diff --git a/compiler/rustc_infer/src/infer/error_reporting/mod.rs b/compiler/rustc_infer/src/infer/error_reporting/mod.rs index e4be435fded30..26d071a01397a 100644 --- a/compiler/rustc_infer/src/infer/error_reporting/mod.rs +++ b/compiler/rustc_infer/src/infer/error_reporting/mod.rs @@ -2444,18 +2444,22 @@ impl<'tcx> TypeErrCtxt<'_, 'tcx> { let suggestion = if has_lifetimes { format!(" + {lt_name}") } else { format!(": {lt_name}") }; suggs.push((sp, suggestion)) - } else { - let generics = self.tcx.hir().get_generics(suggestion_scope).unwrap(); + } else if let Some(generics) = self.tcx.hir().get_generics(suggestion_scope) { let pred = format!("{bound_kind}: {lt_name}"); - let suggestion = format!("{} {}", generics.add_where_or_trailing_comma(), pred,); + let suggestion = format!("{} {}", generics.add_where_or_trailing_comma(), pred); suggs.push((generics.tail_span_for_predicate_suggestion(), suggestion)) + } else { + let consider = format!("{msg} `{bound_kind}: {sub}`..."); + err.help(consider); } - err.multipart_suggestion_verbose( - format!("{msg}"), - suggs, - Applicability::MaybeIncorrect, // Issue #41966 - ); + if !suggs.is_empty() { + err.multipart_suggestion_verbose( + format!("{msg}"), + suggs, + Applicability::MaybeIncorrect, // Issue #41966 + ); + } } err diff --git a/compiler/rustc_middle/src/infer/canonical.rs b/compiler/rustc_middle/src/infer/canonical.rs index 0b5426c3bb1a1..64b63f4c5eb4f 100644 --- a/compiler/rustc_middle/src/infer/canonical.rs +++ b/compiler/rustc_middle/src/infer/canonical.rs @@ -222,7 +222,7 @@ pub enum CanonicalVarKind<'tcx> { Effect, /// A "placeholder" that represents "any const". - PlaceholderConst(ty::PlaceholderConst<'tcx>, Ty<'tcx>), + PlaceholderConst(ty::PlaceholderConst, Ty<'tcx>), } impl<'tcx> CanonicalVarKind<'tcx> { diff --git a/compiler/rustc_middle/src/ty/consts.rs b/compiler/rustc_middle/src/ty/consts.rs index cfacccd26799e..af5ffc20d489c 100644 --- a/compiler/rustc_middle/src/ty/consts.rs +++ b/compiler/rustc_middle/src/ty/consts.rs @@ -84,7 +84,7 @@ impl<'tcx> Const<'tcx> { #[inline] pub fn new_placeholder( tcx: TyCtxt<'tcx>, - placeholder: ty::PlaceholderConst<'tcx>, + placeholder: ty::PlaceholderConst, ty: Ty<'tcx>, ) -> Const<'tcx> { Const::new(tcx, ty::ConstKind::Placeholder(placeholder), ty) diff --git a/compiler/rustc_middle/src/ty/context.rs b/compiler/rustc_middle/src/ty/context.rs index 68812bba42fdd..551c4a15dd0d9 100644 --- a/compiler/rustc_middle/src/ty/context.rs +++ b/compiler/rustc_middle/src/ty/context.rs @@ -106,7 +106,7 @@ impl<'tcx> Interner for TyCtxt<'tcx> { type Const = ty::Const<'tcx>; type InferConst = ty::InferConst; type AliasConst = ty::UnevaluatedConst<'tcx>; - type PlaceholderConst = ty::PlaceholderConst<'tcx>; + type PlaceholderConst = ty::PlaceholderConst; type ParamConst = ty::ParamConst; type BoundConst = ty::BoundVar; type ValueConst = ty::ValTree<'tcx>; diff --git a/compiler/rustc_middle/src/ty/mod.rs b/compiler/rustc_middle/src/ty/mod.rs index 739d4fa886ec3..f6ef1783aa410 100644 --- a/compiler/rustc_middle/src/ty/mod.rs +++ b/compiler/rustc_middle/src/ty/mod.rs @@ -1527,7 +1527,7 @@ pub struct BoundConst<'tcx> { pub ty: Ty<'tcx>, } -pub type PlaceholderConst<'tcx> = Placeholder; +pub type PlaceholderConst = Placeholder; /// When type checking, we use the `ParamEnv` to track /// details about the set of where-clauses that are in scope at this diff --git a/compiler/rustc_trait_selection/src/traits/project.rs b/compiler/rustc_trait_selection/src/traits/project.rs index e4f7592c409e1..471d10dbdbd82 100644 --- a/compiler/rustc_trait_selection/src/traits/project.rs +++ b/compiler/rustc_trait_selection/src/traits/project.rs @@ -783,7 +783,7 @@ pub struct BoundVarReplacer<'me, 'tcx> { // the `var` (but we *could* bring that into scope if we