Handle str literals written with '
lexed as lifetime
Given `'hello world'` and `'1 str', provide a structured suggestion for a valid string literal: ``` error[E0762]: unterminated character literal --> $DIR/lex-bad-str-literal-as-char-3.rs:2:26 | LL | println!('hello world'); | ^^^^ | help: if you meant to write a `str` literal, use double quotes | LL | println!("hello world"); | ~ ~ ``` ``` error[E0762]: unterminated character literal --> $DIR/lex-bad-str-literal-as-char-1.rs:2:20 | LL | println!('1 + 1'); | ^^^^ | help: if you meant to write a `str` literal, use double quotes | LL | println!("1 + 1"); | ~ ~ ``` Fix #119685.
This commit is contained in:
parent
22e241e32e
commit
982918f493
@ -46,7 +46,7 @@ pub(crate) fn prev(&self) -> char {
|
|||||||
/// If requested position doesn't exist, `EOF_CHAR` is returned.
|
/// If requested position doesn't exist, `EOF_CHAR` is returned.
|
||||||
/// However, getting `EOF_CHAR` doesn't always mean actual end of file,
|
/// However, getting `EOF_CHAR` doesn't always mean actual end of file,
|
||||||
/// it should be checked with `is_eof` method.
|
/// it should be checked with `is_eof` method.
|
||||||
pub(crate) fn first(&self) -> char {
|
pub fn first(&self) -> char {
|
||||||
// `.next()` optimizes better than `.nth(0)`
|
// `.next()` optimizes better than `.nth(0)`
|
||||||
self.chars.clone().next().unwrap_or(EOF_CHAR)
|
self.chars.clone().next().unwrap_or(EOF_CHAR)
|
||||||
}
|
}
|
||||||
|
@ -835,6 +835,7 @@ parse_unknown_prefix = prefix `{$prefix}` is unknown
|
|||||||
.label = unknown prefix
|
.label = unknown prefix
|
||||||
.note = prefixed identifiers and literals are reserved since Rust 2021
|
.note = prefixed identifiers and literals are reserved since Rust 2021
|
||||||
.suggestion_br = use `br` for a raw byte string
|
.suggestion_br = use `br` for a raw byte string
|
||||||
|
.suggestion_str = if you meant to write a `str` literal, use double quotes
|
||||||
.suggestion_whitespace = consider inserting whitespace here
|
.suggestion_whitespace = consider inserting whitespace here
|
||||||
|
|
||||||
parse_unknown_start_of_token = unknown start of token: {$escaped}
|
parse_unknown_start_of_token = unknown start of token: {$escaped}
|
||||||
|
@ -1994,6 +1994,17 @@ pub enum UnknownPrefixSugg {
|
|||||||
style = "verbose"
|
style = "verbose"
|
||||||
)]
|
)]
|
||||||
Whitespace(#[primary_span] Span),
|
Whitespace(#[primary_span] Span),
|
||||||
|
#[multipart_suggestion(
|
||||||
|
parse_suggestion_str,
|
||||||
|
applicability = "maybe-incorrect",
|
||||||
|
style = "verbose"
|
||||||
|
)]
|
||||||
|
MeantStr {
|
||||||
|
#[suggestion_part(code = "\"")]
|
||||||
|
start: Span,
|
||||||
|
#[suggestion_part(code = "\"")]
|
||||||
|
end: Span,
|
||||||
|
},
|
||||||
}
|
}
|
||||||
|
|
||||||
#[derive(Diagnostic)]
|
#[derive(Diagnostic)]
|
||||||
|
@ -63,6 +63,7 @@ pub(crate) fn parse_token_trees<'psess, 'src>(
|
|||||||
cursor,
|
cursor,
|
||||||
override_span,
|
override_span,
|
||||||
nbsp_is_whitespace: false,
|
nbsp_is_whitespace: false,
|
||||||
|
last_lifetime: None,
|
||||||
};
|
};
|
||||||
let (stream, res, unmatched_delims) =
|
let (stream, res, unmatched_delims) =
|
||||||
tokentrees::TokenTreesReader::parse_all_token_trees(string_reader);
|
tokentrees::TokenTreesReader::parse_all_token_trees(string_reader);
|
||||||
@ -105,6 +106,10 @@ struct StringReader<'psess, 'src> {
|
|||||||
/// in this file, it's safe to treat further occurrences of the non-breaking
|
/// in this file, it's safe to treat further occurrences of the non-breaking
|
||||||
/// space character as whitespace.
