diff options
| author | ThiƩbaud Weksteen <tweek@google.com> | 2021-06-21 20:59:32 +0200 |
|---|---|---|
| committer | ThiƩbaud Weksteen <tweek@google.com> | 2021-06-22 07:14:19 +0000 |
| commit | f3f7e4ebc152cb84e657db408f4d90fec44c20f5 (patch) | |
| tree | 1890b58e5d2b5643016d6e2e9b99b6f56ac42127 /src | |
| parent | af89f69863b5e89257fde5e0598ba3b1fe2ac06a (diff) | |
| download | nom-f3f7e4ebc152cb84e657db408f4d90fec44c20f5.tar.gz | |
Update to 6.2.0
Add patch to avoid requiring bitvec and its dependencies[1].
[1] https://github.com/Geal/nom/pull/1325
Bug: 178357400
Change-Id: I07c2b3926192d3b31e5971437bba86d5c7be35df
Diffstat (limited to 'src')
35 files changed, 7246 insertions, 5115 deletions
diff --git a/src/bits/complete.rs b/src/bits/complete.rs index f868ec2..bce4df4 100644 --- a/src/bits/complete.rs +++ b/src/bits/complete.rs @@ -1,13 +1,15 @@ -//! bit level parsers +//! Bit level parsers //! use crate::error::{ErrorKind, ParseError}; use crate::internal::{Err, IResult}; -use crate::lib::std::ops::{AddAssign, RangeFrom, Shl, Shr, Div}; +use crate::lib::std::ops::{AddAssign, Div, RangeFrom, Shl, Shr}; use crate::traits::{InputIter, InputLength, Slice, ToUsize}; -/// generates a parser taking `count` bits -pub fn take<I, O, C, E: ParseError<(I, usize)>>(count: C) -> impl Fn((I, usize)) -> IResult<(I, usize), O, E> +/// Generates a parser taking `count` bits +pub fn take<I, O, C, E: ParseError<(I, usize)>>( + count: C, +) -> impl Fn((I, usize)) -> IResult<(I, usize), O, E> where I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, C: ToUsize, @@ -20,11 +22,14 @@ where } else { let cnt = (count + bit_offset).div(8); if input.input_len() * 8 < count + bit_offset { - Err(Err::Error(E::from_error_kind((input, bit_offset), ErrorKind::Eof))) + Err(Err::Error(E::from_error_kind( + (input, bit_offset), + ErrorKind::Eof, + ))) } else { - let mut acc:O = (0 as u8).into(); - let mut offset: usize = bit_offset; - let mut remaining: usize = count; + let mut acc: O = (0 as u8).into(); + let mut offset: usize = bit_offset; + let mut remaining: usize = count; let mut end_offset: usize = 0; for byte in input.iter_elements().take(cnt + 1) { @@ -47,14 +52,17 @@ where offset = 0; } } - Ok(( (input.slice(cnt..), end_offset) , acc)) + Ok(((input.slice(cnt..), end_offset), acc)) } } } } -/// generates a parser taking `count` bits and comparing them to `pattern` -pub fn tag<I, O, C, E: ParseError<(I, usize)>>(pattern: O, count: C) -> impl Fn((I, usize)) -> IResult<(I, usize), O, E> +/// Generates a parser taking `count` bits and comparing them to `pattern` +pub fn tag<I, O, C, E: ParseError<(I, usize)>>( + pattern: O, + count: C, +) -> impl Fn((I, usize)) -> IResult<(I, usize), O, E> where I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength + Clone, C: ToUsize, diff --git a/src/bits/macros.rs b/src/bits/macros.rs index b056503..d39f457 100644 --- a/src/bits/macros.rs +++ b/src/bits/macros.rs @@ -28,7 +28,7 @@ /// let sl = &input[..]; /// /// assert_eq!(take_4_bits( sl ), Ok( (&sl[1..], 0xA) )); -/// assert_eq!(take_4_bits( &b""[..] ), Err(Err::Incomplete(Needed::Size(1)))); +/// assert_eq!(take_4_bits( &b""[..] ), Err(Err::Incomplete(Needed::new(1)))); /// # } #[macro_export(local_inner_macros)] macro_rules! bits ( @@ -40,7 +40,7 @@ macro_rules! bits ( ); ); -/// Counterpart to bits, bytes! transforms its bit stream input into a byte slice for the underlying +/// Counterpart to `bits`, `bytes!` transforms its bit stream input into a byte slice for the underlying /// parser, allowing byte-slice parsers to work on bit streams. /// /// Signature: @@ -52,13 +52,13 @@ macro_rules! bits ( /// ``` /// # #[macro_use] extern crate nom; /// # use nom::combinator::rest; -/// # use nom::error::ErrorKind; +/// # use nom::error::{Error, ErrorKind}; /// # fn main() { /// /// named!( parse<(u8, u8, &[u8])>, bits!( tuple!( /// take_bits!(4u8), /// take_bits!(8u8), -/// bytes!(rest::<_, (_, ErrorKind)>) +/// bytes!(rest::<_, Error<_>>) /// ))); /// /// let input = &[0xde, 0xad, 0xbe, 0xaf]; @@ -134,9 +134,9 @@ macro_rules! tag_bits ( #[cfg(test)] mod tests { - use crate::lib::std::ops::{AddAssign, Shl, Shr}; - use crate::internal::{Err, Needed, IResult}; use crate::error::ErrorKind; + use crate::internal::{Err, IResult, Needed}; + use crate::lib::std::ops::{AddAssign, Shl, Shr}; #[test] fn take_bits() { @@ -157,11 +157,8 @@ mod tests { assert_eq!(take_bits!((sl, 6), 11u8), Ok(((&sl[2..], 1), 1504))); assert_eq!(take_bits!((sl, 0), 20u8), Ok(((&sl[2..], 4), 700_163))); assert_eq!(take_bits!((sl, 4), 20u8), Ok(((&sl[3..], 0), 716_851))); - let r: IResult<_,u32> = take_bits!((sl, 4), 22u8); - assert_eq!( - r, - Err(Err::Incomplete(Needed::Size(22))) - ); + let r: IResult<_, u32> = take_bits!((sl, 4), 22u8); + assert_eq!(r, Err(Err::Incomplete(Needed::new(22)))); } #[test] @@ -188,7 +185,7 @@ mod tests { let sl = &input[..]; assert_eq!(ch((&input[..], 0)), Ok(((&sl[1..], 4), (5, 15)))); assert_eq!(ch((&input[..], 4)), Ok(((&sl[2..], 0), (7, 16)))); - assert_eq!(ch((&input[..1], 0)), Err(Err::Incomplete(Needed::Size(5)))); + assert_eq!(ch((&input[..1], 0)), Err(Err::Incomplete(Needed::new(5)))); } named!(ch_bytes<(u8, u8)>, bits!(ch)); @@ -196,18 +193,26 @@ mod tests { fn bits_to_bytes() { let input = [0b10_10_10_10, 0b11_11_00_00, 0b00_11_00_11]; assert_eq!(ch_bytes(&input[..]), Ok((&input[2..], (5, 15)))); - assert_eq!(ch_bytes(&input[..1]), Err(Err::Incomplete(Needed::Size(1)))); + assert_eq!(ch_bytes(&input[..1]), Err(Err::Incomplete(Needed::new(1)))); assert_eq!( ch_bytes(&input[1..]), Err(Err::Error(error_position!(&input[1..], ErrorKind::TagBits))) ); } - named!(bits_bytes_bs, bits!(bytes!(crate::combinator::rest::<_, (&[u8], ErrorKind)>))); + named!( + bits_bytes_bs, + bits!(bytes!( + crate::combinator::rest::<_, crate::error::Error<&[u8]>> + )) + ); #[test] fn bits_bytes() { let input = [0b10_10_10_10]; - assert_eq!(bits_bytes_bs(&input[..]), Ok((&[][..], &[0b10_10_10_10][..]))); + assert_eq!( + bits_bytes_bs(&input[..]), + Ok((&[][..], &[0b10_10_10_10][..])) + ); } #[derive(PartialEq, Debug)] @@ -255,9 +260,6 @@ mod tests { Ok(((&sl[3..], 0), FakeUint(716_851))) ); let r3: IResult<_, FakeUint> = take_bits!((sl, 4), 22u8); - assert_eq!( - r3, - Err(Err::Incomplete(Needed::Size(22))) - ); + assert_eq!(r3, Err(Err::Incomplete(Needed::new(22)))); } } diff --git a/src/bits/mod.rs b/src/bits/mod.rs index a6b12f1..c5d1ac7 100644 --- a/src/bits/mod.rs +++ b/src/bits/mod.rs @@ -1,17 +1,16 @@ -//! bit level parsers +//! Bit level parsers //! #[macro_use] mod macros; -pub mod streaming; pub mod complete; +pub mod streaming; -use crate::error::{ParseError, ErrorKind}; +use crate::error::{ErrorKind, ParseError}; use crate::internal::{Err, IResult, Needed}; use crate::lib::std::ops::RangeFrom; -use crate::traits::{Slice, ErrorConvert}; - +use crate::traits::{ErrorConvert, Slice}; /// Converts a byte-level input to a bit-level input, for consumption by a parser that uses bits. /// @@ -19,75 +18,92 @@ use crate::traits::{Slice, ErrorConvert}; /// away. /// /// # Example -/// ```ignore -/// # #[macro_use] extern crate nom; -/// # use nom::IResult; -/// use nom::bits::bits; -/// use nom::bits::complete::take; +/// ``` +/// use nom::bits::{bits, streaming::take}; +/// use nom::error::Error; +/// use nom::sequence::tuple; +/// use nom::IResult; /// -/// fn take_4_bits(input: &[u8]) -> IResult<&[u8], u64> { -/// bits(take::<_, _, _, (_, _)>(4usize))(input) +/// fn parse(input: &[u8]) -> IResult<&[u8], (u8, u8)> { +/// bits::<_, _, Error<(&[u8], usize)>, _, _>(tuple((take(4usize), take(8usize))))(input) /// } /// -/// let input = vec![0xAB, 0xCD, 0xEF, 0x12]; -/// let sl = &input[..]; +/// let input = &[0x12, 0x34, 0xff, 0xff]; /// -/// assert_eq!(take_4_bits( sl ), Ok( (&sl[1..], 0xA) )); +/// let output = parse(input).expect("We take 1.5 bytes and the input is longer than 2 bytes"); +/// +/// // The first byte is consumed, the second byte is partially consumed and dropped. +/// let remaining = output.0; +/// assert_eq!(remaining, [0xff, 0xff]); +/// +/// let parsed = output.1; +/// assert_eq!(parsed.0, 0x01); +/// assert_eq!(parsed.1, 0x23); /// ``` -pub fn bits<I, O, E1: ParseError<(I, usize)>+ErrorConvert<E2>, E2: ParseError<I>, P>(parser: P) -> impl Fn(I) -> IResult<I, O, E2> +pub fn bits<I, O, E1, E2, P>(mut parser: P) -> impl FnMut(I) -> IResult<I, O, E2> where + E1: ParseError<(I, usize)> + ErrorConvert<E2>, + E2: ParseError<I>, I: Slice<RangeFrom<usize>>, - P: Fn((I, usize)) -> IResult<(I, usize), O, E1>, + P: FnMut((I, usize)) -> IResult<(I, usize), O, E1>, { move |input: I| match parser((input, 0)) { - Ok(((rest, offset), res)) => { - let byte_index = offset / 8 + if offset % 8 == 0 { 0 } else { 1 }; - Ok((rest.slice(byte_index..), res)) + Ok(((rest, offset), result)) => { + // If the next byte has been partially read, it will be sliced away as well. + // The parser functions might already slice away all fully read bytes. + // That's why `offset / 8` isn't necessarily needed at all times. + let remaining_bytes_index = offset / 8 + if offset % 8 == 0 { 0 } else { 1 }; + Ok((rest.slice(remaining_bytes_index..), result)) } - Err(Err::Incomplete(n)) => Err(Err::Incomplete(n.map(|u| u / 8 + 1))), + Err(Err::Incomplete(n)) => Err(Err::Incomplete(n.map(|u| u.get() / 8 + 1))), Err(Err::Error(e)) => Err(Err::Error(e.convert())), Err(Err::Failure(e)) => Err(Err::Failure(e.convert())), } } #[doc(hidden)] -pub fn bitsc<I, O, E1: ParseError<(I, usize)>+ErrorConvert<E2>, E2: ParseError<I>, P>(input: I, parser: P) -> IResult<I, O, E2> +pub fn bitsc<I, O, E1: ParseError<(I, usize)> + ErrorConvert<E2>, E2: ParseError<I>, P>( + input: I, + parser: P, +) -> IResult<I, O, E2> where I: Slice<RangeFrom<usize>>, - P: Fn((I, usize)) -> IResult<(I, usize), O, E1>, + P: FnMut((I, usize)) -> IResult<(I, usize), O, E1>, { bits(parser)(input) } -/// Counterpart to bits, bytes transforms its bit stream input into a byte slice for the underlying +/// Counterpart to `bits`, `bytes` transforms its bit stream input into a byte slice for the underlying /// parser, allowing byte-slice parsers to work on bit streams. /// /// A partial byte remaining in the input will be ignored and the given parser will start parsing /// at the next full byte. /// -/// ```ignore -/// # #[macro_use] extern crate nom; -/// # use nom::IResult; -/// # use nom::combinator::rest; -/// # use nom::sequence::tuple; -/// use nom::bits::{bits, bytes, streaming::take_bits}; +/// ``` +/// use nom::bits::{bits, bytes, streaming::take}; +/// use nom::combinator::rest; +/// use nom::error::Error; +/// use nom::sequence::tuple; +/// use nom::IResult; /// /// fn parse(input: &[u8]) -> IResult<&[u8], (u8, u8, &[u8])> { -/// bits(tuple(( -/// take_bits(4usize), -/// take_bits(8usize), -/// bytes(rest) +/// bits::<_, _, Error<(&[u8], usize)>, _, _>(tuple(( +/// take(4usize), +/// take(8usize), +/// bytes::<_, _, Error<&[u8]>, _, _>(rest) /// )))(input) /// } /// -/// let input = &[0xde, 0xad, 0xbe, 0xaf]; +/// let input = &[0x12, 0x34, 0xff, 0xff]; /// -/// assert_eq!(parse( input ), Ok(( &[][..], (0xd, 0xea, &[0xbe, 0xaf][..]) ))); +/// assert_eq!(parse( input ), Ok(( &[][..], (0x01, 0x23, &[0xff, 0xff][..]) ))); /// ``` -pub fn bytes<I, O, E1: ParseError<I>+ErrorConvert<E2>, E2: ParseError<(I, usize)>, P>(parser: P) -> impl Fn((I, usize)) -> IResult<(I, usize), O, E2> +pub fn bytes<I, O, E1, E2, P>(mut parser: P) -> impl FnMut((I, usize)) -> IResult<(I, usize), O, E2> where + E1: ParseError<I> + ErrorConvert<E2>, + E2: ParseError<(I, usize)>, I: Slice<RangeFrom<usize>> + Clone, - P: Fn(I) -> IResult<I, O, E1>, + P: FnMut(I) -> IResult<I, O, E1>, { move |(input, offset): (I, usize)| { let inner = if offset % 8 != 0 { @@ -95,12 +111,12 @@ where } else { input.slice((offset / 8)..) }; - let i = (input.clone(), offset); + let i = (input, offset); match parser(inner) { Ok((rest, res)) => Ok(((rest, 0), res)), Err(Err::Incomplete(Needed::Unknown)) => Err(Err::Incomplete(Needed::Unknown)), - Err(Err::Incomplete(Needed::Size(sz))) => Err(match sz.checked_mul(8) { - Some(v) => Err::Incomplete(Needed::Size(v)), + Err(Err::Incomplete(Needed::Size(sz))) => Err(match sz.get().checked_mul(8) { + Some(v) => Err::Incomplete(Needed::new(v)), None => Err::Failure(E2::from_error_kind(i, ErrorKind::TooLarge)), }), Err(Err::Error(e)) => Err(Err::Error(e.convert())), @@ -110,10 +126,81 @@ where } #[doc(hidden)] -pub fn bytesc<I, O, E1: ParseError<I>+ErrorConvert<E2>, E2: ParseError<(I, usize)>, P>(input: (I, usize), parser: P) -> IResult<(I, usize), O, E2> +pub fn bytesc<I, O, E1: ParseError<I> + ErrorConvert<E2>, E2: ParseError<(I, usize)>, P>( + input: (I, usize), + parser: P, +) -> IResult<(I, usize), O, E2> where I: Slice<RangeFrom<usize>> + Clone, - P: Fn(I) -> IResult<I, O, E1>, + P: FnMut(I) -> IResult<I, O, E1>, { bytes(parser)(input) } + +#[cfg(test)] +mod test { + use super::*; + use crate::bits::streaming::take; + use crate::error::Error; + use crate::sequence::tuple; + + #[test] + /// Take the `bits` function and assert that remaining bytes are correctly returned, if the + /// previous bytes are fully consumed + fn test_complete_byte_consumption_bits() { + let input = &[0x12, 0x34, 0x56, 0x78]; + + // Take 3 bit slices with sizes [4, 8, 4]. + let result: IResult<&[u8], (u8, u8, u8)> = + bits::<_, _, Error<(&[u8], usize)>, _, _>(tuple((take(4usize), take(8usize), take(4usize))))( + input, + ); + + let output = result.expect("We take 2 bytes and the input is longer than 2 bytes"); + + let remaining = output.0; + assert_eq!(remaining, [0x56, 0x78]); + + let parsed = output.1; + assert_eq!(parsed.0, 0x01); + assert_eq!(parsed.1, 0x23); + assert_eq!(parsed.2, 0x04); + } + + #[test] + /// Take the `bits` function and assert that remaining bytes are correctly returned, if the + /// previous bytes are NOT fully consumed. Partially consumed bytes are supposed to be dropped. + /// I.e. if we consume 1.5 bytes of 4 bytes, 2 bytes will be returned, bits 13-16 will be + /// dropped. + fn test_partial_byte_consumption_bits() { + let input = &[0x12, 0x34, 0x56, 0x78]; + + // Take bit slices with sizes [4, 8]. + let result: IResult<&[u8], (u8, u8)> = + bits::<_, _, Error<(&[u8], usize)>, _, _>(tuple((take(4usize), take(8usize))))(input); + + let output = result.expect("We take 1.5 bytes and the input is longer than 2 bytes"); + + let remaining = output.0; + assert_eq!(remaining, [0x56, 0x78]); + + let parsed = output.1; + assert_eq!(parsed.0, 0x01); + assert_eq!(parsed.1, 0x23); + } + + #[test] + #[cfg(feature = "std")] + /// Ensure that in Incomplete error is thrown, if too few bytes are passed for a given parser. + fn test_incomplete_bits() { + let input = &[0x12]; + + // Take bit slices with sizes [4, 8]. + let result: IResult<&[u8], (u8, u8)> = + bits::<_, _, Error<(&[u8], usize)>, _, _>(tuple((take(4usize), take(8usize))))(input); + + assert!(result.is_err()); + let error = result.err().unwrap(); + assert_eq!("Parsing requires 2 bytes/chars", error.to_string()); + } +} diff --git a/src/bits/streaming.rs b/src/bits/streaming.rs index 5ab7596..700b903 100644 --- a/src/bits/streaming.rs +++ b/src/bits/streaming.rs @@ -1,13 +1,15 @@ -//! bit level parsers +//! Bit level parsers //! use crate::error::{ErrorKind, ParseError}; use crate::internal::{Err, IResult, Needed}; -use crate::lib::std::ops::{AddAssign, RangeFrom, Shl, Shr, Div}; +use crate::lib::std::ops::{AddAssign, Div, RangeFrom, Shl, Shr}; use crate::traits::{InputIter, InputLength, Slice, ToUsize}; -/// generates a parser taking `count` bits -pub fn take<I, O, C, E: ParseError<(I, usize)>>(count: C) -> impl Fn((I, usize)) -> IResult<(I, usize), O, E> +/// Generates a parser taking `count` bits +pub fn take<I, O, C, E: ParseError<(I, usize)>>( + count: C, +) -> impl Fn((I, usize)) -> IResult<(I, usize), O, E> where I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, C: ToUsize, @@ -20,11 +22,11 @@ where } else { let cnt = (count + bit_offset).div(8); if input.input_len() * 8 < count + bit_offset { - Err(Err::Incomplete(Needed::Size(count as usize))) + Err(Err::Incomplete(Needed::new(count as usize))) } else { - let mut acc:O = (0 as u8).into(); - let mut offset: usize = bit_offset; - let mut remaining: usize = count; + let mut acc: O = (0 as u8).into(); + let mut offset: usize = bit_offset; + let mut remaining: usize = count; let mut end_offset: usize = 0; for byte in input.iter_elements().take(cnt + 1) { @@ -47,14 +49,17 @@ where offset = 0; } } - Ok(( (input.slice(cnt..), end_offset) , acc)) + Ok(((input.slice(cnt..), end_offset), acc)) } } } } -/// generates a parser taking `count` bits and comparing them to `pattern` -pub fn tag<I, O, C, E: ParseError<(I, usize)>>(pattern: O, count: C) -> impl Fn((I, usize)) -> IResult<(I, usize), O, E> +/// Generates a parser taking `count` bits and comparing them to `pattern` +pub fn tag<I, O, C, E: ParseError<(I, usize)>>( + pattern: O, + count: C, +) -> impl Fn((I, usize)) -> IResult<(I, usize), O, E> where I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength + Clone, C: ToUsize, diff --git a/src/branch/macros.rs b/src/branch/macros.rs index 4c02461..a9a3a33 100644 --- a/src/branch/macros.rs +++ b/src/branch/macros.rs @@ -94,7 +94,7 @@ /// /// **BE CAREFUL** there is a case where the behaviour of `alt!` can be confusing: /// -/// when the alternatives have different lengths, like this case: +/// When the alternatives have different lengths, like this case: /// /// ```ignore /// named!( test, alt!( tag!( "abcd" ) | tag!( "ef" ) | tag!( "ghi" ) | tag!( "kl" ) ) ); @@ -138,7 +138,7 @@ /// named!( test, alt!( tag!( "abcd" ) | tag!( "ab" ) | tag!( "ef" ) ) ); /// ``` /// -/// in that case, if you order by size, passing `"abcd"` as input will always be matched by the +/// In that case, if you order by size, passing `"abcd"` as input will always be matched by the /// smallest parser, so the solution using `complete!` is better suited. /// /// You can also nest multiple `alt!`, like this: @@ -382,13 +382,11 @@ macro_rules! switch ( ); ); -/// -/// /// `permutation!(I -> IResult<I,A>, I -> IResult<I,B>, ... I -> IResult<I,X> ) => I -> IResult<I, (A,B,...X)>` /// applies its sub parsers in a sequence, but independent from their order -/// this parser will only succeed if all of its sub parsers succeed +/// this parser will only succeed if all of its sub parsers succeed. /// -/// the tuple of results is in the same order as the parsers are declared +/// The tuple of results is in the same order as the parsers are declared /// /// ``` /// # #[macro_use] extern crate nom; @@ -414,7 +412,7 @@ macro_rules! switch ( /// error_position!(&b"xyzabcdefghi"[..], ErrorKind::Permutation))))); /// /// let e = &b"efgabc"[..]; -/// assert_eq!(perm(e), Err(Err::Incomplete(Needed::Size(4)))); +/// assert_eq!(perm(e), Err(Err::Incomplete(Needed::new(1)))); /// # } /// ``` /// @@ -451,7 +449,7 @@ macro_rules! switch ( /// assert_eq!(perm(c), Ok(("jklm", expected))); /// /// let e = "efgabc"; -/// assert_eq!(perm(e), Err(Err::Incomplete(Needed::Size(4)))); +/// assert_eq!(perm(e), Err(Err::Incomplete(Needed::new(1)))); /// # } /// ``` #[macro_export(local_inner_macros)] @@ -568,6 +566,7 @@ macro_rules! succ ( (17, $submac:ident ! ($($rest:tt)*)) => ($submac!(18, $($rest)*)); (18, $submac:ident ! ($($rest:tt)*)) => ($submac!(19, $($rest)*)); (19, $submac:ident ! ($($rest:tt)*)) => ($submac!(20, $($rest)*)); + (20, $submac:ident ! ($($rest:tt)*)) => ($submac!(21, $($rest)*)); ); #[doc(hidden)] @@ -720,15 +719,15 @@ macro_rules! permutation_iterator ( #[cfg(test)] mod tests { use crate::error::ErrorKind; + use crate::internal::{Err, IResult, Needed}; #[cfg(feature = "alloc")] use crate::{ error::ParseError, lib::std::{ fmt::Debug, - string::{String, ToString} - } + string::{String, ToString}, + }, }; - use crate::internal::{Err, IResult, Needed}; // reproduce the tag and take macros, because of module import order macro_rules! tag ( @@ -762,7 +761,7 @@ mod tests { let e: ErrorKind = ErrorKind::Tag; Err(Err::Error(error_position!($i, e))) } else if m < blen { - Err(Err::Incomplete(Needed::Size(blen))) + Err(Err::Incomplete(Needed::new(blen))) } else { Ok((&$i[blen..], reduced)) }; @@ -776,7 +775,7 @@ mod tests { { let cnt = $count as usize; let res:IResult<&[u8],&[u8],_> = if $i.len() < cnt { - Err(Err::Incomplete(Needed::Size(cnt))) + Err(Err::Incomplete(Needed::new(cnt))) } else { Ok((&$i[cnt..],&$i[0..cnt])) }; @@ -810,7 +809,10 @@ mod tests { } fn append(input: I, kind: ErrorKind, other: Self) -> Self { - ErrorStr(format!("custom error message: ({:?}, {:?}) - {:?}", input, kind, other)) + ErrorStr(format!( + "custom error message: ({:?}, {:?}) - {:?}", + input, kind, other + )) } } @@ -854,7 +856,10 @@ mod tests { assert_eq!(alt4(b), Ok((&b""[..], b))); // test the alternative syntax - named!(alt5<bool>, alt!(tag!("abcd") => { |_| false } | tag!("efgh") => { |_| true })); + named!( + alt5<bool>, + alt!(tag!("abcd") => { |_| false } | tag!("efgh") => { |_| true }) + ); assert_eq!(alt5(a), Ok((&b""[..], false))); assert_eq!(alt5(b), Ok((&b""[..], true))); @@ -869,15 +874,15 @@ mod tests { named!(alt1, alt!(tag!("a") | tag!("bc") | tag!("def"))); let a = &b""[..]; - assert_eq!(alt1(a), Err(Err::Incomplete(Needed::Size(1)))); + assert_eq!(alt1(a), Err(Err::Incomplete(Needed::new(1)))); let a = &b"b"[..]; - assert_eq!(alt1(a), Err(Err::Incomplete(Needed::Size(2)))); + assert_eq!(alt1(a), Err(Err::Incomplete(Needed::new(2)))); let a = &b"bcd"[..]; assert_eq!(alt1(a), Ok((&b"d"[..], &b"bc"[..]))); let a = &b"cde"[..]; assert_eq!(alt1(a), Err(Err::Error(error_position!(a, ErrorKind::Alt)))); let a = &b"de"[..]; - assert_eq!(alt1(a), Err(Err::Incomplete(Needed::Size(3)))); + assert_eq!(alt1(a), Err(Err::Incomplete(Needed::new(3)))); let a = &b"defg"[..]; assert_eq!(alt1(a), Ok((&b"g"[..], &b"def"[..]))); } @@ -899,7 +904,13 @@ mod tests { let b = &b"efghijkl"[..]; assert_eq!(sw(b), Ok((&b""[..], &b"ijkl"[..]))); let c = &b"afghijkl"[..]; - assert_eq!(sw(c), Err(Err::Error(error_position!(&b"afghijkl"[..], ErrorKind::Switch)))); + assert_eq!( + sw(c), + Err(Err::Error(error_position!( + &b"afghijkl"[..], + ErrorKind::Switch + ))) + ); let a = &b"xxxxefgh"[..]; assert_eq!(sw(a), Ok((&b"gh"[..], &b"ef"[..]))); @@ -907,7 +918,10 @@ mod tests { #[test] fn permutation() { - named!(perm<(&[u8], &[u8], &[u8])>, permutation!(tag!("abcd"), tag!("efg"), tag!("hi"))); + named!( + perm<(&[u8], &[u8], &[u8])>, + permutation!(tag!("abcd"), tag!("efg"), tag!("hi")) + ); let expected = (&b"abcd"[..], &b"efg"[..], &b"hi"[..]); @@ -929,7 +943,7 @@ mod tests { ); let e = &b"efgabc"[..]; - assert_eq!(perm(e), Err(Err::Incomplete(Needed::Size(4)))); + assert_eq!(perm(e), Err(Err::Incomplete(Needed::new(4)))); } /* diff --git a/src/branch/mod.rs b/src/branch/mod.rs index 55d3633..6c0adc7 100644 --- a/src/branch/mod.rs +++ b/src/branch/mod.rs @@ -1,23 +1,25 @@ -//! choice combinators +//! Choice combinators #[macro_use] mod macros; use crate::error::ErrorKind; use crate::error::ParseError; -use crate::internal::{Err, IResult}; +use crate::internal::{Err, IResult, Parser}; -/// helper trait for the [alt()] combinator +/// Helper trait for the [alt()] combinator. /// -/// this trait is implemented for tuples of up to 21 elements +/// This trait is implemented for tuples of up to 21 elements pub trait Alt<I, O, E> { - /// tests each parser in the tuple and returns the result of the first one that succeeds - fn choice(&self, input: I) -> IResult<I, O, E>; + /// Tests each parser in the tuple and returns the result of the first one that succeeds + fn choice(&mut self, input: I) -> IResult<I, O, E>; } -/// tests a list of parsers one by one until one succeeds +/// Tests a list of parsers one by one until one succeeds. /// -/// It takes as argument a tuple of parsers. +/// It takes as argument a tuple of parsers. There is a maximum of 21 +/// parsers. If you need more, it is possible to nest them in other `alt` calls, +/// like this: `alt(parser_a, alt(parser_b, parser_c))` /// /// ```rust /// # #[macro_use] extern crate nom; @@ -40,35 +42,37 @@ pub trait Alt<I, O, E> { /// # } /// ``` /// -/// with a custom error type, it is possible to have alt return the error of the parser +/// With a custom error type, it is possible to have alt return the error of the parser /// that went the farthest in the input data -pub fn alt<I: Clone, O, E: ParseError<I>, List: Alt<I, O, E>>(l: List) -> impl Fn(I) -> IResult<I, O, E> { +pub fn alt<I: Clone, O, E: ParseError<I>, List: Alt<I, O, E>>( + mut l: List, +) -> impl FnMut(I) -> IResult<I, O, E> { move |i: I| l.choice(i) } -/// helper trait for the [permutation()] combinator +/// Helper trait for the [permutation()] combinator. /// -/// this trait is implemented for tuples of up to 21 elements +/// This trait is implemented for tuples of up to 21 elements pub trait Permutation<I, O, E> { - /// tries to apply all parsers in the tuple in various orders until all of them succeed - fn permutation(&self, input: I) -> IResult<I, O, E>; + /// Tries to apply all parsers in the tuple in various orders until all of them succeed + fn permutation(&mut self, input: I) -> IResult<I, O, E>; } -/// applies a list of parsers in any order +/// Applies a list of parsers in any order. /// -/// permutation will succeed if all of the child parsers succeeded. +/// Permutation will succeed if all of the child parsers succeeded. /// It takes as argument a tuple of parsers, and returns a /// tuple of the parser results. /// /// ```rust /// # #[macro_use] extern crate nom; -/// # use nom::{Err,error::ErrorKind, Needed, IResult}; +/// # use nom::{Err,error::{Error, ErrorKind}, Needed, IResult}; /// use nom::character::complete::{alpha1, digit1}; /// use nom::branch::permutation; /// # fn main() { /// fn parser(input: &str) -> IResult<&str, (&str, &str)> { /// permutation((alpha1, digit1))(input) -/// }; +/// } /// /// // permutation recognizes alphabetic characters then digit /// assert_eq!