chore: checkpoint before Python removal

This commit is contained in:
2026-03-26 22:33:59 +00:00
parent 683cec9307
commit e568ddf82a
29972 changed files with 11269302 additions and 2 deletions

View File

@@ -0,0 +1,231 @@
//! General purpose combinators
use nom::bytes::streaming::take;
use nom::combinator::map_parser;
use nom::error::{make_error, ErrorKind, ParseError};
use nom::{IResult, Needed, Parser};
use nom::{InputIter, InputTake};
use nom::{InputLength, ToUsize};
#[deprecated(since = "3.0.1", note = "please use `be_var_u64` instead")]
/// Read an entire slice as a big-endian value.
///
/// Returns the value as `u64`. This function checks for integer overflows, and returns a
/// `Result::Err` value if the value is too big.
pub fn bytes_to_u64(s: &[u8]) -> Result<u64, &'static str> {
let mut u: u64 = 0;
if s.is_empty() {
return Err("empty");
};
if s.len() > 8 {
return Err("overflow");
}
for &c in s {
let u1 = u << 8;
u = u1 | (c as u64);
}
Ok(u)
}
/// Read the entire slice as a big endian unsigned integer, up to 8 bytes
#[inline]
pub fn be_var_u64<'a, E: ParseError<&'a [u8]>>(input: &'a [u8]) -> IResult<&'a [u8], u64, E> {
if input.is_empty() {
return Err(nom::Err::Incomplete(Needed::new(1)));
}
if input.len() > 8 {
return Err(nom::Err::Error(make_error(input, ErrorKind::TooLarge)));
}
let mut res = 0u64;
for byte in input {
res = (res << 8) + *byte as u64;
}
Ok((&b""[..], res))
}
/// Read the entire slice as a little endian unsigned integer, up to 8 bytes
#[inline]
pub fn le_var_u64<'a, E: ParseError<&'a [u8]>>(input: &'a [u8]) -> IResult<&'a [u8], u64, E> {
if input.is_empty() {
return Err(nom::Err::Incomplete(Needed::new(1)));
}
if input.len() > 8 {
return Err(nom::Err::Error(make_error(input, ErrorKind::TooLarge)));
}
let mut res = 0u64;
for byte in input.iter().rev() {
res = (res << 8) + *byte as u64;
}
Ok((&b""[..], res))
}
/// Read a slice as a big-endian value.
#[inline]
pub fn parse_hex_to_u64<S>(i: &[u8], size: S) -> IResult<&[u8], u64>
where
S: ToUsize + Copy,
{
map_parser(take(size.to_usize()), be_var_u64)(i)
}
/// Apply combinator, automatically converts between errors if the underlying type supports it
pub fn upgrade_error<I, O, E1: ParseError<I>, E2: ParseError<I>, F>(
mut f: F,
) -> impl FnMut(I) -> IResult<I, O, E2>
where
F: FnMut(I) -> IResult<I, O, E1>,
E2: From<E1>,
{
move |i| f(i).map_err(nom::Err::convert)
}
/// Create a combinator that returns the provided value, and input unchanged
pub fn pure<I, O, E: ParseError<I>>(val: O) -> impl Fn(I) -> IResult<I, O, E>
where
O: Clone,
{
move |input: I| Ok((input, val.clone()))
}
/// Return a closure that takes `len` bytes from input, and applies `parser`.
pub fn flat_take<I, C, O, E: ParseError<I>, F>(
len: C,
mut parser: F,
) -> impl FnMut(I) -> IResult<I, O, E>
where
I: InputTake + InputLength + InputIter,
C: ToUsize + Copy,
F: Parser<I, O, E>,
{
// Note: this is the same as `map_parser(take(len), parser)`
move |input: I| {
let (input, o1) = take(len.to_usize())(input)?;
let (_, o2) = parser.parse(o1)?;
Ok((input, o2))
}
}
/// Take `len` bytes from `input`, and apply `parser`.