were to track them as we pass them). mapped_regions: BTreeMap, mapped_types: BTreeMap, - mapped_consts: BTreeMap, ty::BoundVar>, + mapped_consts: BTreeMap, // The current depth relative to *this* folding, *not* the entire normalization. In other words, // the depth of binders we've passed here. current_index: ty::DebruijnIndex, @@ -843,11 +843,11 @@ impl<'me, 'tcx> BoundVarReplacer<'me, 'tcx> { T, BTreeMap, BTreeMap, - BTreeMap, ty::BoundVar>, + BTreeMap, ) { let mapped_regions: BTreeMap = BTreeMap::new(); let mapped_types: BTreeMap = BTreeMap::new(); - let mapped_consts: BTreeMap, ty::BoundVar> = BTreeMap::new(); + let mapped_consts: BTreeMap = BTreeMap::new(); let mut replacer = BoundVarReplacer { infcx, @@ -966,7 +966,7 @@ pub struct PlaceholderReplacer<'me, 'tcx> { infcx: &'me InferCtxt<'tcx>, mapped_regions: BTreeMap, mapped_types: BTreeMap, - mapped_consts: BTreeMap, ty::BoundVar>, + mapped_consts: BTreeMap, universe_indices: &'me [Option], current_index: ty::DebruijnIndex, } @@ -976,7 +976,7 @@ impl<'me, 'tcx> PlaceholderReplacer<'me, 'tcx> { infcx: &'me InferCtxt<'tcx>, mapped_regions: BTreeMap, mapped_types: BTreeMap, - mapped_consts: BTreeMap, ty::BoundVar>, + mapped_consts: BTreeMap, universe_indices: &'me [Option], value: T, ) -> T { diff --git a/library/core/src/convert/mod.rs b/library/core/src/convert/mod.rs index 89125b7955e02..8c01b0973d65a 100644 --- a/library/core/src/convert/mod.rs +++ b/library/core/src/convert/mod.rs @@ -138,7 +138,7 @@ pub const fn identity(x: T) -> T { /// /// [dereferenceable types]: core::ops::Deref /// [pointed-to value]: core::ops::Deref::Target -/// ['`Deref` coercion']: core::ops::Deref#more-on-deref-coercion +/// ['`Deref` coercion']: core::ops::Deref#deref-coercion /// /// ``` /// let x = Box::new(5i32); @@ -244,7 +244,7 @@ pub trait AsRef { /// /// [mutably dereferenceable types]: core::ops::DerefMut /// [pointed-to value]: core::ops::Deref::Target -/// ['`Deref` coercion']: core::ops::DerefMut#more-on-deref-coercion +/// ['`Deref` coercion']: core::ops::DerefMut#mutable-deref-coercion /// /// ``` /// let mut x = Box::new(5i32); diff --git a/library/core/src/ops/deref.rs b/library/core/src/ops/deref.rs index d6f2f5ca3662b..99adbb91599ac 100644 --- a/library/core/src/ops/deref.rs +++ b/library/core/src/ops/deref.rs @@ -3,40 +3,107 @@ /// In addition to being used for explicit dereferencing operations with the /// (unary) `*` operator in immutable contexts, `Deref` is also used implicitly /// by the compiler in many circumstances. This mechanism is called -/// ['`Deref` coercion'][more]. In mutable contexts, [`DerefMut`] is used. +/// ["`Deref` coercion"][coercion]. In mutable contexts, [`DerefMut`] is used and +/// mutable deref coercion similarly occurs. /// -/// Implementing `Deref` for smart pointers makes accessing the data behind them -/// convenient, which is why they implement `Deref`. On the other hand, the -/// rules regarding `Deref` and [`DerefMut`] were designed specifically to -/// accommodate smart pointers. Because of this, **`Deref` should only be -/// implemented for smart pointers** to avoid confusion. +/// **Warning:** Deref coercion is a powerful language feature which has +/// far-reaching implications for every type that implements `Deref`. The +/// compiler will silently insert calls to `Deref::deref`. For this reason, one +/// should be careful about implementing `Deref` and only do so when deref +/// coercion is desirable. See [below][implementing] for advice on when this is +/// typically desirable or undesirable. /// -/// For similar reasons, **this trait should never fail**. Failure during -/// dereferencing can be extremely confusing when `Deref` is invoked implicitly. +/// Types that implement `Deref` or `DerefMut` are often called "smart +/// pointers" and the mechanism of deref coercion has been specifically designed +/// to facilitate the pointer-like behaviour that name suggests. Often, the +/// purpose of a "smart pointer" type is to change the ownership semantics +/// of a contained value (for example, [`Rc`][rc] or [`Cow`][cow]) or the +/// storage semantics of a contained value (for example, [`Box`][box]). /// -/// Violating these requirements is a logic error. The behavior resulting from a logic error is not -/// specified, but users of the trait must ensure that such logic errors do *not* result in -/// undefined behavior. This means that `unsafe` code **must not** rely on the correctness of this -/// method. +/// # Deref coercion /// -/// # More on `Deref` coercion +/// If `T` implements `Deref`, and `v` is a value of type `T`, then: /// -/// If `T` implements `Deref`, and `x` is a value of type `T`, then: -/// -/// * In immutable contexts, `*x` (where `T` is neither a reference nor a raw pointer) -/// is equivalent to `*Deref::deref(&x)`. +/// * In immutable contexts, `*v` (where `T` is neither a reference nor a raw +/// pointer) is equivalent to `*Deref::deref(&v)`. /// * Values of type `&T` are coerced to values of type `&U` -/// * `T` implicitly implements all the (immutable) methods of the type `U`. +/// * `T` implicitly implements all the methods of the type `U` which take the +/// `&self` receiver. /// /// For more details, visit [the chapter in *The Rust Programming Language*][book] /// as well as the reference sections on [the dereference operator][ref-deref-op], -/// [method resolution] and [type coercions]. +/// [method resolution], and [type coercions]. +/// +/// # When to implement `Deref` or `DerefMut` +/// +/// The same advice applies to both deref traits. In general, deref traits +/// **should** be implemented if: +/// +/// 1. a value of the type transparently behaves like a value of the target +/// type; +/// 1. the implementation of the deref function is cheap; and +/// 1. users of the type will not be surprised by any deref coercion behaviour. +/// +/// In general, deref traits **should not** be implemented if: +/// +/// 1. the deref implementations could fail unexpectedly; or +/// 1. the type has methods that are likely to collide with methods on the +/// target type; or +/// 1. committing to deref coercion as part of the public API is not desirable. +/// +/// Note that there's a large difference between implementing deref traits +/// generically over many target types, and doing so only for specific target +/// types. +/// +/// Generic implementations, such as for [`Box`][box] (which is generic over +/// every type and dereferences to `T`) should be careful to provide few or no +/// methods, since the target type is unknown and therefore every method could +/// collide with one on the target type, causing confusion for users. +/// `impl Box` has no methods (though several associated functions), +/// partly for this reason. +/// +/// Specific implementations, such as for [`String`][string] (whose `Deref` +/// implementation has `Target = str`) can have many methods, since avoiding +/// collision is much easier. `String` and `str` both have many methods, and +/// `String` additionally behaves as if it has every method of `str` because of +/// deref coercion. The implementing type may also be generic while the +/// implementation is still specific in this sense; for example, [`Vec`][vec] +/// dereferences to `[T]`, so methods of `T` are not applicable. +/// +/// Consider also that deref coericion means that deref traits are a much larger +/// part of a type's public API than any other trait as it is implicitly called +/// by the compiler. Therefore, it is advisable to consider whether this is +/// something you are comfortable supporting