(parser("abc123"), Ok(("", ("abc", "123")))); @@ -77,10 +81,32 @@ pub trait Permutation<I, O, E> { /// assert_eq!(parser("123abc"), Ok(("", ("abc", "123")))); /// /// // it will fail if one of the parsers failed -/// assert_eq!(parser("abc;"), Err(Err::Error(error_position!(";", ErrorKind::Permutation)))); +/// assert_eq!(parser("abc;"), Err(Err::Error(Error::new(";", ErrorKind::Digit)))); /// # } /// ``` -pub fn permutation<I: Clone, O, E: ParseError<I>, List: Permutation<I, O, E>>(l: List) -> impl Fn(I) -> IResult<I, O, E> { +/// +/// The parsers are applied greedily: if there are multiple unapplied parsers +/// that could parse the next slice of input, the first one is used. +/// ```rust +/// # use nom::{Err, error::{Error, ErrorKind}, IResult}; +/// use nom::branch::permutation; +/// use nom::character::complete::{anychar, char}; +/// +/// fn parser(input: &str) -> IResult<&str, (char, char)> { +/// permutation((anychar, char('a')))(input) +/// } +/// +/// // anychar parses 'b', then char('a') parses 'a' +/// assert_eq!(parser("ba"), Ok(("", ('b', 'a')))); +/// +/// // anychar parses 'a', then char('a') fails on 'b', +/// // even though char('a') followed by anychar would succeed +/// assert_eq!(parser("ab"), Err(Err::Error(Error::new("b", ErrorKind::Char)))); +/// ``` +/// +pub fn permutation<I: Clone, O, E: ParseError<I>, List: Permutation<I, O, E>>( + mut l: List, +) -> impl FnMut(I) -> IResult<I, O, E> { move |i: I| l.permutation(i) } @@ -103,14 +129,14 @@ macro_rules! alt_trait_impl( ($($id:ident)+) => ( impl< Input: Clone, Output, Error: ParseError<Input>, - $($id: Fn(Input) -> IResult<Input, Output, Error>),+ + $($id: Parser<Input, Output, Error>),+ > Alt<Input, Output, Error> for ( $($id),+ ) { - fn choice(&self, input: Input) -> IResult<Input, Output, Error> { - let mut err: Option<Error> = None; - alt_trait_inner!(0, self, input, err, $($id)+); - - Err(Err::Error(Error::append(input, ErrorKind::Alt, err.unwrap()))) + fn choice(&mut self, input: Input) -> IResult<Input, Output, Error> { + match self.0.parse(input.clone()) { + Err(Err::Error(e)) => alt_trait_inner!(1, self, input, e, $($id)+), + res => res, + } } } ); @@ -118,75 +144,67 @@ macro_rules! alt_trait_impl( macro_rules! alt_trait_inner( ($it:tt, $self:expr, $input:expr, $err:expr, $head:ident $($id:ident)+) => ( - match $self.$it($input.clone()) { + match $self.$it.parse($input.clone()) { Err(Err::Error(e)) => { - $err = Some(match $err.take() { - None => e, - Some(prev) => prev.or(e), - }); - succ!($it, alt_trait_inner!($self, $input, $err, $($id)+)) - }, - res => return res, + let err = $err.or(e); + succ!($it, alt_trait_inner!($self, $input, err, $($id)+)) + } + res => res, } ); ($it:tt, $self:expr, $input:expr, $err:expr, $head:ident) => ( - match $self.$it($input.clone()) { - Err(Err::Error(e)) => { - $err = Some(match $err.take() { - None => e, - Some(prev) => prev.or(e), - }); - }, - res => return res, - } + Err(Err::Error(Error::append($input, ErrorKind::Alt, $err))) ); ); alt_trait!(A B C D E F G H I J K L M N O P Q R S T U); macro_rules! permutation_trait( - ($name1:ident $ty1:ident, $name2:ident $ty2:ident) => ( - permutation_trait_impl!($name1 $ty1, $name2 $ty2); - ); - ($name1:ident $ty1:ident, $name2: ident $ty2:ident, $($name:ident $ty:ident),*) => ( - permutation_trait!(__impl $name1 $ty1, $name2 $ty2; $($name $ty),*); + ( + $name1:ident $ty1:ident $item1:ident + $name2:ident $ty2:ident $item2:ident + $($name3:ident $ty3:ident $item3:ident)* + ) => ( + permutation_trait!(__impl $name1 $ty1 $item1, $name2 $ty2 $item2; $($name3 $ty3 $item3)*); ); - (__impl $($name:ident $ty: ident),+; $name1:ident $ty1:ident, $($name2:ident $ty2:ident),*) => ( - permutation_trait_impl!($($name $ty),+); - permutation_trait!(__impl $($name $ty),+ , $name1 $ty1; $($name2 $ty2),*); + ( + __impl $($name:ident $ty:ident $item:ident),+; + $name1:ident $ty1:ident $item1:ident $($name2:ident $ty2:ident $item2:ident)* + ) => ( + permutation_trait_impl!($($name $ty $item),+); + permutation_trait!(__impl $($name $ty $item),+ , $name1 $ty1 $item1; $($name2 $ty2 $item2)*); ); - (__impl $($name:ident $ty: ident),+; $name1:ident $ty1:ident) => ( - permutation_trait_impl!($($name $ty),+); - permutation_trait_impl!($($name $ty),+, $name1 $ty1); + (__impl $($name:ident $ty:ident $item:ident),+;) => ( + permutation_trait_impl!($($name $ty $item),+); ); ); macro_rules! permutation_trait_impl( - ($($name:ident $ty: ident),+) => ( + ($($name:ident $ty:ident $item:ident),+) => ( impl< Input: Clone, $($ty),+ , Error: ParseError<Input>, - $($name: Fn(Input) -> IResult<Input, $ty, Error>),+ + $($name: Parser<Input, $ty, Error>),+ > Permutation<Input, ( $($ty),+ ), Error> for ( $($name),+ ) { - fn permutation(&self, mut input: Input) -> IResult<Input, ( $($ty),+ ), Error> { - let mut res = permutation_init!((), $($name),+); + fn permutation(&mut self, mut input: Input) -> IResult<Input, ( $($ty),+ ), Error> { + let mut res = ($(Option::<$ty>::None),+); loop { - let mut all_done = true; - permutation_trait_inner!(0, self, input, res, all_done, $($name)+); + let mut err: Option<Error> = None; + permutation_trait_inner!(0, self, input, res, err, $($name)+); - //if we reach that part, it means none of the parsers were able to read anything - if !all_done { - //FIXME: should wrap the error returned by the child parser - return Err(Err::Error(error_position!(input, ErrorKind::Permutation))); + // If we reach here, every iterator has either been applied before, + // or errored on the remaining input + if let Some(err) = err { + // There are remaining parsers, and all errored on the remaining input + return Err(Err::Error(Error::append(input, ErrorKind::Permutation, err))); } - break; - } - if let Some(unwrapped_res) = { permutation_trait_unwrap!(0, (), res, $($name),+) } { - Ok((input, unwrapped_res)) - } else { - Err(Err::Error(error_position!(input, ErrorKind::Permutation))) + // All parsers were applied + match res { + ($(Some($item)),+) => return Ok((input, ($($item),+))), + _ => unreachable!(), + } } } } @@ -194,70 +212,48 @@ macro_rules! permutation_trait_impl( ); macro_rules! permutation_trait_inner( - ($it:tt, $self:expr, $input:ident, $res:expr, $all_done:expr, $head:ident $($id:ident)+) => ({ + ($it:tt, $self:expr, $input:ident, $res:expr, $err:expr, $head:ident $($id:ident)*) => ( if $res.$it.is_none() { - match $self.$it($input.clone()) { - Ok((i,o)) => { + match $self.$it.parse($input.clone()) { + Ok((i, o)) => { $input = i; $res.$it = Some(o); continue; - }, - Err(Err::Error(_)) => { - $all_done = false; - }, - Err(e) => { - return Err(e); } - }; - } - succ!($it, permutation_trait_inner!($self, $input, $res, $all_done, $($id)+)); - }); - ($it:tt, $self:expr, $input:ident, $res:expr, $all_done:expr, $head:ident) => ({ - if $res.$it.is_none() { - match $self.$it($input.clone()) { - Ok((i,o)) => { - $input = i; - $res.$it = Some(o); - continue; - }, - Err(Err::Error(_)) => { - $all_done = false; - }, - Err(e) => { - return Err(e); + Err(Err::Error(e)) => { + $err = Some(match $err { + Some(err) => err.or(e), + None => e, + }); } + Err(e) => return Err(e), }; } - }); + succ!($it, permutation_trait_inner!($self, $input, $res, $err, $($id)*)); + ); + ($it:tt, $self:expr, $input:ident, $res:expr, $err:expr,) => (); ); -macro_rules! permutation_trait_unwrap ( - ($it:tt, (), $res:ident, $e:ident, $($name:ident),+) => ({ - let res = $res.$it; - if res.is_some() { - succ!($it, permutation_trait_unwrap!((res.unwrap()), $res, $($name),+)) - } else { - $crate::lib::std::option::Option::None - } - }); - - ($it:tt, ($($parsed:expr),*), $res:ident, $e:ident, $($name:ident),+) => ({ - let res = $res.$it; - if res.is_some() { - succ!($it, permutation_trait_unwrap!(($($parsed),* , res.unwrap()), $res, $($name),+)) - } else { - $crate::lib::std::option::Option::None - } - }); - - ($it:tt, ($($parsed:expr),*), $res:ident, $name:ident) => ({ - let res = $res.$it; - if res.is_some() { - $crate::lib::std::option::Option::Some(($($parsed),* , res.unwrap() )) - } else { - $crate::lib::std::option::Option::None - } - }); +permutation_trait!( + FnA A a + FnB B b + FnC C c + FnD D d + FnE E e + FnF F f + FnG G g + FnH H h + FnI I i + FnJ J j + FnK K k + FnL L l + FnM M m + FnN N n + FnO O o + FnP P p + FnQ Q q + FnR R r + FnS S s + FnT T t + FnU U u ); - -permutation_trait!(FnA A, FnB B, FnC C, FnD D, FnE E, FnF F, FnG G, FnH H, FnI I, FnJ J, FnK K, FnL L, FnM M, FnN N, FnO O, FnP P, FnQ Q, FnR R, FnS S, FnT T, FnU U); diff --git a/src/bytes/complete.rs b/src/bytes/complete.rs index 8bf4220..37824da 100644 --- a/src/bytes/complete.rs +++ b/src/bytes/complete.rs @@ -1,11 +1,14 @@ -//! parsers recognizing bytes streams, complete input version +//! Parsers recognizing bytes streams, complete input version use crate::error::ErrorKind; use crate::error::ParseError; -use crate::internal::{Err, IResult}; +use crate::internal::{Err, IResult, Parser}; use crate::lib::std::ops::RangeFrom; use crate::lib::std::result::Result::*; -use crate::traits::{Compare, CompareResult, FindSubstring, FindToken, InputIter, InputLength, InputTake, InputTakeAtPosition, Slice, ToUsize}; +use crate::traits::{ + Compare, CompareResult, FindSubstring, FindToken, InputIter, InputLength, InputTake, + InputTakeAtPosition, Slice, ToUsize, +}; /// Recognizes a pattern /// @@ -16,7 +19,7 @@ use crate::traits::{Compare, CompareResult, FindSubstring, FindToken, InputIter, /// # Example /// ```rust /// # #[macro_use] extern crate nom; -/// # use nom::{Err, error::ErrorKind, Needed, IResult}; +/// # use nom::{Err, error::{Error, ErrorKind}, Needed, IResult}; /// use nom::bytes::complete::tag; /// /// fn parser(s: &str) -> IResult<&str, &str> { @@ -24,10 +27,12 @@ use crate::traits::{Compare, CompareResult, FindSubstring, FindToken, InputIter, /// } /// /// assert_eq!(parser("Hello, World!"), Ok((", World!", "Hello"))); -/// assert_eq!(parser("Something"), Err(Err::Error(("Something", ErrorKind::Tag)))); -/// assert_eq!(parser(""), Err(Err::Error(("", ErrorKind::Tag)))); +/// assert_eq!(parser("Something"), Err(Err::Error(Error::new("Something", ErrorKind::Tag)))); +/// assert_eq!(parser(""), Err(Err::Error(Error::new("", ErrorKind::Tag)))); /// ``` -pub fn tag<'a, T: 'a, Input: 'a, Error: ParseError<Input>>(tag: T) -> impl Fn(Input) -> IResult<Input, Input, Error> +pub fn tag<T, Input, Error: ParseError<Input>>( + tag: T, +) -> impl Fn(Input) -> IResult<Input, Input, Error> where Input: InputTake + Compare<T>, T: InputLength + Clone, @@ -46,16 +51,16 @@ where } } -/// Recognizes a case insensitive pattern +/// Recognizes a case insensitive pattern. /// /// The input data will be compared to the tag combinator's argument and will return the part of -/// the input that matches the argument with no regard to case +/// the input that matches the argument with no regard to case. /// -/// It will return `Err(Err::Error((_, ErrorKind::Tag)))` if the input doesn't match the pattern +/// It will return `Err(Err::Error((_, ErrorKind::Tag)))` if the input doesn't match the pattern. /// # Example /// ```rust /// # #[macro_use] extern crate nom; -/// # use nom::{Err, error::ErrorKind, Needed, IResult}; +/// # use nom::{Err, error::{Error, ErrorKind}, Needed, IResult}; /// use nom::bytes::complete::tag_no_case; /// /// fn parser(s: &str) -> IResult<&str, &str> { @@ -65,10 +70,12 @@ where /// assert_eq!(parser("Hello, World!"), Ok((", World!", "Hello"))); /// assert_eq!(parser("hello, World!"), Ok((", World!", "hello"))); /// assert_eq!(parser("HeLlO, World!"), Ok((", World!", "HeLlO"))); -/// assert_eq!(parser("Something"), Err(Err::Error(("Something", ErrorKind::Tag)))); -/// assert_eq!(parser(""), Err(Err::Error(("", ErrorKind::Tag)))); +/// assert_eq!(parser("Something"), Err(Err::Error(Error::new("Something", ErrorKind::Tag)))); +/// assert_eq!(parser(""), Err(Err::Error(Error::new("", ErrorKind::Tag)))); /// ``` -pub fn tag_no_case<T, Input, Error: ParseError<Input>>(tag: T) -> impl Fn(Input) -> IResult<Input, Input, Error> +pub fn tag_no_case<T, Input, Error: ParseError<Input>>( + tag: T, +) -> impl Fn(Input) -> IResult<Input, Input, Error> where Input: InputTake + Compare<T>, T: InputLength + Clone, @@ -88,17 +95,17 @@ where } } -/// Parse till certain characters are met +/// Parse till certain characters are met. /// /// The parser will return the longest slice till one of the characters of the combinator's argument are met. /// -/// It doesn't consume the matched character, +/// It doesn't consume the matched character. /// -/// It will return a `Err::Error(("", ErrorKind::IsNot))` if the pattern wasn't met +/// It will return a `Err::Error(("", ErrorKind::IsNot))` if the pattern wasn't met. /// # Example /// ```rust /// # #[macro_use] extern crate nom; -/// # use nom::{Err, error::ErrorKind, Needed, IResult}; +/// # use nom::{Err, error::{Error, ErrorKind}, Needed, IResult}; /// use nom::bytes::complete::is_not; /// /// fn not_space(s: &str) -> IResult<&str, &str> { @@ -108,12 +115,14 @@ where /// assert_eq!(not_space("Hello, World!"), Ok((" World!", "Hello,"))); /// assert_eq!(not_space("Sometimes\t"), Ok(("\t", "Sometimes"))); /// assert_eq!(not_space("Nospace"), Ok(("", "Nospace"))); -/// assert_eq!(not_space(""), Err(Err::Error(("", ErrorKind::IsNot)))); +/// assert_eq!(not_space(""), Err(Err::Error(Error::new("", ErrorKind::IsNot)))); /// ``` -pub fn is_not<T, Input, Error: ParseError<Input>>(arr: T) -> impl Fn(Input) -> IResult<Input, Input, Error> +pub fn is_not<T, Input, Error: ParseError<Input>>( + arr: T, +) -> impl Fn(Input) -> IResult<Input, Input, Error> where Input: InputTakeAtPosition, - T: InputLength + FindToken<<Input as InputTakeAtPosition>::Item>, + T: FindToken<<Input as InputTakeAtPosition>::Item>, { move |i: Input| { let e: ErrorKind = ErrorKind::IsNot; @@ -121,17 +130,16 @@ where } } -/// Returns the longest slice of the matches the pattern +/// Returns the longest slice of the matches the pattern. /// /// The parser will return the longest slice consisting of the characters in provided in the -/// combinator's argument -/// -/// It will return a `Err(Err::Error((_, ErrorKind::IsA)))` if the pattern wasn't met +/// combinator's argument. /// +/// It will return a `Err(Err::Error((_, ErrorKind::IsA)))` if the pattern wasn't met. /// # Example /// ```rust /// # #[macro_use] extern crate nom; -/// # use nom::{Err, error::ErrorKind, Needed, IResult}; +/// # use nom::{Err, error::{Error, ErrorKind}, Needed, IResult}; /// use nom::bytes::complete::is_a; /// /// fn hex(s: &str) -> IResult<&str, &str> { @@ -142,12 +150,14 @@ where /// assert_eq!(hex("DEADBEEF and others"), Ok((" and others", "DEADBEEF"))); /// assert_eq!(hex("BADBABEsomething"), Ok(("something", "BADBABE"))); /// assert_eq!(hex("D15EA5E"), Ok(("", "D15EA5E"))); -/// assert_eq!(hex(""), Err(Err::Error(("", ErrorKind::IsA)))); +/// assert_eq!(hex(""), Err(Err::Error(Error::new("", ErrorKind::IsA)))); /// ``` -pub fn is_a<T, Input, Error: ParseError<Input>>(arr: T) -> impl Fn(Input) -> IResult<Input, Input, Error> +pub fn is_a<T, Input, Error: ParseError<Input>>( + arr: T, +) -> impl Fn(Input) -> IResult<Input, Input, Error> where Input: InputTakeAtPosition, - T: InputLength + FindToken<<Input as InputTakeAtPosition>::Item>, + T: FindToken<<Input as InputTakeAtPosition>::Item>, { move |i: Input| { let e: ErrorKind = ErrorKind::IsA; @@ -155,11 +165,10 @@ where } } -/// Returns the longest input slice (if any) that matches the predicate +/// Returns the longest input slice (if any) that matches the predicate. /// /// The parser will return the longest slice that matches the given predicate *(a function that -/// takes the input and returns a bool)* -/// +/// takes the input and returns a bool)*. /// # Example /// ```rust /// # #[macro_use] extern crate nom; @@ -176,7 +185,9 @@ where /// assert_eq!(alpha(b"latin"), Ok((&b""[..], &b"latin"[..]))); /// assert_eq!(alpha(b""), Ok((&b""[..], &b""[..]))); /// ``` -pub fn take_while<F, Input, Error: ParseError<Input>>(cond: F) -> impl Fn(Input) -> IResult<Input, Input, Error> +pub fn take_while<F, Input, Error: ParseError<Input>>( + cond: F, +) -> impl Fn(Input) -> IResult<Input, Input, Error> where Input: InputTakeAtPosition, F: Fn(<Input as InputTakeAtPosition>::Item) -> bool, @@ -184,17 +195,16 @@ where move |i: Input| i.split_at_position_complete(|c| !cond(c)) } -/// Returns the longest (atleast 1) input slice that matches the predicate +/// Returns the longest (at least 1) input slice that matches the predicate. /// /// The parser will return the longest slice that matches the given predicate *(a function that -/// takes the input and returns a bool)* -/// -/// It will return an `Err(Err::Error((_, ErrorKind::TakeWhile1)))` if the pattern wasn't met +/// takes the input and returns a bool)*. /// +/// It will return an `Err(Err::Error((_, ErrorKind::TakeWhile1)))` if the pattern wasn't met. /// # Example /// ```rust /// # #[macro_use] extern crate nom; -/// # use nom::{Err, error::ErrorKind, Needed, IResult}; +/// # use nom::{Err, error::{Error, ErrorKind}, Needed, IResult}; /// use nom::bytes::complete::take_while1; /// use nom::character::is_alphabetic; /// @@ -204,9 +214,11 @@ where /// /// assert_eq!(alpha(b"latin123"), Ok((&b"123"[..], &b"latin"[..]))); /// assert_eq!(alpha(b"latin"), Ok((&b""[..], &b"latin"[..]))); -/// assert_eq!(alpha(b"12345"), Err(Err::Error((&b"12345"[..], ErrorKind::TakeWhile1)))); +/// assert_eq!(alpha(b"12345"), Err(Err::Error(Error::new(&b"12345"[..], ErrorKind::TakeWhile1)))); /// ``` -pub fn take_while1<F, Input, Error: ParseError<Input>>(cond: F) -> impl Fn(Input) -> IResult<Input, Input, Error> +pub fn take_while1<F, Input, Error: ParseError<Input>>( + cond: F, +) -> impl Fn(Input) -> IResult<Input, Input, Error> where Input: InputTakeAtPosition, F: Fn(<Input as InputTakeAtPosition>::Item) -> bool, @@ -217,18 +229,17 @@ where } } -/// Returns the longest (m <= len <= n) input slice that matches the predicate +/// Returns the longest (m <= len <= n) input slice that matches the predicate. /// /// The parser will return the longest slice that matches the given predicate *(a function that -/// takes the input and returns a bool)* +/// takes the input and returns a bool)*. /// /// It will return an `Err::Error((_, ErrorKind::TakeWhileMN))` if the pattern wasn't met or is out -/// of range (m <= len <= n) -/// +/// of range (m <= len <= n). /// # Example /// ```rust /// # #[macro_use] extern crate nom; -/// # use nom::{Err, error::ErrorKind, Needed, IResult}; +/// # use nom::{Err, error::{Error, ErrorKind}, Needed, IResult}; /// use nom::bytes::complete::take_while_m_n; /// use nom::character::is_alphabetic; /// @@ -239,10 +250,14 @@ where /// assert_eq!(short_alpha(b"latin123"), Ok((&b"123"[..], &b"latin"[..]))); /// assert_eq!(short_alpha(b"lengthy"), Ok((&b"y"[..], &b"length"[..]))); /// assert_eq!(short_alpha(b"latin"), Ok((&b""[..], &b"latin"[..]))); -/// assert_eq!(short_alpha(b"ed"), Err(Err::Error((&b"ed"[..], ErrorKind::TakeWhileMN)))); -/// assert_eq!(short_alpha(b"12345"), Err(Err::Error((&b"12345"[..], ErrorKind::TakeWhileMN)))); +/// assert_eq!(short_alpha(b"ed"), Err(Err::Error(Error::new(&b"ed"[..], ErrorKind::TakeWhileMN)))); +/// assert_eq!(short_alpha(b"12345"), Err(Err::Error(Error::new(&b"12345"[..], ErrorKind::TakeWhileMN)))); /// ``` -pub fn take_while_m_n<F, Input, Error: ParseError<Input>>(m: usize, n: usize, cond: F) -> impl Fn(Input) -> IResult<Input, Input, Error> +pub fn take_while_m_n<F, Input, Error: ParseError<Input>>( + m: usize, + n: usize, + cond: F, +) -> impl Fn(Input) -> IResult<Input, Input, Error> where Input: InputTake + InputIter + InputLength + Slice<RangeFrom<usize>>, F: Fn(<Input as InputIter>::Item) -> bool, @@ -254,17 +269,23 @@ where Some(idx) => { if idx >= m { if idx <= n { - let res: IResult<_, _, Error> = if let Some(index) = input.slice_index(idx) { + let res: IResult<_, _, Error> = if let Ok(index) = input.slice_index(idx) { Ok(input.take_split(index)) } else { - Err(Err::Error(Error::from_error_kind(input, ErrorKind::TakeWhileMN))) + Err(Err::Error(Error::from_error_kind( + input, + ErrorKind::TakeWhileMN, + ))) }; res } else { - let res: IResult<_, _, Error> = if let Some(index) = input.slice_index(n) { + let res: IResult<_, _, Error> = if let Ok(index) = input.slice_index(n) { Ok(input.take_split(index)) } else { - Err(Err::Error(Error::from_error_kind(input, ErrorKind::TakeWhileMN))) + Err(Err::Error(Error::from_error_kind( + input, + ErrorKind::TakeWhileMN, + ))) }; res } @@ -277,28 +298,28 @@ where let len = input.input_len(); if len >= n { match input.slice_index(n) { - Some(index) => Ok(input.take_split(index)), - None => Err(Err::Error(Error::from_error_kind(input, ErrorKind::TakeWhileMN))) + Ok(index) => Ok(input.take_split(index)), + Err(_needed) => Err(Err::Error(Error::from_error_kind( + input, + ErrorKind::TakeWhileMN, + ))), } + } else if len >= m && len <= n { + let res: IResult<_, _, Error> = Ok((input.slice(len..), input)); + res } else { - if len >= m && len <= n { - let res: IResult<_, _, Error> = Ok((input.slice(len..), input)); - res - } else { - let e = ErrorKind::TakeWhileMN; - Err(Err::Error(Error::from_error_kind(input, e))) - } + let e = ErrorKind::TakeWhileMN; + Err(Err::Error(Error::from_error_kind(input, e))) } } } } } -/// Returns the longest input slice (if any) till a predicate is met +/// Returns the longest input slice (if any) till a predicate is met. /// /// The parser will return the longest slice till the given predicate *(a function that -/// takes the input and returns a bool)* -/// +/// takes the input and returns a bool)*. /// # Example /// ```rust /// # #[macro_use] extern crate nom; @@ -314,7 +335,9 @@ where /// assert_eq!(till_colon("12345"), Ok(("", "12345"))); /// assert_eq!(till_colon(""), Ok(("", ""))); /// ``` -pub fn take_till<F, Input, Error: ParseError<Input>>(cond: F) -> impl Fn(Input) -> IResult<Input, Input, Error> +pub fn take_till<F, Input, Error: ParseError<Input>>( + cond: F, +) -> impl Fn(Input) -> IResult<Input, Input, Error> where Input: InputTakeAtPosition, F: Fn(<Input as InputTakeAtPosition>::Item) -> bool, @@ -322,18 +345,17 @@ where move |i: Input| i.split_at_position_complete(|c| cond(c)) } -/// Returns the longest (atleast 1) input slice till a predicate is met +/// Returns the longest (at least 1) input slice till a predicate is met. /// /// The parser will return the longest slice till the given predicate *(a function that -/// takes the input and returns a bool)* +/// takes the input and returns a bool)*. /// /// It will return `Err(Err::Error((_, ErrorKind::TakeTill1)))` if the input is empty or the -/// predicate matches the first input -/// +/// predicate matches the first input. /// # Example /// ```rust /// # #[macro_use] extern crate nom; -/// # use nom::{Err, error::ErrorKind, Needed, IResult}; +/// # use nom::{Err, error::{Error, ErrorKind}, Needed, IResult}; /// use nom::bytes::complete::take_till1; /// /// fn till_colon(s: &str) -> IResult<&str, &str> { @@ -341,11 +363,13 @@ where /// } /// /// assert_eq!(till_colon("latin:123"), Ok((":123", "latin"))); -/// assert_eq!(till_colon(":empty matched"), Err(Err::Error((":empty matched", ErrorKind::TakeTill1)))); +/// assert_eq!(till_colon(":empty matched"), Err(Err::Error(Error::new(":empty matched", ErrorKind::TakeTill1)))); /// assert_eq!(till_colon("12345"), Ok(("", "12345"))); -/// assert_eq!(till_colon(""), Err(Err::Error(("", ErrorKind::TakeTill1)))); +/// assert_eq!(till_colon(""), Err(Err::Error(Error::new("", ErrorKind::TakeTill1)))); /// ``` -pub fn take_till1<F, Input, Error: ParseError<Input>>(cond: F) -> impl Fn(Input) -> IResult<Input, Input, Error> +pub fn take_till1<F, Input, Error: ParseError<Input>>( + cond: F, +) -> impl Fn(Input) -> IResult<Input, Input, Error> where Input: InputTakeAtPosition, F: Fn(<Input as InputTakeAtPosition>::Item) -> bool, @@ -356,14 +380,13 @@ where } } -/// Returns an input slice containing the first N input elements (Input[..N]) -/// -/// It will return `Err(Err::Error((_, ErrorKind::Eof)))` if the input is shorter than the argument +/// Returns an input slice containing the first N input elements (Input[..N]). /// +/// It will return `Err(Err::Error((_, ErrorKind::Eof)))` if the input is shorter than the argument. /// # Example /// ```rust /// # #[macro_use] extern crate nom; -/// # use nom::{Err, error::ErrorKind, Needed, IResult}; +/// # use nom::{Err, error::{Error, ErrorKind}, Needed, IResult}; /// use nom::bytes::complete::take; /// /// fn take6(s: &str) -> IResult<&str, &str> { @@ -372,30 +395,31 @@ where /// /// assert_eq!