Commit 2c2bb995af398383e3b93b859302bdc447ca7a7c | rust

Cargo.lock+2-2

...	...	@@ -3175,9 +3175,9 @@ checksum = "357703d41365b4b27c590e3ed91eabb1b663f07c4c084095e60cbed4362dff0d"
3175	3175
3176	3176	[[package]]
3177	3177	name = "rustc-literal-escaper"
3178		version = "0.0.2"
	3178	version = "0.0.4"
3179	3179	source = "registry+https://github.com/rust-lang/crates.io-index"
3180		checksum = "0041b6238913c41fe704213a4a9329e2f685a156d1781998128b4149c230ad04"
	3180	checksum = "ab03008eb631b703dd16978282ae36c73282e7922fe101a4bd072a40ecea7b8b"
3181	3181
3182	3182	[[package]]
3183	3183	name = "rustc-main"

Cargo.toml+5

...	...	@@ -89,3 +89,8 @@ codegen-units = 1
89	89	# FIXME: LTO cannot be enabled for binaries in a workspace
90	90	# <https://github.com/rust-lang/cargo/issues/9330>
91	91	# lto = true
	92
	93	# If you want to use a crate with local modifications, you can set a path or git dependency here.
	94	# For git dependencies, also add your source to ALLOWED_SOURCES in src/tools/tidy/src/extdeps.rs.
	95	#[patch.crates-io]
	96

compiler/rustc_ast/Cargo.toml+1-1

...	...	@@ -7,7 +7,7 @@ edition = "2024"
7	7	# tidy-alphabetical-start
8	8	bitflags = "2.4.1"
9	9	memchr = "2.7.4"
10		rustc-literal-escaper = "0.0.2"
	10	rustc-literal-escaper = "0.0.4"
11	11	rustc_ast_ir = { path = "../rustc_ast_ir" }
12	12	rustc_data_structures = { path = "../rustc_data_structures" }
13	13	rustc_index = { path = "../rustc_index" }

compiler/rustc_ast/src/util/literal.rs+7-8

...	...	@@ -3,7 +3,7 @@
3	3	use std::{ascii, fmt, str};
4	4
5	5	use rustc_literal_escaper::{
6		MixedUnit, Mode, byte_from_char, unescape_byte, unescape_char, unescape_mixed, unescape_unicode,
	6	MixedUnit, unescape_byte, unescape_byte_str, unescape_c_str, unescape_char, unescape_str,
7	7	};
8	8	use rustc_span::{Span, Symbol, kw, sym};
9	9	use tracing::debug;
...	...	@@ -87,11 +87,10 @@ impl LitKind {
87	87	// Force-inlining here is aggressive but the closure is
88	88	// called on every char in the string, so it can be hot in
89	89	// programs with many long strings containing escapes.
90		unescape_unicode(
	90	unescape_str(
91	91	s,
92		Mode::Str,
93		&mut #[inline(always)]
94		\|_, c\| match c {
	92	#[inline(always)]
	93	\|_, res\| match res {
95	94	Ok(c) => buf.push(c),
96	95	Err(err) => {
97	96	assert!(!err.is_fatal(), "failed to unescape string literal")
...	...	@@ -111,8 +110,8 @@ impl LitKind {
111	110	token::ByteStr => {
112	111	let s = symbol.as_str();
113	112	let mut buf = Vec::with_capacity(s.len());
114		unescape_unicode(s, Mode::ByteStr, &mut \|_, c\| match c {
115		Ok(c) => buf.push(byte_from_char(c)),
	113	unescape_byte_str(s, \|_, res\| match res {
	114	Ok(b) => buf.push(b),
116	115	Err(err) => {
117	116	assert!(!err.is_fatal(), "failed to unescape string literal")
118	117	}
...	...	@@ -128,7 +127,7 @@ impl LitKind {
128	127	token::CStr => {
129	128	let s = symbol.as_str();
130	129	let mut buf = Vec::with_capacity(s.len());
131		unescape_mixed(s, Mode::CStr, &mut \|_span, c\| match c {
	130	unescape_c_str(s, \|_span, c\| match c {
132	131	Ok(MixedUnit::Char(c)) => {
133	132	buf.extend_from_slice(c.encode_utf8(&mut [0; 4]).as_bytes())
134	133	}

