assert_eq!(string.bytes, b"");
assert_eq!(string.as_bytes(), b"");
assert!(string.is_known_utf8);
}
@ -287,11 +292,11 @@ fn wtf8buf_truncate_around_non_bmp() {
assert!(string.is_known_utf8);
string.truncate(4);
assert_eq!(string.bytes, b"\xF0\x9F\x92\xA9");
assert_eq!(string.as_bytes(), b"\xF0\x9F\x92\xA9");
assert!(string.is_known_utf8);
string.truncate(0);
assert_eq!(string.bytes, b"");
assert_eq!(string.as_bytes(), b"");
assert!(string.is_known_utf8);
}
@ -361,7 +366,7 @@ fn wtf8buf_from_iterator() {
Wtf8Buf { bytes: b"a\xC3\xA9 \xF0\x9F\x92\xA9".to_vec(), is_known_utf8: true }
);
assert_eq!(f(&[0xD83D, 0xDCA9]).bytes, b"\xF0\x9F\x92\xA9"); // Magic!
assert_eq!(f(&[0xD83D, 0xDCA9]).as_bytes(), b"\xF0\x9F\x92\xA9"); // Magic!
assert_eq!(
f(&[0xD83D, 0x20, 0xDCA9]),
Wtf8Buf { bytes: b"\xED\xA0\xBD \xED\xB2\xA9".to_vec(), is_known_utf8: false }
@ -401,7 +406,7 @@ fn wtf8buf_extend() {
Wtf8Buf { bytes: b"a\xC3\xA9 \xF0\x9F\x92\xA9".to_vec(), is_known_utf8: true }
);
assert_eq!(e(&[0xD83D], &[0xDCA9]).bytes, b"\xF0\x9F\x92\xA9"); // Magic!
assert_eq!(e(&[0xD83D], &[0xDCA9]).as_bytes(), b"\xF0\x9F\x92\xA9"); // Magic!
assert_eq!(
e(&[0xD83D, 0x20], &[0xDCA9]),
Wtf8Buf { bytes: b"\xED\xA0\xBD \xED\xB2\xA9".to_vec(), is_known_utf8: false }
@ -449,8 +454,8 @@ fn wtf8buf_show_str() {
#[test]
fn wtf8_from_str() {
assert_eq!(&Wtf8::from_str("").bytes, b"");
assert_eq!(&Wtf8::from_str("aé 💩").bytes, b"a\xC3\xA9 \xF0\x9F\x92\xA9");
assert_eq!(&Wtf8::from_str("").as_bytes(), b"");
assert_eq!(&Wtf8::from_str("aé 💩").as_bytes(), b"a\xC3\xA9 \xF0\x9F\x92\xA9");
}
#[test]
@ -461,7 +466,7 @@ fn wtf8_len() {
#[test]
fn wtf8_slice() {
assert_eq!(&Wtf8::from_str("aé 💩")[1..4].bytes, b"\xC3\xA9 ");
assert_eq!(&Wtf8::from_str("aé 💩")[1..4].as_bytes(), b"\xC3\xA9 ");
}
#[test]
@ -472,7 +477,7 @@ fn wtf8_slice_not_code_point_boundary() {
#[test]
fn wtf8_slice_from() {
assert_eq!(&Wtf8::from_str("aé 💩")[1..].bytes, b"\xC3\xA9 \xF0\x9F\x92\xA9");
assert_eq!(&Wtf8::from_str("aé 💩")[1..].as_bytes(), b"\xC3\xA9 \xF0\x9F\x92\xA9");
}
#[test]
@ -483,7 +488,7 @@ fn wtf8_slice_from_not_code_point_boundary() {
#[test]
fn wtf8_slice_to() {
assert_eq!(&Wtf8::from_str("aé 💩")[..4].bytes, b"a\xC3\xA9 ");
assert_eq!(&Wtf8::from_str("aé 💩")[..4].as_bytes(), b"a\xC3\xA9 ");
}
#[test]
@ -529,12 +534,12 @@ fn wtf8_as_str() {
#[test]
fn wtf8_to_string_lossy() {
assert_eq!(Wtf8::from_str("").to_string_lossy(), Cow::Borrowed(""));
assert_eq!(Wtf8::from_str("aé 💩").to_string_lossy(), Cow::Borrowed("aé 💩"));
assert_eq!(to_string_lossy(Wtf8::from_str("")), Cow::Borrowed(""));
assert_eq!(to_string_lossy(Wtf8::from_str("aé 💩")), Cow::Borrowed("aé 💩"));
let mut string = Wtf8Buf::from_str("aé 💩");
string.push(CodePoint::from_u32(0xD800).unwrap());
let expected: Cow<'_, str> = Cow::Owned(String::from("aé 💩<>"));
assert_eq!(string.to_string_lossy(), expected);
assert_eq!(to_string_lossy(&string), expected);
}
#[test]
@ -548,7 +553,7 @@ fn wtf8_display() {
let mut string = Wtf8Buf::from_str("aé 💩");
string.push(CodePoint::from_u32(0xD800).unwrap());
assert_eq!("aé 💩<>", d(string.as_inner()));
assert_eq!("aé 💩<>", d(string.as_ref()));
}
#[test]
@ -577,67 +582,41 @@ fn wtf8_encode_wide_size_hint() {
#[test]
fn wtf8_clone_into() {
let mut string = Wtf8Buf::new();
Wtf8::from_str("green").clone_into(&mut string);
assert_eq!(string.bytes, b"green");
clone_into(Wtf8::from_str("green"), &mut string);
assert_eq!(string.as_bytes(), b"green");
let mut string = Wtf8Buf::from_str("green");
Wtf8::from_str("").clone_into(&mut string);
assert_eq!(string.bytes, b"");
clone_into(Wtf8::from_str(""), &mut string);
assert_eq!(string.as_bytes(), b"");
let mut string = Wtf8Buf::from_str("red");
Wtf8::from_str("green").clone_into(&mut string);
assert_eq!(string.bytes, b"green");
clone_into(Wtf8::from_str("green"), &mut string);
assert_eq!(string.as_bytes(), b"green");
let mut string = Wtf8Buf::from_str("green");
Wtf8::from_str("red").clone_into(&mut string);
assert_eq!(string.bytes, b"red");
clone_into(Wtf8::from_str("red"), &mut string);
assert_eq!(string.as_bytes(), b"red");
let mut string = Wtf8Buf::from_str("green");
assert!(string.is_known_utf8);
unsafe { Wtf8::from_bytes_unchecked(b"\xED\xA0\x80").clone_into(&mut string) };
assert_eq!(string.bytes, b"\xED\xA0\x80");
clone_into(unsafe { Wtf8::from_bytes_unchecked(b"\xED\xA0\x80") }, &mut string);
assert_eq!(string.as_bytes(), b"\xED\xA0\x80");
assert!(!string.is_known_utf8);
}
#[test]
fn wtf8_to_ascii_lowercase() {
let lowercase = Wtf8::from_str("").to_ascii_lowercase();
assert_eq!(lowercase.bytes, b"");
let lowercase = Wtf8::from_str("GrEeN gRaPeS! 🍇").to_ascii_lowercase();
assert_eq!(lowercase.bytes, b"green grapes! \xf0\x9f\x8d\x87");
let lowercase = unsafe { Wtf8::from_bytes_unchecked(b"\xED\xA0\x80").to_ascii_lowercase() };
assert_eq!(lowercase.bytes, b"\xED\xA0\x80");
assert!(!lowercase.is_known_utf8);
}
#[test]
fn wtf8_to_ascii_uppercase() {
let uppercase = Wtf8::from_str("").to_ascii_uppercase();
assert_eq!(uppercase.bytes, b"");
let uppercase = Wtf8::from_str("GrEeN gRaPeS! 🍇").to_ascii_uppercase();
assert_eq!(uppercase.bytes, b"GREEN GRAPES! \xf0\x9f\x8d\x87");
let uppercase = unsafe { Wtf8::from_bytes_unchecked(b"\xED\xA0\x80").to_ascii_uppercase() };
assert_eq!(uppercase.bytes, b"\xED\xA0\x80");
assert!(!uppercase.is_known_utf8);
}
#[test]
fn wtf8_make_ascii_lowercase() {
let mut lowercase = Wtf8Buf::from_str("");
lowercase.make_ascii_lowercase();
assert_eq!(lowercase.bytes, b"");
assert_eq!(lowercase.as_bytes(), b"");
let mut lowercase = Wtf8Buf::from_str("GrEeN gRaPeS! 🍇");
lowercase.make_ascii_lowercase();
assert_eq!(lowercase.bytes, b"green grapes! \xf0\x9f\x8d\x87");
assert_eq!(lowercase.as_bytes(), b"green grapes! \xf0\x9f\x8d\x87");
let mut lowercase = unsafe { Wtf8::from_bytes_unchecked(b"\xED\xA0\x80").to_owned() };
let mut lowercase = to_owned(unsafe { Wtf8::from_bytes_unchecked(b"\xED\xA0\x80") });
lowercase.make_ascii_lowercase();
assert_eq!(lowercase.bytes, b"\xED\xA0\x80");
assert_eq!(lowercase.as_bytes(), b"\xED\xA0\x80");
assert!(!lowercase.is_known_utf8);
}
@ -645,22 +624,22 @@ fn wtf8_make_ascii_lowercase() {
fn wtf8_make_ascii_uppercase() {
let mut uppercase = Wtf8Buf::from_str("");
uppercase.make_ascii_uppercase();
assert_eq!(uppercase.bytes, b"");
assert_eq!(uppercase.as_bytes(), b"");
let mut uppercase = Wtf8Buf::from_str("GrEeN gRaPeS! 🍇");
uppercase.make_ascii_uppercase();
assert_eq!(uppercase.bytes, b"GREEN GRAPES! \xf0\x9f\x8d\x87");
assert_eq!(uppercase.as_bytes(), b"GREEN GRAPES! \xf0\x9f\x8d\x87");
let mut uppercase = unsafe { Wtf8::from_bytes_unchecked(b"\xED\xA0\x80").to_owned() };
let mut uppercase = to_owned(unsafe { Wtf8::from_bytes_unchecked(b"\xED\xA0\x80") });
uppercase.make_ascii_uppercase();
assert_eq!(uppercase.bytes, b"\xED\xA0\x80");
assert_eq!(uppercase.as_bytes(), b"\xED\xA0\x80");
assert!(!uppercase.is_known_utf8);
}
#[test]
fn wtf8_to_owned() {
let string = unsafe { Wtf8::from_bytes_unchecked(b"\xED\xA0\x80").to_owned() };
assert_eq!(string.bytes, b"\xED\xA0\x80");
let string = to_owned(unsafe { Wtf8::from_bytes_unchecked(b"\xED\xA0\x80") });
assert_eq!(string.as_bytes(), b"\xED\xA0\x80");
assert!(!string.is_known_utf8);
}
@ -669,44 +648,44 @@ fn wtf8_valid_utf8_boundaries() {
let mut string = Wtf8Buf::from_str("aé 💩");
string.push(CodePoint::from_u32(0xD800).unwrap());
string.push(CodePoint::from_u32(0xD800).unwrap());
check_utf8_boundary(&string, 0);
check_utf8_boundary(&string, 1);
check_utf8_boundary(&string, 3);
check_utf8_boundary(&string, 4);
check_utf8_boundary(&string, 8);
check_utf8_boundary(&string, 14);
string.check_utf8_boundary(0);
string.check_utf8_boundary(1);
string.check_utf8_boundary(3);
string.check_utf8_boundary(4);
string.check_utf8_boundary(8);
string.check_utf8_boundary(14);
assert_eq!(string.len(), 14);
string.push_char('a');
check_utf8_boundary(&string, 14);
check_utf8_boundary(&string, 15);
string.check_utf8_boundary(14);
string.check_utf8_boundary(15);
let mut string = Wtf8Buf::from_str("a");
string.push(CodePoint::from_u32(0xD800).unwrap());
check_utf8_boundary(&string, 1);
string.check_utf8_boundary(1);
let mut string = Wtf8Buf::from_str("\u{D7FF}");
string.push(CodePoint::from_u32(0xD800).unwrap());
check_utf8_boundary(&string, 3);
string.check_utf8_boundary(3);
let mut string = Wtf8Buf::new();
string.push(CodePoint::from_u32(0xD800).unwrap());
string.push_char('\u{D7FF}');
check_utf8_boundary(&string, 3);
string.check_utf8_boundary(3);
}
#[test]
#[should_panic(expected = "byte index 4 is out of bounds")]
fn wtf8_utf8_boundary_out_of_bounds() {
let string = Wtf8::from_str("aé");
check_utf8_boundary(&string, 4);
string.check_utf8_boundary(4);
}
#[test]
#[should_panic(expected = "byte index 1 is not a codepoint boundary")]
fn wtf8_utf8_boundary_inside_codepoint() {
let string = Wtf8::from_str("é");
check_utf8_boundary(&string, 1);
string.check_utf8_boundary(1);
}
#[test]
@ -714,7 +693,7 @@ fn wtf8_utf8_boundary_inside_codepoint() {
fn wtf8_utf8_boundary_inside_surrogate() {
let mut string = Wtf8Buf::new();
string.push(CodePoint::from_u32(0xD800).unwrap());
check_utf8_boundary(&string, 1);
string.check_utf8_boundary(1);
}
#[test]
@ -723,20 +702,22 @@ fn wtf8_utf8_boundary_between_surrogates() {
let mut string = Wtf8Buf::new();
string.push(CodePoint::from_u32(0xD800).unwrap());
string.push(CodePoint::from_u32(0xD800).unwrap());
check_utf8_boundary(&string, 3);
string.check_utf8_boundary(3);
}
#[test]
fn wobbled_wtf8_plus_bytes_isnt_utf8() {
let mut string: Wtf8Buf = unsafe { Wtf8::from_bytes_unchecked(b"\xED\xA0\x80").to_owned() };
let mut string: Wtf8Buf = to_owned(unsafe { Wtf8::from_bytes_unchecked(b"\xED\xA0\x80") });
assert!(!string.is_known_utf8);
string.extend_from_slice(b"some utf-8");
unsafe {
string.extend_from_slice_unchecked(b"some utf-8");
}
assert!(!string.is_known_utf8);
}
#[test]
fn wobbled_wtf8_plus_str_isnt_utf8() {
let mut string: Wtf8Buf = unsafe { Wtf8::from_bytes_unchecked(b"\xED\xA0\x80").to_owned() };
let mut string: Wtf8Buf = to_owned(unsafe { Wtf8::from_bytes_unchecked(b"\xED\xA0\x80") });
assert!(!string.is_known_utf8);
string.push_str("some utf-8");
assert!(!string.is_known_utf8);