pub fn flat_takec<I, O, E: ParseError<I>, C, F>(input: I, len: C, parser: F) -> IResult<I, O, E>
where
C: ToUsize + Copy,
F: Parser<I, O, E>,
I: InputTake + InputLength + InputIter,
O: InputLength,
{
flat_take(len, parser)(input)
}
/// Helper macro for nom parsers: run first parser if condition is true, else second parser
pub fn cond_else<I, O, E: ParseError<I>, C, F, G>(
cond: C,
mut first: F,
mut second: G,
) -> impl FnMut(I) -> IResult<I, O, E>
where
C: Fn() -> bool,
F: Parser<I, O, E>,
G: Parser<I, O, E>,
{
move |input: I| {
if cond() {
first.parse(input)
} else {
second.parse(input)
}
}
}
/// Align input value to the next multiple of n bytes
/// Valid only if n is a power of 2
pub const fn align_n2(x: usize, n: usize) -> usize {
(x + (n - 1)) & !(n - 1)
}
/// Align input value to the next multiple of 4 bytes
pub const fn align32(x: usize) -> usize {
(x + 3) & !3
}
#[cfg(test)]
mod tests {
use super::{align32, be_var_u64, cond_else, flat_take, pure};
use nom::bytes::streaming::take;
use nom::number::streaming::{be_u16, be_u32, be_u8};
use nom::{Err, IResult, Needed};
#[test]
fn test_be_var_u64() {
let res: IResult<&[u8], u64> = be_var_u64(b"\x12\x34\x56");
let (_, v) = res.expect("be_var_u64 failed");
assert_eq!(v, 0x123456);
}
#[test]
fn test_flat_take() {
let input = &[0x00, 0x01, 0xff];
// read first 2 bytes and use correct combinator: OK
let res: IResult<&[u8], u16> = flat_take(2u8, be_u16)(input);
assert_eq!(res, Ok((&input[2..], 0x0001)));
// read 3 bytes and use 2: OK (some input is just lost)
let res: IResult<&[u8], u16> = flat_take(3u8, be_u16)(input);
assert_eq!(res, Ok((&b""[..], 0x0001)));
// read 2 bytes and a combinator requiring more bytes
let res: IResult<&[u8], u32> = flat_take(2u8, be_u32)(input);
assert_eq!(res, Err(Err::Incomplete(Needed::new(2))));
}
#[test]
fn test_flat_take_str() {
let input = "abcdef";
// read first 2 bytes and use correct combinator: OK
let res: IResult<&str, &str> = flat_take(2u8, take(2u8))(input);
assert_eq!(res, Ok(("cdef", "ab")));
// read 3 bytes and use 2: OK (some input is just lost)
let res: IResult<&str, &str> = flat_take(3u8, take(2u8))(input);
assert_eq!(res, Ok(("def", "ab")));
// read 2 bytes and a use combinator requiring more bytes
let res: IResult<&str, &str> = flat_take(2u8, take(4u8))(input);
assert_eq!(res, Err(Err::Incomplete(Needed::Unknown)));
}
#[test]
fn test_cond_else() {
let input = &[0x01][..];
let empty = &b""[..];
let a = 1;
fn parse_u8(i: &[u8]) -> IResult<&[u8], u8> {
be_u8(i)
}
assert_eq!(
cond_else(|| a == 1, parse_u8, pure(0x02))(input),
Ok((empty, 0x01))
);
assert_eq!(
cond_else(|| a == 1, parse_u8, pure(0x02))(input),
Ok((empty, 0x01))
);
assert_eq!(
cond_else(|| a == 2, parse_u8, pure(0x02))(input),
Ok((input, 0x02))
);
assert_eq!(
cond_else(|| a == 1, pure(0x02), parse_u8)(input),
Ok((input, 0x02))
);
let res: IResult<&[u8], u8> = cond_else(|| a == 1, parse_u8, parse_u8)(input);
assert_eq!(res, Ok((empty, 0x01)));
}
#[test]
fn test_align32() {
assert_eq!(align32(3), 4);
assert_eq!(align32(4), 4);
assert_eq!(align32(5), 8);
assert_eq!(align32(5usize), 8);
}
}

