Currently it relies on having the right integer for every variant, and
if you add a variant you need to adjust the integers for all subsequent
variants, which is a pain.
This commit introduces a match guard formulation that takes advantage of
the enum-to-integer conversion to avoid specifying the integer for each
variant. And it does this via a macro to avoid lots of boilerplate.
The parser pushes a `TokenType` to `Parser::expected_token_types` on
every call to the various `check`/`eat` methods, and clears it on every
call to `bump`. Some of those `TokenType` values are full tokens that
require cloning and dropping. This is a *lot* of work for something
that is only used in error messages and it accounts for a significant
fraction of parsing execution time.
This commit overhauls `TokenType` so that `Parser::expected_token_types`
can be implemented as a bitset. This requires changing `TokenType` to a
C-style parameterless enum, and adding `TokenTypeSet` which uses a
`u128` for the bits. (The new `TokenType` has 105 variants.)
The new types `ExpTokenPair` and `ExpKeywordPair` are now arguments to
the `check`/`eat` methods. This is for maximum speed. The elements in
the pairs are always statically known; e.g. a
`token::BinOp(token::Star)` is always paired with a `TokenType::Star`.
So we now compute `TokenType`s in advance and pass them in to
`check`/`eat` rather than the current approach of constructing them on
insertion into `expected_token_types`.
Values of these pair types can be produced by the new `exp!` macro,
which is used at every `check`/`eat` call site. The macro is for
convenience, allowing any pair to be generated from a single identifier.
The ident/keyword filtering in `expected_one_of_not_found` is no longer
necessary. It was there to account for some sloppiness in
`TokenKind`/`TokenType` comparisons.
The existing `TokenType` is moved to a new file `token_type.rs`, and all
its new infrastructure is added to that file. There is more boilerplate
code than I would like, but I can't see how to make it shorter.
This is a naming convention used in a handful of spots in the parser for
delimiters. It confused me when I first saw it a long time ago, and I've
never liked it. A web search says "Bra-ket notation" exists in linear
algebra but the terminology has zero prior use in a programming context,
as far as I can tell.
This commit changes it to `open`/`close`, which is consistent with the
rest of the compiler.
The most significant is `check_keyword`: it now only pushes to
`expected_token_types` if the keyword check fails, which matches how all
the other `check` methods work.
The remainder are just tweaks to make these methods more consistent with
each other.
Use field init shorthand where possible
Field init shorthand allows writing initializers like `tcx: tcx` as
`tcx`. The compiler already uses it extensively. Fix the last few places
where it isn't yet used.
EDIT: this PR also updates `rustfmt.toml` to set
`use_field_init_shorthand = true`.
Overhaul keyword handling
The compiler's list of keywords has some problems.
- It contains several items that aren't keywords.
- The order isn't quite right in a couple of places.
- Some of the names of predicates relating to keywords are confusing.
- rustdoc and rustfmt have their own (incorrect) versions of the keyword list.
- `AllKeywords` is unnecessarily complex.
r? ```@jieyouxu```
`rustc_symbol` is the source of truth for keywords.
rustdoc has its own implicit definition of keywords, via the
`is_doc_keyword`. It (presumably) intends to include all keywords, but
it omits `yeet`.
rustfmt has its own explicit list of Rust keywords. It also (presumably)
intends to include all keywords, but it omits `await`, `builtin`, `gen`,
`macro_rules`, `raw`, `reuse`, `safe`, and `yeet`. Also, it does linear
searches through this list, which is inefficient.
This commit fixes all of the above problems by introducing a new
predicate `is_any_keyword` in rustc and using it in rustdoc and rustfmt.
It documents that it's not the right predicate in most cases.
`rustc_span::symbol` defines some things that are re-exported from
`rustc_span`, such as `Symbol` and `sym`. But it doesn't re-export some
closely related things such as `Ident` and `kw`. So you can do `use
rustc_span::{Symbol, sym}` but you have to do `use
rustc_span::symbol::{Ident, kw}`, which is inconsistent for no good
reason.