compiler/rustc_parse/Cargo.toml+1-1

...	...	@@ -6,7 +6,7 @@ edition = "2024"
6	6	[dependencies]
7	7	# tidy-alphabetical-start
8	8	bitflags = "2.4.1"
9		rustc-literal-escaper = "0.0.2"
	9	rustc-literal-escaper = "0.0.4"
10	10	rustc_ast = { path = "../rustc_ast" }
11	11	rustc_ast_pretty = { path = "../rustc_ast_pretty" }
12	12	rustc_data_structures = { path = "../rustc_data_structures" }

compiler/rustc_parse/src/lexer/mod.rs+32-62

...	...	@@ -1,5 +1,3 @@
1		use std::ops::Range;
2
3	1	use diagnostics::make_unclosed_delims_error;
4	2	use rustc_ast::ast::{self, AttrStyle};
5	3	use rustc_ast::token::{self, CommentKind, Delimiter, IdentIsRaw, Token, TokenKind};
...	...	@@ -10,7 +8,7 @@ use rustc_errors::{Applicability, Diag, DiagCtxtHandle, StashKey};
10	8	use rustc_lexer::{
11	9	Base, Cursor, DocStyle, FrontmatterAllowed, LiteralKind, RawStrError, is_whitespace,
12	10	};
13		use rustc_literal_escaper::{EscapeError, Mode, unescape_mixed, unescape_unicode};
	11	use rustc_literal_escaper::{EscapeError, Mode, check_for_errors};
14	12	use rustc_session::lint::BuiltinLintDiag;
15	13	use rustc_session::lint::builtin::{
16	14	RUST_2021_PREFIXES_INCOMPATIBLE_SYNTAX, RUST_2024_GUARDED_STRING_INCOMPATIBLE_SYNTAX,
...	...	@@ -702,7 +700,7 @@ impl<'psess, 'src> Lexer<'psess, 'src> {
702	700	}
703	701	err.emit()
704	702	}
705		self.cook_unicode(token::Char, Mode::Char, start, end, 1, 1) // ' '
	703	self.cook_quoted(token::Char, Mode::Char, start, end, 1, 1) // ' '
706	704	}
707	705	rustc_lexer::LiteralKind::Byte { terminated } => {
708	706	if !terminated {
...	...	@@ -714,7 +712,7 @@ impl<'psess, 'src> Lexer<'psess, 'src> {
714	712	.with_code(E0763)
715	713	.emit()
716	714	}
717		self.cook_unicode(token::Byte, Mode::Byte, start, end, 2, 1) // b' '
	715	self.cook_quoted(token::Byte, Mode::Byte, start, end, 2, 1) // b' '
718	716	}
719	717	rustc_lexer::LiteralKind::Str { terminated } => {
720	718	if !terminated {
...	...	@@ -726,7 +724,7 @@ impl<'psess, 'src> Lexer<'psess, 'src> {
726	724	.with_code(E0765)
727	725	.emit()
728	726	}
729		self.cook_unicode(token::Str, Mode::Str, start, end, 1, 1) // " "
	727	self.cook_quoted(token::Str, Mode::Str, start, end, 1, 1) // " "
730	728	}
731	729	rustc_lexer::LiteralKind::ByteStr { terminated } => {
732	730	if !terminated {
...	...	@@ -738,7 +736,8 @@ impl<'psess, 'src> Lexer<'psess, 'src> {
738	736	.with_code(E0766)
739	737	.emit()
740	738	}
741		self.cook_unicode(token::ByteStr, Mode::ByteStr, start, end, 2, 1) // b" "
	739	self.cook_quoted(token::ByteStr, Mode::ByteStr, start, end, 2, 1)
	740	// b" "
742	741	}
743	742	rustc_lexer::LiteralKind::CStr { terminated } => {
744	743	if !terminated {
...	...	@@ -750,13 +749,14 @@ impl<'psess, 'src> Lexer<'psess, 'src> {
750	749	.with_code(E0767)
751	750	.emit()
752	751	}
753		self.cook_mixed(token::CStr, Mode::CStr, start, end, 2, 1) // c" "
	752	self.cook_quoted(token::CStr, Mode::CStr, start, end, 2, 1) // c" "
754	753	}
755	754	rustc_lexer::LiteralKind::RawStr { n_hashes } => {
756	755	if let Some(n_hashes) = n_hashes {
757	756	let n = u32::from(n_hashes);
758	757	let kind = token::StrRaw(n_hashes);
759		self.cook_unicode(kind, Mode::RawStr, start, end, 2 + n, 1 + n) // r##" "##
	758	self.cook_quoted(kind, Mode::RawStr, start, end, 2 + n, 1 + n)
	759	// r##" "##
760	760	} else {
761	761	self.report_raw_str_error(start, 1);
762	762	}
...	...	@@ -765,7 +765,8 @@ impl<'psess, 'src> Lexer<'psess, 'src> {
765	765	if let Some(n_hashes) = n_hashes {
766	766	let n = u32::from(n_hashes);
767	767	let kind = token::ByteStrRaw(n_hashes);
768		self.cook_unicode(kind, Mode::RawByteStr, start, end, 3 + n, 1 + n) // br##" "##