62
vendor/rusticata-macros/src/debug.rs vendored Normal file
View File

@@ -0,0 +1,62 @@
//! Helper functions and structures for debugging purpose
use nom::combinator::{map, peek, rest};
use nom::HexDisplay;
use nom::IResult;
use std::fmt;
/// Dump the remaining bytes to stderr, formatted as hex
pub fn dbg_dmp_rest(i: &[u8]) -> IResult<&[u8], ()> {
map(peek(rest), |r: &[u8]| eprintln!("\n{}\n", r.to_hex(16)))(i)
}
/// Wrapper for printing value as u8 hex data
pub struct HexU8(pub u8);
impl fmt::Debug for HexU8 {
fn fmt(&self, fmt: &mut fmt::Formatter) -> fmt::Result {
write!(fmt, "0x{:02x}", self.0)
}
}
/// Wrapper for printing value as u16 hex data
pub struct HexU16(pub u16);
impl fmt::Debug for HexU16 {
fn fmt(&self, fmt: &mut fmt::Formatter) -> fmt::Result {
write!(fmt, "0x{:04x}", self.0)
}
}
/// Wrapper for printing slice as hex data
pub struct HexSlice<'a>(pub &'a [u8]);
impl<'a> fmt::Debug for HexSlice<'a> {
fn fmt(&self, fmt: &mut fmt::Formatter) -> fmt::Result {
let s: Vec<_> = self.0.iter().map(|&i| format!("{:02x}", i)).collect();
write!(fmt, "[{}]", s.join(" "))
}
}
#[cfg(test)]
mod tests {
use crate::debug;
#[test]
fn debug_print_hexu8() {
assert_eq!(format!("{:?}", debug::HexU8(18)), "0x12");
}
#[test]
fn debug_print_hexu16() {
assert_eq!(format!("{:?}", debug::HexU16(32769)), "0x8001");
}
#[test]
fn debug_print_hexslice() {
assert_eq!(
format!("{:?}", debug::HexSlice(&[15, 16, 17, 18, 19, 20])),
"[0f 10 11 12 13 14]"
);
}
}

44
vendor/rusticata-macros/src/lib.rs vendored Normal file
View File

@@ -0,0 +1,44 @@
//! # Rusticata-macros
//!
//! Helper macros for the [rusticata](https://github.com/rusticata) project.
//!
//! This crate contains some additions to [nom](https://github.com/Geal/nom).
//!
//! For example, the [`combinator::cond_else`] function allows to apply the first parser if the
//! condition is true, and the second if the condition is false:
//!
//! ```rust
//! # use nom::IResult;
//! # use nom::combinator::map;
//! # use nom::number::streaming::*;
//! use rusticata_macros::combinator::cond_else;
//! # fn parser(s:&[u8]) {
//! let r: IResult<_, _, ()> = cond_else(
//! || s.len() > 1,
//! be_u16,
//! map(be_u8, u16::from)
//! )(s);
//! # }
//! ```
//!
//! See the documentation for more details and examples.
#![deny(
missing_docs,
unsafe_code,
unstable_features,
unused_import_braces,
unused_qualifications
)]
pub mod combinator;
pub mod debug;
pub use macros::*;
#[macro_use]
pub mod macros;
mod traits;
pub use traits::*;
// re-exports
pub use nom;