12
library/alloctests/lib.rs
View file

@ -18,6 +18,8 @@
#![feature(allocator_api)]
#![feature(array_into_iter_constructors)]
#![feature(assert_matches)]
#![feature(char_internals)]
#![feature(char_max_len)]
#![feature(core_intrinsics)]
#![feature(exact_size_is_empty)]
#![feature(extend_one)]
@ -41,6 +43,7 @@
#![feature(trusted_random_access)]
#![feature(try_reserve_kind)]
#![feature(try_trait_v2)]
#![feature(wtf8_internals)]
// tidy-alphabetical-end
//
// Language features:
@ -68,15 +71,18 @@ extern crate test;
mod testing;
use realalloc::*;
// We are directly including collections and raw_vec here as both use non-public
// methods and fields in tests and as such need to have the types to test in the
// same crate as the tests themself.
// We are directly including collections, raw_vec, and wtf8 here as they use non-public
// methods and fields in tests and as such need to have the types to test in the same
// crate as the tests themself.
#[path = "../alloc/src/collections/mod.rs"]
mod collections;
#[path = "../alloc/src/raw_vec/mod.rs"]
mod raw_vec;
#[path = "../alloc/src/wtf8/mod.rs"]
mod wtf8;
#[allow(dead_code)] // Not used in all configurations
pub(crate) mod test_helpers {
/// Copied from `std::test_helpers::test_rng`, since these tests rely on the

1
library/alloctests/tests/lib.rs
View file

@ -3,6 +3,7 @@
#![feature(iter_array_chunks)]
#![feature(assert_matches)]
#![feature(btree_extract_if)]
#![feature(wtf8_internals)]
#![feature(char_max_len)]
#![feature(cow_is_borrowed)]
#![feature(core_intrinsics)]

2
library/core/src/lib.rs
View file

@ -363,6 +363,8 @@ pub mod slice;
pub mod str;
pub mod time;
pub mod wtf8;
pub mod unicode;
/* Async */

15
library/core/src/num/niche_types.rs
View file

@ -178,3 +178,18 @@ impl NotAllOnesHelper for u64 {
impl NotAllOnesHelper for i64 {
type Type = I64NotAllOnes;
}
define_valid_range_type! {
pub struct CodePointInner(u32 as u32 in 0..=0x10ffff);
}
impl CodePointInner {
pub const ZERO: Self = CodePointInner::new(0).unwrap();
}
impl Default for CodePointInner {
#[inline]
fn default() -> Self {
Self::ZERO
}
}

3
library/core/src/ptr/metadata.rs
View file

@ -106,6 +106,9 @@ pub const fn metadata<T: PointeeSized>(ptr: *const T) -> <T as Pointee>::Metadat
/// For slices, see the documentation of [`slice::from_raw_parts`] for safety requirements.
/// For trait objects, the metadata must come from a pointer to the same underlying erased type.
///
/// If you are attempting to deconstruct a DST in a generic context to be reconstructed later,
/// a thin pointer can always be obtained by casting `*const T` to `*const ()`.
///
/// [`slice::from_raw_parts`]: crate::slice::from_raw_parts
#[unstable(feature = "ptr_metadata", issue = "81513")]
#[inline]