|
/// space character as whitespace.
|
||||||
nbsp_is_whitespace: bool,
|
nbsp_is_whitespace: bool,
|
||||||
|
|
||||||
|
/// Track the `Span` for the leading `'` of the last lifetime. Used for
|
||||||
|
/// diagnostics to detect possible typo where `"` was meant.
|
||||||
|
last_lifetime: Option<Span>,
|
||||||
}
|
}
|
||||||
|
|
||||||
impl<'psess, 'src> StringReader<'psess, 'src> {
|
impl<'psess, 'src> StringReader<'psess, 'src> {
|
||||||
@ -130,6 +135,18 @@ fn next_token(&mut self) -> (Token, bool) {
|
|||||||
|
|
||||||
debug!("next_token: {:?}({:?})", token.kind, self.str_from(start));
|
debug!("next_token: {:?}({:?})", token.kind, self.str_from(start));
|
||||||
|
|
||||||
|
if let rustc_lexer::TokenKind::Semi
|
||||||
|
| rustc_lexer::TokenKind::LineComment { .. }
|
||||||
|
| rustc_lexer::TokenKind::BlockComment { .. }
|
||||||
|
| rustc_lexer::TokenKind::CloseParen
|
||||||
|
| rustc_lexer::TokenKind::CloseBrace
|
||||||
|
| rustc_lexer::TokenKind::CloseBracket = token.kind
|
||||||
|
{
|
||||||
|
// Heuristic: we assume that it is unlikely we're dealing with an unterminated
|
||||||
|
// string surrounded by single quotes.
|
||||||
|
self.last_lifetime = None;
|
||||||
|
}
|
||||||
|
|
||||||
// Now "cook" the token, converting the simple `rustc_lexer::TokenKind` enum into a
|
// Now "cook" the token, converting the simple `rustc_lexer::TokenKind` enum into a
|
||||||
// rich `rustc_ast::TokenKind`. This turns strings into interned symbols and runs
|
// rich `rustc_ast::TokenKind`. This turns strings into interned symbols and runs
|
||||||
// additional validation.
|
// additional validation.
|
||||||
@ -247,6 +264,7 @@ fn next_token(&mut self) -> (Token, bool) {
|
|||||||
// expansion purposes. See #12512 for the gory details of why
|
// expansion purposes. See #12512 for the gory details of why
|
||||||
// this is necessary.
|
// this is necessary.
