1
Fork 0
mirror of https://github.com/RGBCube/uutils-coreutils synced 2025-07-28 03:27:44 +00:00

Merge pull request #7227 from cakebaker/bump_nom

Bump `nom` and adapt `tr` to API changes
This commit is contained in:
Sylvestre Ledru 2025-01-27 11:11:52 +01:00 committed by GitHub
commit 7747351cb1
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
4 changed files with 59 additions and 36 deletions

25
Cargo.lock generated
View file

@ -276,7 +276,7 @@ version = "0.6.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "6fac387a98bb7c37292057cffc56d62ecb629900026402633ae9160df93a8766"
dependencies = [
"nom",
"nom 7.1.3",
]
[[package]]
@ -861,7 +861,7 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "33d852cb9b869c2a9b3df2f71a3074817f01e1844f839a144f5fcef059a4eb5d"
dependencies = [
"libc",
"windows-sys 0.52.0",
"windows-sys 0.59.0",
]
[[package]]
@ -1276,7 +1276,7 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "fc2f4eb4bc735547cfed7c0a4922cbd04a4655978c09b54f1f7b228750664c34"
dependencies = [
"cfg-if",
"windows-targets 0.48.5",
"windows-targets 0.52.6",
]
[[package]]
@ -1423,6 +1423,15 @@ dependencies = [
"minimal-lexical",
]
[[package]]
name = "nom"
version = "8.0.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "df9761775871bdef83bee530e60050f7e54b1105350d6884eb0fb4f46c2f9405"
dependencies = [
"memchr",
]
[[package]]
name = "notify"
version = "8.0.0"
@ -1620,7 +1629,7 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "ae130e79b384861c193d6016a46baa2733a6f8f17486eb36a5c098c577ce01e8"
dependencies = [
"chrono",
"nom",
"nom 7.1.3",
"regex",
]
@ -1996,7 +2005,7 @@ dependencies = [
"errno",
"libc",
"linux-raw-sys 0.4.15",
"windows-sys 0.52.0",
"windows-sys 0.59.0",
]
[[package]]
@ -2241,7 +2250,7 @@ dependencies = [
"getrandom",
"once_cell",
"rustix 0.38.43",
"windows-sys 0.52.0",
"windows-sys 0.59.0",
]
[[package]]
@ -3325,7 +3334,7 @@ name = "uu_tr"
version = "0.0.29"
dependencies = [
"clap",
"nom",
"nom 8.0.0",
"uucore",
]
@ -3660,7 +3669,7 @@ version = "0.1.9"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "cf221c93e13a30d793f7645a0e7762c55d169dbb0a49671918a2319d289b10bb"
dependencies = [
"windows-sys 0.48.0",
"windows-sys 0.59.0",
]
[[package]]

View file

@ -308,7 +308,7 @@ lscolors = { version = "0.20.0", default-features = false, features = [
memchr = "2.7.2"
memmap2 = "0.9.4"
nix = { version = "0.29", default-features = false }
nom = "7.1.3"
nom = "8.0.0"
notify = { version = "=8.0.0", features = ["macos_kqueue"] }
num-bigint = "0.4.4"
num-prime = "0.4.4"

View file

@ -92,6 +92,8 @@ skip = [
{ name = "itertools", version = "0.13.0" },
# indexmap
{ name = "hashbrown", version = "0.14.5" },
# parse_datetime, cexpr (via bindgen)
{ name = "nom", version = "7.1.3" },
]
# spell-checker: enable

