| //! 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); |
| } |
| } |