This commit re-exports `Ident`, `kw`, and `MacroRulesNormalizedIdent`,
and changes many `rustc_span::symbol::` qualifiers in `compiler/` to
`rustc_span::`. This is a 200+ net line of code reduction, mostly
because many files with two `use rustc_span` items can be reduced to
one.
- Move it to `rustc_parse`, which is the only crate that uses it. This
lets us remove all the `pub` markers from it.
- Change `next_ref` and `look_ahead` to `get` and `bump`, which work
better for the `rustc_parse` uses.
- This requires adding a `TokenStream::get` method, which is simple.
- In `TokenCursor`, we currently duplicate the
`DelimSpan`/`DelimSpacing`/`Delimiter` from the surrounding
`TokenTree::Delimited` in the stack. This isn't necessary so long as
we don't prematurely move past the `Delimited`, and is a small perf
win on a very hot code path.
- In `parse_token_tree`, we clone the relevant `TokenTree::Delimited`
instead of constructing an identical one from pieces.
Because `TokenStreamIter` is a much better name for a `TokenStream`
iterator. Also rename the `TokenStream::trees` method as
`TokenStream::iter`, and some local variables.
Field init shorthand allows writing initializers like `tcx: tcx` as
`tcx`. The compiler already uses it extensively. Fix the last few places
where it isn't yet used.
Keep track of patterns that could have introduced a binding, but didn't
When we recover from a pattern parse error, or a pattern uses `..`, we keep track of that and affect resolution error for missing bindings that could have been provided by that pattern. We differentiate between `..` and parse recovery. We silence resolution errors likely caused by the pattern parse error.
```
error[E0425]: cannot find value `title` in this scope
--> $DIR/struct-pattern-with-missing-fields-resolve-error.rs:18:30
|
LL | if let Website { url, .. } = website {
| ------------------- this pattern doesn't include `title`, which is available in `Website`
LL | println!("[{}]({})", title, url);
| ^^^^^ not found in this scope
```
Fix#74863.
When we recover from a pattern parse error, or a pattern uses `..`, we keep track of that and affect resolution error for missing bindings that could have been provided by that pattern. We differentiate between `..` and parse recovery. We silence resolution errors likely caused by the pattern parse error.
```
error[E0425]: cannot find value `title` in this scope
--> $DIR/struct-pattern-with-missing-fields-resolve-error.rs:19:30
|
LL | println!("[{}]({})", title, url);
| ^^^^^ not found in this scope
|
note: `Website` has a field `title` which could have been included in this pattern, but it wasn't
--> $DIR/struct-pattern-with-missing-fields-resolve-error.rs:17:12
|
LL | / struct Website {
LL | | url: String,
LL | | title: Option<String> ,
| | ----- defined here
LL | | }
| |_-
...
LL | if let Website { url, .. } = website {
| ^^^^^^^^^^^^^^^^^^^ this pattern doesn't include `title`, which is available in `Website`
```
Fix#74863.
Add AST support for unsafe binders
I'm splitting up #130514 into pieces. It's impossible for me to keep up with a huge PR like that. I'll land type system support for this next, probably w/o MIR lowering, which will come later.
r? `@oli-obk`
cc `@BoxyUwU` and `@lcnr` who also may want to look at this, though this PR doesn't do too much yet
Keep track of parse errors in `mod`s and don't emit resolve errors for paths involving them
When we expand a `mod foo;` and parse `foo.rs`, we now track whether that file had an unrecovered parse error that reached the end of the file. If so, we keep that information around in the HIR and mark its `DefId` in the `Resolver`. When resolving a path like `foo::bar`, we do not emit any errors for "`bar` not found in `foo`", as we know that the parse error might have caused `bar` to not be parsed and accounted for.
When this happens in an existing project, every path referencing `foo` would be an irrelevant compile error. Instead, we now skip emitting anything until `foo.rs` is fixed. Tellingly enough, we didn't have any test for errors caused by expansion of `mod`s with parse errors.