|
||||||
let lifetime_name = self.str_from(start);
|
let lifetime_name = self.str_from(start);
|
||||||
|
self.last_lifetime = Some(self.mk_sp(start, start + BytePos(1)));
|
||||||
if starts_with_number {
|
if starts_with_number {
|
||||||
let span = self.mk_sp(start, self.pos);
|
let span = self.mk_sp(start, self.pos);
|
||||||
self.dcx().struct_err("lifetimes cannot start with a number")
|
self.dcx().struct_err("lifetimes cannot start with a number")
|
||||||
@ -395,10 +413,21 @@ fn cook_lexer_literal(
|
|||||||
match kind {
|
match kind {
|
||||||
rustc_lexer::LiteralKind::Char { terminated } => {
|
rustc_lexer::LiteralKind::Char { terminated } => {
|
||||||
if !terminated {
|
if !terminated {
|
||||||
self.dcx()
|
let mut err = self
|
||||||
|
.dcx()
|
||||||
.struct_span_fatal(self.mk_sp(start, end), "unterminated character literal")
|
.struct_span_fatal(self.mk_sp(start, end), "unterminated character literal")
|
||||||
.with_code(E0762)
|
.with_code(E0762);
|
||||||
.emit()
|
if let Some(lt_sp) = self.last_lifetime {
|
||||||
|
err.multipart_suggestion(
|
||||||
|
"if you meant to write a `str` literal, use double quotes",
|
||||||
|
vec![
|
||||||
|
(lt_sp, "\"".to_string()),
|
||||||
|
(self.mk_sp(start, start + BytePos(1)), "\"".to_string()),
|
||||||
|
],
|
||||||
|
Applicability::MaybeIncorrect,
|
||||||
|
);
|
||||||
|
}
|
||||||
|
err.emit()
|
||||||
}
|
}
|
||||||
self.cook_unicode(token::Char, Mode::Char, start, end, 1, 1) // ' '
|
self.cook_unicode(token::Char, Mode::Char, start, end, 1, 1) // ' '
|
||||||
}
|
}
|
||||||
@ -673,7 +702,16 @@ fn report_unknown_prefix(&self, start: BytePos) {
|
|||||||
let sugg = if prefix == "rb" {
|
let sugg = if prefix == "rb" {
|
||||||
Some(errors::UnknownPrefixSugg::UseBr(prefix_span))
|
Some(errors::UnknownPrefixSugg::UseBr(prefix_span))
|
||||||
} else if expn_data.is_root() {
|
} else if expn_data.is_root() {
|
||||||
Some(errors::UnknownPrefixSugg::Whitespace(prefix_span.shrink_to_hi()))
|
if self.cursor.first() == '\''
|
||||||
|
&& let Some(start) = self.last_lifetime
|
||||||
|
{
|
||||||
|
Some(errors::UnknownPrefixSugg::MeantStr {
|
||||||
|
start,
|
||||||
|
end: self.mk_sp(self.pos, self.pos + BytePos(1)),
|
||||||
|
})
|
||||||
|
} else {
|
||||||
|
Some(errors::UnknownPrefixSugg::Whitespace(prefix_span.shrink_to_hi()))
|
||||||
|
}
|
||||||
} else {
|
} else {
|
||||||
None
|
None
|
||||||
};
|
};
|
||||||
|
6
tests/ui/lexer/lex-bad-str-literal-as-char-1.fixed
Normal file
6
tests/ui/lexer/lex-bad-str-literal-as-char-1.fixed
Normal file
@ -0,0 +1,6 @@
|
|||||||
|
//@ run-rustfix
|
||||||
|
fn main() {
|
||||||
|
println!("1 + 1");
|
||||||
|
//~^ ERROR unterminated character literal
|
||||||
|
//~| ERROR lifetimes cannot start with a number
|
||||||
|
}
|
6
tests/ui/lexer/lex-bad-str-literal-as-char-1.rs
Normal file
6
tests/ui/lexer/lex-bad-str-literal-as-char-1.rs
Normal file
@ -0,0 +1,6 @@
|
|||||||
|
//@ run-rustfix
|
||||||
|
fn main() {
|
||||||
|
println!('1 + 1');
|
||||||
|
//~^ ERROR unterminated character literal
|
||||||
|
//~| ERROR lifetimes cannot start with a number
|
||||||
|
}
|
20
tests/ui/lexer/lex-bad-str-literal-as-char-1.stderr
Normal file
20
tests/ui/lexer/lex-bad-str-literal-as-char-1.stderr
Normal file
@ -0,0 +1,20 @@
|
|||||||
|
error[E0762]: unterminated character literal
|
||||||
|
--> $DIR/lex-bad-str-literal-as-char-1.rs:3:20
|
||||||
|
|
|
||||||
|
LL | println!('1 + 1');
|
||||||
|
| ^^^
|
||||||
|
|
|
||||||
|
help: if you meant to write a `str` literal, use double quotes
|
||||||
|
|
|
||||||
|
LL | println!("1 + 1");
|
||||||
|
| ~ ~
|
||||||
|
|
||||||
|
error: lifetimes cannot start with a number
|
||||||
|
--> $DIR/lex-bad-str-literal-as-char-1.rs:3:14
|
||||||
|
|
|
||||||
|
LL | println!('1 + 1');
|
||||||
|
| ^^
|
||||||
|
|
||||||
|
error: aborting due to 2 previous errors
|
||||||
|
|
||||||
|
For more information about this error, try `rustc --explain E0762`.