	768	self.cook_quoted(kind, Mode::RawByteStr, start, end, 3 + n, 1 + n)
	769	// br##" "##
769	770	} else {
770	771	self.report_raw_str_error(start, 2);
771	772	}
...	...	@@ -774,7 +775,8 @@ impl<'psess, 'src> Lexer<'psess, 'src> {
774	775	if let Some(n_hashes) = n_hashes {
775	776	let n = u32::from(n_hashes);
776	777	let kind = token::CStrRaw(n_hashes);
777		self.cook_unicode(kind, Mode::RawCStr, start, end, 3 + n, 1 + n) // cr##" "##
	778	self.cook_quoted(kind, Mode::RawCStr, start, end, 3 + n, 1 + n)
	779	// cr##" "##
778	780	} else {
779	781	self.report_raw_str_error(start, 2);
780	782	}
...	...	@@ -1091,7 +1093,7 @@ impl<'psess, 'src> Lexer<'psess, 'src> {
1091	1093	self.dcx().emit_fatal(errors::TooManyHashes { span: self.mk_sp(start, self.pos), num });
1092	1094	}
1093	1095
1094		fn cook_common(
	1096	fn cook_quoted(
1095	1097	&self,
1096	1098	mut kind: token::LitKind,
1097	1099	mode: Mode,
...	...	@@ -1099,32 +1101,28 @@ impl<'psess, 'src> Lexer<'psess, 'src> {
1099	1101	end: BytePos,
1100	1102	prefix_len: u32,
1101	1103	postfix_len: u32,
1102		unescape: fn(&str, Mode, &mut dyn FnMut(Range<usize>, Result<(), EscapeError>)),
1103	1104	) -> (token::LitKind, Symbol) {
1104	1105	let content_start = start + BytePos(prefix_len);
1105	1106	let content_end = end - BytePos(postfix_len);
1106	1107	let lit_content = self.str_from_to(content_start, content_end);
1107		unescape(lit_content, mode, &mut \|range, result\| {
1108		// Here we only check for errors. The actual unescaping is done later.
1109		if let Err(err) = result {
1110		let span_with_quotes = self.mk_sp(start, end);
1111		let (start, end) = (range.start as u32, range.end as u32);
1112		let lo = content_start + BytePos(start);
1113		let hi = lo + BytePos(end - start);
1114		let span = self.mk_sp(lo, hi);
1115		let is_fatal = err.is_fatal();
1116		if let Some(guar) = emit_unescape_error(
1117		self.dcx(),
1118		lit_content,
1119		span_with_quotes,
1120		span,
1121		mode,
1122		range,
1123		err,
1124		) {
1125		assert!(is_fatal);
1126		kind = token::Err(guar);
1127		}
	1108	check_for_errors(lit_content, mode, \|range, err\| {
	1109	let span_with_quotes = self.mk_sp(start, end);
	1110	let (start, end) = (range.start as u32, range.end as u32);
	1111	let lo = content_start + BytePos(start);
	1112	let hi = lo + BytePos(end - start);
	1113	let span = self.mk_sp(lo, hi);
	1114	let is_fatal = err.is_fatal();
	1115	if let Some(guar) = emit_unescape_error(
	1116	self.dcx(),
	1117	lit_content,
	1118	span_with_quotes,
	1119	span,
	1120	mode,
	1121	range,
	1122	err,
	1123	) {
	1124	assert!(is_fatal);
	1125	kind = token::Err(guar);
1128	1126	}
1129	1127	});
1130	1128
...	...	@@ -1137,34 +1135,6 @@ impl<'psess, 'src> Lexer<'psess, 'src> {
1137	1135	};
1138	1136	(kind, sym)
1139	1137	}
1140
1141		fn cook_unicode(
1142		&self,
1143		kind: token::LitKind,
1144		mode: Mode,
1145		start: BytePos,
1146		end: BytePos,
1147		prefix_len: u32,
1148		postfix_len: u32,
1149		) -> (token::LitKind, Symbol) {
1150		self.cook_common(kind, mode, start, end, prefix_len, postfix_len, \|src, mode, callback\| {
1151		unescape_unicode(src, mode, &mut \|span, result\| callback(span, result.map(drop)))
1152		})
1153		}
1154
1155		fn cook_mixed(
1156		&self,
1157		kind: token::LitKind,
1158		mode: Mode,
1159		start: BytePos,
1160		end: BytePos,
1161		prefix_len: u32,
1162		postfix_len: u32,
1163		) -> (token::LitKind, Symbol) {
1164		self.cook_common(kind, mode, start, end, prefix_len, postfix_len, \|src, mode, callback\| {
1165		unescape_mixed(src, mode, &mut \|span, result\| callback(span, result.map(drop)))
1166		})
1167		}
1168	1138	}
1169	1139
1170	1140	pub fn nfc_normalize(string: &str) -> Symbol {