300
vendor/rusticata-macros/src/macros.rs vendored Normal file
View File

@@ -0,0 +1,300 @@
//! Helper macros
use nom::bytes::complete::take;
use nom::combinator::map_res;
use nom::IResult;
#[doc(hidden)]
pub mod export {
pub use core::{fmt, mem, ptr};
}
/// Helper macro for newtypes: declare associated constants and implement Display trait
#[macro_export]
macro_rules! newtype_enum (
(@collect_impl, $name:ident, $($key:ident = $val:expr),* $(,)*) => {
$( pub const $key : $name = $name($val); )*
};
(@collect_disp, $name:ident, $f:ident, $m:expr, $($key:ident = $val:expr),* $(,)*) => {
match $m {
$( $val => write!($f, stringify!{$key}), )*
n => write!($f, "{}({} / 0x{:x})", stringify!{$name}, n, n)
}
};
// entry
(impl $name:ident {$($body:tt)*}) => (
#[allow(non_upper_case_globals)]
impl $name {
newtype_enum!{@collect_impl, $name, $($body)*}
}
);
// entry with display
(impl display $name:ident {$($body:tt)*}) => (
newtype_enum!(impl $name { $($body)* });
impl $crate::export::fmt::Display for $name {
fn fmt(&self, f: &mut $crate::export::fmt::Formatter) -> $crate::export::fmt::Result {
newtype_enum!(@collect_disp, $name, f, self.0, $($body)*)
}
}
);
// entry with display and debug
(impl debug $name:ident {$($body:tt)*}) => (
newtype_enum!(impl display $name { $($body)* });
impl $crate::export::fmt::Debug for $name {
fn fmt(&self, f: &mut $crate::export::fmt::Formatter) -> $crate::export::fmt::Result {
write!(f, "{}", self)
}
}
);
);
/// Helper macro for nom parsers: raise error if the condition is true
///
/// This macro is used when using custom errors
#[macro_export]
macro_rules! custom_check (
($i:expr, $cond:expr, $err:expr) => (
{
if $cond {
Err(::nom::Err::Error($err))
} else {
Ok(($i, ()))
}
}
);
);
/// Helper macro for nom parsers: raise error if the condition is true
///
/// This macro is used when using `ErrorKind`
#[macro_export]
macro_rules! error_if (
($i:expr, $cond:expr, $err:expr) => (
{
use nom::error_position;
if $cond {
Err(::nom::Err::Error(error_position!($i, $err)))
} else {
Ok(($i, ()))
}
}
);
);
/// Helper macro for nom parsers: raise error if input is not empty
///
/// Deprecated - use `nom::eof`
#[macro_export]
#[deprecated(since = "2.0.0")]
macro_rules! empty (
($i:expr,) => (
{
use nom::eof;
eof!($i,)
}
);
);
#[deprecated(since = "3.0.1", note = "please use `be_var_u64` instead")]
/// Read an entire slice as a big-endian value.
///
/// Returns the value as `u64`. This function checks for integer overflows, and returns a
/// `Result::Err` value if the value is too big.
pub fn bytes_to_u64(s: &[u8]) -> Result<u64, &'static str> {
let mut u: u64 = 0;
if s.is_empty() {
return Err("empty");
};
if s.len() > 8 {
return Err("overflow");
}
for &c in s {
let u1 = u << 8;
u = u1 | (c as u64);
}
Ok(u)
}
/// Read a slice as a big-endian value.
#[macro_export]
macro_rules! parse_hex_to_u64 (
( $i:expr, $size:expr ) => {
map_res(take($size as usize), $crate::combinator::be_var_u64)($i)
};
);
/// Read 3 bytes as an unsigned integer
#[deprecated(since = "0.5.0", note = "please use `be_u24` instead")]
#[allow(deprecated)]
#[inline]
pub fn parse_uint24(i: &[u8]) -> IResult<&[u8], u64> {
map_res(take(3usize), bytes_to_u64)(i)
}
//named!(parse_hex4<&[u8], u64>, parse_hex_to_u64!(4));
/// Combination and flat_map! and take! as first combinator
#[macro_export]
macro_rules! flat_take (
($i:expr, $len:expr, $f:ident) => ({
if $i.len() < $len { Err(::nom::Err::Incomplete(::nom::Needed::new($len))) }
else {
let taken = &$i[0..$len];
let rem = &$i[$len..];
match $f(taken) {
Ok((_,res)) => Ok((rem,res)),
Err(e) => Err(e)
}
}
});
($i:expr, $len:expr, $submac:ident!( $($args:tt)*)) => ({
if $i.len() < $len { Err(::nom::Err::Incomplete(::nom::Needed::new($len))) }
else {
let taken = &$i[0..$len];
let rem = &$i[$len..];
