Skip to content

update to literal-escaper 0.0.4 for better API without unreachable and faster string parsing #140999

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
wants to merge 1 commit into
base: master
Choose a base branch
from
+272 −273
Open

update to literal-escaper 0.0.4 for better API without unreachable and faster string parsing #140999

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 2 additions & 2 deletions Cargo.lock
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3166,9 +3166,9 @@ checksum = "357703d41365b4b27c590e3ed91eabb1b663f07c4c084095e60cbed4362dff0d"

[[package]]
name = "rustc-literal-escaper"
version = "0.0.2"
version = "0.0.4"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "0041b6238913c41fe704213a4a9329e2f685a156d1781998128b4149c230ad04"
checksum = "ab03008eb631b703dd16978282ae36c73282e7922fe101a4bd072a40ecea7b8b"

[[package]]
name = "rustc-main"
Expand Down
3 changes: 3 additions & 0 deletions Cargo.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -89,3 +89,6 @@ codegen-units = 1
# FIXME: LTO cannot be enabled for binaries in a workspace
# <https://github.com/rust-lang/cargo/issues/9330>
# lto = true

[patch.crates-io]
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Can be removed.

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

True, but it may be nice to have anyway, like in src/tools/rust-analyzer/Cargo.toml

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Why? If it's not in use it should not be there. You can leave a comment explaining what to use in case a similar situation happen in the future though.

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Yeah, should be dropped, but I'd prefer these parts to be filed as a PR in the rust-analyzer repo. I don't think it will cause issues except maybe a small, temporary, build slowdown (since it's compiling two versions).

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Would it make a difference if I comment out "[patch.crates-io]"?

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I guess it's fine but again: please add an explanation why this is kept around.

# rustc-literal-escaper = { path = '../literal-escaper/' }
2 changes: 1 addition & 1 deletion compiler/rustc_ast/Cargo.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -7,7 +7,7 @@ edition = "2024"
# tidy-alphabetical-start
bitflags = "2.4.1"
memchr = "2.7.4"
rustc-literal-escaper = "0.0.2"
rustc-literal-escaper = "0.0.4"
rustc_ast_ir = { path = "../rustc_ast_ir" }
rustc_data_structures = { path = "../rustc_data_structures" }
rustc_index = { path = "../rustc_index" }
Expand Down
15 changes: 7 additions & 8 deletions compiler/rustc_ast/src/util/literal.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3,7 +3,7 @@
use std::{ascii, fmt, str};

use rustc_literal_escaper::{
MixedUnit, Mode, byte_from_char, unescape_byte, unescape_char, unescape_mixed, unescape_unicode,
MixedUnit, unescape_byte, unescape_byte_str, unescape_c_str, unescape_char, unescape_str,
};
use rustc_span::{Span, Symbol, kw, sym};
use tracing::debug;
Expand Down Expand Up @@ -87,11 +87,10 @@ impl LitKind {
// Force-inlining here is aggressive but the closure is
// called on every char in the string, so it can be hot in
// programs with many long strings containing escapes.
unescape_unicode(
unescape_str(
s,
Mode::Str,
&mut #[inline(always)]
|_, c| match c {
#[inline(always)]
|_, res| match res {
Ok(c) => buf.push(c),
Err(err) => {
assert!(!err.is_fatal(), "failed to unescape string literal")
Expand All @@ -111,8 +110,8 @@ impl LitKind {
token::ByteStr => {
let s = symbol.as_str();
let mut buf = Vec::with_capacity(s.len());
unescape_unicode(s, Mode::ByteStr, &mut |_, c| match c {
Ok(c) => buf.push(byte_from_char(c)),
unescape_byte_str(s, |_, res| match res {
Ok(b) => buf.push(b),
Err(err) => {
assert!(!err.is_fatal(), "failed to unescape string literal")
}
Expand All @@ -128,7 +127,7 @@ impl LitKind {
token::CStr => {
let s = symbol.as_str();
let mut buf = Vec::with_capacity(s.len());
unescape_mixed(s, Mode::CStr, &mut |_span, c| match c {
unescape_c_str(s, |_span, c| match c {
Ok(MixedUnit::Char(c)) => {
buf.extend_from_slice(c.encode_utf8(&mut [0; 4]).as_bytes())
}
Expand Down
2 changes: 1 addition & 1 deletion compiler/rustc_parse/Cargo.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -6,7 +6,7 @@ edition = "2024"
[dependencies]
# tidy-alphabetical-start
bitflags = "2.4.1"
rustc-literal-escaper = "0.0.2"
rustc-literal-escaper = "0.0.4"
rustc_ast = { path = "../rustc_ast" }
rustc_ast_pretty = { path = "../rustc_ast_pretty" }
rustc_data_structures = { path = "../rustc_data_structures" }
Expand Down
94 changes: 32 additions & 62 deletions compiler/rustc_parse/src/lexer/mod.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,5 +1,3 @@
use std::ops::Range;