(take6("1234567"), Ok(("7", "123456"))); /// assert_eq!(take6("things"), Ok(("", "things"))); -/// assert_eq!(take6("short"), Err(Err::Error(("short", ErrorKind::Eof)))); -/// assert_eq!(take6(""), Err(Err::Error(("", ErrorKind::Eof)))); +/// assert_eq!(take6("short"), Err(Err::Error(Error::new("short", ErrorKind::Eof)))); +/// assert_eq!(take6(""), Err(Err::Error(Error::new("", ErrorKind::Eof)))); /// ``` -pub fn take<C, Input, Error: ParseError<Input>>(count: C) -> impl Fn(Input) -> IResult<Input, Input, Error> +pub fn take<C, Input, Error: ParseError<Input>>( + count: C, +) -> impl Fn(Input) -> IResult<Input, Input, Error> where Input: InputIter + InputTake, C: ToUsize, { let c = count.to_usize(); move |i: Input| match i.slice_index(c) { - None => Err(Err::Error(Error::from_error_kind(i, ErrorKind::Eof))), - Some(index) => Ok(i.take_split(index)), + Err(_needed) => Err(Err::Error(Error::from_error_kind(i, ErrorKind::Eof))), + Ok(index) => Ok(i.take_split(index)), } } -/// Returns the longest input slice till it matches the pattern. +/// Returns the input slice up to the first occurrence of the pattern. /// /// It doesn't consume the pattern. It will return `Err(Err::Error((_, ErrorKind::TakeUntil)))` -/// if the pattern wasn't met -/// +/// if the pattern wasn't met. /// # Example /// ```rust /// # #[macro_use] extern crate nom; -/// # use nom::{Err, error::ErrorKind, Needed, IResult}; +/// # use nom::{Err, error::{Error, ErrorKind}, Needed, IResult}; /// use nom::bytes::complete::take_until; /// /// fn until_eof(s: &str) -> IResult<&str, &str> { @@ -403,10 +427,13 @@ where /// } /// /// assert_eq!(until_eof("hello, worldeof"), Ok(("eof", "hello, world"))); -/// assert_eq!(until_eof("hello, world"), Err(Err::Error(("hello, world", ErrorKind::TakeUntil)))); -/// assert_eq!(until_eof(""), Err(Err::Error(("", ErrorKind::TakeUntil)))); +/// assert_eq!(until_eof("hello, world"), Err(Err::Error(Error::new("hello, world", ErrorKind::TakeUntil)))); +/// assert_eq!(until_eof(""), Err(Err::Error(Error::new("", ErrorKind::TakeUntil)))); +/// assert_eq!(until_eof("1eof2eof"), Ok(("eof2eof", "1"))); /// ``` -pub fn take_until<T, Input, Error: ParseError<Input>>(tag: T) -> impl Fn(Input) -> IResult<Input, Input, Error> +pub fn take_until<T, Input, Error: ParseError<Input>>( + tag: T, +) -> impl Fn(Input) -> IResult<Input, Input, Error> where Input: InputTake + FindSubstring<T>, T: InputLength + Clone, @@ -423,10 +450,9 @@ where /// Matches a byte string with escaped characters. /// -/// * The first argument matches the normal characters (it must not accept the control character), -/// * the second argument is the control character (like `\` in most languages), -/// * the third argument matches the escaped characters -/// +/// * The first argument matches the normal characters (it must not accept the control character) +/// * The second argument is the control character (like `\` in most languages) +/// * The third argument matches the escaped characters /// # Example /// ``` /// # #[macro_use] extern crate nom; @@ -443,12 +469,22 @@ where /// assert_eq!(esc(r#"12\"34;"#), Ok((";", r#"12\"34"#))); /// ``` /// -pub fn escaped<Input, Error, F, G, O1, O2>(normal: F, control_char: char, escapable: G) -> impl Fn(Input) -> IResult<Input, Input, Error> +pub fn escaped<'a, Input: 'a, Error, F, G, O1, O2>( + mut normal: F, + control_char: char, + mut escapable: G, +) -> impl FnMut(Input) -> IResult<Input, Input, Error> where - Input: Clone + crate::traits::Offset + InputLength + InputTake + InputTakeAtPosition + Slice<RangeFrom<usize>> + InputIter, + Input: Clone + + crate::traits::Offset + + InputLength + + InputTake + + InputTakeAtPosition + + Slice<RangeFrom<usize>> + + InputIter, <Input as InputIter>::Item: crate::traits::AsChar, - F: Fn(Input) -> IResult<Input, O1, Error>, - G: Fn(Input) -> IResult<Input, O2, Error>, + F: Parser<Input, O1, Error>, + G: Parser<Input, O2, Error>, Error: ParseError<Input>, { use crate::traits::AsChar; @@ -457,7 +493,7 @@ where let mut i = input.clone(); while i.input_len() > 0 { - match normal(i.clone()) { + match normal.parse(i.clone()) { Ok((i2, _)) => { if i2.input_len() == 0 { return Ok((input.slice(input.input_len()..), input)); @@ -470,9 +506,12 @@ where if i.iter_elements().next().unwrap().as_char() == control_char { let next = control_char.len_utf8(); if next >= i.input_len() { - return Err(Err::Error(Error::from_error_kind(input, ErrorKind::Escaped))); + return Err(Err::Error(Error::from_error_kind( + input, + ErrorKind::Escaped, + ))); } else { - match escapable(i.slice(next..)) { + match escapable.parse(i.slice(next..)) { Ok((i2, _)) => { if i2.input_len() == 0 { return Ok((input.slice(input.input_len()..), input)); @@ -486,7 +525,10 @@ where } else { let index = input.offset(&i); if index == 0 { - return Err(Err::Error(Error::from_error_kind(input, ErrorKind::Escaped))); + return Err(Err::Error(Error::from_error_kind( + input, + ErrorKind::Escaped, + ))); } return Ok(input.take_split(index)); } @@ -502,9 +544,20 @@ where } #[doc(hidden)] -pub fn escapedc<Input, Error, F, G, O1, O2>(i: Input, normal: F, control_char: char, escapable: G) -> IResult<Input, Input, Error> +pub fn escapedc<Input, Error, F, G, O1, O2>( + i: Input, + normal: F, + control_char: char, + escapable: G, +) -> IResult<Input, Input, Error> where - Input: Clone + crate::traits::Offset + InputLength + InputTake + InputTakeAtPosition + Slice<RangeFrom<usize>> + InputIter, + Input: Clone + + crate::traits::Offset + + InputLength + + InputTake + + InputTakeAtPosition + + Slice<RangeFrom<usize>> + + InputIter, <Input as InputIter>::Item: crate::traits::AsChar, F: Fn(Input) -> IResult<Input, O1, Error>, G: Fn(Input) -> IResult<Input, O2, Error>, @@ -515,9 +568,9 @@ where /// Matches a byte string with escaped characters. /// -/// * The first argument matches the normal characters (it must not match the control character), -/// * the second argument is the control character (like `\` in most languages), -/// * the third argument matches the escaped characters and transforms them. +/// * The first argument matches the normal characters (it must not match the control character) +/// * The second argument is the control character (like `\` in most languages) +/// * The third argument matches the escaped characters and transforms them /// /// As an example, the chain `abc\tdef` could be `abc def` (it also consumes the control character) /// @@ -525,40 +578,47 @@ where /// # #[macro_use] extern crate nom; /// # use nom::{Err, error::ErrorKind, Needed, IResult}; /// # use std::str::from_utf8; -/// use nom::bytes::complete::escaped_transform; +/// use nom::bytes::complete::{escaped_transform, tag}; /// use nom::character::complete::alpha1; +/// use nom::branch::alt; +/// use nom::combinator::value; /// /// fn parser(input: &str) -> IResult<&str, String> { /// escaped_transform( /// alpha1, /// '\\', -/// |i:&str| alt!(i, -/// tag!("\\") => { |_| "\\" } -/// | tag!("\"") => { |_| "\"" } -/// | tag!("n") => { |_| "\n" } -/// ) +/// alt(( +/// value("\\", tag("\\")), +/// value("\"", tag("\"")), +/// value("\n", tag("n")), +/// )) /// )(input) /// } /// /// assert_eq!(parser("ab\\\"cd"), Ok(("", String::from("ab\"cd")))); +/// assert_eq!(parser("ab\\ncd"), Ok(("", String::from("ab\ncd")))); /// ``` #[cfg(feature = "alloc")] +#[cfg_attr(feature = "docsrs", doc(cfg(feature = "alloc")))] pub fn escaped_transform<Input, Error, F, G, O1, O2, ExtendItem, Output>( - normal: F, + mut normal: F, control_char: char, - transform: G, -) -> impl Fn(Input) -> IResult<Input, Output, Error> + mut transform: G, +) -> impl FnMut(Input) -> IResult<Input, Output, Error> where - Input: Clone + crate::traits::Offset + InputLength + InputTake + InputTakeAtPosition + Slice<RangeFrom<usize>> + InputIter, + Input: Clone + + crate::traits::Offset + + InputLength + + InputTake + + InputTakeAtPosition + + Slice<RangeFrom<usize>> + + InputIter, Input: crate::traits::ExtendInto<Item = ExtendItem, Extender = Output>, O1: crate::traits::ExtendInto<Item = ExtendItem, Extender = Output>, O2: crate::traits::ExtendInto<Item = ExtendItem, Extender = Output>, - Output: core::iter::Extend<<Input as crate::traits::ExtendInto>::Item>, - Output: core::iter::Extend<<O1 as crate::traits::ExtendInto>::Item>, - Output: core::iter::Extend<<O2 as crate::traits::ExtendInto>::Item>, <Input as InputIter>::Item: crate::traits::AsChar, - F: Fn(Input) -> IResult<Input, O1, Error>, - G: Fn(Input) -> IResult<Input, O2, Error>, + F: Parser<Input, O1, Error>, + G: Parser<Input, O2, Error>, Error: ParseError<Input>, { use crate::traits::AsChar; @@ -571,7 +631,7 @@ where while index < i.input_len() { let remainder = i.slice(index..); - match normal(remainder.clone()) { + match normal.parse(remainder.clone()) { Ok((i2, o)) => { o.extend_into(&mut res); if i2.input_len() == 0 { @@ -587,9 +647,12 @@ where let input_len = input.input_len(); if next >= input_len { - return Err(Err::Error(Error::from_error_kind(remainder, ErrorKind::EscapedTransform))); + return Err(Err::Error(Error::from_error_kind( + remainder, + ErrorKind::EscapedTransform, + ))); } else { - match transform(i.slice(next..)) { + match transform.parse(i.slice(next..)) { Ok((i2, o)) => { o.extend_into(&mut res); if i2.input_len() == 0 { @@ -603,7 +666,10 @@ where } } else { if index == 0 { - return Err(Err::Error(Error::from_error_kind(remainder, ErrorKind::EscapedTransform))); + return Err(Err::Error(Error::from_error_kind( + remainder, + ErrorKind::EscapedTransform, + ))); } return Ok((remainder, res)); } @@ -617,6 +683,7 @@ where #[doc(hidden)] #[cfg(feature = "alloc")] +#[cfg_attr(feature = "docsrs", doc(cfg(feature = "alloc")))] pub fn escaped_transformc<Input, Error, F, G, O1, O2, ExtendItem, Output>( i: Input, normal: F, @@ -624,20 +691,22 @@ pub fn escaped_transformc<Input, Error, F, G, O1, O2, ExtendItem, Output>( transform: G, ) -> IResult<Input, Output, Error> where - Input: Clone + crate::traits::Offset + InputLength + InputTake + InputTakeAtPosition + Slice<RangeFrom<usize>> + InputIter, + Input: Clone + + crate::traits::Offset + + InputLength + + InputTake + + InputTakeAtPosition + + Slice<RangeFrom<usize>> + + InputIter, Input: crate::traits::ExtendInto<Item = ExtendItem, Extender = Output>, O1: crate::traits::ExtendInto<Item = ExtendItem, Extender = Output>, O2: crate::traits::ExtendInto<Item = ExtendItem, Extender = Output>, - Output: core::iter::Extend<<Input as crate::traits::ExtendInto>::Item>, - Output: core::iter::Extend<<O1 as crate::traits::ExtendInto>::Item>, - Output: core::iter::Extend<<O2 as crate::traits::ExtendInto>::Item>, <Input as InputIter>::Item: crate::traits::AsChar, F: Fn(Input) -> IResult<Input, O1, Error>, G: Fn(Input) -> IResult<Input, O2, Error>, Error: ParseError<Input>, { escaped_transform(normal, control_char, transform)(i) - } #[cfg(test)] @@ -646,13 +715,15 @@ mod tests { #[test] fn complete_take_while_m_n_utf8_all_matching() { - let result: IResult<&str, &str> = super::take_while_m_n(1, 4, |c: char| c.is_alphabetic())("Ćøn"); + let result: IResult<&str, &str> = + super::take_while_m_n(1, 4, |c: char| c.is_alphabetic())("Ćøn"); assert_eq!(result, Ok(("", "Ćøn"))); } #[test] fn complete_take_while_m_n_utf8_all_matching_substring() { - let result: IResult<&str, &str> = super::take_while_m_n(1, 1, |c: char| c.is_alphabetic())("Ćøn"); + let result: IResult<&str, &str> = + super::take_while_m_n(1, 1, |c: char| c.is_alphabetic())("Ćøn"); assert_eq!(result, Ok(("n", "Ćø"))); } } diff --git a/src/bytes/macros.rs b/src/bytes/macros.rs index 1e24ba8..eedb040 100644 --- a/src/bytes/macros.rs +++ b/src/bytes/macros.rs @@ -3,10 +3,9 @@ #[allow(unused_variables)] /// `tag!(&[T]: nom::AsBytes) => &[T] -> IResult<&[T], &[T]>` -/// declares a byte array as a suite to recognize -/// -/// consumes the recognized characters +/// declares a byte array as a suite to recognize. /// +/// Consumes the recognized characters. /// # Example /// ``` /// # #[macro_use] extern crate nom; @@ -24,10 +23,9 @@ macro_rules! tag ( ); /// `tag_no_case!(&[T]) => &[T] -> IResult<&[T], &[T]>` -/// declares a case insensitive ascii string as a suite to recognize -/// -/// consumes the recognized characters +/// declares a case insensitive ascii string as a suite to recognize. /// +/// Consumes the recognized characters. /// # Example /// ``` /// # #[macro_use] extern crate nom; @@ -46,8 +44,7 @@ macro_rules! tag_no_case ( ); /// `is_not!(&[T:AsBytes]) => &[T] -> IResult<&[T], &[T]>` -/// returns the longest list of bytes that do not appear in the provided array -/// +/// returns the longest list of bytes that do not appear in the provided array. /// # Example /// ``` /// # #[macro_use] extern crate nom; @@ -66,8 +63,7 @@ macro_rules! is_not ( ); /// `is_a!(&[T]) => &[T] -> IResult<&[T], &[T]>` -/// returns the longest list of bytes that appear in the provided array -/// +/// returns the longest list of bytes that appear in the provided array. /// # Example /// ``` /// # #[macro_use] extern crate nom; @@ -92,10 +88,9 @@ macro_rules! is_a ( /// U: AsChar` /// matches a byte string with escaped characters. /// -/// The first argument matches the normal characters (it must not accept the control character), -/// the second argument is the control character (like `\` in most languages), -/// the third argument matches the escaped characters -/// +/// * The first argument matches the normal characters (it must not accept the control character) +/// * The second argument is the control character (like `\` in most languages) +/// * The third argument matches the escaped characters /// # Example /// ``` /// # #[macro_use] extern crate nom; @@ -133,12 +128,11 @@ macro_rules! escaped ( /// `escaped_transform!(&[T] -> IResult<&[T], &[T]>, T, &[T] -> IResult<&[T], &[T]>) => &[T] -> IResult<&[T], Vec<T>>` /// matches a byte string with escaped characters. /// -/// The first argument matches the normal characters (it must not match the control character), -/// the second argument is the control character (like `\` in most languages), -/// the third argument matches the escaped characters and transforms them. -/// -/// As an example, the chain `abc\tdef` could be `abc def` (it also consumes the control character) +/// * The first argument matches the normal characters (it must not match the control character) +/// * The second argument is the control character (like `\` in most languages) +/// * The third argument matches the escaped characters and transforms them /// +/// As an example, the chain `abc\tdef` could be `abc def` (it also consumes the control character). /// # Example /// ```rust /// # #[macro_use] extern crate nom; @@ -164,6 +158,7 @@ macro_rules! escaped ( /// # } /// ``` #[cfg(feature = "alloc")] +#[cfg_attr(feature = "docsrs", doc(cfg(feature = "alloc")))] #[macro_export(local_inner_macros)] macro_rules! escaped_transform ( ($i:expr, $submac1:ident!( $($args:tt)* ), $control_char: expr, $submac2:ident!( $($args2:tt)*) ) => ( @@ -192,7 +187,6 @@ macro_rules! escaped_transform ( /// returns the longest list of bytes until the provided function fails. /// /// The argument is either a function `T -> bool` or a macro returning a `bool`. -/// /// # Example /// ``` /// # #[macro_use] extern crate nom; @@ -218,8 +212,7 @@ macro_rules! take_while ( /// `take_while1!(T -> bool) => &[T] -> IResult<&[T], &[T]>` /// returns the longest (non empty) list of bytes until the provided function fails. /// -/// The argument is either a function `&[T] -> bool` or a macro returning a `bool` -/// +/// The argument is either a function `&[T] -> bool` or a macro returning a `bool`. /// # Example /// ``` /// # #[macro_use] extern crate nom; @@ -247,10 +240,9 @@ macro_rules! take_while1 ( /// `take_while_m_n!(m: usize, n: usize, T -> bool) => &[T] -> IResult<&[T], &[T]>` /// returns a list of bytes or characters for which the provided function returns true. -/// the returned list's size will be at least m, and at most n +/// The returned list's size will be at least m, and at most n. /// /// The argument is either a function `T -> bool` or a macro returning a `bool`. -/// /// # Example /// ``` /// # #[macro_use] extern crate nom; @@ -274,10 +266,9 @@ macro_rules! take_while_m_n ( ); /// `take_till!(T -> bool) => &[T] -> IResult<&[T], &[T]>` -/// returns the longest list of bytes until the provided function succeeds +/// returns the longest list of bytes until the provided function succeeds. /// /// The argument is either a function `&[T] -> bool` or a macro returning a `bool`. -/// /// # Example /// ``` /// # #[macro_use] extern crate nom; @@ -302,7 +293,7 @@ macro_rules! take_till ( ); /// `take_till1!(T -> bool) => &[T] -> IResult<&[T], &[T]>` -/// returns the longest non empty list of bytes until the provided function succeeds +/// returns the longest non empty list of bytes until the provided function succeeds. /// /// The argument is either a function `&[T] -> bool` or a macro returning a `bool`. /// @@ -332,8 +323,7 @@ macro_rules! take_till1 ( ); /// `take!(nb) => &[T] -> IResult<&[T], &[T]>` -/// generates a parser consuming the specified number of bytes -/// +/// generates a parser consuming the specified number of bytes. /// # Example /// ``` /// # #[macro_use] extern crate nom; @@ -356,8 +346,7 @@ macro_rules! take ( ); /// `take_str!(nb) => &[T] -> IResult<&[T], &str>` -/// same as take! but returning a &str -/// +/// same as `take!` but returning a `&str`. /// # Example /// ``` /// # #[macro_use] extern crate nom; @@ -384,7 +373,6 @@ macro_rules! take_str ( /// consumes data until it finds the specified tag. /// /// The remainder still contains the tag. -/// /// # Example /// ``` /// # #[macro_use] extern crate nom; @@ -403,10 +391,9 @@ macro_rules! take_until ( ); /// `take_until1!(tag) => &[T] -> IResult<&[T], &[T]>` -/// consumes data (at least one byte) until it finds the specified tag +/// consumes data (at least one byte) until it finds the specified tag. /// /// The remainder still contains the tag. -/// /// # Example /// ``` /// # #[macro_use] extern crate nom; @@ -432,7 +419,7 @@ macro_rules! take_until1 ( let res: IResult<_,_> = match input.find_substring($substr) { None => { - Err(Err::Incomplete(Needed::Size(1 + $substr.input_len()))) + Err(Err::Incomplete(Needed::new(1 + $substr.input_len()))) }, Some(0) => { let e = ErrorKind::TakeUntil; @@ -449,14 +436,17 @@ macro_rules! take_until1 ( #[cfg(test)] mod tests { - use crate::internal::{Err, Needed, IResult}; + use crate::character::is_alphabetic; + use crate::character::streaming::{ + alpha1 as alpha, alphanumeric1 as alphanumeric, digit1 as digit, hex_digit1 as hex_digit, + multispace1 as multispace, oct_digit1 as oct_digit, space1 as space, + }; + use crate::error::ErrorKind; + use crate::internal::{Err, IResult, Needed}; #[cfg(feature = "alloc")] use crate::lib::std::string::String; #[cfg(feature = "alloc")] use crate::lib::std::vec::Vec; - use crate::character::streaming::{alpha1 as alpha, alphanumeric1 as alphanumeric, digit1 as digit, hex_digit1 as hex_digit, multispace1 as multispace, oct_digit1 as oct_digit, space1 as space}; - use crate::error::ErrorKind; - use crate::character::is_alphabetic; #[cfg(feature = "alloc")] macro_rules! one_of ( @@ -471,7 +461,7 @@ mod tests { match ($i).iter_elements().next().map(|c| { $inp.find_token(c) }) { - None => Err::<_,_>(Err::Incomplete(Needed::Size(1))), + None => Err::<_,_>(Err::Incomplete(Needed::new(1))), Some(false) => Err(Err::Error(error_position!($i, ErrorKind::OneOf))), //the unwrap should be safe here Some(true) => Ok(($i.slice(1..), $i.iter_elements().next().unwrap().as_char())) @@ -491,7 +481,10 @@ mod tests { assert_eq!(a_or_b(b), Ok((&b"cde"[..], &b"b"[..]))); let c = &b"cdef"[..]; - assert_eq!(a_or_b(c), Err(Err::Error(error_position!(c, ErrorKind::IsA)))); + assert_eq!( + a_or_b(c), + Err(Err::Error(error_position!(c, ErrorKind::IsA))) + ); let d = &b"bacdef"[..]; assert_eq!(a_or_b(d), Ok((&b"cdef"[..], &b"ba"[..]))); @@ -508,13 +501,16 @@ mod tests { assert_eq!(a_or_b(b), Ok((&b"bde"[..], &b"c"[..]))); let c = &b"abab"[..]; - assert_eq!(a_or_b(c), Err(Err::Error(error_position!(c, ErrorKind::IsNot)))); + assert_eq!( + a_or_b(c), + Err(Err::Error(error_position!(c, ErrorKind::IsNot))) + ); let d = &b"cdefba"[..]; assert_eq!(a_or_b(d), Ok((&b"ba"[..], &b"cdef"[..]))); let e = &b"e"[..]; - assert_eq!(a_or_b(e), Err(Err::Incomplete(Needed::Size(1)))); + assert_eq!(a_or_b(e), Err(Err::Incomplete(Needed::new(1)))); } #[cfg(feature = "alloc")] @@ -527,7 +523,13 @@ mod tests { assert_eq!(esc(&b"\\\"abcd;"[..]), Ok((&b";"[..], &b"\\\"abcd"[..]))); assert_eq!(esc(&b"\\n;"[..]), Ok((&b";"[..], &b"\\n"[..]))); assert_eq!(esc(&b"ab\\\"12"[..]), Ok((&b"12"[..], &b"ab\\\""[..]))); - assert_eq!(esc(&b"AB\\"[..]), Err(Err::Error(error_position!(&b"AB\\"[..], ErrorKind::Escaped)))); + assert_eq!( + esc(&b"AB\\"[..]), + Err(Err::Error(error_position!( + &b"AB\\"[..], + ErrorKind::Escaped + ))) + ); assert_eq!( esc(&b"AB\\A"[..]), Err(Err::Error(error_node_position!( @@ -550,7 +552,10 @@ mod tests { assert_eq!(esc("\\\"abcd;"), Ok((";", "\\\"abcd"))); assert_eq!(esc("\\n;"), Ok((";", "\\n"))); assert_eq!(esc("ab\\\"12"), Ok(("12", "ab\\\""))); - assert_eq!(esc("AB\\"), Err(Err::Error(error_position!("AB\\", ErrorKind::Escaped)))); + assert_eq!( + esc("AB\\"), + Err(Err::Error(error_position!("AB\\", ErrorKind::Escaped))) + ); assert_eq!( esc("AB\\A"), Err(Err::Error(error_node_position!( @@ -594,11 +599,26 @@ mod tests { ); assert_eq!(esc(&b"abcd;"[..]), Ok((&b";"[..], String::from("abcd")))); - assert_eq!(esc(&b"ab\\\"cd;"[..]), Ok((&b";"[..], String::from("ab\"cd")))); - assert_eq!(esc(&b"\\\"abcd;"[..]), Ok((&b";"[..], String::from("\"abcd")))); + assert_eq!( + esc(&b"ab\\\"cd;"[..]), + Ok((&b";"[..], String::from("ab\"cd"))) + ); + assert_eq!( + esc(&b"\\\"abcd;"[..]), + Ok((&b";"[..], String::from("\"abcd"))) + ); assert_eq!(esc(&b"\\n;"[..]), Ok((&b";"[..], String::from("\n")))); - assert_eq!(esc(&b"ab\\\"12"[..]), Ok((&b"12"[..], String::from("ab\"")))); - assert_eq!(esc(&b"AB\\"[..]), Err(Err::Error(error_position!(&b"\\"[..], ErrorKind::EscapedTransform)))); + assert_eq!( + esc(&b"ab\\\"12"[..]), + Ok((&b"12"[..], String::from("ab\""))) + ); + assert_eq!( + esc(&b"AB\\"[..]), + Err(Err::Error(error_position!( + &b"\\"[..], + ErrorKind::EscapedTransform + ))) + ); assert_eq!( esc(&b"AB\\A"[..]), Err(Err::Error(error_node_position!( @@ -622,8 +642,14 @@ mod tests { to_s ) ); - assert_eq!(esc2(&b"abèDEF;"[..]), Ok((&b";"[..], String::from("abĆØDEF")))); - assert_eq!(esc2(&b"abèDàEF;"[..]), Ok((&b";"[..], String::from("abĆØDĆ EF")))); + assert_eq!( + esc2(&b"abèDEF;"[..]), + Ok((&b";"[..], String::from("abĆØDEF"))) + ); + assert_eq!( + esc2(&b"abèDàEF;"[..]), + Ok((&b";"[..], String::from("abĆØDĆ EF"))) + ); } #[cfg(feature = "std")] @@ -642,7 +668,13 @@ mod tests { assert_eq!(esc("\\\"abcd;"), Ok((";", String::from("\"abcd")))); assert_eq!(esc("\\n;"), Ok((";", String::from("\n")))); assert_eq!(esc("ab\\\"12"), Ok(("12", String::from("ab\"")))); - assert_eq!(esc("AB\\"), Err(Err::Error(error_position!("\\", ErrorKind::EscapedTransform)))); + assert_eq!( + esc("AB\\"), + Err(Err::Error(error_position!( + "\\", + ErrorKind::EscapedTransform + ))) + ); assert_eq!( esc("AB\\A"), Err(Err::Error(error_node_position!( @@ -659,7 +691,10 @@ mod tests { )) ); assert_eq!(esc2("abèDEF;"), Ok((";", String::from("abĆØDEF")))); - assert_eq!(esc2("abèDàEF;"), Ok((";", String::from("abĆØDĆ EF")))); + assert_eq!( + esc2("abèDàEF;"), + Ok((";", String::from("abĆØDĆ EF"))) + ); named!(esc3<&str, String>, escaped_transform!(alpha, 'ā', alt!( @@ -672,30 +707,33 @@ mod tests { fn take_str_test() { let a = b"omnomnom"; - let res: IResult<_,_,(&[u8], ErrorKind)> = take_str!(&a[..], 5u32); + let res: IResult<_, _, (&[u8], ErrorKind)> = take_str!(&a[..], 5u32); assert_eq!(res, Ok((&b"nom"[..], "omnom"))); - let res: IResult<_,_,(&[u8], ErrorKind)> = take_str!(&a[..], 9u32); - assert_eq!(res, Err(Err::Incomplete(Needed::Size(9)))); + let res: IResult<_, _, (&[u8], ErrorKind)> = take_str!(&a[..], 9u32); + assert_eq!(res, Err(Err::Incomplete(Needed::new(1)))); } #[test] fn take_until_incomplete() { named!(y, take_until!("end")); - assert_eq!(y(&b"nd"[..]), Err(Err::Incomplete(Needed::Size(3)))); - assert_eq!(y(&b"123"[..]), Err(Err::Incomplete(Needed::Size(3)))); - assert_eq!(y(&b"123en"[..]), Err(Err::Incomplete(Needed::Size(3)))); + assert_eq!(y(&b"nd"[..]), Err(Err::Incomplete(Needed::Unknown))); + assert_eq!(y(&b"123"[..]), Err(Err::Incomplete(Needed::Unknown))); + assert_eq!(y(&b"123en"[..]), Err(Err::Incomplete(Needed::Unknown))); } #[test] fn take_until_incomplete_s() { named!(ys<&str, &str>, take_until!("end")); - assert_eq!(ys("123en"), Err(Err::Incomplete(Needed::Size(3)))); + assert_eq!(ys("123en"), Err(Err::Incomplete(Needed::Unknown))); } #[test] fn recognize() { - named!(x, recognize!(delimited!(tag!("<!--"), take!(5usize), tag!("-->")))); + named!( + x, + recognize!(delimited!(tag!("<!--"), take!(5usize), tag!("-->"))) + ); let r = x(&b"<!-- abc --> aaa"[..]); assert_eq!(r, Ok((&b" aaa"[..], &b"<!-- abc -->"[..]))); @@ -738,8 +776,8 @@ mod tests { let c = b"abcd123"; let d = b"123"; - assert_eq!(f(&a[..]), Err(Err::Incomplete(Needed::Size(1)))); - assert_eq!(f(&b[..]), Err(Err::Incomplete(Needed::Size(1)))); + assert_eq!