597
library/core/src/wtf8.rs Normal file
View file

@ -0,0 +1,597 @@
//! Implementation of [the WTF-8 encoding](https://simonsapin.github.io/wtf-8/).
//!
//! This library uses Rust’s type system to maintain
//! [well-formedness](https://simonsapin.github.io/wtf-8/#well-formed),
//! like the `String` and `&str` types do for UTF-8.
//!
//! Since [WTF-8 must not be used
//! for interchange](https://simonsapin.github.io/wtf-8/#intended-audience),
//! this library deliberately does not provide access to the underlying bytes
//! of WTF-8 strings,
//! nor can it decode WTF-8 from arbitrary bytes.
//! WTF-8 strings can be obtained from UTF-8, UTF-16, or code points.
#![unstable(
feature = "wtf8_internals",
issue = "none",
reason = "this is internal code for representing OsStr on some platforms and not a public API"
)]
// rustdoc bug: doc(hidden) on the module won't stop types in the module from showing up in trait
// implementations, so, we'll have to add more doc(hidden)s anyway
#![doc(hidden)]
use crate::char::{MAX_LEN_UTF16, encode_utf16_raw};
use crate::clone::CloneToUninit;
use crate::fmt::{self, Write};
use crate::hash::{Hash, Hasher};
use crate::iter::FusedIterator;
use crate::num::niche_types::CodePointInner;
use crate::str::next_code_point;
use crate::{ops, slice, str};
/// A Unicode code point: from U+0000 to U+10FFFF.
///
/// Compares with the `char` type,
/// which represents a Unicode scalar value:
/// a code point that is not a surrogate (U+D800 to U+DFFF).
#[derive(Eq, PartialEq, Ord, PartialOrd, Clone, Copy)]
#[doc(hidden)]
pub struct CodePoint(CodePointInner);
/// Format the code point as `U+` followed by four to six hexadecimal digits.
/// Example: `U+1F4A9`
impl fmt::Debug for CodePoint {
#[inline]
fn fmt(&self, formatter: &mut fmt::Formatter<'_>) -> fmt::Result {
write!(formatter, "U+{:04X}", self.0.as_inner())
}
}
impl CodePoint {
/// Unsafely creates a new `CodePoint` without checking the value.
///
/// Only use when `value` is known to be less than or equal to 0x10FFFF.
#[inline]
pub unsafe fn from_u32_unchecked(value: u32) -> CodePoint {
// SAFETY: Guaranteed by caller.
CodePoint(unsafe { CodePointInner::new_unchecked(value) })
}
/// Creates a new `CodePoint` if the value is a valid code point.
///
/// Returns `None` if `value` is above 0x10FFFF.
#[inline]
pub fn from_u32(value: u32) -> Option<CodePoint> {
Some(CodePoint(CodePointInner::new(value)?))
}
/// Creates a new `CodePoint` from a `char`.
///
/// Since all Unicode scalar values are code points, this always succeeds.
#[inline]
pub fn from_char(value: char) -> CodePoint {
// SAFETY: All char are valid for this type.
unsafe { CodePoint::from_u32_unchecked(value as u32) }
}
/// Returns the numeric value of the code point.
#[inline]
pub fn to_u32(&self) -> u32 {
self.0.as_inner()
}
/// Returns the numeric value of the code point if it is a leading surrogate.
#[inline]
pub fn to_lead_surrogate(&self) -> Option<u16> {
match self.to_u32() {
lead @ 0xD800..=0xDBFF => Some(lead as u16),
_ => None,
}
}
/// Returns the numeric value of the code point if it is a trailing surrogate.
#[inline]
pub fn to_trail_surrogate(&self) -> Option<u16> {
match self.to_u32() {
trail @ 0xDC00..=0xDFFF => Some(trail as u16),
_ => None,
}
}
/// Optionally returns a Unicode scalar value for the code point.
///
/// Returns `None` if the code point is a surrogate (from U+D800 to U+DFFF).
#[inline]
pub fn to_char(&self) -> Option<char> {
match self.to_u32() {
0xD800..=0xDFFF => None,
// SAFETY: We explicitly check that the char is valid.
valid => Some(unsafe { char::from_u32_unchecked(valid) }),
}
}
/// Returns a Unicode scalar value for the code point.
///
/// Returns `'\u{FFFD}'` (the replacement character “<>”)
/// if the code point is a surrogate (from U+D800 to U+DFFF).
#[inline]
pub fn to_char_lossy(&self) -> char {
self.to_char().unwrap_or(char::REPLACEMENT_CHARACTER)
}
}
/// A borrowed slice of well-formed WTF-8 data.
///
/// Similar to `&str`, but can additionally contain surrogate code points
/// if they’re not in a surrogate pair.
#[derive(Eq, Ord, PartialEq, PartialOrd)]
#[repr(transparent)]
#[rustc_has_incoherent_inherent_impls]
#[doc(hidden)]
pub struct Wtf8 {
bytes: [u8],
}
impl AsRef<[u8]> for Wtf8 {
#[inline]
fn as_ref(&self) -> &[u8] {
&self.bytes
}
}
/// Formats the string in double quotes, with characters escaped according to
/// [`char::escape_debug`] and unpaired surrogates represented as `\u{xxxx}`,
/// where each `x` is a hexadecimal digit.
impl fmt::Debug for Wtf8 {
fn fmt(&self, formatter: &mut fmt::Formatter<'_>) -> fmt::Result {
fn write_str_escaped(f: &mut fmt::Formatter<'_>, s: &str) -> fmt::Result {
use crate::fmt::Write;
for c in s.chars().flat_map(|c| c.escape_debug()) {
f.write_char(c)?
}
Ok(())
}
formatter.write_str("\"")?;
let mut pos = 0;
while let Some((surrogate_pos, surrogate)) = self.next_surrogate(pos) {
// SAFETY: next_surrogate provides an index for a range of valid UTF-8 bytes.
write_str_escaped(formatter, unsafe {
str::from_utf8_unchecked(&self.bytes[pos..surrogate_pos])
})?;
write!(formatter, "\\u{{{:x}}}", surrogate)?;
pos = surrogate_pos + 3;
}
// SAFETY: after next_surrogate returns None, the remainder is valid UTF-8.
write_str_escaped(formatter, unsafe { str::from_utf8_unchecked(&self.bytes[pos..]) })?;
formatter.write_str("\"")
}
}
/// Formats the string with unpaired surrogates substituted with the replacement
/// character, U+FFFD.
impl fmt::Display for Wtf8 {
fn fmt(&self, formatter: &mut fmt::Formatter<'_>) -> fmt::Result {
let wtf8_bytes = &self.bytes;
let mut pos = 0;
loop {
match self.next_surrogate(pos) {
Some((surrogate_pos, _)) => {
// SAFETY: next_surrogate provides an index for a range of valid UTF-8 bytes.
formatter.write_str(unsafe {
str::from_utf8_unchecked(&wtf8_bytes[pos..surrogate_pos])
})?;
formatter.write_char(char::REPLACEMENT_CHARACTER)?;
pos = surrogate_pos + 3;
}
None => {
// SAFETY: after next_surrogate returns None, the remainder is valid UTF-8.
let s = unsafe { str::from_utf8_unchecked(&wtf8_bytes[pos..]) };
if pos == 0 { return s.fmt(formatter) } else { return formatter.write_str(s) }
}
}
}
}
}
impl Wtf8 {
/// Creates a WTF-8 slice from a UTF-8 `&str` slice.
#[inline]
pub fn from_str(value: &str) -> &Wtf8 {
// SAFETY: Since WTF-8 is a superset of UTF-8, this always is valid.
unsafe { Wtf8::from_bytes_unchecked(value.as_bytes()) }
}
/// Creates a WTF-8 slice from a WTF-8 byte slice.
///
/// Since the byte slice is not checked for valid WTF-8, this functions is
/// marked unsafe.
#[inline]
pub unsafe fn from_bytes_unchecked(value: &[u8]) -> &Wtf8 {
// SAFETY: start with &[u8], end with fancy &[u8]
unsafe { &*(value as *const [u8] as *const Wtf8) }
}
/// Creates a mutable WTF-8 slice from a mutable WTF-8 byte slice.
///
/// Since the byte slice is not checked for valid WTF-8, this functions is
/// marked unsafe.
#[inline]
pub unsafe fn from_mut_bytes_unchecked(value: &mut [u8]) -> &mut Wtf8 {
// SAFETY: start with &mut [u8], end with fancy &mut [u8]
unsafe { &mut *(value as *mut [u8] as *mut Wtf8) }
}
/// Returns the length, in WTF-8 bytes.
#[inline]
pub fn len(&self) -> usize {
self.bytes.len()
}
#[inline]
pub fn is_empty(&self) -> bool {
self.bytes.is_empty()
}
/// Returns the code point at `position` if it is in the ASCII range,
/// or `b'\xFF'` otherwise.
///
/// # Panics
///
/// Panics if `position` is beyond the end of the string.
#[inline]
pub fn ascii_byte_at(&self, position: usize) -> u8 {
match self.bytes[position] {
ascii_byte @ 0x00..=0x7F => ascii_byte,
_ => 0xFF,
}
}
/// Returns an iterator for the string’s code points.
#[inline]
pub fn code_points(&self) -> Wtf8CodePoints<'_> {
Wtf8CodePoints { bytes: self.bytes.iter() }
}
/// Access raw bytes of WTF-8 data
#[inline]
pub fn as_bytes(&self) -> &[u8] {
&self.bytes
}
/// Tries to convert the string to UTF-8 and return a `&str` slice.
///
/// Returns `None` if the string contains surrogates.
///
/// This does not copy the data.
#[inline]
pub fn as_str(&self) -> Result<&str, str::Utf8Error> {
str::from_utf8(&self.bytes)
}
/// Converts the WTF-8 string to potentially ill-formed UTF-16