as a public API. +/// +/// The [`AsRef`] and [`Borrow`][core::borrow::Borrow] traits have very similar +/// signatures to `Deref`. It may be desirable to implement either or both of +/// these, whether in addition to or rather than deref traits. See their +/// documentation for details. +/// +/// # Fallibility +/// +/// **This trait's method should never unexpectedly fail**. Deref coercion means +/// the compiler will often insert calls to `Deref::deref` implicitly. Failure +/// during dereferencing can be extremely confusing when `Deref` is invoked +/// implicitly. In the majority of uses it should be infallible, though it may +/// be acceptable to panic if the type is misused through programmer error, for +/// example. +/// +/// However, infallibility is not enforced and therefore not guaranteed. +/// As such, `unsafe` code should not rely on infallibility in general for +/// soundness. /// /// [book]: ../../book/ch15-02-deref.html -/// [more]: #more-on-deref-coercion +/// [coercion]: #deref-coercion +/// [implementing]: #when-to-implement-deref-or-derefmut /// [ref-deref-op]: ../../reference/expressions/operator-expr.html#the-dereference-operator /// [method resolution]: ../../reference/expressions/method-call-expr.html /// [type coercions]: ../../reference/type-coercions.html +/// [box]: ../../alloc/boxed/struct.Box.html +/// [string]: ../../alloc/string/struct.String.html +/// [vec]: ../../alloc/vec/struct.Vec.html +/// [rc]: ../../alloc/rc/struct.Rc.html +/// [cow]: ../../alloc/borrow/enum.Cow.html /// /// # Examples /// @@ -107,30 +174,29 @@ impl Deref for &mut T { /// In addition to being used for explicit dereferencing operations with the /// (unary) `*` operator in mutable contexts, `DerefMut` is also used implicitly /// by the compiler in many circumstances. This mechanism is called -/// ['`Deref` coercion'][more]. In immutable contexts, [`Deref`] is used. -/// -/// Implementing `DerefMut` for smart pointers makes mutating the data behind -/// them convenient, which is why they implement `DerefMut`. On the other hand, -/// the rules regarding [`Deref`] and `DerefMut` were designed specifically to -/// accommodate smart pointers. Because of this, **`DerefMut` should only be -/// implemented for smart pointers** to avoid confusion. +/// ["mutable deref coercion"][coercion]. In immutable contexts, [`Deref`] is used. /// -/// For similar reasons, **this trait should never fail**. Failure during -/// dereferencing can be extremely confusing when `DerefMut` is invoked -/// implicitly. +/// **Warning:** Deref coercion is a powerful language feature which has +/// far-reaching implications for every type that implements `DerefMut`. The +/// compiler will silently insert calls to `DerefMut::deref_mut`. For this +/// reason, one should be careful about implementing `DerefMut` and only do so +/// when mutable deref coercion is desirable. See [the `Deref` docs][implementing] +/// for advice on when this is typically desirable or undesirable. /// -/// Violating these requirements is a logic error. The behavior resulting from a logic error is not -/// specified, but users of the trait must ensure that such logic errors do *not* result in -/// undefined behavior. This means that `unsafe` code **must not** rely on the correctness of this -/// method. +/// Types that implement `DerefMut` or `Deref` are often called "smart +/// pointers" and the mechanism of deref coercion has been specifically designed +/// to facilitate the pointer-like behaviour that name suggests. Often, the +/// purpose of a "smart pointer" type is to change the ownership semantics +/// of a contained value (for example, [`Rc`][rc] or [`Cow`][cow]) or the +/// storage semantics of a contained value (for example, [`Box`][box]). /// -/// # More on `Deref` coercion +/// # Mutable deref coercion /// -/// If `T` implements `DerefMut`, and `x` is a value of type `T`, +/// If `T` implements `DerefMut`, and `v` is a value of type `T`, /// then: /// -/// * In mutable contexts, `*x` (where `T` is neither a reference nor a raw pointer) -/// is equivalent to `*DerefMut::deref_mut(&mut x)`. +/// * In mutable contexts, `*v` (where `T` is neither a reference nor a raw pointer) +/// is equivalent to `*DerefMut::deref_mut(&mut v)`. /// * Values of type `&mut T` are coerced to values of type `&mut U` /// * `T` implicitly implements all the (mutable) methods of the type `U`. /// @@ -138,11 +204,29 @@ impl Deref for &mut T { /// as well as the reference sections on [the dereference operator][ref-deref-op], /// [method resolution] and [type coercions]. /// +/// # Fallibility +/// +/// **This trait's method should never unexpectedly fail**. Deref coercion means +/// the compiler will often insert calls to `DerefMut::deref_mut` implicitly. +/// Failure during dereferencing can be extremely confusing when `DerefMut` is +/// invoked implicitly. In the majority of uses it should be infallible, though +/// it may be acceptable to panic if the type is misused through programmer +/// error, for example. +/// +/// However, infallibility is not enforced and therefore not guaranteed. +/// As such, `unsafe` code should not rely on infallibility in general for +/// soundness. +/// /// [book]: ../../book/ch15-02-deref.html -/// [more]: #more-on-deref-coercion +/// [coercion]: #mutable-deref-coercion +/// [implementing]: Deref#when-to-implement-deref-or-derefmut /// [ref-deref-op]: ../../reference/expressions/operator-expr.html#the-dereference-operator /// [method resolution]: ../../reference/expressions/method-call-expr.html /// [type coercions]: ../../reference/type-coercions.html +/// [box]: ../../alloc/boxed/struct.Box.html +/// [string]: ../../alloc/string/struct.String.html +/// [rc]: ../../alloc/rc/struct.Rc.html +/// [cow]: ../../alloc/borrow/enum.Cow.html /// /// # Examples /// diff --git a/library/core/src/primitive_docs.rs b/library/core/src/primitive_docs.rs index f3695d16d7a6b..87e4921087492 100644 --- a/library/core/src/primitive_docs.rs +++ b/library/core/src/primitive_docs.rs @@ -283,7 +283,7 @@ mod prim_never {} /// `char` type. For technical reasons, there is additional, separate /// documentation in [the `std::char` module](char/index.html) as well. /// -/// # Validity +/// # Validity and Layout /// /// A `char` is a '[Unicode scalar value]', which is any '[Unicode code point]' /// other than a [surrogate code point]. This has a fixed numerical definition: @@ -291,7 +291,7 @@ mod prim_never {} /// Surrogate code points, used by UTF-16, are in the range 0xD800 to 0xDFFF. /// /// No `char` may be constructed, whether as a literal or at runtime, that is not a -/// Unicode scalar value: +/// Unicode scalar value. Violating this rule causes undefined behavior. /// /// ```compile_fail /// // Each of these is a compiler error @@ -308,9 +308,10 @@ mod prim_never {} /// let _ = unsafe { char::from_u32_unchecked(0x110000) }; /// ``` /// -/// USVs are also the exact set of values that may be encoded in UTF-8. Because -/// `char` values are USVs and `str` values are valid UTF-8, it is safe to store -/// any `char` in a `str` or read any character from a `str` as a `char`. +/// Unicode scalar values are also the exact set of values that may be encoded in UTF-8. Because +/// `char` values are Unicode scalar values and functions may assume [incoming `str` values are +/// valid UTF-8](primitive.str.html#invariant), it is safe to store any `char` in a `str` or read +/// any character from a `str` as a `char`. /// /// The gap in valid `char` values is understood by the compiler, so in the /// below example