match $submac!(taken, $($args)*) {
Ok((_,res)) => Ok((rem,res)),
Err(e) => Err(e)
}
}
});
);
/// Apply combinator, trying to "upgrade" error to next error type (using the `Into` or `From`
/// traits).
#[macro_export]
macro_rules! upgrade_error (
($i:expr, $submac:ident!( $($args:tt)*) ) => ({
upgrade_error!( $submac!( $i, $($args)* ) )
});
($i:expr, $f:expr) => ({
upgrade_error!( call!($i, $f) )
});
($e:expr) => ({
match $e {
Ok(o) => Ok(o),
Err(::nom::Err::Error(e)) => Err(::nom::Err::Error(e.into())),
Err(::nom::Err::Failure(e)) => Err(::nom::Err::Failure(e.into())),
Err(::nom::Err::Incomplete(i)) => Err(::nom::Err::Incomplete(i)),
}
});
);
/// Apply combinator, trying to "upgrade" error to next error type (using the `Into` or `From`
/// traits).
#[macro_export]
macro_rules! upgrade_error_to (
($i:expr, $ty:ty, $submac:ident!( $($args:tt)*) ) => ({
upgrade_error_to!( $ty, $submac!( $i, $($args)* ) )
});
($i:expr, $ty:ty, $f:expr) => ({
upgrade_error_to!( $ty, call!($i, $f) )
});
($ty:ty, $e:expr) => ({
match $e {
Ok(o) => Ok(o),
Err(::nom::Err::Error(e)) => Err(::nom::Err::Error(e.into::<$ty>())),
Err(::nom::Err::Failure(e)) => Err(::nom::Err::Failure(e.into::<$ty>())),
Err(::nom::Err::Incomplete(i)) => Err(::nom::Err::Incomplete(i)),
}
});
);
/// Nom combinator that returns the given expression unchanged
#[macro_export]
macro_rules! q {
($i:expr, $x:expr) => {{
Ok(($i, $x))
}};
}
/// Align input value to the next multiple of n bytes
/// Valid only if n is a power of 2
#[macro_export]
macro_rules! align_n2 {
($x:expr, $n:expr) => {
($x + ($n - 1)) & !($n - 1)
};
}
/// Align input value to the next multiple of 4 bytes
#[macro_export]
macro_rules! align32 {
($x:expr) => {
$crate::align_n2!($x, 4)
};
}
#[cfg(test)]
mod tests {
use nom::error::ErrorKind;
use nom::number::streaming::{be_u16, be_u32};
use nom::{error_position, Err, IResult, Needed};
#[test]
fn test_error_if() {
let empty = &b""[..];
let res: IResult<&[u8], ()> = error_if!(empty, true, ErrorKind::Tag);
assert_eq!(res, Err(Err::Error(error_position!(empty, ErrorKind::Tag))));
}
#[test]
fn test_newtype_enum() {
#[derive(Debug, PartialEq, Eq)]
struct MyType(pub u8);
newtype_enum! {
impl display MyType {
Val1 = 0,
Val2 = 1
}
}
assert_eq!(MyType(0), MyType::Val1);
assert_eq!(MyType(1), MyType::Val2);
assert_eq!(format!("{}", MyType(0)), "Val1");
assert_eq!(format!("{}", MyType(4)), "MyType(4 / 0x4)");
}
#[test]
fn test_flat_take() {
let input = &[0x00, 0x01, 0xff];
// read first 2 bytes and use correct combinator: OK
let res: IResult<&[u8], u16> = flat_take!(input, 2, be_u16);
assert_eq!(res, Ok((&input[2..], 0x0001)));
// read 3 bytes and use 2: OK (some input is just lost)
let res: IResult<&[u8], u16> = flat_take!(input, 3, be_u16);
assert_eq!(res, Ok((&b""[..], 0x0001)));
// read 2 bytes and a combinator requiring more bytes
let res: IResult<&[u8], u32> = flat_take!(input, 2, be_u32);
assert_eq!(res, Err(Err::Incomplete(Needed::new(2))));
// test with macro as sub-combinator
let res: IResult<&[u8], u16> = flat_take!(input, 2, be_u16);
assert_eq!(res, Ok((&input[2..], 0x0001)));
}
#[test]
fn test_q() {
let empty = &b""[..];
let res: IResult<&[u8], &str, ErrorKind> = q!(empty, "test");
assert_eq!(res, Ok((empty, "test")));
}
#[test]
fn test_align32() {
assert_eq!(align32!(3), 4);
assert_eq!(align32!(4), 4);
assert_eq!(align32!(5), 8);
assert_eq!(align32!(5u32), 8);
assert_eq!(align32!(5i32), 8);
assert_eq!(align32!(5usize), 8);
}
}

9
vendor/rusticata-macros/src/traits.rs vendored Normal file
View File

@@ -0,0 +1,9 @@
//! Common traits
/// Common trait for structures serialization
pub trait Serialize<O = Vec<u8>> {
/// Type of serialization error
type Error;
/// Try to serialize object
fn serialize(&self) -> Result<O, Self::Error>;
}