(f(&a[..]), Err(Err::Incomplete(Needed::new(1)))); + assert_eq!(f(&b[..]), Err(Err::Incomplete(Needed::new(1)))); assert_eq!(f(&c[..]), Ok((&d[..], &b[..]))); assert_eq!(f(&d[..]), Ok((&d[..], &a[..]))); } @@ -752,10 +790,13 @@ mod tests { let c = b"abcd123"; let d = b"123"; - assert_eq!(f(&a[..]), Err(Err::Incomplete(Needed::Size(1)))); - assert_eq!(f(&b[..]), Err(Err::Incomplete(Needed::Size(1)))); + assert_eq!(f(&a[..]), Err(Err::Incomplete(Needed::new(1)))); + assert_eq!(f(&b[..]), Err(Err::Incomplete(Needed::new(1)))); assert_eq!(f(&c[..]), Ok((&b"123"[..], &b[..]))); - assert_eq!(f(&d[..]), Err(Err::Error(error_position!(&d[..], ErrorKind::TakeWhile1)))); + assert_eq!( + f(&d[..]), + Err(Err::Error(error_position!(&d[..], ErrorKind::TakeWhile1))) + ); } #[test] @@ -768,56 +809,60 @@ mod tests { let e = b"abcde"; let f = b"123"; - assert_eq!(x(&a[..]), Err(Err::Incomplete(Needed::Size(2)))); - assert_eq!(x(&b[..]), Err(Err::Incomplete(Needed::Size(1)))); - assert_eq!(x(&c[..]), Err(Err::Incomplete(Needed::Size(1)))); + assert_eq!(x(&a[..]), Err(Err::Incomplete(Needed::new(2)))); + assert_eq!(x(&b[..]), Err(Err::Incomplete(Needed::new(1)))); + assert_eq!(x(&c[..]), Err(Err::Incomplete(Needed::new(1)))); assert_eq!(x(&d[..]), Ok((&b"123"[..], &c[..]))); assert_eq!(x(&e[..]), Ok((&b"e"[..], &b"abcd"[..]))); - assert_eq!(x(&f[..]), Err(Err::Error(error_position!(&f[..], ErrorKind::TakeWhileMN)))); + assert_eq!( + x(&f[..]), + Err(Err::Error(error_position!(&f[..], ErrorKind::TakeWhileMN))) + ); } #[test] fn take_till() { - named!(f, take_till!(is_alphabetic)); let a = b""; let b = b"abcd"; let c = b"123abcd"; let d = b"123"; - assert_eq!(f(&a[..]), Err(Err::Incomplete(Needed::Size(1)))); + assert_eq!(f(&a[..]), Err(Err::Incomplete(Needed::new(1)))); assert_eq!(f(&b[..]), Ok((&b"abcd"[..], &b""[..]))); assert_eq!(f(&c[..]), Ok((&b"abcd"[..], &b"123"[..]))); - assert_eq!(f(&d[..]), Err(Err::Incomplete(Needed::Size(1)))); + assert_eq!(f(&d[..]), Err(Err::Incomplete(Needed::new(1)))); } #[test] fn take_till1() { - named!(f, take_till1!(is_alphabetic)); let a = b""; let b = b"abcd"; let c = b"123abcd"; let d = b"123"; - assert_eq!(f(&a[..]), Err(Err::Incomplete(Needed::Size(1)))); - assert_eq!(f(&b[..]), Err(Err::Error(error_position!(&b[..], ErrorKind::TakeTill1)))); + assert_eq!(f(&a[..]), Err(Err::Incomplete(Needed::new(1)))); + assert_eq!( + f(&b[..]), + Err(Err::Error(error_position!(&b[..], ErrorKind::TakeTill1))) + ); assert_eq!(f(&c[..]), Ok((&b"abcd"[..], &b"123"[..]))); - assert_eq!(f(&d[..]), Err(Err::Incomplete(Needed::Size(1)))); + assert_eq!(f(&d[..]), Err(Err::Incomplete(Needed::new(1)))); } #[test] fn take_while_utf8() { named!(f<&str,&str>, take_while!(|c:char| { c != 'é»' })); - assert_eq!(f(""), Err(Err::Incomplete(Needed::Size(1)))); - assert_eq!(f("abcd"), Err(Err::Incomplete(Needed::Size(1)))); + assert_eq!(f(""), Err(Err::Incomplete(Needed::new(1)))); + assert_eq!(f("abcd"), Err(Err::Incomplete(Needed::new(1)))); assert_eq!(f("abcdé»"), Ok(("é»", "abcd"))); assert_eq!(f("abcdé»a"), Ok(("é»a", "abcd"))); named!(g<&str,&str>, take_while!(|c:char| { c == 'é»' })); - assert_eq!(g(""), Err(Err::Incomplete(Needed::Size(1)))); + assert_eq!(g(""), Err(Err::Incomplete(Needed::new(1)))); assert_eq!(g("é»abcd"), Ok(("abcd", "é»"))); assert_eq!(g("é»é»é»a"), Ok(("a", "é»é»é»"))); } @@ -826,14 +871,14 @@ mod tests { fn take_till_utf8() { named!(f<&str,&str>, take_till!(|c:char| { c == 'é»' })); - assert_eq!(f(""), Err(Err::Incomplete(Needed::Size(1)))); - assert_eq!(f("abcd"), Err(Err::Incomplete(Needed::Size(1)))); + assert_eq!(f(""), Err(Err::Incomplete(Needed::new(1)))); + assert_eq!(f("abcd"), Err(Err::Incomplete(Needed::new(1)))); assert_eq!(f("abcdé»"), Ok(("é»", "abcd"))); assert_eq!(f("abcdé»a"), Ok(("é»a", "abcd"))); named!(g<&str,&str>, take_till!(|c:char| { c != 'é»' })); - assert_eq!(g(""), Err(Err::Incomplete(Needed::Size(1)))); + assert_eq!(g(""), Err(Err::Incomplete(Needed::new(1)))); assert_eq!(g("é»abcd"), Ok(("abcd", "é»"))); assert_eq!(g("é»é»é»a"), Ok(("a", "é»é»é»"))); } @@ -842,16 +887,16 @@ mod tests { fn take_utf8() { named!(f<&str,&str>, take!(3)); - assert_eq!(f(""), Err(Err::Incomplete(Needed::Size(3)))); - assert_eq!(f("ab"), Err(Err::Incomplete(Needed::Size(3)))); - assert_eq!(f("é»"), Err(Err::Incomplete(Needed::Size(3)))); + assert_eq!(f(""), Err(Err::Incomplete(Needed::Unknown))); + assert_eq!(f("ab"), Err(Err::Incomplete(Needed::Unknown))); + assert_eq!(f("é»"), Err(Err::Incomplete(Needed::Unknown))); assert_eq!(f("abé»cd"), Ok(("cd", "abé»"))); assert_eq!(f("aé»bcd"), Ok(("cd", "aé»b"))); assert_eq!(f("aé»b"), Ok(("", "aé»b"))); named!(g<&str,&str>, take_while!(|c:char| { c == 'é»' })); - assert_eq!(g(""), Err(Err::Incomplete(Needed::Size(1)))); + assert_eq!(g(""), Err(Err::Incomplete(Needed::new(1)))); assert_eq!(g("é»abcd"), Ok(("abcd", "é»"))); assert_eq!(g("é»é»é»a"), Ok(("a", "é»é»é»"))); } @@ -875,7 +920,6 @@ mod tests { #[cfg(nightly)] #[bench] fn take_while_bench(b: &mut Bencher) { - named!(f, take_while!(is_alphabetic)); b.iter(|| f(&b"abcdefghijklABCDEejfrfrjgro12aa"[..])); } @@ -897,14 +941,14 @@ mod tests { named!(x, length_data!(le_u8)); assert_eq!(x(b"\x02..>>"), Ok((&b">>"[..], &b".."[..]))); assert_eq!(x(b"\x02.."), Ok((&[][..], &b".."[..]))); - assert_eq!(x(b"\x02."), Err(Err::Incomplete(Needed::Size(2)))); - assert_eq!(x(b"\x02"), Err(Err::Incomplete(Needed::Size(2)))); + assert_eq!(x(b"\x02."), Err(Err::Incomplete(Needed::new(1)))); + assert_eq!(x(b"\x02"), Err(Err::Incomplete(Needed::new(2)))); named!(y, do_parse!(tag!("magic") >> b: length_data!(le_u8) >> (b))); assert_eq!(y(b"magic\x02..>>"), Ok((&b">>"[..], &b".."[..]))); assert_eq!(y(b"magic\x02.."), Ok((&[][..], &b".."[..]))); - assert_eq!(y(b"magic\x02."), Err(Err::Incomplete(Needed::Size(2)))); - assert_eq!(y(b"magic\x02"), Err(Err::Incomplete(Needed::Size(2)))); + assert_eq!(y(b"magic\x02."), Err(Err::Incomplete(Needed::new(1)))); + assert_eq!(y(b"magic\x02"), Err(Err::Incomplete(Needed::new(2)))); } #[cfg(feature = "alloc")] @@ -914,17 +958,29 @@ mod tests { assert_eq!(test(&b"aBCdefgh"[..]), Ok((&b"efgh"[..], &b"aBCd"[..]))); assert_eq!(test(&b"abcdefgh"[..]), Ok((&b"efgh"[..], &b"abcd"[..]))); assert_eq!(test(&b"ABCDefgh"[..]), Ok((&b"efgh"[..], &b"ABCD"[..]))); - assert_eq!(test(&b"ab"[..]), Err(Err::Incomplete(Needed::Size(4)))); - assert_eq!(test(&b"Hello"[..]), Err(Err::Error(error_position!(&b"Hello"[..], ErrorKind::Tag)))); - assert_eq!(test(&b"Hel"[..]), Err(Err::Error(error_position!(&b"Hel"[..], ErrorKind::Tag)))); + assert_eq!(test(&b"ab"[..]), Err(Err::Incomplete(Needed::new(2)))); + assert_eq!( + test(&b"Hello"[..]), + Err(Err::Error(error_position!(&b"Hello"[..], ErrorKind::Tag))) + ); + assert_eq!( + test(&b"Hel"[..]), + Err(Err::Error(error_position!(&b"Hel"[..], ErrorKind::Tag))) + ); named!(test2<&str, &str>, tag_no_case!("ABcd")); assert_eq!(test2("aBCdefgh"), Ok(("efgh", "aBCd"))); assert_eq!(test2("abcdefgh"), Ok(("efgh", "abcd"))); assert_eq!(test2("ABCDefgh"), Ok(("efgh", "ABCD"))); - assert_eq!(test2("ab"), Err(Err::Incomplete(Needed::Size(4)))); - assert_eq!(test2("Hello"), Err(Err::Error(error_position!(&"Hello"[..], ErrorKind::Tag)))); - assert_eq!(test2("Hel"), Err(Err::Error(error_position!(&"Hel"[..], ErrorKind::Tag)))); + assert_eq!(test2("ab"), Err(Err::Incomplete(Needed::new(2)))); + assert_eq!( + test2("Hello"), + Err(Err::Error(error_position!(&"Hello"[..], ErrorKind::Tag))) + ); + assert_eq!( + test2("Hel"), + Err(Err::Error(error_position!(&"Hel"[..], ErrorKind::Tag))) + ); } #[test] diff --git a/src/bytes/mod.rs b/src/bytes/mod.rs index 13b2012..64fb00d 100644 --- a/src/bytes/mod.rs +++ b/src/bytes/mod.rs @@ -1,7 +1,6 @@ -//! parsers recognizing bytes streams +//! Parsers recognizing bytes streams #[macro_use] mod macros; -pub mod streaming; pub mod complete; - +pub mod streaming; diff --git a/src/bytes/streaming.rs b/src/bytes/streaming.rs index f6e6706..f3f8b2c 100644 --- a/src/bytes/streaming.rs +++ b/src/bytes/streaming.rs @@ -1,20 +1,23 @@ -//! parsers recognizing bytes streams, streaming version +//! Parsers recognizing bytes streams, streaming version use crate::error::ErrorKind; use crate::error::ParseError; -use crate::internal::{Err, IResult, Needed}; +use crate::internal::{Err, IResult, Needed, Parser}; use crate::lib::std::ops::RangeFrom; use crate::lib::std::result::Result::*; -use crate::traits::{Compare, CompareResult, FindSubstring, FindToken, InputIter, InputLength, InputTake, InputTakeAtPosition, Slice, ToUsize}; +use crate::traits::{ + Compare, CompareResult, FindSubstring, FindToken, InputIter, InputLength, InputTake, + InputTakeAtPosition, Slice, ToUsize, +}; -/// Recognizes a pattern +/// Recognizes a pattern. /// /// The input data will be compared to the tag combinator's argument and will return the part of -/// the input that matches the argument +/// the input that matches the argument. /// # Example /// ```rust /// # #[macro_use] extern crate nom; -/// # use nom::{Err, error::ErrorKind, Needed, IResult}; +/// # use nom::{Err, error::{Error, ErrorKind}, Needed, IResult}; /// use nom::bytes::streaming::tag; /// /// fn parser(s: &str) -> IResult<&str, &str> { @@ -22,12 +25,15 @@ use crate::traits::{Compare, CompareResult, FindSubstring, FindToken, InputIter, /// } /// /// assert_eq!(parser("Hello, World!"), Ok((", World!", "Hello"))); -/// assert_eq!(parser("Something"), Err(Err::Error(("Something", ErrorKind::Tag)))); -/// assert_eq!(parser(""), Err(Err::Incomplete(Needed::Size(5)))); +/// assert_eq!(parser("Something"), Err(Err::Error(Error::new("Something", ErrorKind::Tag)))); +/// assert_eq!(parser("S"), Err(Err::Error(Error::new("S", ErrorKind::Tag)))); +/// assert_eq!(parser("H"), Err(Err::Incomplete(Needed::new(4)))); /// ``` -pub fn tag<'a, T: 'a, Input: 'a, Error: ParseError<Input>>(tag: T) -> impl Fn(Input) -> IResult<Input, Input, Error> +pub fn tag<T, Input, Error: ParseError<Input>>( + tag: T, +) -> impl Fn(Input) -> IResult<Input, Input, Error> where - Input: InputTake + Compare<T>, + Input: InputTake + InputLength + Compare<T>, T: InputLength + Clone, { move |i: Input| { @@ -36,7 +42,7 @@ where let res: IResult<_, _, Error> = match i.compare(t) { CompareResult::Ok => Ok(i.take_split(tag_len)), - CompareResult::Incomplete => Err(Err::Incomplete(Needed::Size(tag_len))), + CompareResult::Incomplete => Err(Err::Incomplete(Needed::new(tag_len - i.input_len()))), CompareResult::Error => { let e: ErrorKind = ErrorKind::Tag; Err(Err::Error(Error::from_error_kind(i, e))) @@ -46,14 +52,14 @@ where } } -/// Recognizes a case insensitive pattern +/// Recognizes a case insensitive pattern. /// /// The input data will be compared to the tag combinator's argument and will return the part of -/// the input that matches the argument with no regard to case +/// the input that matches the argument with no regard to case. /// # Example /// ```rust /// # #[macro_use] extern crate nom; -/// # use nom::{Err, error::ErrorKind, Needed, IResult}; +/// # use nom::{Err, error::{Error, ErrorKind}, Needed, IResult}; /// use nom::bytes::streaming::tag_no_case; /// /// fn parser(s: &str) -> IResult<&str, &str> { @@ -63,12 +69,14 @@ where /// assert_eq!(parser("Hello, World!"), Ok((", World!", "Hello"))); /// assert_eq!(parser("hello, World!"), Ok((", World!", "hello"))); /// assert_eq!(parser("HeLlO, World!"), Ok((", World!", "HeLlO"))); -/// assert_eq!(parser("Something"), Err(Err::Error(("Something", ErrorKind::Tag)))); -/// assert_eq!(parser(""), Err(Err::Incomplete(Needed::Size(5)))); +/// assert_eq!(parser("Something"), Err(Err::Error(Error::new("Something", ErrorKind::Tag)))); +/// assert_eq!(parser(""), Err(Err::Incomplete(Needed::new(5)))); /// ``` -pub fn tag_no_case<T, Input, Error: ParseError<Input>>(tag: T) -> impl Fn(Input) -> IResult<Input, Input, Error> +pub fn tag_no_case<T, Input, Error: ParseError<Input>>( + tag: T, +) -> impl Fn(Input) -> IResult<Input, Input, Error> where - Input: InputTake + Compare<T>, + Input: InputTake + InputLength + Compare<T>, T: InputLength + Clone, { move |i: Input| { @@ -77,7 +85,7 @@ where let res: IResult<_, _, Error> = match (i).compare_no_case(t) { CompareResult::Ok => Ok(i.take_split(tag_len)), - CompareResult::Incomplete => Err(Err::Incomplete(Needed::Size(tag_len))), + CompareResult::Incomplete => Err(Err::Incomplete(Needed::new(tag_len - i.input_len()))), CompareResult::Error => { let e: ErrorKind = ErrorKind::Tag; Err(Err::Error(Error::from_error_kind(i, e))) @@ -87,13 +95,13 @@ where } } -/// Parse till certain characters are met +/// Parse till certain characters are met. /// /// The parser will return the longest slice till one of the characters of the combinator's argument are met. /// -/// It doesn't consume the matched character, +/// It doesn't consume the matched character. /// -/// It will return a `Err::Incomplete(Needed::Size(1))` if the pattern wasn't met +/// It will return a `Err::Incomplete(Needed::new(1))` if the pattern wasn't met. /// # Example /// ```rust /// # #[macro_use] extern crate nom; @@ -106,13 +114,15 @@ where /// /// assert_eq!(not_space("Hello, World!"), Ok((" World!", "Hello,"))); /// assert_eq!(not_space("Sometimes\t"), Ok(("\t", "Sometimes"))); -/// assert_eq!(not_space("Nospace"), Err(Err::Incomplete(Needed::Size(1)))); -/// assert_eq!(not_space(""), Err(Err::Incomplete(Needed::Size(1)))); +/// assert_eq!(not_space("Nospace"), Err(Err::Incomplete(Needed::new(1)))); +/// assert_eq!(not_space(""), Err(Err::Incomplete(Needed::new(1)))); /// ``` -pub fn is_not<T, Input, Error: ParseError<Input>>(arr: T) -> impl Fn(Input) -> IResult<Input, Input, Error> +pub fn is_not<T, Input, Error: ParseError<Input>>( + arr: T, +) -> impl Fn(Input) -> IResult<Input, Input, Error> where Input: InputTakeAtPosition, - T: InputLength + FindToken<<Input as InputTakeAtPosition>::Item>, + T: FindToken<<Input as InputTakeAtPosition>::Item>, { move |i: Input| { let e: ErrorKind = ErrorKind::IsNot; @@ -120,14 +130,14 @@ where } } -/// Returns the longest slice of the matches the pattern +/// Returns the longest slice of the matches the pattern. /// /// The parser will return the longest slice consisting of the characters in provided in the -/// combinator's argument +/// combinator's argument. /// /// # Streaming specific -/// *Streaming version* will return a `Err::Incomplete(Needed::Size(1))` if the pattern wasn't met -/// or if the pattern reaches the end of the input +/// *Streaming version* will return a `Err::Incomplete(Needed::new(1))` if the pattern wasn't met +/// or if the pattern reaches the end of the input. /// # Example /// ```rust /// # #[macro_use] extern crate nom; @@ -141,13 +151,15 @@ where /// assert_eq!(hex("123 and voila"), Ok((" and voila", "123"))); /// assert_eq!(hex("DEADBEEF and others"), Ok((" and others", "DEADBEEF"))); /// assert_eq!(hex("BADBABEsomething"), Ok(("something", "BADBABE"))); -/// assert_eq!(hex("D15EA5E"), Err(Err::Incomplete(Needed::Size(1)))); -/// assert_eq!(hex(""), Err(Err::Incomplete(Needed::Size(1)))); +/// assert_eq!(hex("D15EA5E"), Err(Err::Incomplete(Needed::new(1)))); +/// assert_eq!(hex(""), Err(Err::Incomplete(Needed::new(1)))); /// ``` -pub fn is_a<T, Input, Error: ParseError<Input>>(arr: T) -> impl Fn(Input) -> IResult<Input, Input, Error> +pub fn is_a<T, Input, Error: ParseError<Input>>( + arr: T, +) -> impl Fn(Input) -> IResult<Input, Input, Error> where Input: InputTakeAtPosition, - T: InputLength + FindToken<<Input as InputTakeAtPosition>::Item>, + T: FindToken<<Input as InputTakeAtPosition>::Item>, { move |i: Input| { let e: ErrorKind = ErrorKind::IsA; @@ -155,13 +167,13 @@ where } } -/// Returns the longest input slice (if any) that matches the predicate +/// Returns the longest input slice (if any) that matches the predicate. /// /// The parser will return the longest slice that matches the given predicate *(a function that -/// takes the input and returns a bool)* +/// takes the input and returns a bool)*. /// /// # Streaming Specific -/// *Streaming version* will return a `Err::Incomplete(Needed::Size(1))` if the pattern reaches the end of the input +/// *Streaming version* will return a `Err::Incomplete(Needed::new(1))` if the pattern reaches the end of the input. /// # Example /// ```rust /// # #[macro_use] extern crate nom; @@ -175,10 +187,12 @@ where /// /// assert_eq!(alpha(b"latin123"), Ok((&b"123"[..], &b"latin"[..]))); /// assert_eq!(alpha(b"12345"), Ok((&b"12345"[..], &b""[..]))); -/// assert_eq!(alpha(b"latin"), Err(Err::Incomplete(Needed::Size(1)))); -/// assert_eq!(alpha(b""), Err(Err::Incomplete(Needed::Size(1)))); +/// assert_eq!(alpha(b"latin"), Err(Err::Incomplete(Needed::new(1)))); +/// assert_eq!(alpha(b""), Err(Err::Incomplete(Needed::new(1)))); /// ``` -pub fn take_while<F, Input, Error: ParseError<Input>>(cond: F) -> impl Fn(Input) -> IResult<Input, Input, Error> +pub fn take_while<F, Input, Error: ParseError<Input>>( + cond: F, +) -> impl Fn(Input) -> IResult<Input, Input, Error> where Input: InputTakeAtPosition, F: Fn(<Input as InputTakeAtPosition>::Item) -> bool, @@ -186,20 +200,20 @@ where move |i: Input| i.split_at_position(|c| !cond(c)) } -/// Returns the longest (atleast 1) input slice that matches the predicate +/// Returns the longest (at least 1) input slice that matches the predicate. /// /// The parser will return the longest slice that matches the given predicate *(a function that -/// takes the input and returns a bool)* +/// takes the input and returns a bool)*. /// -/// It will return an `Err(Err::Error((_, ErrorKind::TakeWhile1)))` if the pattern wasn't met +/// It will return an `Err(Err::Error((_, ErrorKind::TakeWhile1)))` if the pattern wasn't met. /// /// # Streaming Specific -/// *Streaming version* will return a `Err::Incomplete(Needed::Size(1))` or if the pattern reaches the end of the input. +/// *Streaming version* will return a `Err::Incomplete(Needed::new(1))` or if the pattern reaches the end of the input. /// /// # Example /// ```rust /// # #[macro_use] extern crate nom; -/// # use nom::{Err, error::ErrorKind, Needed, IResult}; +/// # use nom::{Err, error::{Error, ErrorKind}, Needed, IResult}; /// use nom::bytes::streaming::take_while1; /// use nom::character::is_alphabetic; /// @@ -208,10 +222,12 @@ where /// } /// /// assert_eq!(alpha(b"latin123"), Ok((&b"123"[..], &b"latin"[..]))); -/// assert_eq!(alpha(b"latin"), Err(Err::Incomplete(Needed::Size(1)))); -/// assert_eq!(alpha(b"12345"), Err(Err::Error((&b"12345"[..], ErrorKind::TakeWhile1)))); +/// assert_eq!(alpha(b"latin"), Err(Err::Incomplete(Needed::new(1)))); +/// assert_eq!(alpha(b"12345"), Err(Err::Error(Error::new(&b"12345"[..], ErrorKind::TakeWhile1)))); /// ``` -pub fn take_while1<F, Input, Error: ParseError<Input>>(cond: F) -> impl Fn(Input) -> IResult<Input, Input, Error> +pub fn take_while1<F, Input, Error: ParseError<Input>>( + cond: F, +) -> impl Fn(Input) -> IResult<Input, Input, Error> where Input: InputTakeAtPosition, F: Fn(<Input as InputTakeAtPosition>::Item) -> bool, @@ -222,19 +238,19 @@ where } } -/// Returns the longest (m <= len <= n) input slice that matches the predicate +/// Returns the longest (m <= len <= n) input slice that matches the predicate. /// /// The parser will return the longest slice that matches the given predicate *(a function that -/// takes the input and returns a bool)* +/// takes the input and returns a bool)*. /// -/// It will return an `Err::Error((_, ErrorKind::TakeWhileMN))` if the pattern wasn't met +/// It will return an `Err::Error((_, ErrorKind::TakeWhileMN))` if the pattern wasn't met. /// # Streaming Specific -/// *Streaming version* will return a `Err::Incomplete(Needed::Size(1))` if the pattern reaches the end of the input or is too short. +/// *Streaming version* will return a `Err::Incomplete(Needed::new(1))` if the pattern reaches the end of the input or is too short. /// /// # Example /// ```rust /// # #[macro_use] extern crate nom; -/// # use nom::{Err, error::ErrorKind, Needed, IResult}; +/// # use nom::{Err, error::{Error, ErrorKind}, Needed, IResult}; /// use nom::bytes::streaming::take_while_m_n; /// use nom::character::is_alphabetic; /// @@ -244,13 +260,17 @@ where /// /// assert_eq!(short_alpha(b"latin123"), Ok((&b"123"[..], &b"latin"[..]))); /// assert_eq!(short_alpha(b"lengthy"), Ok((&b"y"[..], &b"length"[..]))); -/// assert_eq!(short_alpha(b"latin"), Err(Err::Incomplete(Needed::Size(1)))); -/// assert_eq!(short_alpha(b"ed"), Err(Err::Incomplete(Needed::Size(1)))); -/// assert_eq!(short_alpha(b"12345"), Err(Err::Error((&b"12345"[..], ErrorKind::TakeWhileMN)))); +/// assert_eq!(short_alpha(b"latin"), Err(Err::Incomplete(Needed::new(1)))); +/// assert_eq!(short_alpha(b"ed"), Err(Err::Incomplete(Needed::new(1)))); +/// assert_eq!(short_alpha(b"12345"), Err(Err::Error(Error::new(&b"12345"[..], ErrorKind::TakeWhileMN)))); /// ``` -pub fn take_while_m_n<F, Input, Error: ParseError<Input>>(m: usize, n: usize, cond: F) -> impl Fn(Input) -> IResult<Input, Input, Error> +pub fn take_while_m_n<F, Input, Error: ParseError<Input>>( + m: usize, + n: usize, + cond: F, +) -> impl Fn(Input) -> IResult<Input, Input, Error> where - Input: InputTake + InputIter + InputLength + Slice<RangeFrom<usize>>, + Input: InputTake + InputIter + InputLength, F: Fn(<Input as InputIter>::Item) -> bool, { move |i: Input| { @@ -260,17 +280,23 @@ where Some(idx) => { if idx >= m { if idx <= n { - let res: IResult<_, _, Error> = if let Some(index) = input.slice_index(idx) { + let res: IResult<_, _, Error> = if let Ok(index) = input.slice_index(idx) { Ok(input.take_split(index)) } else { - Err(Err::Error(Error::from_error_kind(input, ErrorKind::TakeWhileMN))) + Err(Err::Error(Error::from_error_kind( + input, + ErrorKind::TakeWhileMN, + ))) }; res } else { - let res: IResult<_, _, Error> = if let Some(index) = input.slice_index(n) { + let res: IResult<_, _, Error> = if let Ok(index) = input.slice_index(n) { Ok(input.take_split(index)) } else { - Err(Err::Error(Error::from_error_kind(input, ErrorKind::TakeWhileMN))) + Err(Err::Error(Error::from_error_kind( + input, + ErrorKind::TakeWhileMN, + ))) }; res } @@ -283,26 +309,29 @@ where let len = input.input_len(); if len >= n { match input.slice_index(n) { - Some(index) => Ok(input.take_split(index)), - None => Err(Err::Error(Error::from_error_kind(input, ErrorKind::TakeWhileMN))) + Ok(index) => Ok(input.take_split(index)), + Err(_needed) => Err(Err::Error(Error::from_error_kind( + input, + ErrorKind::TakeWhileMN, + ))), } } else { let needed = if m > len { m - len } else { 1 }; - Err(Err::Incomplete(Needed::Size(needed))) + Err(Err::Incomplete(Needed::new(needed))) } } } } } -/// Returns the longest input slice (if any) till a predicate is met +/// Returns the longest input slice (if any) till a predicate is met. /// /// The parser will return the longest slice till the given predicate *(a function that -/// takes the input and returns a bool)* +/// takes the input and returns a bool)*. /// /// # Streaming Specific -/// *Streaming version* will return a `Err::Incomplete(Needed::Size(1))` if the match reaches the -/// end of input or if there was not match +/// *Streaming version* will return a `Err::Incomplete(Needed::new(1))` if the match reaches the +/// end of input or if there was not match. /// /// # Example /// ```rust @@ -316,10 +345,12 @@ where /// /// assert_eq!(till_colon("latin:123"), Ok((":123", "latin"))); /// assert_eq!(till_colon(":empty matched"), Ok((":empty matched", ""))); //allowed -/// assert_eq!(till_colon("12345"), Err(Err::Incomplete(Needed::Size(1)))); -/// assert_eq!(till_colon(""), Err(Err::Incomplete(Needed::Size(1)))); +/// assert_eq!(till_colon("12345"), Err(Err::Incomplete(Needed::new(1)))); +/// assert_eq!(till_colon(""), Err(Err::Incomplete(Needed::new(1)))); /// ``` -pub fn take_till<F, Input, Error: ParseError<Input>>(cond: F) -> impl Fn(Input) -> IResult<Input, Input, Error> +pub fn take_till<F, Input, Error: ParseError<Input>>( + cond: F, +) -> impl Fn(Input) -> IResult<Input, Input, Error> where Input: InputTakeAtPosition, F: Fn(<Input as InputTakeAtPosition>::Item) -> bool, @@ -327,18 +358,18 @@ where move |i: Input| i.split_at_position(|c| cond(c)) } -/// Returns the longest (atleast 1) input slice till a predicate is met +/// Returns the longest (at least 1) input slice till a predicate is met. /// /// The parser will return the longest slice till the given predicate *(a function that -/// takes the input and returns a bool)* +/// takes the input and returns a bool)*. /// /// # Streaming Specific -/// *Streaming version* will return a `Err::Incomplete(Needed::Size(1))` if the match reaches the -/// end of input or if there was not match +/// *Streaming version* will return a `Err::Incomplete(Needed::new(1))` if the match reaches the +/// end of input or if there was not match. /// # Example /// ```rust /// # #[macro_use] extern crate nom; -/// # use nom::{Err, error::ErrorKind, Needed, IResult}; +/// # use nom::{Err, error::{Error, ErrorKind}, Needed, IResult}; /// use nom::bytes::streaming::take_till1; /// /// fn till_colon(s: &str) -> IResult<&str, &str> { @@ -346,11 +377,13 @@ where /// } /// /// assert_eq!(till_colon("latin:123"), Ok((":123", "latin"))); -/// assert_eq!(till_colon(":empty matched"), Err(Err::Error((":empty matched", ErrorKind::TakeTill1)))); -/// assert_eq!