use diagnostics::make_unclosed_delims_error;
use rustc_ast::ast::{self, AttrStyle};
use rustc_ast::token::{self, CommentKind, Delimiter, IdentIsRaw, Token, TokenKind};
Expand All @@ -10,7 +8,7 @@ use rustc_errors::{Applicability, Diag, DiagCtxtHandle, StashKey};
use rustc_lexer::{
Base, Cursor, DocStyle, FrontmatterAllowed, LiteralKind, RawStrError, is_whitespace,
};
use rustc_literal_escaper::{EscapeError, Mode, unescape_mixed, unescape_unicode};
use rustc_literal_escaper::{EscapeError, Mode, check_for_errors};
use rustc_session::lint::BuiltinLintDiag;
use rustc_session::lint::builtin::{
RUST_2021_PREFIXES_INCOMPATIBLE_SYNTAX, RUST_2024_GUARDED_STRING_INCOMPATIBLE_SYNTAX,
Expand Down Expand Up @@ -702,7 +700,7 @@ impl<'psess, 'src> Lexer<'psess, 'src> {
}
err.emit()
}
self.cook_unicode(token::Char, Mode::Char, start, end, 1, 1) // ' '
self.cook_quoted(token::Char, Mode::Char, start, end, 1, 1) // ' '
}
rustc_lexer::LiteralKind::Byte { terminated } => {
if !terminated {
Expand All @@ -714,7 +712,7 @@ impl<'psess, 'src> Lexer<'psess, 'src> {
.with_code(E0763)
.emit()
}
self.cook_unicode(token::Byte, Mode::Byte, start, end, 2, 1) // b' '
self.cook_quoted(token::Byte, Mode::Byte, start, end, 2, 1) // b' '
}
rustc_lexer::LiteralKind::Str { terminated } => {
if !terminated {
Expand All @@ -726,7 +724,7 @@ impl<'psess, 'src> Lexer<'psess, 'src> {
.with_code(E0765)
.emit()
}
self.cook_unicode(token::Str, Mode::Str, start, end, 1, 1) // " "
self.cook_quoted(token::Str, Mode::Str, start, end, 1, 1) // " "
}
rustc_lexer::LiteralKind::ByteStr { terminated } => {
if !terminated {
Expand All @@ -738,7 +736,8 @@ impl<'psess, 'src> Lexer<'psess, 'src> {
.with_code(E0766)
.emit()
}
self.cook_unicode(token::ByteStr, Mode::ByteStr, start, end, 2, 1) // b" "
self.cook_quoted(token::ByteStr, Mode::ByteStr, start, end, 2, 1)
// b" "
}
rustc_lexer::LiteralKind::CStr { terminated } => {
if !terminated {
Expand All @@ -750,13 +749,14 @@ impl<'psess, 'src> Lexer<'psess, 'src> {
.with_code(E0767)
.emit()
}
self.cook_mixed(token::CStr, Mode::CStr, start, end, 2, 1) // c" "
self.cook_quoted(token::CStr, Mode::CStr, start, end, 2, 1) // c" "
}
rustc_lexer::LiteralKind::RawStr { n_hashes } => {
if let Some(n_hashes) = n_hashes {
let n = u32::from(n_hashes);
let kind = token::StrRaw(n_hashes);
self.cook_unicode(kind, Mode::RawStr, start, end, 2 + n, 1 + n) // r##" "##
self.cook_quoted(kind, Mode::RawStr, start, end, 2 + n, 1 + n)
// r##" "##
} else {
self.report_raw_str_error(start, 1);
}
Expand All @@ -765,7 +765,8 @@ impl<'psess, 'src> Lexer<'psess, 'src> {
if let Some(n_hashes) = n_hashes {
let n = u32::from(n_hashes);
let kind = token::ByteStrRaw(n_hashes);
self.cook_unicode(kind, Mode::RawByteStr, start, end, 3 + n, 1 + n) // br##" "##
self.cook_quoted(kind, Mode::RawByteStr, start, end, 3 + n, 1 + n)
// br##" "##
} else {
self.report_raw_str_error(start, 2);
}
Expand All @@ -774,7 +775,8 @@ impl<'psess, 'src> Lexer<'psess, 'src> {
if let Some(n_hashes) = n_hashes {
let n = u32::from(n_hashes);
let kind = token::CStrRaw(n_hashes);
self.cook_unicode(kind, Mode::RawCStr, start, end, 3 + n, 1 + n) // cr##" "##
self.cook_quoted(kind, Mode::RawCStr, start, end, 3 + n, 1 + n)
// cr##" "##
} else {
self.report_raw_str_error(start, 2);
}
Expand Down Expand Up @@ -1091,40 +1093,36 @@ impl<'psess, 'src> Lexer<'psess, 'src> {
self.dcx().emit_fatal(errors::TooManyHashes { span: self.mk_sp(start, self.pos), num });
}