Fix https://github.com/rust-lang/rust/issues/97734.
allow `symbol_intern_string_literal` lint in test modules
Since #133545, `x check compiler --stage 1` no longer works because compiler test modules trigger `symbol_intern_string_literal` lint errors. Bootstrap shouldn't control when to ignore or enable this lint in the compiler tree (using `Kind != Test` was ineffective for obvious reasons).
Also, conditionally adding this rustflag invalidates the build cache between `x test` and other commands.
This PR removes the `Kind` check from bootstrap and handles it directly in the compiler tree in a more natural way.
When we expand a `mod foo;` and parse `foo.rs`, we now track whether that file had an unrecovered parse error that reached the end of the file. If so, we keep that information around. When resolving a path like `foo::bar`, we do not emit any errors for "`bar` not found in `foo`", as we know that the parse error might have caused `bar` to not be parsed and accounted for.
When this happens in an existing project, every path referencing `foo` would be an irrelevant compile error. Instead, we now skip emitting anything until `foo.rs` is fixed. Tellingly enough, we didn't have any test for errors caused by `mod` expansion.
Fix#97734.
Initial implementation of `#[feature(default_field_values]`, proposed in https://github.com/rust-lang/rfcs/pull/3681.
Support default fields in enum struct variant
Allow default values in an enum struct variant definition:
```rust
pub enum Bar {
Foo {
bar: S = S,
baz: i32 = 42 + 3,
}
}
```
Allow using `..` without a base on an enum struct variant
```rust
Bar::Foo { .. }
```
`#[derive(Default)]` doesn't account for these as it is still gating `#[default]` only being allowed on unit variants.
Support `#[derive(Default)]` on enum struct variants with all defaulted fields
```rust
pub enum Bar {
#[default]
Foo {
bar: S = S,
baz: i32 = 42 + 3,
}
}
```
Check for missing fields in typeck instead of mir_build.
Expand test with `const` param case (needs `generic_const_exprs` enabled).
Properly instantiate MIR const
The following works:
```rust
struct S<A> {
a: Vec<A> = Vec::new(),
}
S::<i32> { .. }
```
Add lint for default fields that will always fail const-eval
We *allow* this to happen for API writers that might want to rely on users'
getting a compile error when using the default field, different to the error
that they would get when the field isn't default. We could change this to
*always* error instead of being a lint, if we wanted.
This will *not* catch errors for partially evaluated consts, like when the
expression relies on a const parameter.
Suggestions when encountering `Foo { .. }` without `#[feature(default_field_values)]`:
- Suggest adding a base expression if there are missing fields.
- Suggest enabling the feature if all the missing fields have optional values.
- Suggest removing `..` if there are no missing fields.
Gate async fn trait bound modifier on `async_trait_bounds`
This PR moves `async Fn()` trait bounds into a new feature gate: `feature(async_trait_bounds)`. The general vibe is that we will most likely stabilize the `feature(async_closure)` *without* the `async Fn()` trait bound modifier, so we need to gate that separately.
We're trying to work on the general vision of `async` trait bound modifier general in: https://github.com/rust-lang/rfcs/pull/3710, however that RFC still needs more time for consensus to converge, and we've decided that the value that users get from calling the bound `async Fn()` is *not really* worth blocking landing async closures in general.
Change `AttrArgs::Eq` to a struct variant
Cleanups for simplifying https://github.com/rust-lang/rust/pull/131808
Basically changes `AttrArgs::Eq` to a struct variant and then avoids several matches on `AttrArgsEq` in favor of methods on it. This will make future refactorings simpler, as they can either keep methods or switch to field accesses without having to restructure code
Improve span handling in `parse_expr_bottom`.
`parse_expr_bottom` stores `this.token.span` in `lo`, but then fails to use it in many places where it could. This commit fixes that, and likewise (to a smaller extent) in `parse_ty_common`.
r? ``@spastorino``
`parse_expr_bottom` stores `this.token.span` in `lo`, but then fails to
use it in many places where it could. This commit fixes that, and
likewise (to a smaller extent) in `parse_ty_common`.