/// and return an iterator of 16-bit code units.
///
/// This is lossless:
/// calling `Wtf8Buf::from_ill_formed_utf16` on the resulting code units
/// would always return the original WTF-8 string.
#[inline]
pub fn encode_wide(&self) -> EncodeWide<'_> {
EncodeWide { code_points: self.code_points(), extra: 0 }
}
#[inline]
pub fn next_surrogate(&self, mut pos: usize) -> Option<(usize, u16)> {
let mut iter = self.bytes[pos..].iter();
loop {
let b = *iter.next()?;
if b < 0x80 {
pos += 1;
} else if b < 0xE0 {
iter.next();
pos += 2;
} else if b == 0xED {
match (iter.next(), iter.next()) {
(Some(&b2), Some(&b3)) if b2 >= 0xA0 => {
return Some((pos, decode_surrogate(b2, b3)));
}
_ => pos += 3,
}
} else if b < 0xF0 {
iter.next();
iter.next();
pos += 3;
} else {
iter.next();
iter.next();
iter.next();
pos += 4;
}
}
}
#[inline]
pub fn final_lead_surrogate(&self) -> Option<u16> {
match self.bytes {
[.., 0xED, b2 @ 0xA0..=0xAF, b3] => Some(decode_surrogate(b2, b3)),
_ => None,
}
}
#[inline]
pub fn initial_trail_surrogate(&self) -> Option<u16> {
match self.bytes {
[0xED, b2 @ 0xB0..=0xBF, b3, ..] => Some(decode_surrogate(b2, b3)),
_ => None,
}
}
#[inline]
pub fn make_ascii_lowercase(&mut self) {
self.bytes.make_ascii_lowercase()
}
#[inline]
pub fn make_ascii_uppercase(&mut self) {
self.bytes.make_ascii_uppercase()
}
#[inline]
pub fn is_ascii(&self) -> bool {
self.bytes.is_ascii()
}
#[inline]
pub fn eq_ignore_ascii_case(&self, other: &Self) -> bool {
self.bytes.eq_ignore_ascii_case(&other.bytes)
}
}
/// Returns a slice of the given string for the byte range \[`begin`..`end`).
///
/// # Panics
///
/// Panics when `begin` and `end` do not point to code point boundaries,
/// or point beyond the end of the string.
impl ops::Index<ops::Range<usize>> for Wtf8 {
type Output = Wtf8;
#[inline]
fn index(&self, range: ops::Range<usize>) -> &Wtf8 {
if range.start <= range.end
&& self.is_code_point_boundary(range.start)
&& self.is_code_point_boundary(range.end)
{
// SAFETY: is_code_point_boundary checks that the index is valid
unsafe { slice_unchecked(self, range.start, range.end) }
} else {
slice_error_fail(self, range.start, range.end)
}
}
}
/// Returns a slice of the given string from byte `begin` to its end.
///
/// # Panics
///
/// Panics when `begin` is not at a code point boundary,
/// or is beyond the end of the string.
impl ops::Index<ops::RangeFrom<usize>> for Wtf8 {
type Output = Wtf8;
#[inline]
fn index(&self, range: ops::RangeFrom<usize>) -> &Wtf8 {
if self.is_code_point_boundary(range.start) {
// SAFETY: is_code_point_boundary checks that the index is valid
unsafe { slice_unchecked(self, range.start, self.len()) }
} else {
slice_error_fail(self, range.start, self.len())
}
}
}
/// Returns a slice of the given string from its beginning to byte `end`.
///
/// # Panics
///
/// Panics when `end` is not at a code point boundary,
/// or is beyond the end of the string.
impl ops::Index<ops::RangeTo<usize>> for Wtf8 {
type Output = Wtf8;
#[inline]
fn index(&self, range: ops::RangeTo<usize>) -> &Wtf8 {
if self.is_code_point_boundary(range.end) {
// SAFETY: is_code_point_boundary checks that the index is valid
unsafe { slice_unchecked(self, 0, range.end) }
} else {
slice_error_fail(self, 0, range.end)
}
}
}
impl ops::Index<ops::RangeFull> for Wtf8 {
type Output = Wtf8;
#[inline]
fn index(&self, _range: ops::RangeFull) -> &Wtf8 {
self
}
}
#[inline]
fn decode_surrogate(second_byte: u8, third_byte: u8) -> u16 {
// The first byte is assumed to be 0xED
0xD800 | (second_byte as u16 & 0x3F) << 6 | third_byte as u16 & 0x3F
}
impl Wtf8 {
/// Copied from str::is_char_boundary
#[inline]
pub fn is_code_point_boundary(&self, index: usize) -> bool {
if index == 0 {
return true;
}
match self.bytes.get(index) {
None => index == self.len(),
Some(&b) => (b as i8) >= -0x40,
}
}
/// Verify that `index` is at the edge of either a valid UTF-8 codepoint
/// (i.e. a codepoint that's not a surrogate) or of the whole string.
///
/// These are the cases currently permitted by `OsStr::self_encoded_bytes`.
/// Splitting between surrogates is valid as far as WTF-8 is concerned, but
/// we do not permit it in the public API because WTF-8 is considered an
/// implementation detail.
#[track_caller]
#[inline]
pub fn check_utf8_boundary(&self, index: usize) {
if index == 0 {
return;
}
match self.bytes.get(index) {
Some(0xED) => (), // Might be a surrogate
Some(&b) if (b as i8) >= -0x40 => return,
Some(_) => panic!("byte index {index} is not a codepoint boundary"),
None if index == self.len() => return,
None => panic!("byte index {index} is out of bounds"),
}
if self.bytes[index + 1] >= 0xA0 {
// There's a surrogate after index. Now check before index.
if index >= 3 && self.bytes[index - 3] == 0xED && self.bytes[index - 2] >= 0xA0 {
panic!("byte index {index} lies between surrogate codepoints");
}
}
}
}
/// Copied from core::str::raw::slice_unchecked
#[inline]
unsafe fn slice_unchecked(s: &Wtf8, begin: usize, end: usize) -> &Wtf8 {
// SAFETY: memory layout of a &[u8] and &Wtf8 are the same
unsafe {
let len = end - begin;
let start = s.as_bytes().as_ptr().add(begin);
Wtf8::from_bytes_unchecked(slice::from_raw_parts(start, len))
}
}
/// Copied from core::str::raw::slice_error_fail
#[inline(never)]
fn slice_error_fail(s: &Wtf8, begin: usize, end: usize) -> ! {
assert!(begin <= end);
panic!("index {begin} and/or {end} in `{s:?}` do not lie on character boundary");
}
/// Iterator for the code points of a WTF-8 string.
///
/// Created with the method `.code_points()`.
#[derive(Clone)]
#[doc(hidden)]
pub struct Wtf8CodePoints<'a> {
bytes: slice::Iter<'a, u8>,
}
impl Iterator for Wtf8CodePoints<'_> {
type Item = CodePoint;
#[inline]
fn next(&mut self) -> Option<CodePoint> {
// SAFETY: `self.bytes` has been created from a WTF-8 string
unsafe { next_code_point(&mut self.bytes).map(|c| CodePoint::from_u32_unchecked(c)) }
}
#[inline]
fn size_hint(&self) -> (usize, Option<usize>) {
let len = self.bytes.len();
(len.saturating_add(3) / 4, Some(len))
}
}
impl fmt::Debug for Wtf8CodePoints<'_> {
fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
f.debug_tuple("Wtf8CodePoints")
// SAFETY: We always leave the string in a valid state after each iteration.
.field(&unsafe { Wtf8::from_bytes_unchecked(self.bytes.as_slice()) })
.finish()
}
}
/// Generates a wide character sequence for potentially ill-formed UTF-16.
#[stable(feature = "rust1", since = "1.0.0")]
#[derive(Clone)]
#[doc(hidden)]
pub struct EncodeWide<'a> {
code_points: Wtf8CodePoints<'a>,
extra: u16,
}
// Copied from libunicode/u_str.rs
#[stable(feature = "rust1", since = "1.0.0")]
impl Iterator for EncodeWide<'_> {
type Item = u16;
#[inline]
fn next(&mut self) -> Option<u16> {
if self.extra != 0 {
let tmp = self.extra;
self.extra = 0;
return Some(tmp);
}
let mut buf = [0; MAX_LEN_UTF16];
self.code_points.next().map(|code_point| {
let n = encode_utf16_raw(code_point.to_u32(), &mut buf).len();
if n == 2 {
self.extra = buf[1];
}
buf[0]
})
}
#[inline]
fn size_hint(&self) -> (usize, Option<usize>) {
let (low, high) = self.code_points.size_hint();
let ext = (self.extra != 0) as usize;
// every code point gets either one u16 or two u16,
// so this iterator is between 1 or 2 times as
// long as the underlying iterator.
(low + ext, high.and_then(|n| n.checked_mul(2)).and_then(|n| n.checked_add(ext)))
}
}
impl fmt::Debug for EncodeWide<'_> {
fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
f.debug_struct("EncodeWide").finish_non_exhaustive()
}
}
#[stable(feature = "encode_wide_fused_iterator", since = "1.62.0")]
impl FusedIterator for EncodeWide<'_> {}
impl Hash for CodePoint {
#[inline]
fn hash<H: Hasher>(&self, state: &mut H) {
self.0.hash(state)
}
}
impl Hash for Wtf8 {
#[inline]
fn hash<H: Hasher>(&self, state: &mut H) {
state.write(&self.bytes);
0xfeu8.hash(state)
}
}
#[unstable(feature = "clone_to_uninit", issue = "126799")]
unsafe impl CloneToUninit for Wtf8 {
#[inline]
#[cfg_attr(debug_assertions, track_caller)]
unsafe fn clone_to_uninit(&self, dst: *mut u8) {
// SAFETY: we're just a transparent wrapper around [u8]
unsafe { self.bytes.clone_to_uninit(dst) }
}
}