fn cook_common(
fn cook_quoted(
&self,
mut kind: token::LitKind,
mode: Mode,
start: BytePos,
end: BytePos,
prefix_len: u32,
postfix_len: u32,
unescape: fn(&str, Mode, &mut dyn FnMut(Range<usize>, Result<(), EscapeError>)),
) -> (token::LitKind, Symbol) {
let content_start = start + BytePos(prefix_len);
let content_end = end - BytePos(postfix_len);
let lit_content = self.str_from_to(content_start, content_end);
unescape(lit_content, mode, &mut |range, result| {
// Here we only check for errors. The actual unescaping is done later.
if let Err(err) = result {
let span_with_quotes = self.mk_sp(start, end);
let (start, end) = (range.start as u32, range.end as u32);
let lo = content_start + BytePos(start);
let hi = lo + BytePos(end - start);
let span = self.mk_sp(lo, hi);
let is_fatal = err.is_fatal();
if let Some(guar) = emit_unescape_error(
self.dcx(),
lit_content,
span_with_quotes,
span,
mode,
range,
err,
) {
assert!(is_fatal);
kind = token::Err(guar);
}
check_for_errors(lit_content, mode, |range, err| {
let span_with_quotes = self.mk_sp(start, end);
let (start, end) = (range.start as u32, range.end as u32);
let lo = content_start + BytePos(start);
let hi = lo + BytePos(end - start);
let span = self.mk_sp(lo, hi);
let is_fatal = err.is_fatal();
if let Some(guar) = emit_unescape_error(
self.dcx(),
lit_content,
span_with_quotes,
span,
mode,
range,
err,
) {
assert!(is_fatal);
kind = token::Err(guar);
}
});

Expand All @@ -1137,34 +1135,6 @@ impl<'psess, 'src> Lexer<'psess, 'src> {
};
(kind, sym)
}

fn cook_unicode(
&self,
kind: token::LitKind,
mode: Mode,
start: BytePos,
end: BytePos,
prefix_len: u32,
postfix_len: u32,
) -> (token::LitKind, Symbol) {
self.cook_common(kind, mode, start, end, prefix_len, postfix_len, |src, mode, callback| {
unescape_unicode(src, mode, &mut |span, result| callback(span, result.map(drop)))
})
}

fn cook_mixed(
&self,
kind: token::LitKind,
mode: Mode,
start: BytePos,
end: BytePos,
prefix_len: u32,
postfix_len: u32,
) -> (token::LitKind, Symbol) {
self.cook_common(kind, mode, start, end, prefix_len, postfix_len, |src, mode, callback| {
unescape_mixed(src, mode, &mut |span, result| callback(span, result.map(drop)))
})
}
}

pub fn nfc_normalize(string: &str) -> Symbol {
Expand Down
2 changes: 1 addition & 1 deletion compiler/rustc_parse_format/Cargo.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -5,7 +5,7 @@ edition = "2024"

[dependencies]
# tidy-alphabetical-start
rustc-literal-escaper = "0.0.2"
rustc-literal-escaper = "0.0.4"
rustc_lexer = { path = "../rustc_lexer" }
# tidy-alphabetical-end

Expand Down
3 changes: 1 addition & 2 deletions compiler/rustc_parse_format/src/lib.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -20,7 +20,6 @@ use std::ops::Range;
pub use Alignment::*;
pub use Count::*;
pub use Position::*;
use rustc_literal_escaper::{Mode, unescape_unicode};

/// The type of format string that we are parsing.
#[derive(Copy, Clone, Debug, Eq, PartialEq)]
Expand Down Expand Up @@ -320,7 +319,7 @@ impl<'input> Parser<'input> {
let without_quotes = &snippet[1..snippet.len() - 1];
let (mut ok, mut vec) = (true, vec![]);
let mut chars = input.chars();
unescape_unicode(without_quotes, Mode::Str, &mut |range, res| match res {
rustc_literal_escaper::unescape_str(without_quotes, |range, res| match res {
Ok(ch) if ok && chars.next().is_some_and(|c| ch == c) => {
vec.push((range, ch));
}
Expand Down
2 changes: 1 addition & 1 deletion compiler/rustc_proc_macro/Cargo.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -15,7 +15,7 @@ test = false
doctest = false

[dependencies]
rustc-literal-escaper = "0.0.2"
rustc-literal-escaper = "0.0.4"

[features]
rustc-dep-of-std = []
5 changes: 3 additions & 2 deletions library/Cargo.lock
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -283,10 +283,11 @@ dependencies = [

