mikros/rust
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Rollup merge of #120221 - compiler-errors:statements-are-not-patterns, r=nnethercote

Browse Source 
Don't make statement nonterminals match pattern nonterminals

Right now, the heuristic we use to check if a token may begin a pattern nonterminal falls back to `may_be_ident`:
ef71f1047e/compiler/rustc_parse/src/parser/nonterminal.rs (L21-L37)

This has the unfortunate side effect that a `stmt` nonterminal eagerly matches against a `pat` nonterminal, leading to a parse error:
```rust
macro_rules! m {
    ($pat:pat) => {};
    ($stmt:stmt) => {};
}

macro_rules! m2 {
    ($stmt:stmt) => {
        m! { $stmt }
    };
}

m2! { let x = 1 }
```

This PR fixes it by more accurately reflecting the set of nonterminals that may begin a pattern nonterminal.

As a side-effect, I modified `Token::can_begin_pattern` to work correctly and used that in `Parser::nonterminal_may_begin_with`.
This commit is contained in:
Matthias Krüger 2024-08-31 10:08:51 +02:00 committed by GitHub
commit 1fd0c71818
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
5 changed files with 65 additions and 42 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

57
compiler/rustc_ast/src/token.rs
View File

@ -486,6 +486,9 @@ pub fn is_like_plus(&self) -> bool {
}
/// Returns `true` if the token can appear at the start of an expression.
///
/// **NB**: Take care when modifying this function, since it will change
/// the stable set of tokens that are allowed to match an expr nonterminal.
pub fn can_begin_expr(&self) -> bool {
match self.uninterpolate().kind {
Ident(name, is_raw) =>
@ -504,10 +507,13 @@ pub fn can_begin_expr(&self) -> bool {
PathSep | // global path
Lifetime(..) | // labeled loop
Pound => true, // expression attributes
Interpolated(ref nt) => matches!(&**nt, NtLiteral(..) |
NtExpr(..) |
NtBlock(..) |
NtPath(..)),
Interpolated(ref nt) =>
matches!(&**nt,
NtBlock(..) |
NtExpr(..) |
NtLiteral(..) |
NtPath(..)
),
_ => false,
}
}
@ -515,23 +521,32 @@ pub fn can_begin_expr(&self) -> bool {
/// Returns `true` if the token can appear at the start of a pattern.
///
/// Shamelessly borrowed from `can_begin_expr`, only used for diagnostics right now.
pub fn can_begin_pattern(&self) -> bool {
match self.uninterpolate().kind {
Ident(name, is_raw) =>
ident_can_begin_expr(name, self.span, is_raw), // value name or keyword
| OpenDelim(Delimiter::Bracket | Delimiter::Parenthesis) // tuple or array
| Literal(..) // literal
| BinOp(Minus) // unary minus
| BinOp(And) // reference
| AndAnd // double reference
// DotDotDot is no longer supported
| DotDot | DotDotDot | DotDotEq // ranges
| Lt | BinOp(Shl) // associated path
| PathSep => true, // global path
Interpolated(ref nt) => matches!(&**nt, NtLiteral(..) |
NtPat(..) |
NtBlock(..) |
NtPath(..)),
pub fn can_begin_pattern(&self, pat_kind: NtPatKind) -> bool {
match &self.uninterpolate().kind {
// box, ref, mut, and other identifiers (can stricten)
Ident(..) | NtIdent(..) |
OpenDelim(Delimiter::Parenthesis) | // tuple pattern
OpenDelim(Delimiter::Bracket) | // slice pattern
BinOp(And) | // reference
BinOp(Minus) | // negative literal
AndAnd | // double reference
Literal(_) | // literal
DotDot | // range pattern (future compat)
DotDotDot | // range pattern (future compat)
PathSep | // path
Lt | // path (UFCS constant)
BinOp(Shl) => true, // path (double UFCS)
// leading vert `|` or-pattern
BinOp(Or) => matches!(pat_kind, PatWithOr),
Interpolated(nt) =>
matches!(&**nt,
| NtExpr(..)
| NtLiteral(..)
| NtMeta(..)
| NtPat(..)
| NtPath(..)
| NtTy(..)
),
_ => false,
}
}

20
compiler/rustc_parse/src/parser/nonterminal.rs
View File

@ -86,25 +86,7 @@ fn may_be_ident(nt: &token::Nonterminal) -> bool {
token::Interpolated(nt) => may_be_ident(nt),
_ => false,
},
NonterminalKind::Pat(pat_kind) => match &token.kind {
// box, ref, mut, and other identifiers (can stricten)
token::Ident(..) | token::NtIdent(..) |
token::OpenDelim(Delimiter::Parenthesis) | // tuple pattern
token::OpenDelim(Delimiter::Bracket) | // slice pattern
token::BinOp(token::And) | // reference
token::BinOp(token::Minus) | // negative literal
token::AndAnd | // double reference
token::Literal(_) | // literal
token::DotDot | // range pattern (future compat)
token::DotDotDot | // range pattern (future compat)
token::PathSep | // path
token::Lt | // path (UFCS constant)
token::BinOp(token::Shl) => true, // path (double UFCS)
// leading vert `|` or-pattern
token::BinOp(token::Or) => matches!(pat_kind, PatWithOr),
token::Interpolated(nt) => may_be_ident(nt),
_ => false,
},
NonterminalKind::Pat(pat_kind) => token.can_begin_pattern(pat_kind),
NonterminalKind::Lifetime => match &token.kind {
token::Lifetime(_) | token::NtLifetime(..) => true,
_ => false,

6
compiler/rustc_parse/src/parser/pat.rs
View File

@ -444,7 +444,11 @@ fn parse_pat_with_range_pat(
let mut lo = self.token.span;
if self.token.is_keyword(kw::Let) && self.look_ahead(1, |tok| tok.can_begin_pattern()) {
if self.token.is_keyword(kw::Let)
&& self.look_ahead(1, |tok| {
tok.can_begin_pattern(token::NtPatKind::PatParam { inferred: false })
})
{
self.bump();
self.dcx().emit_err(RemoveLet { span: lo });
lo = self.token.span;

5
compiler/rustc_parse/src/parser/path.rs
View File

@ -378,7 +378,10 @@ pub(super) fn parse_path_segment(
if self.may_recover()
&& prev_token_before_parsing == token::PathSep
&& (style == PathStyle::Expr && self.token.can_begin_expr()
|| style == PathStyle::Pat && self.token.can_begin_pattern())
|| style == PathStyle::Pat
&& self.token.can_begin_pattern(token::NtPatKind::PatParam {
inferred: false,
}))
{
snapshot = Some(self.create_snapshot_for_diagnostic());
}

19
tests/ui/pattern/patterns-dont-match-nt-statement.rs Normal file
View File

@ -0,0 +1,19 @@
//@ check-pass
// Make sure that a `stmt` nonterminal does not eagerly match against
// a `pat`, since this will always cause a parse error...
macro_rules! m {
($pat:pat) => {};
($stmt:stmt) => {};
}
macro_rules! m2 {
($stmt:stmt) => {
m! { $stmt }
};
}
m2! { let x = 1 }
fn main() {}