Inline ExprPrecedence::order into Expr::precedence
The representation of expression precedence in rustc_ast has been an obstacle to further improvements in the pretty-printer (continuing from #119105 and #119427).
Previously the operation of *"does this expression have lower precedence than that one"* (relevant for parenthesis insertion in macro-generated syntax trees) consisted of 3 steps:
1. Convert `Expr` to `ExprPrecedence` using `.precedence()`
2. Convert `ExprPrecedence` to `i8` using `.order()`
3. Compare using `<`
As far as I can guess, the reason for the separation between `precedence()` and `order()` was so that both `rustc_ast::Expr` and `rustc_hir::Expr` could convert as straightforwardly as possible to the same `ExprPrecedence` enum, and then the more finicky logic performed by `order` could be present just once.
The mapping between `Expr` and `ExprPrecedence` was intended to be as straightforward as possible:
```rust
match self.kind {
ExprKind::Closure(..) => ExprPrecedence::Closure,
...
}
```
although there were exceptions of both many-to-one, and one-to-many:
```rust
ExprKind::Underscore => ExprPrecedence::Path,
ExprKind::Path(..) => ExprPrecedence::Path,
...
ExprKind::Match(_, _, MatchKind::Prefix) => ExprPrecedence::Match,
ExprKind::Match(_, _, MatchKind::Postfix) => ExprPrecedence::PostfixMatch,
```
Where the nature of `ExprPrecedence` becomes problematic is when a single expression kind might be associated with multiple different precedence levels depending on context (outside the expression) and contents (inside the expression). For example consider what is the precedence of an ExprKind::Closure `$closure`. Well, on the left-hand side of a binary operator it would need parentheses in order to avoid the trailing binary operator being absorbed into the closure body: `($closure) + Rhs`, so the precedence is something lower than that of `+`. But on the right-hand side of a binary operator, a closure is just a straightforward prefix expression like a unary op, which is a relatively high precedence level, higher than binops but lower than method calls: `Lhs + $closure` is fine without parens but `($closure).method()` needs them. But as a third case, if the closure contains an explicit return type, then the precedence is an even higher level than that, never needing parenthesization even in a binop left-hand side or method call: `|| -> bool { false } + Rhs` or `|| -> bool { false }.method()`.
You can see that trying to capture all of this resolution about expressions into `ExprPrecedence` violates the intention of `ExprPrecedence` being a straightforward one-to-one correspondence from each AST and HIR `ExprKind` variant. It would be possible to attempt that by doing stuff like `ExprPrecedence::Closure(Side::Leading, ReturnType::No)`, but I don't foresee the original envisioned benefit of the `precedence()`/`order()` distinction being retained in this approach. Instead I want to move toward a model that Syn has been using successfully. In Syn, there is a Precedence enum but it differs from rustc in the following ways:
- There are [relatively few variants](https://github.com/dtolnay/syn/blob/2.0.87/src/precedence.rs#L11-L47) compared to rustc's `ExprPrecedence`. For example there is no distinction at the precedence level between returns and closures, or between loops and method calls.
- We distinguish between [leading](https://github.com/dtolnay/syn/blob/2.0.87/src/fixup.rs#L293) and [trailing](https://github.com/dtolnay/syn/blob/2.0.87/src/fixup.rs#L309) precedence, taking into account an expression's context such as what token follows it (for various syntactic bail-outs in Rust's grammar, like ambiguities around break-with-value) and how it relates to operators from the surrounding syntax tree.
- There are no hardcoded mysterious integer quantities like rustc's `PREC_CLOSURE = -40`. All precedence comparisons are performed via PartialOrd on a C-like enum.
This PR is just a first step in these changes. As you can tell from Syn, I definitely think there is value in having a dedicated type to represent precedence, instead of what `order()` is doing with `i8`. But that is a whole separate adventure because rustc_ast doesn't even agree consistently on `i8` being the type for precedence order; `AssocOp::precedence` instead uses `usize` and there are casts in both directions. It is likely that a type called `ExprPrecedence` will re-appear, but it will look substantially different from the one that existed before this PR.