compiler/rustc_parse_format/Cargo.toml+1-1

...	...	@@ -5,7 +5,7 @@ edition = "2024"
5	5
6	6	[dependencies]
7	7	# tidy-alphabetical-start
8		rustc-literal-escaper = "0.0.2"
	8	rustc-literal-escaper = "0.0.4"
9	9	rustc_lexer = { path = "../rustc_lexer" }
10	10	# tidy-alphabetical-end
11	11

compiler/rustc_parse_format/src/lib.rs+1-2

...	...	@@ -20,7 +20,6 @@ use std::ops::Range;
20	20	pub use Alignment::*;
21	21	pub use Count::*;
22	22	pub use Position::*;
23		use rustc_literal_escaper::{Mode, unescape_unicode};
24	23
25	24	/// The type of format string that we are parsing.
26	25	#[derive(Copy, Clone, Debug, Eq, PartialEq)]
...	...	@@ -320,7 +319,7 @@ impl<'input> Parser<'input> {
320	319	let without_quotes = &snippet[1..snippet.len() - 1];
321	320	let (mut ok, mut vec) = (true, vec![]);
322	321	let mut chars = input.chars();
323		unescape_unicode(without_quotes, Mode::Str, &mut \|range, res\| match res {
	322	rustc_literal_escaper::unescape_str(without_quotes, \|range, res\| match res {
324	323	Ok(ch) if ok && chars.next().is_some_and(\|c\| ch == c) => {
325	324	vec.push((range, ch));
326	325	}