1
library/coretests/tests/lib.rs
View file

@ -194,6 +194,7 @@ mod time;
mod tuple;
mod unicode;
mod waker;
mod wtf8;
/// Copied from `std::test_helpers::test_rng`, see that function for rationale.
#[track_caller]

1
library/coretests/tests/wtf8.rs Normal file
View file

@ -0,0 +1 @@
// All `wtf8` tests live in library/alloctests/tests/wtf8.rs

3
library/std/src/fs.rs
View file

@ -3036,6 +3036,9 @@ pub fn remove_dir_all<P: AsRef<Path>>(path: P) -> io::Result<()> {
/// Entries for the current and parent directories (typically `.` and `..`) are
/// skipped.
///
/// The order in which `read_dir` returns entries can change between calls. If reproducible
/// ordering is required, the entries should be explicitly sorted.
///
/// # Platform-specific behavior
///
/// This function currently corresponds to the `opendir` function on Unix

1
library/std/src/lib.rs
View file

@ -391,6 +391,7 @@
#![feature(try_with_capacity)]
#![feature(unique_rc_arc)]
#![feature(vec_into_raw_parts)]
#![feature(wtf8_internals)]
// tidy-alphabetical-end
//
// Library features (unwind):

38
library/std/src/os/windows/ffi.rs
View file

@ -53,12 +53,13 @@
#![stable(feature = "rust1", since = "1.0.0")]
use alloc::wtf8::Wtf8Buf;
use crate::ffi::{OsStr, OsString};
use crate::fmt;
use crate::iter::FusedIterator;
use crate::sealed::Sealed;
use crate::sys::os_str::Buf;
#[stable(feature = "rust1", since = "1.0.0")]
pub use crate::sys_common::wtf8::EncodeWide;
use crate::sys_common::wtf8::Wtf8Buf;
use crate::sys_common::{AsInner, FromInner};
/// Windows-specific extensions to [`OsString`].
@ -130,6 +131,35 @@ pub trait OsStrExt: Sealed {
impl OsStrExt for OsStr {
#[inline]
fn encode_wide(&self) -> EncodeWide<'_> {
self.as_inner().inner.encode_wide()
EncodeWide { inner: self.as_inner().inner.encode_wide() }
}
}
/// Iterator returned by [`OsStrExt::encode_wide`].
#[stable(feature = "rust1", since = "1.0.0")]
#[derive(Clone)]
pub struct EncodeWide<'a> {
inner: alloc::wtf8::EncodeWide<'a>,
}
#[stable(feature = "encode_wide_debug", since = "CURRENT_RUSTC_VERSION")]
impl fmt::Debug for EncodeWide<'_> {
fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
fmt::Debug::fmt(&self.inner, f)
}
}
#[stable(feature = "rust1", since = "1.0.0")]
impl Iterator for EncodeWide<'_> {
type Item = u16;
#[inline]
fn next(&mut self) -> Option<u16> {
self.inner.next()
}
#[inline]
fn size_hint(&self) -> (usize, Option<usize>) {
self.inner.size_hint()
}
}
#[stable(feature = "encode_wide_fused_iterator", since = "1.62.0")]
impl FusedIterator for EncodeWide<'_> {}

8
library/std/src/sync/mpmc/utils.rs
View file

@ -23,14 +23,13 @@ use crate::ops::{Deref, DerefMut};
any(target_arch = "x86_64", target_arch = "aarch64", target_arch = "powerpc64",),
repr(align(128))
)]
// arm, mips, mips64, and riscv64 have 32-byte cache line size.
// arm, mips and mips64 have 32-byte cache line size.
//
// Sources:
// - https://github.com/golang/go/blob/3dd58676054223962cd915bb0934d1f9f489d4d2/src/internal/cpu/cpu_arm.go#L7
// - https://github.com/golang/go/blob/3dd58676054223962cd915bb0934d1f9f489d4d2/src/internal/cpu/cpu_mips.go#L7
// - https://github.com/golang/go/blob/3dd58676054223962cd915bb0934d1f9f489d4d2/src/internal/cpu/cpu_mipsle.go#L7
// - https://github.com/golang/go/blob/3dd58676054223962cd915bb0934d1f9f489d4d2/src/internal/cpu/cpu_mips64x.go#L9
// - https://github.com/golang/go/blob/3dd58676054223962cd915bb0934d1f9f489d4d2/src/internal/cpu/cpu_riscv64.go#L7
#[cfg_attr(
any(
target_arch = "arm",
@ -38,7 +37,6 @@ use crate::ops::{Deref, DerefMut};
target_arch = "mips32r6",
target_arch = "mips64",
target_arch = "mips64r6",
target_arch = "riscv64",
),
repr(align(32))
)]
@ -47,11 +45,12 @@ use crate::ops::{Deref, DerefMut};
// Sources:
// - https://github.com/golang/go/blob/3dd58676054223962cd915bb0934d1f9f489d4d2/src/internal/cpu/cpu_s390x.go#L7
#[cfg_attr(target_arch = "s390x", repr(align(256)))]
// x86 and wasm have 64-byte cache line size.
// x86, wasm and riscv have 64-byte cache line size.
//
// Sources:
// - https://github.com/golang/go/blob/dda2991c2ea0c5914714469c4defc2562a907230/src/internal/cpu/cpu_x86.go#L9
// - https://github.com/golang/go/blob/3dd58676054223962cd915bb0934d1f9f489d4d2/src/internal/cpu/cpu_wasm.go#L7
// - https://github.com/golang/go/blob/5e31f78c8a4ed1b872ddc194f0cd1ae931b37d7e/src/internal/cpu/cpu_riscv64.go#L7
//
// All others are assumed to have 64-byte cache line size.
#[cfg_attr(
@ -64,7 +63,6 @@ use crate::ops::{Deref, DerefMut};
target_arch = "mips32r6",
target_arch = "mips64",
target_arch = "mips64r6",
target_arch = "riscv64",
target_arch = "s390x",
)),
repr(align(64))

8
library/std/src/sys/os_str/wtf8.rs
View file

@ -1,12 +1,12 @@
//! The underlying OsString/OsStr implementation on Windows is a
//! wrapper around the "WTF-8" encoding; see the `wtf8` module for more.
use alloc::wtf8::{Wtf8, Wtf8Buf};
use core::clone::CloneToUninit;
use crate::borrow::Cow;
use crate::collections::TryReserveError;
use crate::rc::Rc;
use crate::sync::Arc;
use crate::sys_common::wtf8::{Wtf8, Wtf8Buf, check_utf8_boundary};
use crate::sys_common::{AsInner, FromInner, IntoInner};
use crate::{fmt, mem};
@ -220,7 +220,9 @@ impl Buf {
/// trailing surrogate half.
#[inline]
pub unsafe fn extend_from_slice_unchecked(&mut self, other: &[u8]) {
self.inner.extend_from_slice(other);
unsafe {
self.inner.extend_from_slice_unchecked(other);
}
}
}
@ -238,7 +240,7 @@ impl Slice {
#[track_caller]
#[inline]
pub fn check_public_boundary(&self, index: usize) {
check_utf8_boundary(&self.inner, index);
self.inner.check_utf8_boundary(index);
}
#[inline]

1
library/std/src/sys_common/mod.rs
View file

@ -21,7 +21,6 @@
mod tests;
pub mod wstr;
pub mod wtf8;
// common error constructors