|
4
tests/ui/lexer/lex-bad-str-literal-as-char-2.fixed
Normal file
4
tests/ui/lexer/lex-bad-str-literal-as-char-2.fixed
Normal file
@ -0,0 +1,4 @@
|
|||||||
|
//@ run-rustfix
|
||||||
|
fn main() {
|
||||||
|
println!(" 1 + 1"); //~ ERROR character literal may only contain one codepoint
|
||||||
|
}
|
4
tests/ui/lexer/lex-bad-str-literal-as-char-2.rs
Normal file
4
tests/ui/lexer/lex-bad-str-literal-as-char-2.rs
Normal file
@ -0,0 +1,4 @@
|
|||||||
|
//@ run-rustfix
|
||||||
|
fn main() {
|
||||||
|
println!(' 1 + 1'); //~ ERROR character literal may only contain one codepoint
|
||||||
|
}
|
13
tests/ui/lexer/lex-bad-str-literal-as-char-2.stderr
Normal file
13
tests/ui/lexer/lex-bad-str-literal-as-char-2.stderr
Normal file
@ -0,0 +1,13 @@
|
|||||||
|
error: character literal may only contain one codepoint
|
||||||
|
--> $DIR/lex-bad-str-literal-as-char-2.rs:3:14
|
||||||
|
|
|
||||||
|
LL | println!(' 1 + 1');
|
||||||
|
| ^^^^^^^^
|
||||||
|
|
|
||||||
|
help: if you meant to write a `str` literal, use double quotes
|
||||||
|
|
|
||||||
|
LL | println!(" 1 + 1");
|
||||||
|
| ~~~~~~~~
|
||||||
|
|
||||||
|
error: aborting due to 1 previous error
|
||||||
|
|
4
tests/ui/lexer/lex-bad-str-literal-as-char-3.fixed
Normal file
4
tests/ui/lexer/lex-bad-str-literal-as-char-3.fixed
Normal file
@ -0,0 +1,4 @@
|
|||||||
|
//@ run-rustfix
|
||||||
|
fn main() {
|
||||||
|
println!("hello world"); //~ ERROR unterminated character literal
|
||||||
|
}
|
4
tests/ui/lexer/lex-bad-str-literal-as-char-3.rs
Normal file
4
tests/ui/lexer/lex-bad-str-literal-as-char-3.rs
Normal file
@ -0,0 +1,4 @@
|
|||||||
|
//@ run-rustfix
|
||||||
|
fn main() {
|
||||||
|
println!('hello world'); //~ ERROR unterminated character literal
|
||||||
|
}
|
14
tests/ui/lexer/lex-bad-str-literal-as-char-3.stderr
Normal file
14
tests/ui/lexer/lex-bad-str-literal-as-char-3.stderr
Normal file
@ -0,0 +1,14 @@
|
|||||||
|
error[E0762]: unterminated character literal
|
||||||
|
--> $DIR/lex-bad-str-literal-as-char-3.rs:3:26
|
||||||
|
|
|
||||||
|
LL | println!('hello world');
|
||||||
|
| ^^^^
|
||||||
|
|
|
||||||
|
help: if you meant to write a `str` literal, use double quotes
|
||||||
|
|
|
||||||
|
LL | println!("hello world");
|
||||||
|
| ~ ~
|
||||||
|
|
||||||
|
error: aborting due to 1 previous error
|
||||||
|
|
||||||
|
For more information about this error, try `rustc --explain E0762`.
|
Loading…
Reference in New Issue
Block a user