compiler/rustc_proc_macro/Cargo.toml+1-1

...	...	@@ -15,7 +15,7 @@ test = false
15	15	doctest = false
16	16
17	17	[dependencies]
18		rustc-literal-escaper = "0.0.2"
	18	rustc-literal-escaper = "0.0.4"
19	19
20	20	[features]
21	21	rustc-dep-of-std = []

library/Cargo.lock+3-2

...	...	@@ -271,10 +271,11 @@ dependencies = [
271	271
272	272	[[package]]
273	273	name = "rustc-literal-escaper"
274		version = "0.0.2"
	274	version = "0.0.4"
275	275	source = "registry+https://github.com/rust-lang/crates.io-index"
276		checksum = "0041b6238913c41fe704213a4a9329e2f685a156d1781998128b4149c230ad04"
	276	checksum = "ab03008eb631b703dd16978282ae36c73282e7922fe101a4bd072a40ecea7b8b"
277	277	dependencies = [
	278	"rustc-std-workspace-core",
278	279	"rustc-std-workspace-std",
279	280	]
280	281

library/Cargo.toml+1-2

...	...	@@ -55,8 +55,7 @@ rustflags = ["-Cpanic=abort"]
55	55	rustflags = ["-Cpanic=abort"]
56	56
57	57	[patch.crates-io]
58		# See comments in `library/rustc-std-workspace-core/README.md` for what's going on
59		# here
	58	# See comments in `library/rustc-std-workspace-core/README.md` for what's going on here
60	59	rustc-std-workspace-core = { path = 'rustc-std-workspace-core' }
61	60	rustc-std-workspace-alloc = { path = 'rustc-std-workspace-alloc' }
62	61	rustc-std-workspace-std = { path = 'rustc-std-workspace-std' }

library/proc_macro/Cargo.toml+1-1

...	...	@@ -9,7 +9,7 @@ std = { path = "../std" }
9	9	# `core` when resolving doc links. Without this line a different `core` will be
10	10	# loaded from sysroot causing duplicate lang items and other similar errors.
11	11	core = { path = "../core" }
12		rustc-literal-escaper = { version = "0.0.2", features = ["rustc-dep-of-std"] }
	12	rustc-literal-escaper = { version = "0.0.4", features = ["rustc-dep-of-std"] }
13	13
14	14	[features]
15	15	default = ["rustc-dep-of-std"]

library/proc_macro/src/lib.rs+6-7

...	...	@@ -56,7 +56,7 @@ use std::{error, fmt};
56	56	pub use diagnostic::{Diagnostic, Level, MultiSpan};
57	57	#[unstable(feature = "proc_macro_value", issue = "136652")]
58	58	pub use rustc_literal_escaper::EscapeError;
59		use rustc_literal_escaper::{MixedUnit, Mode, byte_from_char, unescape_mixed, unescape_unicode};
	59	use rustc_literal_escaper::{MixedUnit, unescape_byte_str, unescape_c_str, unescape_str};
60	60	#[unstable(feature = "proc_macro_totokens", issue = "130977")]
61	61	pub use to_tokens::ToTokens;
62	62
...	...	@@ -1440,10 +1440,9 @@ impl Literal {
1440	1440	// Force-inlining here is aggressive but the closure is
1441	1441	// called on every char in the string, so it can be hot in
1442	1442	// programs with many long strings containing escapes.
1443		unescape_unicode(
	1443	unescape_str(
1444	1444	symbol,
1445		Mode::Str,
1446		&mut #[inline(always)]
	1445	#[inline(always)]
1447	1446	\|_, c\| match c {
1448	1447	Ok(c) => buf.push(c),
1449	1448	Err(err) => {
...	...	@@ -1472,7 +1471,7 @@ impl Literal {
1472	1471	let mut error = None;
1473	1472	let mut buf = Vec::with_capacity(symbol.len());
1474	1473
1475		unescape_mixed(symbol, Mode::CStr, &mut \|_span, c\| match c {
	1474	unescape_c_str(symbol, \|_span, c\| match c {
1476	1475	Ok(MixedUnit::Char(c)) => {
1477	1476	buf.extend_from_slice(c.encode_utf8(&mut [0; 4]).as_bytes())
1478	1477	}
...	...	@@ -1511,8 +1510,8 @@ impl Literal {
1511	1510	let mut buf = Vec::with_capacity(symbol.len());
1512	1511	let mut error = None;
1513	1512
1514		unescape_unicode(symbol, Mode::ByteStr, &mut \|_, c\| match c {
1515		Ok(c) => buf.push(byte_from_char(c)),
	1513	unescape_byte_str(symbol, \|_, res\| match res {
	1514	Ok(b) => buf.push(b),
1516	1515	Err(err) => {
1517	1516	if err.is_fatal() {
1518	1517	error = Some(ConversionErrorKind::FailedToUnescape(err));