1083
library/std/src/sys_common/wtf8.rs
View file

File diff suppressed because it is too large Load diff

2
src/doc/rustc/src/platform-support.md
View file

@ -103,7 +103,6 @@ target | notes
[`powerpc64le-unknown-linux-gnu`](platform-support/powerpc64le-unknown-linux-gnu.md) | PPC64LE Linux (kernel 3.10+, glibc 2.17)
[`powerpc64le-unknown-linux-musl`](platform-support/powerpc64le-unknown-linux-musl.md) | PPC64LE Linux (kernel 4.19+, musl 1.2.3)
[`riscv64gc-unknown-linux-gnu`](platform-support/riscv64gc-unknown-linux-gnu.md) | RISC-V Linux (kernel 4.20+, glibc 2.29)
[`riscv64gc-unknown-linux-musl`](platform-support/riscv64gc-unknown-linux-musl.md) | RISC-V Linux (kernel 4.20+, musl 1.2.3)
[`s390x-unknown-linux-gnu`](platform-support/s390x-unknown-linux-gnu.md) | S390x Linux (kernel 3.2+, glibc 2.17)
[`x86_64-apple-darwin`](platform-support/apple-darwin.md) | 64-bit macOS (10.12+, Sierra+)
[`x86_64-pc-windows-gnullvm`](platform-support/windows-gnullvm.md) | 64-bit x86 MinGW (Windows 10+), LLVM ABI
@ -183,6 +182,7 @@ target | std | notes
[`riscv32imac-unknown-none-elf`](platform-support/riscv32-unknown-none-elf.md) | * | Bare RISC-V (RV32IMAC ISA)
[`riscv32imafc-unknown-none-elf`](platform-support/riscv32-unknown-none-elf.md) | * | Bare RISC-V (RV32IMAFC ISA)
[`riscv32imc-unknown-none-elf`](platform-support/riscv32-unknown-none-elf.md) | * | Bare RISC-V (RV32IMC ISA)
[`riscv64gc-unknown-linux-musl`](platform-support/riscv64gc-unknown-linux-musl.md) | RISC-V Linux (kernel 4.20+, musl 1.2.3)
`riscv64gc-unknown-none-elf` | * | Bare RISC-V (RV64IMAFDC ISA)
`riscv64imac-unknown-none-elf` | * | Bare RISC-V (RV64IMAC ISA)
`sparc64-unknown-linux-gnu` | ✓ | SPARC Linux (kernel 4.4+, glibc 2.23)

1
src/tools/tidy/src/unit_tests.rs
View file

@ -61,6 +61,7 @@ pub fn check(root_path: &Path, stdlib: bool, bad: &mut bool) {
|| path.ends_with("library/alloc/src/collections/linked_list/tests.rs")
|| path.ends_with("library/alloc/src/collections/vec_deque/tests.rs")
|| path.ends_with("library/alloc/src/raw_vec/tests.rs")
|| path.ends_with("library/alloc/src/wtf8/tests.rs")
}
};

2
tests/mir-opt/building/async_await.rs
View file

@ -4,7 +4,7 @@
// related to `yield` are `&mut Context`, and its return type is `Poll`.
//@ edition:2018
//@ compile-flags: -C panic=abort
//@ compile-flags: -Zmir-opt-level=0 -C panic=abort
#![crate_type = "lib"]

2
tests/mir-opt/building/dump_mir_cycle.rs
View file

@ -1,3 +1,5 @@
//@ compile-flags: -Zmir-opt-level=0
#[derive(Debug)]
pub struct Thing {
pub next: &'static Thing,

1
tests/mir-opt/building/enum_cast.rs
View file

@ -1,3 +1,4 @@
//@ compile-flags: -Zmir-opt-level=0
// skip-filecheck
// EMIT_MIR enum_cast.foo.built.after.mir
// EMIT_MIR enum_cast.bar.built.after.mir

1
tests/mir-opt/building/eq_never_type.rs
View file

@ -1,3 +1,4 @@
//@ compile-flags: -Zmir-opt-level=0
// skip-filecheck
#![feature(never_type)]
#![allow(unreachable_code)]

47
tests/mir-opt/building/index_array_and_slice.rs
View file

@ -1,36 +1,42 @@
//@ compile-flags: -C opt-level=0
//@ compile-flags: -Zmir-opt-level=0 -C opt-level=0
// EMIT_MIR index_array_and_slice.index_array.built.after.mir
fn index_array(array: &[i32; 7], index: usize) -> &i32 {
// CHECK: bb0:
// CHECK: [[LT:_.+]] = Lt(copy _2, const 7_usize);
// CHECK: assert(move [[LT]], "index out of bounds{{.+}}", const 7_usize, copy _2) -> [success: bb1, unwind
// CHECK: _3 = copy _2;
// CHECK: [[LT:_.+]] = Lt(copy _3, const 7_usize);
// CHECK: assert(move [[LT]], "index out of bounds{{.+}}", const 7_usize, copy _3) -> [success: bb1, unwind
// CHECK: bb1:
// CHECK: _0 = &(*_1)[_2];
// CHECK: _5 = &(*_1)[_3];
// CHECK: _0 = &(*_5);
&array[index]
}
// EMIT_MIR index_array_and_slice.index_const_generic_array.built.after.mir
fn index_const_generic_array<const N: usize>(array: &[i32; N], index: usize) -> &i32 {
// CHECK: bb0:
// CHECK: [[LT:_.+]] = Lt(copy _2, const N);
// CHECK: assert(move [[LT]], "index out of bounds{{.+}}", const N, copy _2) -> [success: bb1, unwind
// CHECK: _3 = copy _2;
// CHECK: [[LT:_.+]] = Lt(copy _3, const N);
// CHECK: assert(move [[LT]], "index out of bounds{{.+}}", const N, copy _3) -> [success: bb1, unwind
// CHECK: bb1:
// CHECK: _0 = &(*_1)[_2];
// CHECK: _5 = &(*_1)[_3];
// CHECK: _0 = &(*_5);
&array[index]
}
// EMIT_MIR index_array_and_slice.index_slice.built.after.mir
fn index_slice(slice: &[i32], index: usize) -> &i32 {
// CHECK: bb0:
// CHECK: _3 = copy _2;
// CHECK: [[LEN:_.+]] = PtrMetadata(copy _1);
// CHECK: [[LT:_.+]] = Lt(copy _2, copy [[LEN]]);
// CHECK: assert(move [[LT]], "index out of bounds{{.+}}", move [[LEN]], copy _2) -> [success: bb1,
// CHECK: [[LT:_.+]] = Lt(copy _3, copy [[LEN]]);
// CHECK: assert(move [[LT]], "index out of bounds{{.+}}", move [[LEN]], copy _3) -> [success: bb1,
// CHECK: bb1:
// CHECK: _0 = &(*_1)[_2];
// CHECK: _6 = &(*_1)[_3];
// CHECK: _0 = &(*_6);
&slice[index]
}
@ -40,12 +46,15 @@ fn index_mut_slice(slice: &mut [i32], index: usize) -> &i32 {
// This cannot `copy _1` in the *built* MIR, only in the *runtime* MIR.
// CHECK: bb0:
// CHECK: [[LEN:_.+]] = PtrMetadata(copy _1);
// CHECK: [[LT:_.+]] = Lt(copy _2, copy [[LEN]]);
// CHECK: assert(move [[LT]], "index out of bounds{{.+}}", move [[LEN]], copy _2) -> [success: bb1,
// CHECK: _3 = copy _2;
// CHECK: _4 = &raw const (fake) (*_1);
// CHECK: [[LEN:_.+]] = PtrMetadata(move _4);
// CHECK: [[LT:_.+]] = Lt(copy _3, copy [[LEN]]);
// CHECK: assert(move [[LT]], "index out of bounds{{.+}}", move [[LEN]], copy _3) -> [success: bb1,
// CHECK: bb1:
// CHECK: _0 = &(*_1)[_2];
// CHECK: _7 = &(*_1)[_3];
// CHECK: _0 = &(*_7);
&slice[index]
}
@ -54,13 +63,15 @@ struct WithSliceTail(f64, [i32]);
// EMIT_MIR index_array_and_slice.index_custom.built.after.mir
fn index_custom(custom: &WithSliceTail, index: usize) -> &i32 {
// CHECK: bb0:
// CHECK: _3 = copy _2;
// CHECK: [[PTR:_.+]] = &raw const (fake) ((*_1).1: [i32]);
// CHECK: [[LEN:_.+]] = PtrMetadata(copy [[PTR]]);
// CHECK: [[LT:_.+]] = Lt(copy _2, copy [[LEN]]);
// CHECK: assert(move [[LT]], "index out of bounds{{.+}}", move [[LEN]], copy _2) -> [success: bb1,
// CHECK: [[LEN:_.+]] = PtrMetadata(move [[PTR]]);
// CHECK: [[LT:_.+]] = Lt(copy _3, copy [[LEN]]);
// CHECK: assert(move [[LT]], "index out of bounds{{.+}}", move [[LEN]], copy _3) -> [success: bb1,
// CHECK: bb1:
// CHECK: _0 = &((*_1).1: [i32])[_2];
// CHECK: _7 = &((*_1).1: [i32])[_3];
// CHECK: _0 = &(*_7);
&custom.1[index]
}