[[package]]
name = "rustc-literal-escaper"
version = "0.0.2"
version = "0.0.4"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "0041b6238913c41fe704213a4a9329e2f685a156d1781998128b4149c230ad04"
checksum = "ab03008eb631b703dd16978282ae36c73282e7922fe101a4bd072a40ecea7b8b"
dependencies = [
"rustc-std-workspace-core",
"rustc-std-workspace-std",
]

Expand Down
5 changes: 3 additions & 2 deletions library/Cargo.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -45,9 +45,10 @@ rustc-demangle.debug = 0
rustc-demangle.opt-level = "s"

[patch.crates-io]
# See comments in `library/rustc-std-workspace-core/README.md` for what's going on
# here
# See comments in `library/rustc-std-workspace-core/README.md` for what's going on here
rustc-std-workspace-core = { path = 'rustc-std-workspace-core' }
rustc-std-workspace-alloc = { path = 'rustc-std-workspace-alloc' }
rustc-std-workspace-std = { path = 'rustc-std-workspace-std' }
compiler_builtins = { path = "compiler-builtins/compiler-builtins" }

# rustc-literal-escaper = { path = '../../literal-escaper/' }
2 changes: 1 addition & 1 deletion library/proc_macro/Cargo.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -9,7 +9,7 @@ std = { path = "../std" }
# `core` when resolving doc links. Without this line a different `core` will be
# loaded from sysroot causing duplicate lang items and other similar errors.
core = { path = "../core" }
rustc-literal-escaper = { version = "0.0.2", features = ["rustc-dep-of-std"] }
rustc-literal-escaper = { version = "0.0.4", features = ["rustc-dep-of-std"] }

[features]
default = ["rustc-dep-of-std"]
Expand Down
13 changes: 6 additions & 7 deletions library/proc_macro/src/lib.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -55,7 +55,7 @@ use std::{error, fmt};
pub use diagnostic::{Diagnostic, Level, MultiSpan};
#[unstable(feature = "proc_macro_value", issue = "136652")]
pub use rustc_literal_escaper::EscapeError;
use rustc_literal_escaper::{MixedUnit, Mode, byte_from_char, unescape_mixed, unescape_unicode};
use rustc_literal_escaper::{MixedUnit, unescape_byte_str, unescape_c_str, unescape_str};
#[unstable(feature = "proc_macro_totokens", issue = "130977")]
pub use to_tokens::ToTokens;

Expand Down Expand Up @@ -1439,10 +1439,9 @@ impl Literal {
// Force-inlining here is aggressive but the closure is
// called on every char in the string, so it can be hot in
// programs with many long strings containing escapes.
unescape_unicode(
unescape_str(
symbol,
Mode::Str,
&mut #[inline(always)]
#[inline(always)]
|_, c| match c {
Ok(c) => buf.push(c),
Err(err) => {
Expand Down Expand Up @@ -1471,7 +1470,7 @@ impl Literal {
let mut error = None;
let mut buf = Vec::with_capacity(symbol.len());

unescape_mixed(symbol, Mode::CStr, &mut |_span, c| match c {
unescape_c_str(symbol, |_span, c| match c {
Ok(MixedUnit::Char(c)) => {
buf.extend_from_slice(c.encode_utf8(&mut [0; 4]).as_bytes())
}
Expand Down Expand Up @@ -1510,8 +1509,8 @@ impl Literal {
let mut buf = Vec::with_capacity(symbol.len());
let mut error = None;

unescape_unicode(symbol, Mode::ByteStr, &mut |_, c| match c {
Ok(c) => buf.push(byte_from_char(c)),
unescape_byte_str(symbol, |_, res| match res {
Ok(b) => buf.push(b),
Err(err) => {
if err.is_fatal() {
error = Some(ConversionErrorKind::FailedToUnescape(err));
Expand Down
3 changes: 2 additions & 1 deletion src/tools/clippy/clippy_dev/src/update_lints.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,6 +2,7 @@ use crate::utils::{
ErrAction, File, FileUpdater, RustSearcher, Token, UpdateMode, UpdateStatus, expect_action, update_text_region_fn,
};
use itertools::Itertools;
use rustc_lexer::{LiteralKind, TokenKind, tokenize};
use std::collections::HashSet;
use std::fmt::Write;
use std::ops::Range;
Expand Down Expand Up @@ -342,7 +343,7 @@ fn parse_str_lit(s: &str) -> String {
.and_then(|s| s.strip_suffix('"'))
.unwrap_or_else(|| panic!("expected quoted string, found `{s}`"));
let mut res = String::with_capacity(s.len());
rustc_literal_escaper::unescape_unicode(s, mode, &mut |_, ch| {
rustc_literal_escaper::unescape_str(s, |range, ch| {
if let Ok(ch) = ch {
res.push(ch);
}
Expand Down
Loading
Loading