(till_colon("12345"), Err(Err::Incomplete(Needed::Size(1)))); -/// assert_eq!(till_colon(""), Err(Err::Incomplete(Needed::Size(1)))); +/// assert_eq!(till_colon(":empty matched"), Err(Err::Error(Error::new(":empty matched", ErrorKind::TakeTill1)))); +/// assert_eq!(till_colon("12345"), Err(Err::Incomplete(Needed::new(1)))); +/// assert_eq!(till_colon(""), Err(Err::Incomplete(Needed::new(1)))); /// ``` -pub fn take_till1<F, Input, Error: ParseError<Input>>(cond: F) -> impl Fn(Input) -> IResult<Input, Input, Error> +pub fn take_till1<F, Input, Error: ParseError<Input>>( + cond: F, +) -> impl Fn(Input) -> IResult<Input, Input, Error> where Input: InputTakeAtPosition, F: Fn(<Input as InputTakeAtPosition>::Item) -> bool, @@ -361,11 +394,16 @@ where } } -/// Returns an input slice containing the first N input elements (Input[..N]) +/// Returns an input slice containing the first N input elements (Input[..N]). /// /// # Streaming Specific -/// *Streaming version* will return a `Err::Incomplete(Needed::Size(N))` where N is the -/// argument if the input is less than the length provided +/// *Streaming version* if the input has less than N elements, `take` will +/// return a `Err::Incomplete(Needed::new(M))` where M is the number of +/// additional bytes the parser would need to succeed. +/// It is well defined for `&[u8]` as the number of elements is the byte size, +/// but for types like `&str`, we cannot know how many bytes correspond for +/// the next few chars, so the result will be `Err::Incomplete(Needed::Unknown)` +/// /// # Example /// ```rust /// # #[macro_use] extern crate nom; @@ -378,28 +416,29 @@ where /// /// assert_eq!(take6("1234567"), Ok(("7", "123456"))); /// assert_eq!(take6("things"), Ok(("", "things"))); -/// assert_eq!(take6("short"), Err(Err::Incomplete(Needed::Size(6)))); //N doesn't change -/// assert_eq!(take6(""), Err(Err::Incomplete(Needed::Size(6)))); +/// assert_eq!(take6("short"), Err(Err::Incomplete(Needed::Unknown))); /// ``` -pub fn take<C, Input, Error: ParseError<Input>>(count: C) -> impl Fn(Input) -> IResult<Input, Input, Error> +pub fn take<C, Input, Error: ParseError<Input>>( + count: C, +) -> impl Fn(Input) -> IResult<Input, Input, Error> where - Input: InputIter + InputTake, + Input: InputIter + InputTake + InputLength, C: ToUsize, { let c = count.to_usize(); move |i: Input| match i.slice_index(c) { - None => Err(Err::Incomplete(Needed::Size(c))), - Some(index) => Ok(i.take_split(index)), + Err(i) => Err(Err::Incomplete(i)), + Ok(index) => Ok(i.take_split(index)), } } -/// Returns the longest input slice till it matches the pattern. +/// Returns the input slice up to the first occurrence of the pattern. /// -/// It doesn't consume the pattern +/// It doesn't consume the pattern. /// /// # Streaming Specific -/// *Streaming version* will return a `Err::Incomplete(Needed::Size(N))` if the input doesn't -/// contain the pattern or if the input is smaller than the pattern +/// *Streaming version* will return a `Err::Incomplete(Needed::new(N))` if the input doesn't +/// contain the pattern or if the input is smaller than the pattern. /// # Example /// ```rust /// # #[macro_use] extern crate nom; @@ -411,20 +450,22 @@ where /// } /// /// assert_eq!(until_eof("hello, worldeof"), Ok(("eof", "hello, world"))); -/// assert_eq!(until_eof("hello, world"), Err(Err::Incomplete(Needed::Size(3)))); -/// assert_eq!(until_eof(""), Err(Err::Incomplete(Needed::Size(3)))); +/// assert_eq!(until_eof("hello, world"), Err(Err::Incomplete(Needed::Unknown))); +/// assert_eq!(until_eof("hello, worldeo"), Err(Err::Incomplete(Needed::Unknown))); +/// assert_eq!(until_eof("1eof2eof"), Ok(("eof2eof", "1"))); /// ``` -pub fn take_until<T, Input, Error: ParseError<Input>>(tag: T) -> impl Fn(Input) -> IResult<Input, Input, Error> +pub fn take_until<T, Input, Error: ParseError<Input>>( + tag: T, +) -> impl Fn(Input) -> IResult<Input, Input, Error> where - Input: InputTake + FindSubstring<T>, - T: InputLength + Clone, + Input: InputTake + InputLength + FindSubstring<T>, + T: Clone, { move |i: Input| { - let len = tag.input_len(); let t = tag.clone(); let res: IResult<_, _, Error> = match i.find_substring(t) { - None => Err(Err::Incomplete(Needed::Size(len))), + None => Err(Err::Incomplete(Needed::Unknown)), Some(index) => Ok(i.take_split(index)), }; res @@ -433,10 +474,9 @@ where /// Matches a byte string with escaped characters. /// -/// * The first argument matches the normal characters (it must not accept the control character), -/// * the second argument is the control character (like `\` in most languages), -/// * the third argument matches the escaped characters -/// +/// * The first argument matches the normal characters (it must not accept the control character) +/// * The second argument is the control character (like `\` in most languages) +/// * The third argument matches the escaped characters /// # Example /// ``` /// # #[macro_use] extern crate nom; @@ -453,12 +493,22 @@ where /// assert_eq!(esc("12\\\"34;"), Ok((";", "12\\\"34"))); /// ``` /// -pub fn escaped<Input, Error, F, G, O1, O2>(normal: F, control_char: char, escapable: G) -> impl Fn(Input) -> IResult<Input, Input, Error> +pub fn escaped<Input, Error, F, G, O1, O2>( + mut normal: F, + control_char: char, + mut escapable: G, +) -> impl FnMut(Input) -> IResult<Input, Input, Error> where - Input: Clone + crate::traits::Offset + InputLength + InputTake + InputTakeAtPosition + Slice<RangeFrom<usize>> + InputIter, + Input: Clone + + crate::traits::Offset + + InputLength + + InputTake + + InputTakeAtPosition + + Slice<RangeFrom<usize>> + + InputIter, <Input as InputIter>::Item: crate::traits::AsChar, - F: Fn(Input) -> IResult<Input, O1, Error>, - G: Fn(Input) -> IResult<Input, O2, Error>, + F: Parser<Input, O1, Error>, + G: Parser<Input, O2, Error>, Error: ParseError<Input>, { use crate::traits::AsChar; @@ -467,7 +517,7 @@ where let mut i = input.clone(); while i.input_len() > 0 { - match normal(i.clone()) { + match normal.parse(i.clone()) { Ok((i2, _)) => { if i2.input_len() == 0 { return Err(Err::Incomplete(Needed::Unknown)); @@ -480,9 +530,9 @@ where if i.iter_elements().next().unwrap().as_char() == control_char { let next = control_char.len_utf8(); if next >= i.input_len() { - return Err(Err::Incomplete(Needed::Size(1))); + return Err(Err::Incomplete(Needed::new(1))); } else { - match escapable(i.slice(next..)) { + match escapable.parse(i.slice(next..)) { Ok((i2, _)) => { if i2.input_len() == 0 { return Err(Err::Incomplete(Needed::Unknown)); @@ -509,9 +559,20 @@ where } #[doc(hidden)] -pub fn escapedc<Input, Error, F, G, O1, O2>(i: Input, normal: F, control_char: char, escapable: G) -> IResult<Input, Input, Error> +pub fn escapedc<Input, Error, F, G, O1, O2>( + i: Input, + normal: F, + control_char: char, + escapable: G, +) -> IResult<Input, Input, Error> where - Input: Clone + crate::traits::Offset + InputLength + InputTake + InputTakeAtPosition + Slice<RangeFrom<usize>> + InputIter, + Input: Clone + + crate::traits::Offset + + InputLength + + InputTake + + InputTakeAtPosition + + Slice<RangeFrom<usize>> + + InputIter, <Input as InputIter>::Item: crate::traits::AsChar, F: Fn(Input) -> IResult<Input, O1, Error>, G: Fn(Input) -> IResult<Input, O2, Error>, @@ -522,9 +583,9 @@ where /// Matches a byte string with escaped characters. /// -/// * The first argument matches the normal characters (it must not match the control character), -/// * the second argument is the control character (like `\` in most languages), -/// * the third argument matches the escaped characters and transforms them. +/// * The first argument matches the normal characters (it must not match the control character) +/// * The second argument is the control character (like `\` in most languages) +/// * The third argument matches the escaped characters and transforms them /// /// As an example, the chain `abc\tdef` could be `abc def` (it also consumes the control character) /// @@ -532,40 +593,46 @@ where /// # #[macro_use] extern crate nom; /// # use nom::{Err, error::ErrorKind, Needed, IResult}; /// # use std::str::from_utf8; -/// use nom::bytes::streaming::escaped_transform; +/// use nom::bytes::streaming::{escaped_transform, tag}; /// use nom::character::streaming::alpha1; +/// use nom::branch::alt; +/// use nom::combinator::value; /// /// fn parser(input: &str) -> IResult<&str, String> { /// escaped_transform( /// alpha1, /// '\\', -/// |i:&str| alt!(i, -/// tag!("\\") => { |_| "\\" } -/// | tag!("\"") => { |_| "\"" } -/// | tag!("n") => { |_| "\n" } -/// ) +/// alt(( +/// value("\\", tag("\\")), +/// value("\"", tag("\"")), +/// value("n", tag("\n")), +/// )) /// )(input) /// } /// /// assert_eq!(parser("ab\\\"cd\""), Ok(("\"", String::from("ab\"cd")))); /// ``` #[cfg(feature = "alloc")] +#[cfg_attr(feature = "docsrs", doc(cfg(feature = "alloc")))] pub fn escaped_transform<Input, Error, F, G, O1, O2, ExtendItem, Output>( - normal: F, + mut normal: F, control_char: char, - transform: G, -) -> impl Fn(Input) -> IResult<Input, Output, Error> + mut transform: G, +) -> impl FnMut(Input) -> IResult<Input, Output, Error> where - Input: Clone + crate::traits::Offset + InputLength + InputTake + InputTakeAtPosition + Slice<RangeFrom<usize>> + InputIter, + Input: Clone + + crate::traits::Offset + + InputLength + + InputTake + + InputTakeAtPosition + + Slice<RangeFrom<usize>> + + InputIter, Input: crate::traits::ExtendInto<Item = ExtendItem, Extender = Output>, O1: crate::traits::ExtendInto<Item = ExtendItem, Extender = Output>, O2: crate::traits::ExtendInto<Item = ExtendItem, Extender = Output>, - Output: core::iter::Extend<<Input as crate::traits::ExtendInto>::Item>, - Output: core::iter::Extend<<O1 as crate::traits::ExtendInto>::Item>, - Output: core::iter::Extend<<O2 as crate::traits::ExtendInto>::Item>, <Input as InputIter>::Item: crate::traits::AsChar, - F: Fn(Input) -> IResult<Input, O1, Error>, - G: Fn(Input) -> IResult<Input, O2, Error>, + F: Parser<Input, O1, Error>, + G: Parser<Input, O2, Error>, Error: ParseError<Input>, { use crate::traits::AsChar; @@ -578,7 +645,7 @@ where while index < i.input_len() { let remainder = i.slice(index..); - match normal(remainder.clone()) { + match normal.parse(remainder.clone()) { Ok((i2, o)) => { o.extend_into(&mut res); if i2.input_len() == 0 { @@ -596,7 +663,7 @@ where if next >= input_len { return Err(Err::Incomplete(Needed::Unknown)); } else { - match transform(i.slice(next..)) { + match transform.parse(i.slice(next..)) { Ok((i2, o)) => { o.extend_into(&mut res); if i2.input_len() == 0 { @@ -621,6 +688,7 @@ where #[doc(hidden)] #[cfg(feature = "alloc")] +#[cfg_attr(feature = "docsrs", doc(cfg(feature = "alloc")))] pub fn escaped_transformc<Input, Error, F, G, O1, O2, ExtendItem, Output>( i: Input, normal: F, @@ -628,18 +696,20 @@ pub fn escaped_transformc<Input, Error, F, G, O1, O2, ExtendItem, Output>( transform: G, ) -> IResult<Input, Output, Error> where - Input: Clone + crate::traits::Offset + InputLength + InputTake + InputTakeAtPosition + Slice<RangeFrom<usize>> + InputIter, + Input: Clone + + crate::traits::Offset + + InputLength + + InputTake + + InputTakeAtPosition + + Slice<RangeFrom<usize>> + + InputIter, Input: crate::traits::ExtendInto<Item = ExtendItem, Extender = Output>, O1: crate::traits::ExtendInto<Item = ExtendItem, Extender = Output>, O2: crate::traits::ExtendInto<Item = ExtendItem, Extender = Output>, - Output: core::iter::Extend<<Input as crate::traits::ExtendInto>::Item>, - Output: core::iter::Extend<<O1 as crate::traits::ExtendInto>::Item>, - Output: core::iter::Extend<<O2 as crate::traits::ExtendInto>::Item>, <Input as InputIter>::Item: crate::traits::AsChar, F: Fn(Input) -> IResult<Input, O1, Error>, G: Fn(Input) -> IResult<Input, O2, Error>, Error: ParseError<Input>, { escaped_transform(normal, control_char, transform)(i) - } diff --git a/src/character/complete.rs b/src/character/complete.rs index beddc3f..cbcff46 100644 --- a/src/character/complete.rs +++ b/src/character/complete.rs @@ -2,27 +2,28 @@ //! //! Functions recognizing specific characters. -use crate::internal::{Err, IResult}; +use crate::error::ErrorKind; use crate::error::ParseError; +use crate::internal::{Err, IResult}; use crate::lib::std::ops::{Range, RangeFrom, RangeTo}; use crate::traits::{AsChar, FindToken, InputIter, InputLength, InputTakeAtPosition, Slice}; use crate::traits::{Compare, CompareResult}; -use crate::error::ErrorKind; /// Recognizes one character. /// -/// *complete version*: Will return an error if there's not enough input data. -/// +/// *Complete version*: Will return an error if there's not enough input data. /// # Example /// /// ``` -/// # use nom::{Err, error::ErrorKind}; +/// # use nom::{Err, error::{ErrorKind, Error}, IResult}; /// # use nom::character::complete::char; -/// # fn main() { -/// assert_eq!(char::<_, (&str, ErrorKind)>('a')("abc"), Ok(("bc", 'a'))); -/// assert_eq!(char::<_, (&str, ErrorKind)>('a')("bc"), Err(Err::Error(("bc", ErrorKind::Char)))); -/// assert_eq!(char::<_, (&str, ErrorKind)>('a')(""), Err(Err::Error(("", ErrorKind::Char)))); -/// # } +/// fn parser(i: &str) -> IResult<&str, char> { +/// char('a')(i) +/// } +/// assert_eq!(parser("abc"), Ok(("bc", 'a'))); +/// assert_eq!(parser(" abc"), Err(Err::Error(Error::new(" abc", ErrorKind::Char)))); +/// assert_eq!(parser("bc"), Err(Err::Error(Error::new("bc", ErrorKind::Char)))); +/// assert_eq!(parser(""), Err(Err::Error(Error::new("", ErrorKind::Char)))); /// ``` pub fn char<I, Error: ParseError<I>>(c: char) -> impl Fn(I) -> IResult<I, char, Error> where @@ -38,20 +39,48 @@ where } } -/// Recognizes one of the provided characters. +/// Recognizes one character and checks that it satisfies a predicate /// -/// *complete version*: Will return an error if there's not enough input data. +/// *Complete version*: Will return an error if there's not enough input data. +/// # Example /// +/// ``` +/// # use nom::{Err, error::{ErrorKind, Error}, Needed, IResult}; +/// # use nom::character::complete::satisfy; +/// fn parser(i: &str) -> IResult<&str, char> { +/// satisfy(|c| c == 'a' || c == 'b')(i) +/// } +/// assert_eq!(parser("abc"), Ok(("bc", 'a'))); +/// assert_eq!(parser("cd"), Err(Err::Error(Error::new("cd", ErrorKind::Satisfy)))); +/// assert_eq!(parser(""), Err(Err::Error(Error::new("", ErrorKind::Satisfy)))); +/// ``` +pub fn satisfy<F, I, Error: ParseError<I>>(cond: F) -> impl Fn(I) -> IResult<I, char, Error> +where + I: Slice<RangeFrom<usize>> + InputIter, + <I as InputIter>::Item: AsChar, + F: Fn(char) -> bool, +{ + move |i: I| match (i).iter_elements().next().map(|t| { + let c = t.as_char(); + let b = cond(c); + (c, b) + }) { + Some((c, true)) => Ok((i.slice(c.len()..), c)), + _ => Err(Err::Error(Error::from_error_kind(i, ErrorKind::Satisfy))), + } +} + +/// Recognizes one of the provided characters. +/// +/// *Complete version*: Will return an error if there's not enough input data. /// # Example /// /// ``` /// # use nom::{Err, error::ErrorKind}; /// # use nom::character::complete::one_of; -/// # fn main() { /// assert_eq!(one_of::<_, _, (&str, ErrorKind)>("abc")("b"), Ok(("", 'b'))); /// assert_eq!(one_of::<_, _, (&str, ErrorKind)>("a")("bc"), Err(Err::Error(("bc", ErrorKind::OneOf)))); /// assert_eq!(one_of::<_, _, (&str, ErrorKind)>("a")(""), Err(Err::Error(("", ErrorKind::OneOf)))); -/// # } /// ``` pub fn one_of<I, T, Error: ParseError<I>>(list: T) -> impl Fn(I) -> IResult<I, char, Error> where @@ -67,18 +96,15 @@ where /// Recognizes a character that is not in the provided characters. /// -/// *complete version*: Will return an error if there's not enough input data. -/// +/// *Complete version*: Will return an error if there's not enough input data. /// # Example /// /// ``` /// # use nom::{Err, error::ErrorKind}; /// # use nom::character::complete::none_of; -/// # fn main() { /// assert_eq!(none_of::<_, _, (&str, ErrorKind)>("abc")("z"), Ok(("", 'z'))); /// assert_eq!(none_of::<_, _, (&str, ErrorKind)>("ab")("a"), Err(Err::Error(("a", ErrorKind::NoneOf)))); /// assert_eq!(none_of::<_, _, (&str, ErrorKind)>("a")(""), Err(Err::Error(("", ErrorKind::NoneOf)))); -/// # } /// ``` pub fn none_of<I, T, Error: ParseError<I>>(list: T) -> impl Fn(I) -> IResult<I, char, Error> where @@ -94,26 +120,23 @@ where /// Recognizes the string "\r\n". /// -/// *complete version*: Will return an error if there's not enough input data. -/// +/// *Complete version*: Will return an error if there's not enough input data. /// # Example /// /// ``` -/// # use nom::{Err, error::ErrorKind, IResult}; +/// # use nom::{Err, error::{Error, ErrorKind}, IResult}; /// # use nom::character::complete::crlf; -/// # fn main() { /// fn parser(input: &str) -> IResult<&str, &str> { /// crlf(input) /// } /// /// assert_eq!(parser("\r\nc"), Ok(("c", "\r\n"))); -/// assert_eq!(parser("ab\r\nc"), Err(Err::Error(("ab\r\nc", ErrorKind::CrLf)))); -/// assert_eq!(parser(""), Err(Err::Error(("", ErrorKind::CrLf)))); -/// # } +/// assert_eq!(parser("ab\r\nc"), Err(Err::Error(Error::new("ab\r\nc", ErrorKind::CrLf)))); +/// assert_eq!(parser(""), Err(Err::Error(Error::new("", ErrorKind::CrLf)))); /// ``` pub fn crlf<T, E: ParseError<T>>(input: T) -> IResult<T, T, E> where - T: Slice<Range<usize>> + Slice<RangeFrom<usize>> + Slice<RangeTo<usize>>, + T: Slice<Range<usize>> + Slice<RangeFrom<usize>>, T: InputIter, T: Compare<&'static str>, { @@ -128,24 +151,24 @@ where } //FIXME: there's still an incomplete -/// Recognizes a string of any char except '\r' or '\n'. -/// -/// *complete version*: Will return an error if there's not enough input data. +/// Recognizes a string of any char except '\r\n' or '\n'. /// +/// *Complete version*: Will return an error if there's not enough input data. /// # Example /// /// ``` -/// # use nom::{Err, error::ErrorKind, IResult, Needed}; +/// # use nom::{Err, error::{Error, ErrorKind}, IResult, Needed}; /// # use nom::character::complete::not_line_ending; -/// # fn main() { /// fn parser(input: &str) -> IResult<&str, &str> { /// not_line_ending(input) /// } /// /// assert_eq!(parser("ab\r\nc"), Ok(("\r\nc", "ab"))); +/// assert_eq!(parser("ab\nc"), Ok(("\nc", "ab"))); /// assert_eq!(parser("abc"), Ok(("", "abc"))); /// assert_eq!(parser(""), Ok(("", ""))); -/// # } +/// assert_eq!(parser("a\rb\nc"), Err(Err::Error(Error { input: "a\rb\nc", code: ErrorKind::Tag }))); +/// assert_eq!(parser("a\rbc"), Err(Err::Error(Error { input: "a\rbc", code: ErrorKind::Tag }))); /// ``` pub fn not_line_ending<T, E: ParseError<T>>(input: T) -> IResult<T, T, E> where @@ -159,9 +182,7 @@ where let c = item.as_char(); c == '\r' || c == '\n' }) { - None => { - Ok((input.slice(input.input_len()..), input)) - } + None => Ok((input.slice(input.input_len()..), input)), Some(index) => { let mut it = input.slice(index..).iter_elements(); let nth = it.next().unwrap().as_char(); @@ -185,22 +206,19 @@ where /// Recognizes an end of line (both '\n' and '\r\n'). /// -/// *complete version*: Will return an error if there's not enough input data. -/// +/// *Complete version*: Will return an error if there's not enough input data. /// # Example /// /// ``` -/// # use nom::{Err, error::ErrorKind, IResult, Needed}; +/// # use nom::{Err, error::{Error, ErrorKind}, IResult, Needed}; /// # use nom::character::complete::line_ending; -/// # fn main() { /// fn parser(input: &str) -> IResult<&str, &str> { /// line_ending(input) /// } /// /// assert_eq!(parser("\r\nc"), Ok(("c", "\r\n"))); -/// assert_eq!(parser("ab\r\nc"), Err(Err::Error(("ab\r\nc", ErrorKind::CrLf)))); -/// assert_eq!(parser(""), Err(Err::Error(("", ErrorKind::CrLf)))); -/// # } +/// assert_eq!(parser("ab\r\nc"), Err(Err::Error(Error::new("ab\r\nc", ErrorKind::CrLf)))); +/// assert_eq!(parser(""), Err(Err::Error(Error::new("", ErrorKind::CrLf)))); /// ``` pub fn line_ending<T, E: ParseError<T>>(input: T) -> IResult<T, T, E> where @@ -223,22 +241,19 @@ where /// Matches a newline character '\n'. /// -/// *complete version*: Will return an error if there's not enough input data. -/// +/// *Complete version*: Will return an error if there's not enough input data. /// # Example /// /// ``` -/// # use nom::{Err, error::ErrorKind, IResult, Needed}; +/// # use nom::{Err, error::{Error, ErrorKind}, IResult, Needed}; /// # use nom::character::complete::newline; -/// # fn main() { /// fn parser(input: &str) -> IResult<&str, char> { /// newline(input) /// } /// /// assert_eq!(parser("\nc"), Ok(("c", '\n'))); -/// assert_eq!(parser("\r\nc"), Err(Err::Error(("\r\nc", ErrorKind::Char)))); -/// assert_eq!(parser(""), Err(Err::Error(("", ErrorKind::Char)))); -/// # } +/// assert_eq!(parser("\r\nc"), Err(Err::Error(Error::new("\r\nc", ErrorKind::Char)))); +/// assert_eq!(parser(""), Err(Err::Error(Error::new("", ErrorKind::Char)))); /// ``` pub fn newline<I, Error: ParseError<I>>(input: I) -> IResult<I, char, Error> where @@ -250,22 +265,19 @@ where /// Matches a tab character '\t'. /// -/// *complete version*: Will return an error if there's not enough input data. -/// +/// *Complete version*: Will return an error if there's not enough input data. /// # Example /// /// ``` -/// # use nom::{Err, error::ErrorKind, IResult, Needed}; +/// # use nom::{Err, error::{Error, ErrorKind}, IResult, Needed}; /// # use nom::character::complete::tab; -/// # fn main() { /// fn parser(input: &str) -> IResult<&str, char> { /// tab(input) /// } /// /// assert_eq!(parser("\tc"), Ok(("c", '\t'))); -/// assert_eq!(parser("\r\nc"), Err(Err::Error(("\r\nc", ErrorKind::Char)))); -/// assert_eq!(parser(""), Err(Err::Error(("", ErrorKind::Char)))); -/// # } +/// assert_eq!(parser("\r\nc"), Err(Err::Error(Error::new("\r\nc", ErrorKind::Char)))); +/// assert_eq!(parser(""), Err(Err::Error(Error::new("", ErrorKind::Char)))); /// ``` pub fn tab<I, Error: ParseError<I>>(input: I) -> IResult<I, char, Error> where @@ -278,20 +290,17 @@ where /// Matches one byte as a character. Note that the input type will /// accept a `str`, but not a `&[u8]`, unlike many other nom parsers. /// -/// *complete version*: Will return an error if there's not enough input data. -/// +/// *Complete version*: Will return an error if there's not enough input data. /// # Example /// /// ``` -/// # use nom::{character::complete::anychar, Err, error::ErrorKind, IResult}; -/// # fn main() { +/// # use nom::{character::complete::anychar, Err, error::{Error, ErrorKind}, IResult}; /// fn parser(input: &str) -> IResult<&str, char> { /// anychar(input) /// } /// /// assert_eq!(parser("abc"), Ok(("bc",'a'))); -/// assert_eq!(parser(""), Err(Err::Error(("", ErrorKind::Eof)))); -/// # } +/// assert_eq!(parser(""), Err(Err::Error(Error::new("", ErrorKind::Eof)))); /// ``` pub fn anychar<T, E: ParseError<T>>(input: T) -> IResult<T, char, E> where @@ -310,15 +319,13 @@ where /// Recognizes zero or more lowercase and uppercase ASCII alphabetic characters: a-z, A-Z /// -/// *complete version*: Will return the whole input if no terminating token is found (a non +/// *Complete version*: Will return the whole input if no terminating token is found (a non /// alphabetic character). -/// /// # Example /// /// ``` /// # use nom::{Err, error::ErrorKind, IResult, Needed}; /// # use nom::character::complete::alpha0; -/// # fn main() { /// fn parser(input: &str) -> IResult<&str, &str> { /// alpha0(input) /// } @@ -326,7 +333,6 @@ where /// assert_eq!(parser("ab1c"), Ok(("1c", "ab"))); /// assert_eq!(parser("1c"), Ok(("1c", ""))); /// assert_eq!(parser(""), Ok(("", ""))); -/// # } /// ``` pub fn alpha0<T, E: ParseError<T>>(input: T) -> IResult<T, T, E> where @@ -338,23 +344,20 @@ where /// Recognizes one or more lowercase and uppercase ASCII alphabetic characters: a-z, A-Z /// -/// *complete version*: Will return an error if there's not enough input data, +/// *Complete version*: Will return an error if there's not enough input data, /// or the whole input if no terminating token is found (a non alphabetic character). -/// /// # Example /// /// ``` -/// # use nom::{Err, error::ErrorKind, IResult, Needed}; +/// # use nom::{Err, error::{Error, ErrorKind}, IResult, Needed}; /// # use nom::character::complete::alpha1; -/// # fn main() { /// fn parser(input: &str) -> IResult<&str, &str> { /// alpha1(input) /// } /// /// assert_eq!(parser("aB1c"), Ok(("1c", "aB"))); -/// assert_eq!(parser("1c"), Err(Err::Error(("1c", ErrorKind::Alpha)))); -/// assert_eq!(parser(""), Err(Err::Error(("", ErrorKind::Alpha)))); -/// # } +/// assert_eq!(parser("1c"), Err(Err::Error(Error::new("1c", ErrorKind::Alpha)))); +/// assert_eq!(parser(""), Err(Err::Error(Error::new("", ErrorKind::Alpha)))); /// ``` pub fn alpha1<T, E: ParseError<T>>(input: T) -> IResult<T, T, E> where @@ -366,15 +369,13 @@ where /// Recognizes zero or more ASCII numerical characters: 0-9 /// -/// *complete version*: Will return an error if there's not enough input data, +/// *Complete version*: Will return an error if there's not enough input data, /// or the whole input if no terminating token is found (a non digit character). -/// /// # Example /// /// ``` /// # use nom::{Err, error::ErrorKind, IResult, Needed}; /// # use nom::character::complete::digit0; -/// # fn main() { /// fn parser(input: &str) -> IResult<&str, &str> { /// digit0(input) /// } @@ -383,7 +384,6 @@ where /// assert_eq!