4
tests/mir-opt/building/issue_101867.main.built.after.mir
View file

@ -1,8 +1,8 @@
// MIR for `main` after built
| User Type Annotations
| 0: user_ty: Canonical { value: Ty(std::option::Option<u8>), max_universe: U0, variables: [] }, span: $DIR/issue_101867.rs:4:12: 4:22, inferred_ty: std::option::Option<u8>
| 1: user_ty: Canonical { value: Ty(std::option::Option<u8>), max_universe: U0, variables: [] }, span: $DIR/issue_101867.rs:4:12: 4:22, inferred_ty: std::option::Option<u8>
| 0: user_ty: Canonical { value: Ty(std::option::Option<u8>), max_universe: U0, variables: [] }, span: $DIR/issue_101867.rs:5:12: 5:22, inferred_ty: std::option::Option<u8>
| 1: user_ty: Canonical { value: Ty(std::option::Option<u8>), max_universe: U0, variables: [] }, span: $DIR/issue_101867.rs:5:12: 5:22, inferred_ty: std::option::Option<u8>
|
fn main() -> () {
let mut _0: ();

1
tests/mir-opt/building/issue_101867.rs
View file

@ -1,3 +1,4 @@
//@ compile-flags: -Zmir-opt-level=0
// skip-filecheck
// EMIT_MIR issue_101867.main.built.after.mir
fn main() {

1
tests/mir-opt/building/issue_110508.rs
View file

@ -1,3 +1,4 @@
//@ compile-flags: -Zmir-opt-level=0
// skip-filecheck
// EMIT_MIR issue_110508.{impl#0}-BAR.built.after.mir
// EMIT_MIR issue_110508.{impl#0}-SELF_BAR.built.after.mir

4
tests/mir-opt/building/issue_110508.{impl#0}-BAR.built.after.mir
View file

@ -1,6 +1,6 @@
// MIR for `<impl at $DIR/issue_110508.rs:9:1: 9:9>::BAR` after built
// MIR for `<impl at $DIR/issue_110508.rs:10:1: 10:9>::BAR` after built
const <impl at $DIR/issue_110508.rs:9:1: 9:9>::BAR: Foo = {
const <impl at $DIR/issue_110508.rs:10:1: 10:9>::BAR: Foo = {
let mut _0: Foo;
let mut _1: ();

4
tests/mir-opt/building/issue_110508.{impl#0}-SELF_BAR.built.after.mir
View file

@ -1,6 +1,6 @@
// MIR for `<impl at $DIR/issue_110508.rs:9:1: 9:9>::SELF_BAR` after built
// MIR for `<impl at $DIR/issue_110508.rs:10:1: 10:9>::SELF_BAR` after built
const <impl at $DIR/issue_110508.rs:9:1: 9:9>::SELF_BAR: Foo = {
const <impl at $DIR/issue_110508.rs:10:1: 10:9>::SELF_BAR: Foo = {
let mut _0: Foo;
let mut _1: ();

1
tests/mir-opt/building/issue_49232.rs
View file

@ -1,3 +1,4 @@
//@ compile-flags: -Zmir-opt-level=0
// skip-filecheck
// We must mark a variable whose initialization fails due to an
// abort statement as StorageDead.

2
tests/mir-opt/building/logical_or_in_conditional.rs
View file

@ -1,5 +1,5 @@
// skip-filecheck
//@ compile-flags: -Z validate-mir
//@ compile-flags: -Zmir-opt-level=0 -Z validate-mir
//@ edition: 2024
struct Droppy(u8);
impl Drop for Droppy {

8
tests/mir-opt/building/receiver_ptr_mutability.main.built.after.mir
View file

@ -1,10 +1,10 @@
// MIR for `main` after built
| User Type Annotations
| 0: user_ty: Canonical { value: Ty(*mut Test), max_universe: U0, variables: [] }, span: $DIR/receiver_ptr_mutability.rs:15:14: 15:23, inferred_ty: *mut Test
| 1: user_ty: Canonical { value: Ty(*mut Test), max_universe: U0, variables: [] }, span: $DIR/receiver_ptr_mutability.rs:15:14: 15:23, inferred_ty: *mut Test
| 2: user_ty: Canonical { value: Ty(&&&&*mut Test), max_universe: U0, variables: [Region(U0), Region(U0), Region(U0), Region(U0)] }, span: $DIR/receiver_ptr_mutability.rs:19:18: 19:31, inferred_ty: &&&&*mut Test
| 3: user_ty: Canonical { value: Ty(&&&&*mut Test), max_universe: U0, variables: [Region(U0), Region(U0), Region(U0), Region(U0)] }, span: $DIR/receiver_ptr_mutability.rs:19:18: 19:31, inferred_ty: &&&&*mut Test
| 0: user_ty: Canonical { value: Ty(*mut Test), max_universe: U0, variables: [] }, span: $DIR/receiver_ptr_mutability.rs:16:14: 16:23, inferred_ty: *mut Test
| 1: user_ty: Canonical { value: Ty(*mut Test), max_universe: U0, variables: [] }, span: $DIR/receiver_ptr_mutability.rs:16:14: 16:23, inferred_ty: *mut Test
| 2: user_ty: Canonical { value: Ty(&&&&*mut Test), max_universe: U0, variables: [Region(U0), Region(U0), Region(U0), Region(U0)] }, span: $DIR/receiver_ptr_mutability.rs:20:18: 20:31, inferred_ty: &&&&*mut Test
| 3: user_ty: Canonical { value: Ty(&&&&*mut Test), max_universe: U0, variables: [Region(U0), Region(U0), Region(U0), Region(U0)] }, span: $DIR/receiver_ptr_mutability.rs:20:18: 20:31, inferred_ty: &&&&*mut Test
|
fn main() -> () {
let mut _0: ();

1
tests/mir-opt/building/receiver_ptr_mutability.rs
View file

@ -1,3 +1,4 @@
//@ compile-flags: -Zmir-opt-level=0
// skip-filecheck
// EMIT_MIR receiver_ptr_mutability.main.built.after.mir

2
tests/mir-opt/building/shifts.rs
View file

@ -1,5 +1,5 @@
// skip-filecheck
//@ compile-flags: -C debug-assertions=yes
//@ compile-flags: -Zmir-opt-level=0 -C debug-assertions=yes
// EMIT_MIR shifts.shift_signed.built.after.mir
fn shift_signed(small: i8, big: u128, a: i8, b: i32, c: i128) -> ([i8; 3], [u128; 3]) {

1
tests/mir-opt/building/storage_live_dead_in_statics.rs
View file

@ -1,3 +1,4 @@
//@ compile-flags: -Zmir-opt-level=0
// skip-filecheck
// Check that when we compile the static `XXX` into MIR, we do not
// generate `StorageStart` or `StorageEnd` statements.

1
tests/mir-opt/building/uniform_array_move_out.rs
View file

@ -1,3 +1,4 @@
//@ compile-flags: -Zmir-opt-level=0
// skip-filecheck
#![feature(liballoc_internals, rustc_attrs)]

4
tests/mir-opt/building/user_type_annotations.let_else.built.after.mir
View file

@ -1,8 +1,8 @@
// MIR for `let_else` after built
| User Type Annotations
| 0: user_ty: Canonical { value: Ty((u32, u64, &'static char)), max_universe: U0, variables: [] }, span: $DIR/user_type_annotations.rs:35:20: 35:45, inferred_ty: (u32, u64, &char)
| 1: user_ty: Canonical { value: Ty((u32, u64, &'static char)), max_universe: U0, variables: [] }, span: $DIR/user_type_annotations.rs:35:20: 35:45, inferred_ty: (u32, u64, &char)
| 0: user_ty: Canonical { value: Ty((u32, u64, &'static char)), max_universe: U0, variables: [] }, span: $DIR/user_type_annotations.rs:36:20: 36:45, inferred_ty: (u32, u64, &char)
| 1: user_ty: Canonical { value: Ty((u32, u64, &'static char)), max_universe: U0, variables: [] }, span: $DIR/user_type_annotations.rs:36:20: 36:45, inferred_ty: (u32, u64, &char)
|
fn let_else() -> () {
let mut _0: ();

4
tests/mir-opt/building/user_type_annotations.let_else_bindless.built.after.mir
View file

@ -1,8 +1,8 @@
// MIR for `let_else_bindless` after built
| User Type Annotations
| 0: user_ty: Canonical { value: Ty((u32, u64, &'static char)), max_universe: U0, variables: [] }, span: $DIR/user_type_annotations.rs:40:20: 40:45, inferred_ty: (u32, u64, &char)
| 1: user_ty: Canonical { value: Ty((u32, u64, &'static char)), max_universe: U0, variables: [] }, span: $DIR/user_type_annotations.rs:40:20: 40:45, inferred_ty: (u32, u64, &char)
| 0: user_ty: Canonical { value: Ty((u32, u64, &'static char)), max_universe: U0, variables: [] }, span: $DIR/user_type_annotations.rs:41:20: 41:45, inferred_ty: (u32, u64, &char)
| 1: user_ty: Canonical { value: Ty((u32, u64, &'static char)), max_universe: U0, variables: [] }, span: $DIR/user_type_annotations.rs:41:20: 41:45, inferred_ty: (u32, u64, &char)
|
fn let_else_bindless() -> () {
let mut _0: ();