the two ranges are understood to cover the whole range of @@ -324,11 +325,17 @@ mod prim_never {} /// }; /// ``` /// -/// All USVs are valid `char` values, but not all of them represent a real -/// character. Many USVs are not currently assigned to a character, but may be -/// in the future ("reserved"); some will never be a character -/// ("noncharacters"); and some may be given different meanings by different -/// users ("private use"). +/// All Unicode scalar values are valid `char` values, but not all of them represent a real +/// character. Many Unicode scalar values are not currently assigned to a character, but may be in +/// the future ("reserved"); some will never be a character ("noncharacters"); and some may be given +/// different meanings by different users ("private use"). +/// +/// `char` is guaranteed to have the same size and alignment as `u32` on all +/// platforms. +/// ``` +/// use std::alloc::Layout; +/// assert_eq!(Layout::new::(), Layout::new::()); +/// ``` /// /// [Unicode code point]: https://www.unicode.org/glossary/#code_point /// [Unicode scalar value]: https://www.unicode.org/glossary/#unicode_scalar_value @@ -887,8 +894,6 @@ mod prim_slice {} /// type. It is usually seen in its borrowed form, `&str`. It is also the type /// of string literals, `&'static str`. /// -/// String slices are always valid UTF-8. -/// /// # Basic Usage /// /// String literals are string slices: @@ -942,6 +947,14 @@ mod prim_slice {} /// Note: This example shows the internals of `&str`. `unsafe` should not be /// used to get a string slice under normal circumstances. Use `as_str` /// instead. +/// +/// # Invariant +/// +/// Rust libraries may assume that string slices are always valid UTF-8. +/// +/// Constructing a non-UTF-8 string slice is not immediate undefined behavior, but any function +/// called on a string slice may assume that it is valid UTF-8, which means that a non-UTF-8 string +/// slice can lead to undefined behavior down the road. #[stable(feature = "rust1", since = "1.0.0")] mod prim_str {} diff --git a/tests/ui/impl-trait/in-trait/async-and-ret-ref.rs b/tests/ui/impl-trait/in-trait/async-and-ret-ref.rs new file mode 100644 index 0000000000000..af6ffe8339452 --- /dev/null +++ b/tests/ui/impl-trait/in-trait/async-and-ret-ref.rs @@ -0,0 +1,11 @@ +// edition:2021 +// https://github.com/rust-lang/rust/issues/117547 + +trait T {} + +trait MyTrait { + async fn foo() -> &'static impl T; + //~^ ERROR the associated type `::{opaque#0}` may not live long enough +} + +fn main() {} diff --git a/tests/ui/impl-trait/in-trait/async-and-ret-ref.stderr b/tests/ui/impl-trait/in-trait/async-and-ret-ref.stderr new file mode 100644 index 0000000000000..7c9028a8cd53c --- /dev/null +++ b/tests/ui/impl-trait/in-trait/async-and-ret-ref.stderr @@ -0,0 +1,14 @@ +error[E0310]: the associated type `::{opaque#0}` may not live long enough + --> $DIR/async-and-ret-ref.rs:7:5 + | +LL | async fn foo() -> &'static impl T; + | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ + | | + | the associated type `::{opaque#0}` must be valid for the static lifetime... + | ...so that the reference type `&'static impl T` does not outlive the data it points at + | + = help: consider adding an explicit lifetime bound `::{opaque#0}: 'static`... + +error: aborting due to previous error + +For more information about this error, try `rustc --explain E0310`. diff --git a/triagebot.toml b/triagebot.toml index b419727bcfdd6..b846edca98b39 100644 --- a/triagebot.toml +++ b/triagebot.toml @@ -415,7 +415,7 @@ Issue #{number} "{title}" has been added. """ [no-merges] -exclude_titles = ["Rollup of", "subtree update"] +exclude_titles = ["Rollup of", "subtree update", "Subtree update"] labels = ["has-merge-commits", "S-waiting-on-author"] [github-releases]