(parser("21"), Ok(("", "21"))); /// assert_eq!(parser("a21c"), Ok(("a21c", ""))); /// assert_eq!(parser(""), Ok(("", ""))); -/// # } /// ``` pub fn digit0<T, E: ParseError<T>>(input: T) -> IResult<T, T, E> where @@ -395,23 +395,20 @@ where /// Recognizes one or more ASCII numerical characters: 0-9 /// -/// *complete version*: Will return an error if there's not enough input data, +/// *Complete version*: Will return an error if there's not enough input data, /// or the whole input if no terminating token is found (a non digit character). -/// /// # Example /// /// ``` -/// # use nom::{Err, error::ErrorKind, IResult, Needed}; +/// # use nom::{Err, error::{Error, ErrorKind}, IResult, Needed}; /// # use nom::character::complete::digit1; -/// # fn main() { /// fn parser(input: &str) -> IResult<&str, &str> { /// digit1(input) /// } /// /// assert_eq!(parser("21c"), Ok(("c", "21"))); -/// assert_eq!(parser("c1"), Err(Err::Error(("c1", ErrorKind::Digit)))); -/// assert_eq!(parser(""), Err(Err::Error(("", ErrorKind::Digit)))); -/// # } +/// assert_eq!(parser("c1"), Err(Err::Error(Error::new("c1", ErrorKind::Digit)))); +/// assert_eq!(parser(""), Err(Err::Error(Error::new("", ErrorKind::Digit)))); /// ``` pub fn digit1<T, E: ParseError<T>>(input: T) -> IResult<T, T, E> where @@ -423,14 +420,12 @@ where /// Recognizes zero or more ASCII hexadecimal numerical characters: 0-9, A-F, a-f /// -/// *complete version*: Will return the whole input if no terminating token is found (a non hexadecimal digit character). -/// +/// *Complete version*: Will return the whole input if no terminating token is found (a non hexadecimal digit character). /// # Example /// /// ``` /// # use nom::{Err, error::ErrorKind, IResult, Needed}; /// # use nom::character::complete::hex_digit0; -/// # fn main() { /// fn parser(input: &str) -> IResult<&str, &str> { /// hex_digit0(input) /// } @@ -438,7 +433,6 @@ where /// assert_eq!(parser("21cZ"), Ok(("Z", "21c"))); /// assert_eq!(parser("Z21c"), Ok(("Z21c", ""))); /// assert_eq!(parser(""), Ok(("", ""))); -/// # } /// ``` pub fn hex_digit0<T, E: ParseError<T>>(input: T) -> IResult<T, T, E> where @@ -449,23 +443,20 @@ where } /// Recognizes one or more ASCII hexadecimal numerical characters: 0-9, A-F, a-f /// -/// *complete version*: Will return an error if there's not enough input data, +/// *Complete version*: Will return an error if there's not enough input data, /// or the whole input if no terminating token is found (a non hexadecimal digit character). -/// /// # Example /// /// ``` -/// # use nom::{Err, error::ErrorKind, IResult, Needed}; +/// # use nom::{Err, error::{Error, ErrorKind}, IResult, Needed}; /// # use nom::character::complete::hex_digit1; -/// # fn main() { /// fn parser(input: &str) -> IResult<&str, &str> { /// hex_digit1(input) /// } /// /// assert_eq!(parser("21cZ"), Ok(("Z", "21c"))); -/// assert_eq!(parser("H2"), Err(Err::Error(("H2", ErrorKind::HexDigit)))); -/// assert_eq!(parser(""), Err(Err::Error(("", ErrorKind::HexDigit)))); -/// # } +/// assert_eq!(parser("H2"), Err(Err::Error(Error::new("H2", ErrorKind::HexDigit)))); +/// assert_eq!(parser(""), Err(Err::Error(Error::new("", ErrorKind::HexDigit)))); /// ``` pub fn hex_digit1<T, E: ParseError<T>>(input: T) -> IResult<T, T, E> where @@ -477,15 +468,13 @@ where /// Recognizes zero or more octal characters: 0-7 /// -/// *complete version*: Will return the whole input if no terminating token is found (a non octal +/// *Complete version*: Will return the whole input if no terminating token is found (a non octal /// digit character). -/// /// # Example /// /// ``` /// # use nom::{Err, error::ErrorKind, IResult, Needed}; /// # use nom::character::complete::oct_digit0; -/// # fn main() { /// fn parser(input: &str) -> IResult<&str, &str> { /// oct_digit0(input) /// } @@ -493,7 +482,6 @@ where /// assert_eq!(parser("21cZ"), Ok(("cZ", "21"))); /// assert_eq!(parser("Z21c"), Ok(("Z21c", ""))); /// assert_eq!(parser(""), Ok(("", ""))); -/// # } /// ``` pub fn oct_digit0<T, E: ParseError<T>>(input: T) -> IResult<T, T, E> where @@ -505,23 +493,20 @@ where /// Recognizes one or more octal characters: 0-7 /// -/// *complete version*: Will return an error if there's not enough input data, +/// *Complete version*: Will return an error if there's not enough input data, /// or the whole input if no terminating token is found (a non octal digit character). -/// /// # Example /// /// ``` -/// # use nom::{Err, error::ErrorKind, IResult, Needed}; +/// # use nom::{Err, error::{Error, ErrorKind}, IResult, Needed}; /// # use nom::character::complete::oct_digit1; -/// # fn main() { /// fn parser(input: &str) -> IResult<&str, &str> { /// oct_digit1(input) /// } /// /// assert_eq!(parser("21cZ"), Ok(("cZ", "21"))); -/// assert_eq!(parser("H2"), Err(Err::Error(("H2", ErrorKind::OctDigit)))); -/// assert_eq!(parser(""), Err(Err::Error(("", ErrorKind::OctDigit)))); -/// # } +/// assert_eq!(parser("H2"), Err(Err::Error(Error::new("H2", ErrorKind::OctDigit)))); +/// assert_eq!(parser(""), Err(Err::Error(Error::new("", ErrorKind::OctDigit)))); /// ``` pub fn oct_digit1<T, E: ParseError<T>>(input: T) -> IResult<T, T, E> where @@ -533,15 +518,13 @@ where /// Recognizes zero or more ASCII numerical and alphabetic characters: 0-9, a-z, A-Z /// -/// *complete version*: Will return the whole input if no terminating token is found (a non +/// *Complete version*: Will return the whole input if no terminating token is found (a non /// alphanumerical character). -/// /// # Example /// /// ``` /// # use nom::{Err, error::ErrorKind, IResult, Needed}; /// # use nom::character::complete::alphanumeric0; -/// # fn main() { /// fn parser(input: &str) -> IResult<&str, &str> { /// alphanumeric0(input) /// } @@ -549,7 +532,6 @@ where /// assert_eq!(parser("21cZ%1"), Ok(("%1", "21cZ"))); /// assert_eq!(parser("&Z21c"), Ok(("&Z21c", ""))); /// assert_eq!(parser(""), Ok(("", ""))); -/// # } /// ``` pub fn alphanumeric0<T, E: ParseError<T>>(input: T) -> IResult<T, T, E> where @@ -561,23 +543,20 @@ where /// Recognizes one or more ASCII numerical and alphabetic characters: 0-9, a-z, A-Z /// -/// *complete version*: Will return an error if there's not enough input data, +/// *Complete version*: Will return an error if there's not enough input data, /// or the whole input if no terminating token is found (a non alphanumerical character). -/// /// # Example /// /// ``` -/// # use nom::{Err, error::ErrorKind, IResult, Needed}; +/// # use nom::{Err, error::{Error, ErrorKind}, IResult, Needed}; /// # use nom::character::complete::alphanumeric1; -/// # fn main() { /// fn parser(input: &str) -> IResult<&str, &str> { /// alphanumeric1(input) /// } /// /// assert_eq!(parser("21cZ%1"), Ok(("%1", "21cZ"))); -/// assert_eq!(parser("&H2"), Err(Err::Error(("&H2", ErrorKind::AlphaNumeric)))); -/// assert_eq!(parser(""), Err(Err::Error(("", ErrorKind::AlphaNumeric)))); -/// # } +/// assert_eq!(parser("&H2"), Err(Err::Error(Error::new("&H2", ErrorKind::AlphaNumeric)))); +/// assert_eq!(parser(""), Err(Err::Error(Error::new("", ErrorKind::AlphaNumeric)))); /// ``` pub fn alphanumeric1<T, E: ParseError<T>>(input: T) -> IResult<T, T, E> where @@ -589,15 +568,13 @@ where /// Recognizes zero or more spaces and tabs. /// -/// *complete version*: Will return the whole input if no terminating token is found (a non space +/// *Complete version*: Will return the whole input if no terminating token is found (a non space /// character). -/// /// # Example /// /// ``` /// # use nom::{Err, error::ErrorKind, IResult, Needed}; /// # use nom::character::complete::space0; -/// # fn main() { /// fn parser(input: &str) -> IResult<&str, &str> { /// space0(input) /// } @@ -605,7 +582,6 @@ where /// assert_eq!(parser(" \t21c"), Ok(("21c", " \t"))); /// assert_eq!(parser("Z21c"), Ok(("Z21c", ""))); /// assert_eq!(parser(""), Ok(("", ""))); -/// # } /// ``` pub fn space0<T, E: ParseError<T>>(input: T) -> IResult<T, T, E> where @@ -613,30 +589,27 @@ where <T as InputTakeAtPosition>::Item: AsChar + Clone, { input.split_at_position_complete(|item| { - let c = item.clone().as_char(); + let c = item.as_char(); !(c == ' ' || c == '\t') }) } /// Recognizes one or more spaces and tabs. /// -/// *complete version*: Will return an error if there's not enough input data, +/// *Complete version*: Will return an error if there's not enough input data, /// or the whole input if no terminating token is found (a non space character). -/// /// # Example /// /// ``` -/// # use nom::{Err, error::ErrorKind, IResult, Needed}; +/// # use nom::{Err, error::{Error, ErrorKind}, IResult, Needed}; /// # use nom::character::complete::space1; -/// # fn main() { /// fn parser(input: &str) -> IResult<&str, &str> { /// space1(input) /// } /// /// assert_eq!(parser(" \t21c"), Ok(("21c", " \t"))); -/// assert_eq!(parser("H2"), Err(Err::Error(("H2", ErrorKind::Space)))); -/// assert_eq!(parser(""), Err(Err::Error(("", ErrorKind::Space)))); -/// # } +/// assert_eq!(parser("H2"), Err(Err::Error(Error::new("H2", ErrorKind::Space)))); +/// assert_eq!(parser(""), Err(Err::Error(Error::new("", ErrorKind::Space)))); /// ``` pub fn space1<T, E: ParseError<T>>(input: T) -> IResult<T, T, E> where @@ -645,7 +618,7 @@ where { input.split_at_position1_complete( |item| { - let c = item.clone().as_char(); + let c = item.as_char(); !(c == ' ' || c == '\t') }, ErrorKind::Space, @@ -654,15 +627,13 @@ where /// Recognizes zero or more spaces, tabs, carriage returns and line feeds. /// -/// *complete version*: will return the whole input if no terminating token is found (a non space +/// *Complete version*: will return the whole input if no terminating token is found (a non space /// character). -/// /// # Example /// /// ``` /// # use nom::{Err, error::ErrorKind, IResult, Needed}; /// # use nom::character::complete::multispace0; -/// # fn main() { /// fn parser(input: &str) -> IResult<&str, &str> { /// multispace0(input) /// } @@ -670,7 +641,6 @@ where /// assert_eq!(parser(" \t\n\r21c"), Ok(("21c", " \t\n\r"))); /// assert_eq!(parser("Z21c"), Ok(("Z21c", ""))); /// assert_eq!(parser(""), Ok(("", ""))); -/// # } /// ``` pub fn multispace0<T, E: ParseError<T>>(input: T) -> IResult<T, T, E> where @@ -678,30 +648,27 @@ where <T as InputTakeAtPosition>::Item: AsChar + Clone, { input.split_at_position_complete(|item| { - let c = item.clone().as_char(); + let c = item.as_char(); !(c == ' ' || c == '\t' || c == '\r' || c == '\n') }) } /// Recognizes one or more spaces, tabs, carriage returns and line feeds. /// -/// *complete version*: will return an error if there's not enough input data, +/// *Complete version*: will return an error if there's not enough input data, /// or the whole input if no terminating token is found (a non space character). -/// /// # Example /// /// ``` -/// # use nom::{Err, error::ErrorKind, IResult, Needed}; +/// # use nom::{Err, error::{Error, ErrorKind}, IResult, Needed}; /// # use nom::character::complete::multispace1; -/// # fn main() { /// fn parser(input: &str) -> IResult<&str, &str> { /// multispace1(input) /// } /// /// assert_eq!(parser(" \t\n\r21c"), Ok(("21c", " \t\n\r"))); -/// assert_eq!(parser("H2"), Err(Err::Error(("H2", ErrorKind::MultiSpace)))); -/// assert_eq!(parser(""), Err(Err::Error(("", ErrorKind::MultiSpace)))); -/// # } +/// assert_eq!(parser("H2"), Err(Err::Error(Error::new("H2", ErrorKind::MultiSpace)))); +/// assert_eq!(parser(""), Err(Err::Error(Error::new("", ErrorKind::MultiSpace)))); /// ``` pub fn multispace1<T, E: ParseError<T>>(input: T) -> IResult<T, T, E> where @@ -710,7 +677,7 @@ where { input.split_at_position1_complete( |item| { - let c = item.clone().as_char(); + let c = item.as_char(); !(c == ' ' || c == '\t' || c == '\r' || c == '\n') }, ErrorKind::MultiSpace, @@ -729,7 +696,6 @@ mod tests { }; ); - #[test] fn character() { let empty: &[u8] = b""; @@ -741,50 +707,35 @@ mod tests { let f: &[u8] = b" ;"; //assert_eq!(alpha1::<_, (_, ErrorKind)>(a), Err(Err::Incomplete(Needed::Size(1)))); assert_parse!(alpha1(a), Ok((empty, a))); - assert_eq!( - alpha1(b), - Err(Err::Error((b, ErrorKind::Alpha))) - ); + assert_eq!(alpha1(b), Err(Err::Error((b, ErrorKind::Alpha)))); assert_eq!(alpha1::<_, (_, ErrorKind)>(c), Ok((&c[1..], &b"a"[..]))); - assert_eq!(alpha1::<_, (_, ErrorKind)>(d), Ok(("Ć©12".as_bytes(), &b"az"[..]))); assert_eq!( - digit1(a), - Err(Err::Error((a, ErrorKind::Digit))) + alpha1::<_, (_, ErrorKind)>(d), + Ok(("Ć©12".as_bytes(), &b"az"[..])) ); + assert_eq!(digit1(a), Err(Err::Error((a, ErrorKind::Digit)))); assert_eq!(digit1::<_, (_, ErrorKind)>(b), Ok((empty, b))); - assert_eq!( - digit1(c), - Err(Err::Error((c, ErrorKind::Digit))) - ); - assert_eq!( - digit1(d), - Err(Err::Error((d, ErrorKind::Digit))) - ); + assert_eq!(digit1(c), Err(Err::Error((c, ErrorKind::Digit)))); + assert_eq!(digit1(d), Err(Err::Error((d, ErrorKind::Digit)))); assert_eq!(hex_digit1::<_, (_, ErrorKind)>(a), Ok((empty, a))); assert_eq!(hex_digit1::<_, (_, ErrorKind)>(b), Ok((empty, b))); assert_eq!(hex_digit1::<_, (_, ErrorKind)>(c), Ok((empty, c))); - assert_eq!(hex_digit1::<_, (_, ErrorKind)>(d), Ok(("zĆ©12".as_bytes(), &b"a"[..]))); assert_eq!( - hex_digit1(e), - Err(Err::Error((e, ErrorKind::HexDigit))) - ); - assert_eq!( - oct_digit1(a), - Err(Err::Error((a, ErrorKind::OctDigit))) + hex_digit1::<_, (_, ErrorKind)>(d), + Ok(("zĆ©12".as_bytes(), &b"a"[..])) ); + assert_eq!(hex_digit1(e), Err(Err::Error((e, ErrorKind::HexDigit)))); + assert_eq!(oct_digit1(a), Err(Err::Error((a, ErrorKind::OctDigit)))); assert_eq!(oct_digit1::<_, (_, ErrorKind)>(b), Ok((empty, b))); - assert_eq!( - oct_digit1(c), - Err(Err::Error((c, ErrorKind::OctDigit))) - ); - assert_eq!( - oct_digit1(d), - Err(Err::Error((d, ErrorKind::OctDigit))) - ); + assert_eq!(oct_digit1(c), Err(Err::Error((c, ErrorKind::OctDigit)))); + assert_eq!(oct_digit1(d), Err(Err::Error((d, ErrorKind::OctDigit)))); assert_eq!(alphanumeric1::<_, (_, ErrorKind)>(a), Ok((empty, a))); //assert_eq!(fix_error!(b,(), alphanumeric), Ok((empty, b))); assert_eq!(alphanumeric1::<_, (_, ErrorKind)>(c), Ok((empty, c))); - assert_eq!(alphanumeric1::<_, (_, ErrorKind)>(d), Ok(("Ć©12".as_bytes(), &b"az"[..]))); + assert_eq!( + alphanumeric1::<_, (_, ErrorKind)>(d), + Ok(("Ć©12".as_bytes(), &b"az"[..])) + ); assert_eq!(space1::<_, (_, ErrorKind)>(e), Ok((empty, e))); assert_eq!(space1::<_, (_, ErrorKind)>(f), Ok((&b";"[..], &b" "[..]))); } @@ -799,46 +750,22 @@ mod tests { let d = "azĆ©12"; let e = " "; assert_eq!(alpha1::<_, (_, ErrorKind)>(a), Ok((empty, a))); - assert_eq!( - alpha1(b), - Err(Err::Error((b, ErrorKind::Alpha))) - ); + assert_eq!(alpha1(b), Err(Err::Error((b, ErrorKind::Alpha)))); assert_eq!(alpha1::<_, (_, ErrorKind)>(c), Ok((&c[1..], &"a"[..]))); assert_eq!(alpha1::<_, (_, ErrorKind)>(d), Ok(("Ć©12", &"az"[..]))); - assert_eq!( - digit1(a), - Err(Err::Error((a, ErrorKind::Digit))) - ); + assert_eq!(digit1(a), Err(Err::Error((a, ErrorKind::Digit)))); assert_eq!(digit1::<_, (_, ErrorKind)>(b), Ok((empty, b))); - assert_eq!( - digit1(c), - Err(Err::Error((c, ErrorKind::Digit))) - ); - assert_eq!( - digit1(d), - Err(Err::Error((d, ErrorKind::Digit))) - ); + assert_eq!(digit1(c), Err(Err::Error((c, ErrorKind::Digit)))); + assert_eq!(digit1(d), Err(Err::Error((d, ErrorKind::Digit)))); assert_eq!(hex_digit1::<_, (_, ErrorKind)>(a), Ok((empty, a))); assert_eq!(hex_digit1::<_, (_, ErrorKind)>(b), Ok((empty, b))); assert_eq!(hex_digit1::<_, (_, ErrorKind)>(c), Ok((empty, c))); assert_eq!(hex_digit1::<_, (_, ErrorKind)>(d), Ok(("zĆ©12", &"a"[..]))); - assert_eq!( - hex_digit1(e), - Err(Err::Error((e, ErrorKind::HexDigit))) - ); - assert_eq!( - oct_digit1(a), - Err(Err::Error((a, ErrorKind::OctDigit))) - ); + assert_eq!(hex_digit1(e), Err(Err::Error((e, ErrorKind::HexDigit)))); + assert_eq!(oct_digit1(a), Err(Err::Error((a, ErrorKind::OctDigit)))); assert_eq!(oct_digit1::<_, (_, ErrorKind)>(b), Ok((empty, b))); - assert_eq!( - oct_digit1(c), - Err(Err::Error((c, ErrorKind::OctDigit))) - ); - assert_eq!( - oct_digit1(d), - Err(Err::Error((d, ErrorKind::OctDigit))) - ); + assert_eq!(oct_digit1(c), Err(Err::Error((c, ErrorKind::OctDigit)))); + assert_eq!(oct_digit1(d), Err(Err::Error((d, ErrorKind::OctDigit)))); assert_eq!(alphanumeric1::<_, (_, ErrorKind)>(a), Ok((empty, a))); //assert_eq!(fix_error!(b,(), alphanumeric), Ok((empty, b))); assert_eq!(alphanumeric1::<_, (_, ErrorKind)>(c), Ok((empty, c))); @@ -903,7 +830,10 @@ mod tests { #[test] fn is_not_line_ending_bytes() { let a: &[u8] = b"ab12cd\nefgh"; - assert_eq!(not_line_ending::<_, (_, ErrorKind)>(a), Ok((&b"\nefgh"[..], &b"ab12cd"[..]))); + assert_eq!( + not_line_ending::<_, (_, ErrorKind)>(a), + Ok((&b"\nefgh"[..], &b"ab12cd"[..])) + ); let b: &[u8] = b"ab12cd\nefgh\nijkl"; assert_eq!( @@ -918,12 +848,14 @@ mod tests { ); let d: &[u8] = b"ab12cd"; - assert_eq!(not_line_ending::<_, (_, ErrorKind)>(d), Ok((&[][..], &d[..]))); + assert_eq!( + not_line_ending::<_, (_, ErrorKind)>(d), + Ok((&[][..], &d[..])) + ); } #[test] fn is_not_line_ending_str() { - /* let a: &str = "ab12cd\nefgh"; assert_eq!(not_line_ending(a), Ok((&"\nefgh"[..], &"ab12cd"[..]))); @@ -942,10 +874,7 @@ mod tests { */ let f = "Ī²ĆØĘĆ“ÅĆØ\rĆĆĆĆ”ĘĘĆØÅ"; - assert_eq!( - not_line_ending(f), - Err(Err::Error((f, ErrorKind::Tag))) - ); + assert_eq!(not_line_ending(f), Err(Err::Error((f, ErrorKind::Tag)))); let g2: &str = "ab12cd"; assert_eq!(not_line_ending::<_, (_, ErrorKind)>(g2), Ok(("", g2))); @@ -1047,14 +976,20 @@ mod tests { #[test] fn cr_lf() { assert_parse!(crlf(&b"\r\na"[..]), Ok((&b"a"[..], &b"\r\n"[..]))); - assert_parse!(crlf(&b"\r"[..]), Err(Err::Error(error_position!(&b"\r"[..], ErrorKind::CrLf)))); + assert_parse!( + crlf(&b"\r"[..]), + Err(Err::Error(error_position!(&b"\r"[..], ErrorKind::CrLf))) + ); assert_parse!( crlf(&b"\ra"[..]), Err(Err::Error(error_position!(&b"\ra"[..], ErrorKind::CrLf))) ); assert_parse!(crlf("\r\na"), Ok(("a", "\r\n"))); - assert_parse!(crlf("\r"), Err(Err::Error(error_position!(&"\r"[..], ErrorKind::CrLf)))); + assert_parse!( + crlf("\r"), + Err(Err::Error(error_position!(&"\r"[..], ErrorKind::CrLf))) + ); assert_parse!( crlf("\ra"), Err(Err::Error(error_position!("\ra", ErrorKind::CrLf))) @@ -1065,7 +1000,10 @@ mod tests { fn end_of_line() { assert_parse!(line_ending(&b"\na"[..]), Ok((&b"a"[..], &b"\n"[..]))); assert_parse!(line_ending(&b"\r\na"[..]), Ok((&b"a"[..], &b"\r\n"[..]))); - assert_parse!(line_ending(&b"\r"[..]), Err(Err::Error(error_position!(&b"\r"[..], ErrorKind::CrLf)))); + assert_parse!( + line_ending(&b"\r"[..]), + Err(Err::Error(error_position!(&b"\r"[..], ErrorKind::CrLf))) + ); assert_parse!( line_ending(&b"\ra"[..]), Err(Err::Error(error_position!(&b"\ra"[..], ErrorKind::CrLf))) @@ -1073,7 +1011,10 @@ mod tests { assert_parse!(line_ending("\na"), Ok(("a", "\n"))); assert_parse!(line_ending("\r\na"), Ok(("a", "\r\n"))); - assert_parse!(line_ending("\r"), Err(Err::Error(error_position!(&"\r"[..], ErrorKind::CrLf)))); + assert_parse!( + line_ending("\r"), + Err(Err::Error(error_position!(&"\r"[..], ErrorKind::CrLf))) + ); assert_parse!( line_ending("\ra"), Err(Err::Error(error_position!("\ra", ErrorKind::CrLf))) diff --git a/src/character/macros.rs b/src/character/macros.rs index 097ee40..ba99645 100644 --- a/src/character/macros.rs +++ b/src/character/macros.rs @@ -1,6 +1,6 @@ /// Character level parsers -/// matches one of the provided characters +/// Matches one of the provided characters. /// /// # Example /// ``` @@ -18,7 +18,7 @@ macro_rules! one_of ( ($i:expr, $inp: expr) => ( $crate::character::streaming::one_of($inp)($i) ); ); -/// matches anything but the provided characters +/// Matches anything but the provided characters. /// /// # Example /// ``` @@ -37,7 +37,7 @@ macro_rules! none_of ( ($i:expr, $inp: expr) => ( $crate::character::streaming::none_of($inp)($i) ); ); -/// matches one character: `char!(char) => &[u8] -> IResult<&[u8], char> +/// Matches one character: `char!(char) => &[u8] -> IResult<&[u8], char>`. /// /// # Example /// ``` @@ -57,8 +57,8 @@ macro_rules! char ( #[cfg(test)] mod tests { - use crate::internal::Err; use crate::error::ErrorKind; + use crate::internal::Err; #[test] fn one_of() { diff --git a/src/character/mod.rs b/src/character/mod.rs index 3f0899a..4c70f6e 100644 --- a/src/character/mod.rs +++ b/src/character/mod.rs @@ -1,12 +1,12 @@ -//! character specific parsers and combinators +//! Character specific parsers and combinators //! -//! functions recognizing specific characters +//! Functions recognizing specific characters #[macro_use] mod macros; -pub mod streaming; pub mod complete; +pub mod streaming; /// Tests if byte is ASCII alphabetic: A-Z, a-z /// @@ -99,3 +99,18 @@ pub fn is_space(chr: u8) -> bool { chr == b' ' || chr == b'\t' } +/// Tests if byte is ASCII newline: \n +/// +/// # Example +/// +/// ``` +/// # use nom::character::is_newline; +/// assert_eq!(is_newline(b'\n'), true); +/// assert_eq!(is_newline(b'\r'), false); +/// assert_eq!(is_newline(b' '), false); +/// assert_eq!(is_newline(b'\t'), false); +/// ``` +#[inline] +pub fn is_newline(chr: u8) -> bool { + chr == b'\n' +} diff --git a/src/character/streaming.rs b/src/character/streaming.rs index f62dace..4c33ee9 100644 --- a/src/character/streaming.rs +++ b/src/character/streaming.rs @@ -1,9 +1,9 @@ -//! character specific parsers and combinators, streaming version +//! Character specific parsers and combinators, streaming version //! -//! functions recognizing specific characters +//! Functions recognizing specific characters -use crate::internal::{Err, IResult, Needed}; use crate::error::ParseError; +use crate::internal::{Err, IResult, Needed}; use crate::lib::std::ops::{Range, RangeFrom, RangeTo}; use crate::traits::{AsChar, FindToken, InputIter, InputLength, InputTakeAtPosition, Slice}; use crate::traits::{Compare, CompareResult}; @@ -12,50 +12,77 @@ use crate::error::ErrorKind; /// Recognizes one character. /// -/// *streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data. -/// +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data. /// # Example /// /// ``` -/// # use nom::{Err, error::ErrorKind, Needed}; +/// # use nom::{Err, error::{ErrorKind, Error}, Needed, IResult}; /// # use nom::character::streaming::char; -/// # fn main() { -/// assert_eq!(char::<_, (_, ErrorKind)>('a')(&b"abc"[..]), Ok((&b"bc"[..], 'a'))); -/// assert_eq!(char::<_, (_, ErrorKind)>('a')(&b"bc"[..]), Err(Err::Error((&b"bc"[..], ErrorKind::Char)))); -/// assert_eq!(char::<_, (_, ErrorKind)>('a')(&b""[..]), Err(Err::Incomplete(Needed::Size(1)))); -/// # } +/// fn parser(i: &str) -> IResult<&str, char> { +/// char('a')(i) +/// } +/// assert_eq!(parser("abc"), Ok(("bc", 'a'))); +/// assert_eq!(parser("bc"), Err(Err::Error(Error::new("bc", ErrorKind::Char)))); +/// assert_eq!(parser(""), Err(Err::Incomplete(Needed::new(1)))); /// ``` pub fn char<I, Error: ParseError<I>>(c: char) -> impl Fn(I) -> IResult<I, char, Error> where - I: Slice<RangeFrom<usize>> + InputIter, + I: Slice<RangeFrom<usize>> + InputIter + InputLength, <I as InputIter>::Item: AsChar, { move |i: I| match (i).iter_elements().next().map(|t| { let b = t.as_char() == c; (&c, b) }) { - None => Err(Err::Incomplete(Needed::Size(1))), - Some((_, false)) => { - Err(Err::Error(Error::from_char(i, c))) - } + None => Err(Err::Incomplete(Needed::new(c.len() - i.input_len()))), + Some((_, false)) => Err(Err::Error(Error::from_char(i, c))), Some((c, true)) => Ok((i.slice(c.len()..), c.as_char())), } } -/// Recognizes one of the provided characters. +/// Recognizes one character and checks that it satisfies a predicate /// -/// *streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data. +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data. +/// # Example /// +/// ``` +/// # use nom::{Err, error::{ErrorKind, Error}, Needed, IResult}; +/// # use nom::character::streaming::satisfy; +/// fn parser(i: &str) -> IResult<&str, char> { +/// satisfy(|c| c == 'a' || c == 'b')(i) +/// } +/// assert_eq!(parser("abc"), Ok(("bc", 'a'))); +/// assert_eq!(parser("cd"), Err(Err::Error(Error::new("cd", ErrorKind::Satisfy)))); +/// assert_eq!(parser(""), Err(Err::Incomplete(Needed::Unknown))); +/// ``` +pub fn satisfy<F, I, Error: ParseError<I>>(cond: F) -> impl Fn(I) -> IResult<I, char, Error> +where + I: Slice<RangeFrom<usize>> + InputIter, + <I as InputIter>::Item: AsChar, + F: Fn(char) -> bool, +{ + move |i: I| match (i).iter_elements().next().map(|t| { + let c = t.as_char(); + let b = cond(c); + (c, b) + }) { + None => Err(Err::Incomplete(Needed::Unknown)), + Some((_, false)) => Err(Err::Error(Error::from_error_kind(i, ErrorKind::Satisfy))), + Some((c, true)) => Ok((i.slice(c.len()..), c)), + } +} + +/// Recognizes one of the provided characters. +/// +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data. /// # Example /// /// ``` /// # use nom::{Err, error::ErrorKind, Needed}; /// # use nom::character::streaming::one_of; -/// # fn main() { /// assert_eq!(one_of::<_, _, (_, ErrorKind)>("abc")("b"), Ok(("", 'b'))); /// assert_eq!(one_of::<_, _, (_, ErrorKind)>("a")("bc"), Err(Err::Error(("bc", ErrorKind::OneOf)))); -/// assert_eq!(one_of::<_, _, (_, ErrorKind)>("a")(""), Err(Err::Incomplete(Needed::Size(1)))); -/// # } +/// assert_eq!