View file

@ -13,7 +13,7 @@ use nom::{
combinator::{map, map_opt, peek, recognize, value},
multi::{many0, many_m_n},
sequence::{delimited, preceded, separated_pair},
IResult,
IResult, Parser,
};
use std::{
char,
@ -298,7 +298,8 @@ impl Sequence {
map(Self::parse_backslash_or_char_with_warning, |s| {
Ok(Self::Char(s))
}),
)))(input)
)))
.parse(input)
.map(|(_, r)| r)
.unwrap()
.into_iter()
@ -308,7 +309,7 @@ impl Sequence {
fn parse_octal(input: &[u8]) -> IResult<&[u8], u8> {
// For `parse_char_range`, `parse_char_star`, `parse_char_repeat`, `parse_char_equal`.
// Because in these patterns, there's no ambiguous cases.
preceded(tag("\\"), Self::parse_octal_up_to_three_digits)(input)
preceded(tag("\\"), Self::parse_octal_up_to_three_digits).parse(input)
}
fn parse_octal_with_warning(input: &[u8]) -> IResult<&[u8], u8> {
@ -321,7 +322,8 @@ impl Sequence {
// See test `test_multibyte_octal_sequence`
Self::parse_octal_two_digits,
)),
)(input)
)
.parse(input)
}
fn parse_octal_up_to_three_digits(input: &[u8]) -> IResult<&[u8], u8> {
@ -331,7 +333,8 @@ impl Sequence {
let str_to_parse = std::str::from_utf8(out).unwrap();
u8::from_str_radix(str_to_parse, 8).ok()
},
)(input)
)
.parse(input)
}
fn parse_octal_up_to_three_digits_with_warning(input: &[u8]) -> IResult<&[u8], u8> {
@ -353,34 +356,37 @@ impl Sequence {
}
result
},
)(input)
).parse(input)
}
fn parse_octal_two_digits(input: &[u8]) -> IResult<&[u8], u8> {
map_opt(
recognize(many_m_n(2, 2, one_of("01234567"))),
|out: &[u8]| u8::from_str_radix(std::str::from_utf8(out).unwrap(), 8).ok(),
)(input)
)
.parse(input)
}
fn parse_backslash(input: &[u8]) -> IResult<&[u8], u8> {
preceded(tag("\\"), Self::single_char)(input).map(|(l, a)| {
let c = match a {
b'a' => unicode_table::BEL,
b'b' => unicode_table::BS,
b'f' => unicode_table::FF,
b'n' => unicode_table::LF,
b'r' => unicode_table::CR,
b't' => unicode_table::HT,
b'v' => unicode_table::VT,
x => x,
};
(l, c)
})
preceded(tag("\\"), Self::single_char)
.parse(input)
.map(|(l, a)| {
let c = match a {
b'a' => unicode_table::BEL,
b'b' => unicode_table::BS,
b'f' => unicode_table::FF,
b'n' => unicode_table::LF,
b'r' => unicode_table::CR,
b't' => unicode_table::HT,
b'v' => unicode_table::VT,
x => x,
};
(l, c)
})
}
fn parse_backslash_or_char(input: &[u8]) -> IResult<&[u8], u8> {
alt((Self::parse_octal, Self::parse_backslash, Self::single_char))(input)
alt((Self::parse_octal, Self::parse_backslash, Self::single_char)).parse(input)
}
fn parse_backslash_or_char_with_warning(input: &[u8]) -> IResult<&[u8], u8> {
@ -388,7 +394,8 @@ impl Sequence {
Self::parse_octal_with_warning,
Self::parse_backslash,
Self::single_char,
))(input)
))
.parse(input)
}
fn single_char(input: &[u8]) -> IResult<&[u8], u8> {
@ -400,7 +407,8 @@ impl Sequence {
Self::parse_backslash_or_char,
tag("-"),
Self::parse_backslash_or_char,
)(input)
)
.parse(input)
.map(|(l, (a, b))| {
(l, {
let (start, end) = (u32::from(a), u32::from(b));
@ -417,7 +425,8 @@ impl Sequence {
}
fn parse_char_star(input: &[u8]) -> IResult<&[u8], Result<Self, BadSequence>> {
delimited(tag("["), Self::parse_backslash_or_char, tag("*]"))(input)
delimited(tag("["), Self::parse_backslash_or_char, tag("*]"))
.parse(input)
.map(|(l, a)| (l, Ok(Self::CharStar(a))))
}
@ -433,7 +442,8 @@ impl Sequence {
take_till(|ue| matches!(ue, b']' | b'\\')),
),
tag("]"),
)(input)
)
.parse(input)
.map(|(l, (c, cnt_str))| {
let s = String::from_utf8_lossy(cnt_str);
let result = if cnt_str.starts_with(b"0") {
@ -477,7 +487,8 @@ impl Sequence {
value(Err(BadSequence::MissingCharClassName), tag("")),
)),
tag(":]"),
)(input)
)
.parse(input)
}
fn parse_char_equal(input: &[u8]) -> IResult<&[u8], Result<Self, BadSequence>> {
@ -491,7 +502,8 @@ impl Sequence {
map(Self::parse_backslash_or_char, |c| Ok(Self::Char(c))),
)),
tag("=]"),
)(input)
)
.parse(input)
}
}