4
tests/mir-opt/building/user_type_annotations.let_init.built.after.mir
View file

@ -1,8 +1,8 @@
// MIR for `let_init` after built
| User Type Annotations
| 0: user_ty: Canonical { value: Ty((u32, u64, &'static char)), max_universe: U0, variables: [] }, span: $DIR/user_type_annotations.rs:25:20: 25:45, inferred_ty: (u32, u64, &char)
| 1: user_ty: Canonical { value: Ty((u32, u64, &'static char)), max_universe: U0, variables: [] }, span: $DIR/user_type_annotations.rs:25:20: 25:45, inferred_ty: (u32, u64, &char)
| 0: user_ty: Canonical { value: Ty((u32, u64, &'static char)), max_universe: U0, variables: [] }, span: $DIR/user_type_annotations.rs:26:20: 26:45, inferred_ty: (u32, u64, &char)
| 1: user_ty: Canonical { value: Ty((u32, u64, &'static char)), max_universe: U0, variables: [] }, span: $DIR/user_type_annotations.rs:26:20: 26:45, inferred_ty: (u32, u64, &char)
|
fn let_init() -> () {
let mut _0: ();

4
tests/mir-opt/building/user_type_annotations.let_init_bindless.built.after.mir
View file

@ -1,8 +1,8 @@
// MIR for `let_init_bindless` after built
| User Type Annotations
| 0: user_ty: Canonical { value: Ty((u32, u64, &'static char)), max_universe: U0, variables: [] }, span: $DIR/user_type_annotations.rs:30:20: 30:45, inferred_ty: (u32, u64, &char)
| 1: user_ty: Canonical { value: Ty((u32, u64, &'static char)), max_universe: U0, variables: [] }, span: $DIR/user_type_annotations.rs:30:20: 30:45, inferred_ty: (u32, u64, &char)
| 0: user_ty: Canonical { value: Ty((u32, u64, &'static char)), max_universe: U0, variables: [] }, span: $DIR/user_type_annotations.rs:31:20: 31:45, inferred_ty: (u32, u64, &char)
| 1: user_ty: Canonical { value: Ty((u32, u64, &'static char)), max_universe: U0, variables: [] }, span: $DIR/user_type_annotations.rs:31:20: 31:45, inferred_ty: (u32, u64, &char)
|
fn let_init_bindless() -> () {
let mut _0: ();

2
tests/mir-opt/building/user_type_annotations.let_uninit.built.after.mir
View file

@ -1,7 +1,7 @@
// MIR for `let_uninit` after built
| User Type Annotations
| 0: user_ty: Canonical { value: Ty((u32, u64, &'static char)), max_universe: U0, variables: [] }, span: $DIR/user_type_annotations.rs:15:20: 15:45, inferred_ty: (u32, u64, &char)
| 0: user_ty: Canonical { value: Ty((u32, u64, &'static char)), max_universe: U0, variables: [] }, span: $DIR/user_type_annotations.rs:16:20: 16:45, inferred_ty: (u32, u64, &char)
|
fn let_uninit() -> () {
let mut _0: ();

2
tests/mir-opt/building/user_type_annotations.let_uninit_bindless.built.after.mir
View file

@ -1,7 +1,7 @@
// MIR for `let_uninit_bindless` after built
| User Type Annotations
| 0: user_ty: Canonical { value: Ty((u32, u64, &'static char)), max_universe: U0, variables: [] }, span: $DIR/user_type_annotations.rs:20:20: 20:45, inferred_ty: (u32, u64, &char)
| 0: user_ty: Canonical { value: Ty((u32, u64, &'static char)), max_universe: U0, variables: [] }, span: $DIR/user_type_annotations.rs:21:20: 21:45, inferred_ty: (u32, u64, &char)
|
fn let_uninit_bindless() -> () {
let mut _0: ();

4
tests/mir-opt/building/user_type_annotations.match_assoc_const.built.after.mir
View file

@ -1,8 +1,8 @@
// MIR for `match_assoc_const` after built
| User Type Annotations
| 0: user_ty: Canonical { value: TypeOf(DefId(0:11 ~ user_type_annotations[ee8e]::MyTrait::FOO), UserArgs { args: [MyStruct, 'static], user_self_ty: None }), max_universe: U0, variables: [] }, span: $DIR/user_type_annotations.rs:54:9: 54:44, inferred_ty: u32
| 1: user_ty: Canonical { value: TypeOf(DefId(0:11 ~ user_type_annotations[ee8e]::MyTrait::FOO), UserArgs { args: [MyStruct, 'static], user_self_ty: None }), max_universe: U0, variables: [] }, span: $DIR/user_type_annotations.rs:54:9: 54:44, inferred_ty: u32
| 0: user_ty: Canonical { value: TypeOf(DefId(0:11 ~ user_type_annotations[ee8e]::MyTrait::FOO), UserArgs { args: [MyStruct, 'static], user_self_ty: None }), max_universe: U0, variables: [] }, span: $DIR/user_type_annotations.rs:55:9: 55:44, inferred_ty: u32
| 1: user_ty: Canonical { value: TypeOf(DefId(0:11 ~ user_type_annotations[ee8e]::MyTrait::FOO), UserArgs { args: [MyStruct, 'static], user_self_ty: None }), max_universe: U0, variables: [] }, span: $DIR/user_type_annotations.rs:55:9: 55:44, inferred_ty: u32
|
fn match_assoc_const() -> () {
let mut _0: ();

8
tests/mir-opt/building/user_type_annotations.match_assoc_const_range.built.after.mir
View file

@ -1,10 +1,10 @@
// MIR for `match_assoc_const_range` after built
| User Type Annotations
| 0: user_ty: Canonical { value: TypeOf(DefId(0:11 ~ user_type_annotations[ee8e]::MyTrait::FOO), UserArgs { args: [MyStruct, 'static], user_self_ty: None }), max_universe: U0, variables: [] }, span: $DIR/user_type_annotations.rs:62:11: 62:46, inferred_ty: u32
| 1: user_ty: Canonical { value: TypeOf(DefId(0:11 ~ user_type_annotations[ee8e]::MyTrait::FOO), UserArgs { args: [MyStruct, 'static], user_self_ty: None }), max_universe: U0, variables: [] }, span: $DIR/user_type_annotations.rs:62:11: 62:46, inferred_ty: u32
| 2: user_ty: Canonical { value: TypeOf(DefId(0:11 ~ user_type_annotations[ee8e]::MyTrait::FOO), UserArgs { args: [MyStruct, 'static], user_self_ty: None }), max_universe: U0, variables: [] }, span: $DIR/user_type_annotations.rs:63:9: 63:44, inferred_ty: u32
| 3: user_ty: Canonical { value: TypeOf(DefId(0:11 ~ user_type_annotations[ee8e]::MyTrait::FOO), UserArgs { args: [MyStruct, 'static], user_self_ty: None }), max_universe: U0, variables: [] }, span: $DIR/user_type_annotations.rs:63:9: 63:44, inferred_ty: u32
| 0: user_ty: Canonical { value: TypeOf(DefId(0:11 ~ user_type_annotations[ee8e]::MyTrait::FOO), UserArgs { args: [MyStruct, 'static], user_self_ty: None }), max_universe: U0, variables: [] }, span: $DIR/user_type_annotations.rs:63:11: 63:46, inferred_ty: u32
| 1: user_ty: Canonical { value: TypeOf(DefId(0:11 ~ user_type_annotations[ee8e]::MyTrait::FOO), UserArgs { args: [MyStruct, 'static], user_self_ty: None }), max_universe: U0, variables: [] }, span: $DIR/user_type_annotations.rs:63:11: 63:46, inferred_ty: u32
| 2: user_ty: Canonical { value: TypeOf(DefId(0:11 ~ user_type_annotations[ee8e]::MyTrait::FOO), UserArgs { args: [MyStruct, 'static], user_self_ty: None }), max_universe: U0, variables: [] }, span: $DIR/user_type_annotations.rs:64:9: 64:44, inferred_ty: u32
| 3: user_ty: Canonical { value: TypeOf(DefId(0:11 ~ user_type_annotations[ee8e]::MyTrait::FOO), UserArgs { args: [MyStruct, 'static], user_self_ty: None }), max_universe: U0, variables: [] }, span: $DIR/user_type_annotations.rs:64:9: 64:44, inferred_ty: u32
|
fn match_assoc_const_range() -> () {
let mut _0: ();

1
tests/mir-opt/building/user_type_annotations.rs
View file

@ -1,3 +1,4 @@
//@ compile-flags: -Zmir-opt-level=0
//@ edition: 2024
// skip-filecheck