(one_of::<_, _, (_, ErrorKind)>("a")(""), Err(Err::Incomplete(Needed::new(1)))); /// ``` pub fn one_of<I, T, Error: ParseError<I>>(list: T) -> impl Fn(I) -> IResult<I, char, Error> where @@ -64,7 +91,7 @@ where T: FindToken<<I as InputIter>::Item>, { move |i: I| match (i).iter_elements().next().map(|c| (c, list.find_token(c))) { - None => Err(Err::Incomplete(Needed::Size(1))), + None => Err(Err::Incomplete(Needed::new(1))), Some((_, false)) => Err(Err::Error(Error::from_error_kind(i, ErrorKind::OneOf))), Some((c, true)) => Ok((i.slice(c.len()..), c.as_char())), } @@ -72,18 +99,15 @@ where /// Recognizes a character that is not in the provided characters. /// -/// *streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data. -/// +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data. /// # Example /// /// ``` /// # use nom::{Err, error::ErrorKind, Needed}; /// # use nom::character::streaming::none_of; -/// # fn main() { /// assert_eq!(none_of::<_, _, (_, ErrorKind)>("abc")("z"), Ok(("", 'z'))); /// assert_eq!(none_of::<_, _, (_, ErrorKind)>("ab")("a"), Err(Err::Error(("a", ErrorKind::NoneOf)))); -/// assert_eq!(none_of::<_, _, (_, ErrorKind)>("a")(""), Err(Err::Incomplete(Needed::Size(1)))); -/// # } +/// assert_eq!(none_of::<_, _, (_, ErrorKind)>("a")(""), Err(Err::Incomplete(Needed::new(1)))); /// ``` pub fn none_of<I, T, Error: ParseError<I>>(list: T) -> impl Fn(I) -> IResult<I, char, Error> where @@ -92,7 +116,7 @@ where T: FindToken<<I as InputIter>::Item>, { move |i: I| match (i).iter_elements().next().map(|c| (c, !list.find_token(c))) { - None => Err(Err::Incomplete(Needed::Size(1))), + None => Err(Err::Incomplete(Needed::new(1))), Some((_, false)) => Err(Err::Error(Error::from_error_kind(i, ErrorKind::NoneOf))), Some((c, true)) => Ok((i.slice(c.len()..), c.as_char())), } @@ -100,18 +124,15 @@ where /// Recognizes the string "\r\n". /// -/// *streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data. -/// +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data. /// # Example /// /// ``` /// # use nom::{Err, error::ErrorKind, IResult, Needed}; /// # use nom::character::streaming::crlf; -/// # fn main() { /// assert_eq!(crlf::<_, (_, ErrorKind)>("\r\nc"), Ok(("c", "\r\n"))); /// assert_eq!(crlf::<_, (_, ErrorKind)>("ab\r\nc"), Err(Err::Error(("ab\r\nc", ErrorKind::CrLf)))); -/// assert_eq!(crlf::<_, (_, ErrorKind)>(""), Err(Err::Incomplete(Needed::Size(2)))); -/// # } +/// assert_eq!(crlf::<_, (_, ErrorKind)>(""), Err(Err::Incomplete(Needed::new(2)))); /// ``` pub fn crlf<T, E: ParseError<T>>(input: T) -> IResult<T, T, E> where @@ -122,7 +143,7 @@ where match input.compare("\r\n") { //FIXME: is this the right index? CompareResult::Ok => Ok((input.slice(2..), input.slice(0..2))), - CompareResult::Incomplete => Err(Err::Incomplete(Needed::Size(2))), + CompareResult::Incomplete => Err(Err::Incomplete(Needed::new(2))), CompareResult::Error => { let e: ErrorKind = ErrorKind::CrLf; Err(Err::Error(E::from_error_kind(input, e))) @@ -130,20 +151,19 @@ where } } -/// Recognizes a string of any char except '\r' or '\n'. -/// -/// *streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data. +/// Recognizes a string of any char except '\r\n' or '\n'. /// +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data. /// # Example /// /// ``` -/// # use nom::{Err, error::ErrorKind, IResult, Needed}; +/// # use nom::{Err, error::{Error, ErrorKind}, IResult, Needed}; /// # use nom::character::streaming::not_line_ending; -/// # fn main() { /// assert_eq!(not_line_ending::<_, (_, ErrorKind)>("ab\r\nc"), Ok(("\r\nc", "ab"))); /// assert_eq!(not_line_ending::<_, (_, ErrorKind)>("abc"), Err(Err::Incomplete(Needed::Unknown))); /// assert_eq!(not_line_ending::<_, (_, ErrorKind)>(""), Err(Err::Incomplete(Needed::Unknown))); -/// # } +/// assert_eq!(not_line_ending::<_, (_, ErrorKind)>("a\rb\nc"), Err(Err::Error(("a\rb\nc", ErrorKind::Tag )))); +/// assert_eq!(not_line_ending::<_, (_, ErrorKind)>("a\rbc"), Err(Err::Error(("a\rbc", ErrorKind::Tag )))); /// ``` pub fn not_line_ending<T, E: ParseError<T>>(input: T) -> IResult<T, T, E> where @@ -157,9 +177,7 @@ where let c = item.as_char(); c == '\r' || c == '\n' }) { - None => { - Err(Err::Incomplete(Needed::Unknown)) - } + None => Err(Err::Incomplete(Needed::Unknown)), Some(index) => { let mut it = input.slice(index..).iter_elements(); let nth = it.next().unwrap().as_char(); @@ -184,18 +202,15 @@ where /// Recognizes an end of line (both '\n' and '\r\n'). /// -/// *streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data. -/// +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data. /// # Example /// /// ``` /// # use nom::{Err, error::ErrorKind, IResult, Needed}; /// # use nom::character::streaming::line_ending; -/// # fn main() { /// assert_eq!(line_ending::<_, (_, ErrorKind)>("\r\nc"), Ok(("c", "\r\n"))); /// assert_eq!(line_ending::<_, (_, ErrorKind)>("ab\r\nc"), Err(Err::Error(("ab\r\nc", ErrorKind::CrLf)))); -/// assert_eq!(line_ending::<_, (_, ErrorKind)>(""), Err(Err::Incomplete(Needed::Size(1)))); -/// # } +/// assert_eq!(line_ending::<_, (_, ErrorKind)>(""), Err(Err::Incomplete(Needed::new(1)))); /// ``` pub fn line_ending<T, E: ParseError<T>>(input: T) -> IResult<T, T, E> where @@ -205,12 +220,12 @@ where { match input.compare("\n") { CompareResult::Ok => Ok((input.slice(1..), input.slice(0..1))), - CompareResult::Incomplete => Err(Err::Incomplete(Needed::Size(1))), + CompareResult::Incomplete => Err(Err::Incomplete(Needed::new(1))), CompareResult::Error => { match input.compare("\r\n") { //FIXME: is this the right index? CompareResult::Ok => Ok((input.slice(2..), input.slice(0..2))), - CompareResult::Incomplete => Err(Err::Incomplete(Needed::Size(2))), + CompareResult::Incomplete => Err(Err::Incomplete(Needed::new(2))), CompareResult::Error => Err(Err::Error(E::from_error_kind(input, ErrorKind::CrLf))), } } @@ -219,22 +234,19 @@ where /// Matches a newline character '\\n'. /// -/// *streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data. -/// +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data. /// # Example /// /// ``` /// # use nom::{Err, error::ErrorKind, IResult, Needed}; /// # use nom::character::streaming::newline; -/// # fn main() { /// assert_eq!(newline::<_, (_, ErrorKind)>("\nc"), Ok(("c", '\n'))); /// assert_eq!(newline::<_, (_, ErrorKind)>("\r\nc"), Err(Err::Error(("\r\nc", ErrorKind::Char)))); -/// assert_eq!(newline::<_, (_, ErrorKind)>(""), Err(Err::Incomplete(Needed::Size(1)))); -/// # } +/// assert_eq!(newline::<_, (_, ErrorKind)>(""), Err(Err::Incomplete(Needed::new(1)))); /// ``` pub fn newline<I, Error: ParseError<I>>(input: I) -> IResult<I, char, Error> where - I: Slice<RangeFrom<usize>> + InputIter, + I: Slice<RangeFrom<usize>> + InputIter + InputLength, <I as InputIter>::Item: AsChar, { char('\n')(input) @@ -242,22 +254,19 @@ where /// Matches a tab character '\t'. /// -/// *streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data. -/// +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data. /// # Example /// /// ``` /// # use nom::{Err, error::ErrorKind, IResult, Needed}; /// # use nom::character::streaming::tab; -/// # fn main() { /// assert_eq!(tab::<_, (_, ErrorKind)>("\tc"), Ok(("c", '\t'))); /// assert_eq!(tab::<_, (_, ErrorKind)>("\r\nc"), Err(Err::Error(("\r\nc", ErrorKind::Char)))); -/// assert_eq!(tab::<_, (_, ErrorKind)>(""), Err(Err::Incomplete(Needed::Size(1)))); -/// # } +/// assert_eq!(tab::<_, (_, ErrorKind)>(""), Err(Err::Incomplete(Needed::new(1)))); /// ``` pub fn tab<I, Error: ParseError<I>>(input: I) -> IResult<I, char, Error> where - I: Slice<RangeFrom<usize>> + InputIter, + I: Slice<RangeFrom<usize>> + InputIter + InputLength, <I as InputIter>::Item: AsChar, { char('\t')(input) @@ -266,16 +275,13 @@ where /// Matches one byte as a character. Note that the input type will /// accept a `str`, but not a `&[u8]`, unlike many other nom parsers. /// -/// *streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data. -/// +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data. /// # Example /// /// ``` /// # use nom::{character::streaming::anychar, Err, error::ErrorKind, IResult, Needed}; -/// # fn main() { /// assert_eq!(anychar::<_, (_, ErrorKind)>("abc"), Ok(("bc",'a'))); -/// assert_eq!(anychar::<_, (_, ErrorKind)>(""), Err(Err::Incomplete(Needed::Size(1)))); -/// # } +/// assert_eq!(anychar::<_, (_, ErrorKind)>(""), Err(Err::Incomplete(Needed::new(1)))); /// ``` pub fn anychar<T, E: ParseError<T>>(input: T) -> IResult<T, char, E> where @@ -284,7 +290,7 @@ where { let mut it = input.iter_indices(); match it.next() { - None => Err(Err::Incomplete(Needed::Size(1))), + None => Err(Err::Incomplete(Needed::new(1))), Some((_, c)) => match it.next() { None => Ok((input.slice(input.input_len()..), c.as_char())), Some((idx, _)) => Ok((input.slice(idx..), c.as_char())), @@ -294,19 +300,16 @@ where /// Recognizes zero or more lowercase and uppercase ASCII alphabetic characters: a-z, A-Z /// -/// *streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data, +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data, /// or if no terminating token is found (a non alphabetic character). -/// /// # Example /// /// ``` /// # use nom::{Err, error::ErrorKind, IResult, Needed}; /// # use nom::character::streaming::alpha0; -/// # fn main() { /// assert_eq!(alpha0::<_, (_, ErrorKind)>("ab1c"), Ok(("1c", "ab"))); /// assert_eq!(alpha0::<_, (_, ErrorKind)>("1c"), Ok(("1c", ""))); -/// assert_eq!(alpha0::<_, (_, ErrorKind)>(""), Err(Err::Incomplete(Needed::Size(1)))); -/// # } +/// assert_eq!(alpha0::<_, (_, ErrorKind)>(""), Err(Err::Incomplete(Needed::new(1)))); /// ``` pub fn alpha0<T, E: ParseError<T>>(input: T) -> IResult<T, T, E> where @@ -318,19 +321,16 @@ where /// Recognizes one or more lowercase and uppercase ASCII alphabetic characters: a-z, A-Z /// -/// *streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data, +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data, /// or if no terminating token is found (a non alphabetic character). -/// /// # Example /// /// ``` /// # use nom::{Err, error::ErrorKind, IResult, Needed}; /// # use nom::character::streaming::alpha1; -/// # fn main() { /// assert_eq!(alpha1::<_, (_, ErrorKind)>("aB1c"), Ok(("1c", "aB"))); /// assert_eq!(alpha1::<_, (_, ErrorKind)>("1c"), Err(Err::Error(("1c", ErrorKind::Alpha)))); -/// assert_eq!(alpha1::<_, (_, ErrorKind)>(""), Err(Err::Incomplete(Needed::Size(1)))); -/// # } +/// assert_eq!(alpha1::<_, (_, ErrorKind)>(""), Err(Err::Incomplete(Needed::new(1)))); /// ``` pub fn alpha1<T, E: ParseError<T>>(input: T) -> IResult<T, T, E> where @@ -342,19 +342,16 @@ where /// Recognizes zero or more ASCII numerical characters: 0-9 /// -/// *streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data, +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data, /// or if no terminating token is found (a non digit character). -/// /// # Example /// /// ``` /// # use nom::{Err, error::ErrorKind, IResult, Needed}; /// # use nom::character::streaming::digit0; -/// # fn main() { /// assert_eq!(digit0::<_, (_, ErrorKind)>("21c"), Ok(("c", "21"))); /// assert_eq!(digit0::<_, (_, ErrorKind)>("a21c"), Ok(("a21c", ""))); -/// assert_eq!(digit0::<_, (_, ErrorKind)>(""), Err(Err::Incomplete(Needed::Size(1)))); -/// # } +/// assert_eq!(digit0::<_, (_, ErrorKind)>(""), Err(Err::Incomplete(Needed::new(1)))); /// ``` pub fn digit0<T, E: ParseError<T>>(input: T) -> IResult<T, T, E> where @@ -366,19 +363,16 @@ where /// Recognizes one or more ASCII numerical characters: 0-9 /// -/// *streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data, +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data, /// or if no terminating token is found (a non digit character). -/// /// # Example /// /// ``` /// # use nom::{Err, error::ErrorKind, IResult, Needed}; /// # use nom::character::streaming::digit1; -/// # fn main() { /// assert_eq!(digit1::<_, (_, ErrorKind)>("21c"), Ok(("c", "21"))); /// assert_eq!(digit1::<_, (_, ErrorKind)>("c1"), Err(Err::Error(("c1", ErrorKind::Digit)))); -/// assert_eq!(digit1::<_, (_, ErrorKind)>(""), Err(Err::Incomplete(Needed::Size(1)))); -/// # } +/// assert_eq!(digit1::<_, (_, ErrorKind)>(""), Err(Err::Incomplete(Needed::new(1)))); /// ``` pub fn digit1<T, E: ParseError<T>>(input: T) -> IResult<T, T, E> where @@ -390,19 +384,16 @@ where /// Recognizes zero or more ASCII hexadecimal numerical characters: 0-9, A-F, a-f /// -/// *streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data, +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data, /// or if no terminating token is found (a non hexadecimal digit character). -/// /// # Example /// /// ``` /// # use nom::{Err, error::ErrorKind, IResult, Needed}; /// # use nom::character::streaming::hex_digit0; -/// # fn main() { /// assert_eq!(hex_digit0::<_, (_, ErrorKind)>("21cZ"), Ok(("Z", "21c"))); /// assert_eq!(hex_digit0::<_, (_, ErrorKind)>("Z21c"), Ok(("Z21c", ""))); -/// assert_eq!(hex_digit0::<_, (_, ErrorKind)>(""), Err(Err::Incomplete(Needed::Size(1)))); -/// # } +/// assert_eq!(hex_digit0::<_, (_, ErrorKind)>(""), Err(Err::Incomplete(Needed::new(1)))); /// ``` pub fn hex_digit0<T, E: ParseError<T>>(input: T) -> IResult<T, T, E> where @@ -414,19 +405,16 @@ where /// Recognizes one or more ASCII hexadecimal numerical characters: 0-9, A-F, a-f /// -/// *streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data, +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data, /// or if no terminating token is found (a non hexadecimal digit character). -/// /// # Example /// /// ``` /// # use nom::{Err, error::ErrorKind, IResult, Needed}; /// # use nom::character::streaming::hex_digit1; -/// # fn main() { /// assert_eq!(hex_digit1::<_, (_, ErrorKind)>("21cZ"), Ok(("Z", "21c"))); /// assert_eq!(hex_digit1::<_, (_, ErrorKind)>("H2"), Err(Err::Error(("H2", ErrorKind::HexDigit)))); -/// assert_eq!(hex_digit1::<_, (_, ErrorKind)>(""), Err(Err::Incomplete(Needed::Size(1)))); -/// # } +/// assert_eq!(hex_digit1::<_, (_, ErrorKind)>(""), Err(Err::Incomplete(Needed::new(1)))); /// ``` pub fn hex_digit1<T, E: ParseError<T>>(input: T) -> IResult<T, T, E> where @@ -438,19 +426,16 @@ where /// Recognizes zero or more octal characters: 0-7 /// -/// *streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data, +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data, /// or if no terminating token is found (a non octal digit character). -/// /// # Example /// /// ``` /// # use nom::{Err, error::ErrorKind, IResult, Needed}; /// # use nom::character::streaming::oct_digit0; -/// # fn main() { /// assert_eq!(oct_digit0::<_, (_, ErrorKind)>("21cZ"), Ok(("cZ", "21"))); /// assert_eq!(oct_digit0::<_, (_, ErrorKind)>("Z21c"), Ok(("Z21c", ""))); -/// assert_eq!(oct_digit0::<_, (_, ErrorKind)>(""), Err(Err::Incomplete(Needed::Size(1)))); -/// # } +/// assert_eq!(oct_digit0::<_, (_, ErrorKind)>(""), Err(Err::Incomplete(Needed::new(1)))); /// ``` pub fn oct_digit0<T, E: ParseError<T>>(input: T) -> IResult<T, T, E> where @@ -462,19 +447,16 @@ where /// Recognizes one or more octal characters: 0-7 /// -/// *streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data, +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data, /// or if no terminating token is found (a non octal digit character). -/// /// # Example /// /// ``` /// # use nom::{Err, error::ErrorKind, IResult, Needed}; /// # use nom::character::streaming::oct_digit1; -/// # fn main() { /// assert_eq!(oct_digit1::<_, (_, ErrorKind)>("21cZ"), Ok(("cZ", "21"))); /// assert_eq!(oct_digit1::<_, (_, ErrorKind)>("H2"), Err(Err::Error(("H2", ErrorKind::OctDigit)))); -/// assert_eq!(oct_digit1::<_, (_, ErrorKind)>(""), Err(Err::Incomplete(Needed::Size(1)))); -/// # } +/// assert_eq!(oct_digit1::<_, (_, ErrorKind)>(""), Err(Err::Incomplete(Needed::new(1)))); /// ``` pub fn oct_digit1<T, E: ParseError<T>>(input: T) -> IResult<T, T, E> where @@ -486,19 +468,16 @@ where /// Recognizes zero or more ASCII numerical and alphabetic characters: 0-9, a-z, A-Z /// -/// *streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data, +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data, /// or if no terminating token is found (a non alphanumerical character). -/// /// # Example /// /// ``` /// # use nom::{Err, error::ErrorKind, IResult, Needed}; /// # use nom::character::streaming::alphanumeric0; -/// # fn main() { /// assert_eq!(alphanumeric0::<_, (_, ErrorKind)>("21cZ%1"), Ok(("%1", "21cZ"))); /// assert_eq!(alphanumeric0::<_, (_, ErrorKind)>("&Z21c"), Ok(("&Z21c", ""))); -/// assert_eq!(alphanumeric0::<_, (_, ErrorKind)>(""), Err(Err::Incomplete(Needed::Size(1)))); -/// # } +/// assert_eq!(alphanumeric0::<_, (_, ErrorKind)>(""), Err(Err::Incomplete(Needed::new(1)))); /// ``` pub fn alphanumeric0<T, E: ParseError<T>>(input: T) -> IResult<T, T, E> where @@ -510,19 +489,16 @@ where /// Recognizes one or more ASCII numerical and alphabetic characters: 0-9, a-z, A-Z /// -/// *streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data, +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data, /// or if no terminating token is found (a non alphanumerical character). -/// /// # Example /// /// ``` /// # use nom::{Err, error::ErrorKind, IResult, Needed}; /// # use nom::character::streaming::alphanumeric1; -/// # fn main() { /// assert_eq!(alphanumeric1::<_, (_, ErrorKind)>("21cZ%1"), Ok(("%1", "21cZ"))); /// assert_eq!(alphanumeric1::<_, (_, ErrorKind)>("&H2"), Err(Err::Error(("&H2", ErrorKind::AlphaNumeric)))); -/// assert_eq!(alphanumeric1::<_, (_, ErrorKind)>(""), Err(Err::Incomplete(Needed::Size(1)))); -/// # } +/// assert_eq!(alphanumeric1::<_, (_, ErrorKind)>(""), Err(Err::Incomplete(Needed::new(1)))); /// ``` pub fn alphanumeric1<T, E: ParseError<T>>(input: T) -> IResult<T, T, E> where @@ -534,19 +510,16 @@ where /// Recognizes zero or more spaces and tabs. /// -/// *streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data, +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data, /// or if no terminating token is found (a non space character). -/// /// # Example /// /// ``` /// # use nom::{Err, error::ErrorKind, IResult, Needed}; /// # use nom::character::streaming::space0; -/// # fn main() { /// assert_eq!(space0::<_, (_, ErrorKind)>(" \t21c"), Ok(("21c", " \t"))); /// assert_eq!(space0::<_, (_, ErrorKind)>("Z21c"), Ok(("Z21c", ""))); -/// assert_eq!(space0::<_, (_, ErrorKind)>(""), Err(Err::Incomplete(Needed::Size(1)))); -/// # } +/// assert_eq!(space0::<_, (_, ErrorKind)>(""), Err(Err::Incomplete(Needed::new(1)))); /// ``` pub fn space0<T, E: ParseError<T>>(input: T) -> IResult<T, T, E> where @@ -554,25 +527,22 @@ where <T as InputTakeAtPosition>::Item: AsChar + Clone, { input.split_at_position(|item| { - let c = item.clone().as_char(); + let c = item.as_char(); !(c == ' ' || c == '\t') }) } /// Recognizes one or more spaces and tabs. /// -/// *streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data, +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data, /// or if no terminating token is found (a non space character). -/// /// # Example /// /// ``` /// # use nom::{Err, error::ErrorKind, IResult, Needed}; /// # use nom::character::streaming::space1; -/// # fn main() { /// assert_eq!(space1::<_, (_, ErrorKind)>(" \t21c"), Ok(("21c", " \t"))); /// assert_eq!(space1::<_, (_, ErrorKind)>("H2"), Err(Err::Error(("H2", ErrorKind::Space)))); -/// assert_eq!(space1::<_, (_, ErrorKind)>(""), Err(Err::Incomplete(Needed::Size(1)))); -/// # } +/// assert_eq!(space1::<_, (_, ErrorKind)>(""), Err(Err::Incomplete(Needed::new(1)))); /// ``` pub fn space1<T, E: ParseError<T>>(input: T) -> IResult<T, T, E> where @@ -581,7 +551,7 @@ where { input.split_at_position1( |item| { - let c = item.clone().as_char(); + let c = item.as_char(); !(c == ' ' || c == '\t') }, ErrorKind::Space, @@ -590,19 +560,16 @@ where /// Recognizes zero or more spaces, tabs, carriage returns and line feeds. /// -/// *streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data, +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data, /// or if no terminating token is found (a non space character). -/// /// # Example /// /// ``` /// # use nom::{Err, error::ErrorKind, IResult, Needed}; /// # use nom::character::streaming::multispace0; -/// # fn main() { /// assert_eq!(multispace0::<_, (_, ErrorKind)>(" \t\n\r21c"), Ok(("21c", " \t\n\r"))); /// assert_eq!(multispace0::<_, (_, ErrorKind)>("Z21c"), Ok(("Z21c", ""))); -/// assert_eq!(multispace0::<_, (_, ErrorKind)>(""), Err(Err::Incomplete(Needed::Size(1)))); -/// # } +/// assert_eq!(multispace0::<_, (_, ErrorKind)>(""), Err(Err::Incomplete(Needed::new(1)))); /// ``` pub fn multispace0<T, E: ParseError<T>>(input: T) -> IResult<T, T, E> where @@ -610,26 +577,23 @@ where <T as InputTakeAtPosition>::Item: AsChar + Clone, { input.split_at_position(|item| { - let c = item.clone().as_char(); + let c = item.as_char(); !(c == ' ' || c == '\t' || c == '\r' || c == '\n') }) } /// Recognizes one or more spaces, tabs, carriage returns and line feeds. /// -/// *streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data, +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data, /// or if no terminating token is found (a non space character). -/// /// # Example /// /// ``` /// # use nom::{Err, error::ErrorKind, IResult, Needed}; /// # use nom::character::streaming::multispace1; -/// # fn main() { /// assert_eq!(multispace1::<_, (_, ErrorKind)>(" \t\n\r21c"), Ok(("21c", " \t\n\r"))); /// assert_eq!(multispace1::<_, (_, ErrorKind)>("H2"), Err(Err::Error(("H2", ErrorKind::MultiSpace)))); -/// assert_eq!(multispace1::<_, (_, ErrorKind)>(""), Err(Err::Incomplete(Needed::Size(1)))); -/// # } +/// assert_eq!(multispace1::<_, (_, ErrorKind)>(""), Err(Err::Incomplete(Needed::new(1)))); /// ``` pub fn multispace1<T, E: ParseError<T>>(input: T) -> IResult<T, T, E> where @@ -638,7 +602,7 @@ where { input.split_at_position1( |item| { - let c = item.clone().as_char(); + let c = item.as_char(); !(c == ' ' || c == '\t' || c == '\r' || c == '\n') }, ErrorKind::MultiSpace, @@ -648,8 +612,8 @@ where #[cfg(test)] mod tests { use super::*; - use crate::internal::{Err, Needed}; use crate::error::ErrorKind; + use crate::internal::{Err, Needed}; macro_rules! assert_parse( ($left: expr, $right: expr) => { @@ -672,53 +636,62 @@ mod tests { let d: &[u8] = "azĆ©12".as_bytes(); let e: &[u8] = b" "; let f: &[u8] = b" ;"; - //assert_eq!(alpha1::<_, (_, ErrorKind)>(a), Err(Err::Incomplete(Needed::Size(1)))); - assert_parse!(alpha1(a), Err(Err::Incomplete(Needed::Size(1)))); + //assert_eq!(alpha1::<_, (_, ErrorKind)>(a), Err(Err::Incomplete(Needed::new(1)))); + assert_parse!(alpha1(a), Err(Err::Incomplete(Needed::new(1)))); + assert_eq!(alpha1(b), Err(Err::Error((b, ErrorKind::Alpha)))); + assert_eq!(alpha1::<_, (_, ErrorKind)>(c), Ok((&c[1..], &b"a"[..]))); assert_eq!( - alpha1(b), - Err(Err::Error((b, ErrorKind::Alpha))) + alpha1::<_, (_, ErrorKind)>(d), + Ok(("Ć©12".as_bytes(), &b"az"[..])) ); - assert_eq!(alpha1::<_, (_, ErrorKind)>(c), Ok((&c[1..], &b"a"[..]))); - assert_eq!(alpha1::<_, (_, ErrorKind)>(d), Ok(("Ć©12".as_bytes(), &b"az"[..]))); + assert_eq!(digit1(a), Err(Err::Error((a, ErrorKind::Digit)))); assert_eq!( - digit1(a), - Err(Err::Error((a, ErrorKind::Digit))) + digit1::<_, (_, ErrorKind)>(b), + Err(Err::Incomplete(Needed::new(1))) ); - assert_eq!(digit1::<_, (_, ErrorKind)>(b), Err(Err::Incomplete(Needed::Size(1)))); + assert_eq!(digit1(c), Err(Err::Error((c, ErrorKind::Digit)))); + assert_eq!(digit1(d), Err(Err::Error((d, ErrorKind::Digit)))); assert_eq!( - digit1(c), - Err(Err::Error((c, ErrorKind::Digit))) + hex_digit1::<_, (_, ErrorKind)>(a), + Err(Err::Incomplete(Needed::new(1))) ); assert_eq!( - digit1(d), - Err(Err::Error((d, ErrorKind::Digit))) + hex_digit1::<_, (_, ErrorKind)>(b), + Err(Err::Incomplete(Needed::new(1))) ); - assert_eq!(hex_digit1::<_, (_, ErrorKind)>(a), Err(Err::Incomplete(Needed::Size(1)))); - assert_eq!(hex_digit1::<_, (_, ErrorKind)>(b), Err(Err::Incomplete(Needed::Size(1)))); - assert_eq!(hex_digit1::<_, (_, ErrorKind)>(c), Err(Err::Incomplete(Needed::Size(1)))); - assert_eq!(hex_digit1::<_, (_, ErrorKind)>(d), Ok(("zĆ©12".as_bytes(), &b"a"[..]))); assert_eq!( - hex_digit1(e), - Err(Err::Error((e, ErrorKind::HexDigit))) + hex_digit1::<_, (_, ErrorKind)>(c), + Err(Err::Incomplete(Needed::new(1))) ); assert_eq!( - oct_digit1(a), - Err(Err::Error((a, ErrorKind::OctDigit))) + hex_digit1::<_, (_, ErrorKind)>(d), + Ok(("zĆ©12".as_bytes(), &b"a"[..])) ); - assert_eq!(oct_digit1::<_, (_, ErrorKind)>(b), Err(Err::Incomplete(Needed::Size(1)))); + assert_eq!(hex_digit1(e), Err(Err::Error((e, ErrorKind::HexDigit)))); + assert_eq!(oct_digit1(a), Err(Err::Error((a, ErrorKind::OctDigit)))); assert_eq!( - oct_digit1(c), - Err(Err::Error((c, ErrorKind::OctDigit))) + oct_digit1::<_, (_, ErrorKind)>(b), + Err(Err::Incomplete(Needed::new(1))) ); + assert_eq!(oct_digit1(c), Err(Err::Error((c, ErrorKind::OctDigit)))); + assert_eq!(oct_digit1(d), Err(Err::Error((d, ErrorKind::OctDigit)))); assert_eq!( - oct_digit1(d), - Err(Err::Error((d, ErrorKind::OctDigit))) + alphanumeric1::<_, (_, ErrorKind)>(a), + Err(Err::Incomplete(Needed::new(1))) ); - assert_eq!(alphanumeric1::<_, (_, ErrorKind)>(a), Err(Err::Incomplete(Needed::Size(1)))); //assert_eq!(fix_error!(b,(), alphanumeric1), Ok((empty, b))); - assert_eq!