src/tools/clippy/clippy_dev/src/update_lints.rs+2-1

...	...	@@ -2,6 +2,7 @@ use crate::utils::{
2	2	ErrAction, File, FileUpdater, RustSearcher, Token, UpdateMode, UpdateStatus, expect_action, update_text_region_fn,
3	3	};
4	4	use itertools::Itertools;
	5	use rustc_lexer::{LiteralKind, TokenKind, tokenize};
5	6	use std::collections::HashSet;
6	7	use std::fmt::Write;
7	8	use std::ops::Range;
...	...	@@ -342,7 +343,7 @@ fn parse_str_lit(s: &str) -> String {
342	343	.and_then(\|s\| s.strip_suffix('"'))
343	344	.unwrap_or_else(\|\| panic!("expected quoted string, found `{s}`"));
344	345	let mut res = String::with_capacity(s.len());
345		rustc_literal_escaper::unescape_unicode(s, mode, &mut \|_, ch\| {
	346	rustc_literal_escaper::unescape_str(s, \|range, ch\| {
346	347	if let Ok(ch) = ch {
347	348	res.push(ch);
348	349	}

src/tools/lint-docs/Cargo.toml+1-1

...	...	@@ -7,7 +7,7 @@ description = "A script to extract the lint documentation for the rustc book."
7	7	# See more keys and their definitions at https://doc.rust-lang.org/cargo/reference/manifest.html
8	8
9	9	[dependencies]
10		rustc-literal-escaper = "0.0.2"
	10	rustc-literal-escaper = "0.0.4"
11	11	serde_json = "1.0.57"
12	12	tempfile = "3.1.0"
13	13	walkdir = "2.3.1"

src/tools/lint-docs/src/lib.rs+2-2

...	...	@@ -4,7 +4,7 @@ use std::fs;
4	4	use std::path::{Path, PathBuf};
5	5	use std::process::Command;
6	6
7		use rustc_literal_escaper::{Mode, unescape_unicode};
	7	use rustc_literal_escaper::unescape_str;
8	8	use walkdir::WalkDir;
9	9
10	10	mod groups;
...	...	@@ -218,7 +218,7 @@ impl<'a> LintExtractor<'a> {
218	218	} else if let Some(text) = line.strip_prefix("#[doc = \"") {
219	219	let escaped = text.strip_suffix("\"]").unwrap();
220	220	let mut buf = String::new();
221		unescape_unicode(escaped, Mode::Str, &mut \|_, c\| match c {
	221	unescape_str(escaped, \|_, res\| match res {
222	222	Ok(c) => buf.push(c),
223	223	Err(err) => {
224	224	assert!(!err.is_fatal(), "failed to unescape string literal")

author	bors <bors@rust-lang.org> 2025-06-25 01:03:30 UTC
committer	bors <bors@rust-lang.org> 2025-06-25 01:03:30 UTC
log	2c2bb995af398383e3b93b859302bdc447ca7a7c
tree	b038512b5d792f44f554b823a9db293ba93c7697
parent	3de5b08ef6b260277dd4c77f7472fe6904bd6002
parent	707a6f54631c322e8c8ccff363fe024d67d93aa2

Auto merge of #140999 - hkBst:update-escaper, r=nnethercote