(alphanumeric1::<_, (_, ErrorKind)>(c), Err(Err::Incomplete(Needed::Size(1)))); - assert_eq!(alphanumeric1::<_, (_, ErrorKind)>(d), Ok(("Ć©12".as_bytes(), &b"az"[..]))); - assert_eq!(space1::<_, (_, ErrorKind)>(e), Err(Err::Incomplete(Needed::Size(1)))); + assert_eq!( + alphanumeric1::<_, (_, ErrorKind)>(c), + Err(Err::Incomplete(Needed::new(1))) + ); + assert_eq!( + alphanumeric1::<_, (_, ErrorKind)>(d), + Ok(("Ć©12".as_bytes(), &b"az"[..])) + ); + assert_eq!( + space1::<_, (_, ErrorKind)>(e), + Err(Err::Incomplete(Needed::new(1))) + ); assert_eq!(space1::<_, (_, ErrorKind)>(f), Ok((&b";"[..], &b" "[..]))); } @@ -730,52 +703,55 @@ mod tests { let c = "a123"; let d = "azĆ©12"; let e = " "; - assert_eq!(alpha1::<_, (_, ErrorKind)>(a), Err(Err::Incomplete(Needed::Size(1)))); assert_eq!( - alpha1(b), - Err(Err::Error((b, ErrorKind::Alpha))) + alpha1::<_, (_, ErrorKind)>(a), + Err(Err::Incomplete(Needed::new(1))) ); + assert_eq!(alpha1(b), Err(Err::Error((b, ErrorKind::Alpha)))); assert_eq!(alpha1::<_, (_, ErrorKind)>(c), Ok((&c[1..], &"a"[..]))); assert_eq!(alpha1::<_, (_, ErrorKind)>(d), Ok(("Ć©12", &"az"[..]))); + assert_eq!(digit1(a), Err(Err::Error((a, ErrorKind::Digit)))); assert_eq!( - digit1(a), - Err(Err::Error((a, ErrorKind::Digit))) + digit1::<_, (_, ErrorKind)>(b), + Err(Err::Incomplete(Needed::new(1))) ); - assert_eq!(digit1::<_, (_, ErrorKind)>(b), Err(Err::Incomplete(Needed::Size(1)))); + assert_eq!(digit1(c), Err(Err::Error((c, ErrorKind::Digit)))); + assert_eq!(digit1(d), Err(Err::Error((d, ErrorKind::Digit)))); assert_eq!( - digit1(c), - Err(Err::Error((c, ErrorKind::Digit))) + hex_digit1::<_, (_, ErrorKind)>(a), + Err(Err::Incomplete(Needed::new(1))) ); assert_eq!( - digit1(d), - Err(Err::Error((d, ErrorKind::Digit))) + hex_digit1::<_, (_, ErrorKind)>(b), + Err(Err::Incomplete(Needed::new(1))) ); - assert_eq!(hex_digit1::<_, (_, ErrorKind)>(a), Err(Err::Incomplete(Needed::Size(1)))); - assert_eq!(hex_digit1::<_, (_, ErrorKind)>(b), Err(Err::Incomplete(Needed::Size(1)))); - assert_eq!(hex_digit1::<_, (_, ErrorKind)>(c), Err(Err::Incomplete(Needed::Size(1)))); - assert_eq!(hex_digit1::<_, (_, ErrorKind)>(d), Ok(("zĆ©12", &"a"[..]))); assert_eq!( - hex_digit1(e), - Err(Err::Error((e, ErrorKind::HexDigit))) + hex_digit1::<_, (_, ErrorKind)>(c), + Err(Err::Incomplete(Needed::new(1))) ); + assert_eq!(hex_digit1::<_, (_, ErrorKind)>(d), Ok(("zĆ©12", &"a"[..]))); + assert_eq!(hex_digit1(e), Err(Err::Error((e, ErrorKind::HexDigit)))); + assert_eq!(oct_digit1(a), Err(Err::Error((a, ErrorKind::OctDigit)))); assert_eq!( - oct_digit1(a), - Err(Err::Error((a, ErrorKind::OctDigit))) + oct_digit1::<_, (_, ErrorKind)>(b), + Err(Err::Incomplete(Needed::new(1))) ); - assert_eq!(oct_digit1::<_, (_, ErrorKind)>(b), Err(Err::Incomplete(Needed::Size(1)))); + assert_eq!(oct_digit1(c), Err(Err::Error((c, ErrorKind::OctDigit)))); + assert_eq!(oct_digit1(d), Err(Err::Error((d, ErrorKind::OctDigit)))); assert_eq!( - oct_digit1(c), - Err(Err::Error((c, ErrorKind::OctDigit))) + alphanumeric1::<_, (_, ErrorKind)>(a), + Err(Err::Incomplete(Needed::new(1))) ); + //assert_eq!(fix_error!(b,(), alphanumeric1), Ok((empty, b))); assert_eq!( - oct_digit1(d), - Err(Err::Error((d, ErrorKind::OctDigit))) + alphanumeric1::<_, (_, ErrorKind)>(c), + Err(Err::Incomplete(Needed::new(1))) ); - assert_eq!(alphanumeric1::<_, (_, ErrorKind)>(a), Err(Err::Incomplete(Needed::Size(1)))); - //assert_eq!(fix_error!(b,(), alphanumeric1), Ok((empty, b))); - assert_eq!(alphanumeric1::<_, (_, ErrorKind)>(c), Err(Err::Incomplete(Needed::Size(1)))); assert_eq!(alphanumeric1::<_, (_, ErrorKind)>(d), Ok(("Ć©12", "az"))); - assert_eq!(space1::<_, (_, ErrorKind)>(e), Err(Err::Incomplete(Needed::Size(1)))); + assert_eq!( + space1::<_, (_, ErrorKind)>(e), + Err(Err::Incomplete(Needed::new(1))) + ); } use crate::traits::Offset; @@ -835,7 +811,10 @@ mod tests { #[test] fn is_not_line_ending_bytes() { let a: &[u8] = b"ab12cd\nefgh"; - assert_eq!(not_line_ending::<_, (_, ErrorKind)>(a), Ok((&b"\nefgh"[..], &b"ab12cd"[..]))); + assert_eq!( + not_line_ending::<_, (_, ErrorKind)>(a), + Ok((&b"\nefgh"[..], &b"ab12cd"[..])) + ); let b: &[u8] = b"ab12cd\nefgh\nijkl"; assert_eq!( @@ -850,7 +829,10 @@ mod tests { ); let d: &[u8] = b"ab12cd"; - assert_eq!(not_line_ending::<_, (_, ErrorKind)>(d), Err(Err::Incomplete(Needed::Unknown))); + assert_eq!( + not_line_ending::<_, (_, ErrorKind)>(d), + Err(Err::Incomplete(Needed::Unknown)) + ); } #[test] @@ -873,13 +855,13 @@ mod tests { */ let f = "Ī²ĆØĘĆ“ÅĆØ\rĆĆĆĆ”ĘĘĆØÅ"; - assert_eq!( - not_line_ending(f), - Err(Err::Error((f, ErrorKind::Tag))) - ); + assert_eq!(not_line_ending(f), Err(Err::Error((f, ErrorKind::Tag)))); let g2: &str = "ab12cd"; - assert_eq!(not_line_ending::<_, (_, ErrorKind)>(g2), Err(Err::Incomplete(Needed::Unknown))); + assert_eq!( + not_line_ending::<_, (_, ErrorKind)>(g2), + Err(Err::Incomplete(Needed::Unknown)) + ); } #[test] @@ -975,14 +957,14 @@ mod tests { #[test] fn cr_lf() { assert_parse!(crlf(&b"\r\na"[..]), Ok((&b"a"[..], &b"\r\n"[..]))); - assert_parse!(crlf(&b"\r"[..]), Err(Err::Incomplete(Needed::Size(2)))); + assert_parse!(crlf(&b"\r"[..]), Err(Err::Incomplete(Needed::new(2)))); assert_parse!( crlf(&b"\ra"[..]), Err(Err::Error(error_position!(&b"\ra"[..], ErrorKind::CrLf))) ); assert_parse!(crlf("\r\na"), Ok(("a", "\r\n"))); - assert_parse!(crlf("\r"), Err(Err::Incomplete(Needed::Size(2)))); + assert_parse!(crlf("\r"), Err(Err::Incomplete(Needed::new(2)))); assert_parse!( crlf("\ra"), Err(Err::Error(error_position!("\ra", ErrorKind::CrLf))) @@ -993,7 +975,10 @@ mod tests { fn end_of_line() { assert_parse!(line_ending(&b"\na"[..]), Ok((&b"a"[..], &b"\n"[..]))); assert_parse!(line_ending(&b"\r\na"[..]), Ok((&b"a"[..], &b"\r\n"[..]))); - assert_parse!(line_ending(&b"\r"[..]), Err(Err::Incomplete(Needed::Size(2)))); + assert_parse!( + line_ending(&b"\r"[..]), + Err(Err::Incomplete(Needed::new(2))) + ); assert_parse!( line_ending(&b"\ra"[..]), Err(Err::Error(error_position!(&b"\ra"[..], ErrorKind::CrLf))) @@ -1001,7 +986,7 @@ mod tests { assert_parse!(line_ending("\na"), Ok(("a", "\n"))); assert_parse!(line_ending("\r\na"), Ok(("a", "\r\n"))); - assert_parse!(line_ending("\r"), Err(Err::Incomplete(Needed::Size(2)))); + assert_parse!(line_ending("\r"), Err(Err::Incomplete(Needed::new(2)))); assert_parse!( line_ending("\ra"), Err(Err::Error(error_position!("\ra", ErrorKind::CrLf))) diff --git a/src/combinator/macros.rs b/src/combinator/macros.rs index 59df06f..11ece16 100644 --- a/src/combinator/macros.rs +++ b/src/combinator/macros.rs @@ -17,7 +17,7 @@ //! ); //! ``` //! -//! But when used in other combinators, are Used +//! But when used in other combinators, are used //! like this: //! //! ```ignore @@ -45,8 +45,8 @@ //! ``` //! //! Combinators must have a specific variant for -//! non-macro arguments. Example: passing a function -//! to take_while! instead of another combinator. +//! non-macro arguments. Example: Passing a function +//! to `take_while!` instead of another combinator. //! //! ```ignore //! macro_rules! take_while( @@ -67,7 +67,7 @@ /// Makes a function from a parser combination /// /// The type can be set up if the compiler needs -/// more information +/// more information. /// /// Function-like declaration: /// ``` @@ -105,7 +105,7 @@ macro_rules! named ( named_attr!(#$($args)*); ); ($vis:vis $name:ident( $i:ty ) -> $o:ty, $submac:ident!( $($args:tt)* )) => ( - $vis fn $name( i: $i ) -> $crate::IResult<$i, $o, ($i, $crate::error::ErrorKind)> { + $vis fn $name( i: $i ) -> $crate::IResult<$i, $o, $crate::error::Error<$i>> { $submac!(i, $($args)*) } ); @@ -115,17 +115,17 @@ macro_rules! named ( } ); ($vis:vis $name:ident<$i:ty,$o:ty>, $submac:ident!( $($args:tt)* )) => ( - $vis fn $name( i: $i ) -> $crate::IResult<$i, $o, ($i, $crate::error::ErrorKind)> { + $vis fn $name( i: $i ) -> $crate::IResult<$i, $o, $crate::error::Error<$i>> { $submac!(i, $($args)*) } ); ($vis:vis $name:ident<$o:ty>, $submac:ident!( $($args:tt)* )) => ( - $vis fn $name( i: &[u8] ) -> $crate::IResult<&[u8], $o, (&[u8], $crate::error::ErrorKind)> { + $vis fn $name( i: &[u8] ) -> $crate::IResult<&[u8], $o, $crate::error::Error<&[u8]>> { $submac!(i, $($args)*) } ); ($vis:vis $name:ident, $submac:ident!( $($args:tt)* )) => ( - $vis fn $name( i: &[u8] ) -> $crate::IResult<&[u8], &[u8], (&[u8], $crate::error::ErrorKind)> { + $vis fn $name( i: &[u8] ) -> $crate::IResult<&[u8], &[u8], $crate::error::Error<&[u8]>> { $submac!(i, $($args)*) } ); @@ -145,7 +145,7 @@ macro_rules! named ( /// ); /// ``` /// -/// Note: if using arguments that way gets hard to read, it is always +/// Note: If using arguments that way gets hard to read, it is always /// possible to write the equivalent parser definition manually, like /// this: /// @@ -192,7 +192,7 @@ macro_rules! named_args { }; } -/// Makes a function from a parser combination, with attributes +/// Makes a function from a parser combination, with attributes. /// /// The usage of this macro is almost identical to `named!`, except that /// you also pass attributes to be attached to the generated function. @@ -217,7 +217,7 @@ macro_rules! named_args { macro_rules! named_attr ( ($(#[$attr:meta])*, $vis:vis $name:ident( $i:ty ) -> $o:ty, $submac:ident!( $($args:tt)* )) => ( $(#[$attr])* - $vis fn $name( i: $i ) -> $crate::IResult<$i,$o, ($i, $crate::error::ErrorKind)> { + $vis fn $name( i: $i ) -> $crate::IResult<$i,$o, $crate::error::Error<$i>> { $submac!(i, $($args)*) } ); @@ -229,25 +229,25 @@ macro_rules! named_attr ( ); ($(#[$attr:meta])*, $vis:vis $name:ident<$i:ty,$o:ty>, $submac:ident!( $($args:tt)* )) => ( $(#[$attr])* - $vis fn $name( i: $i ) -> $crate::IResult<$i, $o, ($i, $crate::error::ErrorKind)> { + $vis fn $name( i: $i ) -> $crate::IResult<$i, $o, $crate::error::Error<$i>> { $submac!(i, $($args)*) } ); ($(#[$attr:meta])*, $vis:vis $name:ident<$o:ty>, $submac:ident!( $($args:tt)* )) => ( $(#[$attr])* - $vis fn $name( i: &[u8] ) -> $crate::IResult<&[u8], $o, (&[u8], $crate::error::ErrorKind)> { + $vis fn $name( i: &[u8] ) -> $crate::IResult<&[u8], $o, $crate::error::Error<&[u8]>> { $submac!(i, $($args)*) } ); ($(#[$attr:meta])*, $vis:vis $name:ident, $submac:ident!( $($args:tt)* )) => ( $(#[$attr])* - $vis fn $name<'a>( i: &'a [u8] ) -> $crate::IResult<&[u8], &[u8], (&[u8], $crate::error::ErrorKind)> { + $vis fn $name<'a>( i: &'a [u8] ) -> $crate::IResult<&[u8], &[u8], $crate::error::Error<&[u8]>> { $submac!(i, $($args)*) } ); ); -/// Used to wrap common expressions and function as macros +/// Used to wrap common expressions and function as macros. /// /// ``` /// # #[macro_use] extern crate nom; @@ -266,7 +266,7 @@ macro_rules! call ( ); //FIXME: error rewrite -/// Prevents backtracking if the child parser fails +/// Prevents backtracking if the child parser fails. /// /// This parser will do an early return instead of sending /// its result to the parent parser. @@ -363,11 +363,11 @@ macro_rules! return_error ( ); //FIXME: error rewrite -/// Add an error if the child parser fails +/// Add an error if the child parser fails. /// /// While `return_error!` does an early return and avoids backtracking, -/// add_return_error! backtracks normally. It just provides more context -/// for an error +/// `add_return_error!` backtracks normally. It just provides more context +/// for an error. /// /// ``` /// # #[macro_use] extern crate nom; @@ -407,8 +407,8 @@ macro_rules! add_return_error ( ); ); -/// replaces a `Incomplete` returned by the child parser -/// with an `Error` +/// Replaces a `Incomplete` returned by the child parser +/// with an `Error`. /// /// ``` /// # #[macro_use] extern crate nom; @@ -438,7 +438,7 @@ macro_rules! complete ( /// parsed value if the child parser returned `Ok`, and will do an early /// return for the `Err` side. /// -/// this can provide more flexibility than `do_parse!` if needed +/// This can provide more flexibility than `do_parse!` if needed. /// /// ``` /// # #[macro_use] extern crate nom; @@ -479,7 +479,7 @@ macro_rules! try_parse ( /// `map!(I -> IResult<I, O>, O -> P) => I -> IResult<I, P>` /// -/// maps a function on the result of a parser +/// Maps a function on the result of a parser. /// /// ```rust /// # #[macro_use] extern crate nom; @@ -511,7 +511,7 @@ macro_rules! map( ); /// `map_res!(I -> IResult<I, O>, O -> Result<P>) => I -> IResult<I, P>` -/// maps a function returning a Result on the output of a parser +/// maps a function returning a `Result` on the output of a parser. /// /// ```rust /// # #[macro_use] extern crate nom; @@ -552,11 +552,11 @@ macro_rules! map_res ( ); /// `map_opt!(I -> IResult<I, O>, O -> Option<P>) => I -> IResult<I, P>` -/// maps a function returning an Option on the output of a parser +/// maps a function returning an `Option` on the output of a parser. /// /// ```rust /// # #[macro_use] extern crate nom; -/// # use nom::{Err,error::ErrorKind, IResult}; +/// # use nom::{Err,error::{Error, ErrorKind}, IResult}; /// use nom::character::complete::digit1; /// # fn main() { /// @@ -566,10 +566,10 @@ macro_rules! map_res ( /// assert_eq!(parser("123"), Ok(("", 123))); /// /// // this will fail if digit1 fails -/// assert_eq!(parser("abc"), Err(Err::Error(("abc", ErrorKind::Digit)))); +/// assert_eq!(parser("abc"), Err(Err::Error(Error::new("abc", ErrorKind::Digit)))); /// /// // this will fail if the mapped function fails (a `u8` is too small to hold `123456`) -/// assert_eq!(parser("123456"), Err(Err::Error(("123456", ErrorKind::MapOpt)))); +/// assert_eq!(parser("123456"), Err(Err::Error(Error::new("123456", ErrorKind::MapOpt)))); /// # } /// ``` #[macro_export(local_inner_macros)] @@ -593,14 +593,14 @@ macro_rules! map_opt ( ); /// `parse_to!(O) => I -> IResult<I, O>` -/// uses the `parse` method from `std::str::FromStr` to convert the current -/// input to the specified type +/// Uses the `parse` method from `std::str::FromStr` to convert the current +/// input to the specified type. /// -/// this will completely consume the input +/// This will completely consume the input. /// /// ```rust /// # #[macro_use] extern crate nom; -/// # use nom::{Err,error::ErrorKind, IResult}; +/// # use nom::{Err,error::{Error, ErrorKind}, IResult}; /// use nom::character::complete::digit1; /// # fn main() { /// @@ -608,10 +608,10 @@ macro_rules! map_opt ( /// /// assert_eq!(parser("123"), Ok(("", 123))); /// -/// assert_eq!(parser("abc"), Err(Err::Error(("abc", ErrorKind::ParseTo)))); +/// assert_eq!(parser("abc"), Err(Err::Error(Error::new("abc", ErrorKind::ParseTo)))); /// /// // this will fail if the mapped function fails (a `u8` is too small to hold `123456`) -/// assert_eq!(parser("123456"), Err(Err::Error(("123456", ErrorKind::ParseTo)))); +/// assert_eq!(parser("123456"), Err(Err::Error(Error::new("123456", ErrorKind::ParseTo)))); /// # } /// ``` #[macro_export(local_inner_macros)] @@ -637,7 +637,7 @@ macro_rules! parse_to ( ); /// `verify!(I -> IResult<I, O>, O -> bool) => I -> IResult<I, O>` -/// returns the result of the child parser if it satisfies a verification function +/// returns the result of the child parser if it satisfies a verification function. /// /// ``` /// # #[macro_use] extern crate nom; @@ -666,7 +666,7 @@ macro_rules! verify ( /// or `value!(T) => R -> IResult<R, T>` /// /// If the child parser was successful, return the value. -/// If no child parser is provided, always return the value +/// If no child parser is provided, always return the value. /// /// ``` /// # #[macro_use] extern crate nom; @@ -694,10 +694,10 @@ macro_rules! value ( ); /// `opt!(I -> IResult<I,O>) => I -> IResult<I, Option<O>>` -/// make the underlying parser optional +/// make the underlying parser optional. /// -/// returns an Option of the returned type. This parser returns `Some(result)` if the child parser -/// succeeds,`None` if it fails, and `Incomplete` if it did not have enough data to decide +/// Returns an `Option` of the returned type. This parser returns `Some(result)` if the child parser +/// succeeds, `None` if it fails, and `Incomplete` if it did not have enough data to decide. /// /// *Warning*: if you are using `opt` for some kind of optional ending token (like an end of line), /// you should combine it with `complete` to make sure it works. @@ -731,9 +731,9 @@ macro_rules! opt( ); /// `opt_res!(I -> IResult<I,O>) => I -> IResult<I, Result<nom::Err,O>>` -/// make the underlying parser optional +/// make the underlying parser optional. /// -/// returns a Result, with Err containing the parsing error +/// Returns a `Result`, with `Err` containing the parsing error. /// /// ```ignore /// # #[macro_use] extern crate nom; @@ -810,9 +810,9 @@ macro_rules! cond( ); /// `peek!(I -> IResult<I,O>) => I -> IResult<I, O>` -/// returns a result without consuming the input +/// returns a result without consuming the input. /// -/// the embedded parser may return Err(Err::Incomplete +/// The embedded parser may return `Err(Err::Incomplete)`. /// /// ``` /// # #[macro_use] extern crate nom; @@ -834,8 +834,8 @@ macro_rules! peek( ); /// `not!(I -> IResult<I,O>) => I -> IResult<I, ()>` -/// returns a result only if the embedded parser returns Error or Err(Err::Incomplete) -/// does not consume the input +/// returns a result only if the embedded parser returns `Error` or `Err(Err::Incomplete)`. +/// Does not consume the input. /// /// ``` /// # #[macro_use] extern crate nom; @@ -866,7 +866,7 @@ macro_rules! not( ); /// `tap!(name: I -> IResult<I,O> => { block }) => I -> IResult<I, O>` -/// allows access to the parser's result without affecting it +/// allows access to the parser's result without affecting it. /// /// ``` /// # #[macro_use] extern crate nom; @@ -900,20 +900,20 @@ macro_rules! tap ( ); ); -/// `eof!()` returns its input if it is at the end of input data +/// `eof!()` returns its input if it is at the end of input data. /// /// When we're at the end of the data, this combinator -/// will succeed +/// will succeed. /// /// /// ``` /// # #[macro_use] extern crate nom; /// # use std::str; -/// # use nom::{Err, error::ErrorKind}; +/// # use nom::{Err, error::{Error ,ErrorKind}}; /// # fn main() { /// named!(parser, eof!()); /// -/// assert_eq!(parser(&b"abc"[..]), Err(Err::Error((&b"abc"[..], ErrorKind::Eof)))); +/// assert_eq!(parser(&b"abc"[..]), Err(Err::Error(Error::new(&b"abc"[..], ErrorKind::Eof)))); /// assert_eq!(parser(&b""[..]), Ok((&b""[..], &b""[..]))); /// # } /// ``` @@ -926,7 +926,8 @@ macro_rules! eof ( use $crate::InputLength; if ($i).input_len() == 0 { - Ok(($i, $i)) + let clone = $i.clone(); + Ok(($i, clone)) } else { Err(Err::Error(error_position!($i, ErrorKind::Eof))) } @@ -934,7 +935,7 @@ macro_rules! eof ( ); ); -/// `exact!()` will fail if the child parser does not consume the whole data +/// `exact!()` will fail if the child parser does not consume the whole data. /// /// TODO: example #[macro_export(local_inner_macros)] @@ -948,7 +949,7 @@ macro_rules! exact ( ); /// `recognize!(I -> IResult<I, O> ) => I -> IResult<I, I>` -/// if the child parser was successful, return the consumed input as produced value +/// if the child parser was successful, return the consumed input as produced value. /// /// ``` /// # #[macro_use] extern crate nom; @@ -968,11 +969,35 @@ macro_rules! recognize ( ); ); +/// `into!(I -> IResult<I, O1, E1>) => I -> IResult<I, O2, E2>` +/// automatically converts the child parser's result to another type +/// +/// it will be able to convert the output value and the error value +/// as long as the `Into` implementations are available +/// +/// ```rust +/// # #[macro_use] extern crate nom; +/// # use nom::IResult; +/// # fn main() { +/// named!(parse_to_str<&str, &str>, take!(4)); +/// named!(parse_to_vec<&str, Vec<u8>>, into!(parse_to_str)); +/// # } +/// ``` +#[macro_export(local_inner_macros)] +macro_rules! into ( + ($i:expr, $submac:ident!( $($args:tt)* )) => ( + $crate::combinator::intoc($i, |i| $submac!(i, $($args)*)) + ); + ($i:expr, $f:expr) => ( + $crate::combinator::intoc($i, $f) + ); +); + #[cfg(test)] mod tests { - use crate::internal::{Err, IResult, Needed}; - use crate::error::ParseError; use crate::error::ErrorKind; + use crate::error::ParseError; + use crate::internal::{Err, IResult, Needed}; #[cfg(feature = "alloc")] use crate::lib::std::boxed::Box; @@ -989,7 +1014,7 @@ mod tests { Ok(($i.slice(blen..), $i.slice(..blen))) }, CompareResult::Incomplete => { - Err(Err::Incomplete(Needed::Size($tag.input_len()))) + Err(Err::Incomplete(Needed::new($tag.input_len() - $i.input_len()))) }, CompareResult::Error => { let e:ErrorKind = ErrorKind::Tag; @@ -1005,7 +1030,7 @@ mod tests { { let cnt = $count as usize; let res:IResult<&[u8],&[u8]> = if $i.len() < cnt { - Err($crate::Err::Incomplete($crate::Needed::Size(cnt))) + Err($crate::Err::Incomplete($crate::Needed::new(cnt - $i.len()))) } else { Ok((&$i[cnt..],&$i[0..cnt])) }; @@ -1060,12 +1085,12 @@ mod tests { let c = &b"ab"[..]; assert_eq!(opt_abcd(a), Ok((&b"ef"[..], Some(&b"abcd"[..])))); assert_eq!(opt_abcd(b), Ok((&b"bcdefg"[..], None))); - assert_eq!(opt_abcd(c), Err(Err::Incomplete(Needed::Size(4)))); + assert_eq!(opt_abcd(c), Err(Err::Incomplete(Needed::new(2)))); } #[test] fn opt_res() { - named!(opt_res_abcd<&[u8], Result<&[u8], Err<(&[u8], ErrorKind)>> >, opt_res!(tag!("abcd"))); + named!(opt_res_abcd<&[u8], Result<&[u8], Err<crate::error::Error<&[u8]>>> >, opt_res!(tag!("abcd"))); let a = &b"abcdef"[..]; let b = &b"bcdefg"[..]; @@ -1078,7 +1103,7 @@ mod tests { Err(Err::Error(error_position!(b, ErrorKind::Tag))) )) ); - assert_eq!(opt_res_abcd(c), Err(Err::Incomplete(Needed::Size(4)))); + assert_eq!(opt_res_abcd(c), Err(Err::Incomplete(Needed::new(2)))); } use crate::lib::std::convert::From; @@ -1089,6 +1114,11 @@ mod tests { CustomError("test") } } + impl<I> From<crate::error::Error<I>> for CustomError { + fn from(_: crate::error::Error<I>) -> Self { + CustomError("test") + } + } impl<I> ParseError<I> for CustomError { fn from_error_kind(_: I, _: ErrorKind) -> Self { @@ -1100,7 +1130,6 @@ mod tests { } } - #[test] #[cfg(feature = "alloc")] fn cond() { @@ -1113,7 +1142,7 @@ mod tests { } assert_eq!(f_true(&b"abcdef"[..]), Ok((&b"ef"[..], Some(&b"abcd"[..])))); - assert_eq!(f_true(&b"ab"[..]), Err(Err::Incomplete(Needed::Size(4)))); + assert_eq!(f_true(&b"ab"[..]), Err(Err::Incomplete(Needed::new(2)))); assert_eq!(f_true(&b"xxx"[..]), Err(Err::Error(CustomError("test")))); assert_eq!(f_false(&b"abcdef"[..]), Ok((&b"abcdef"[..], None))); @@ -1135,7 +1164,7 @@ mod tests { } assert_eq!(f_true(&b"abcdef"[..]), Ok((&b"ef"[..], Some(&b"abcd"[..])))); - assert_eq!(f_true(&b"ab"[..]), Err(Err::Incomplete(Needed::Size(4)))); + assert_eq!(f_true(&b"ab"[..]), Err(Err::Incomplete(Needed::new(2)))); assert_eq!(f_true(&b"xxx"[..]), Err(Err::Error(CustomError("test")))); assert_eq!(f_false(&b"abcdef"[..]), Ok((&b"abcdef"[..], None))); @@ -1148,7 +1177,7 @@ mod tests { named!(peek_tag<&[u8],&[u8]>, peek!(tag!("abcd"))); assert_eq!(peek_tag(&b"abcdef"[..]), Ok((&b"abcdef"[..], &b"abcd"[..]))); - assert_eq!(peek_tag(&b"ab"[..]), Err(Err::Incomplete(Needed::Size(4)))); + assert_eq!(peek_tag(&b"ab"[..]), Err(Err::Incomplete(Needed::new(2)))); assert_eq!( peek_tag(&b"xxx"[..]), Err(Err::Error(error_position!(&b"xxx"[..], ErrorKind::Tag))) @@ -1162,14 +1191,14 @@ mod tests { not_aaa(&b"aaa"[..]), Err(Err::Error(error_position!(&b"aaa"[..], ErrorKind::Not))) ); - assert_eq!(not_aaa(&b"aa"[..]), Err(Err::Incomplete(Needed::Size(3)))); + assert_eq!(not_aaa(&b"aa"[..]), Err(Err::Incomplete(Needed::new(1)))); assert_eq!(not_aaa(&b"abcd"[..]), Ok((&b"abcd"[..], ()))); } #[test] fn verify() { named!(test, verify!(take!(5), |slice: &[u8]| slice[0] == b'a')); - assert_eq!(test(&b"bcd"[..]), Err(Err::Incomplete(Needed::Size(5)))); + assert_eq!(test(&b"bcd"[..]), Err(Err::Incomplete(Needed::new(2)))); assert_eq!( test(&b"bcdefg"[..]), Err(Err::Error(error_position!( @@ -1186,10 +1215,7 @@ mod tests { assert_eq!( res, - Err(Err::Error(error_position!( - "ab", - ErrorKind::ParseTo - ))) + Err(Err::Error(error_position!("ab", ErrorKind::ParseTo))) ); let res: IResult<_, _, (&str, ErrorKind)> = parse_to!("42", usize); @@ -1197,5 +1223,4 @@ mod tests { assert_eq!(res, Ok(("", 42))); //assert_eq!(ErrorKind::convert(ErrorKind::ParseTo), ErrorKind::ParseTo::<u64>); } - } diff --git a/src/combinator/mod.rs b/src/combinator/mod.rs index 933bed8..4792d0b 100644 --- a/src/combinator/mod.rs +++ b/src/combinator/mod.rs @@ -1,25 +1,25 @@ -//! general purpose combinators +//! General purpose combinators #![allow(unused_imports)] #[cfg(feature = "alloc")] use crate::lib::std::boxed::Box; +use crate::error::{ErrorKind, FromExternalError, ParseError}; +use crate::internal::*; +use crate::lib::std::borrow::Borrow; +use crate::lib::std::convert::Into; #[cfg(feature = "std")] use crate::lib::std::fmt::Debug; -use crate::internal::*; -use crate::error::ParseError; -use crate::traits::{AsChar, InputIter, InputLength, InputTakeAtPosition, ParseTo}; +use crate::lib::std::mem::transmute; use crate::lib::std::ops::{Range, RangeFrom, RangeTo}; -use crate::lib::std::borrow::Borrow; +use crate::traits::{AsChar, InputIter, InputLength, InputTakeAtPosition, ParseTo}; use crate::traits::{Compare, CompareResult, Offset, Slice}; -use crate::error::ErrorKind; -use crate::lib::std::mem::transmute; #[macro_use] mod macros; -/// Return the remaining input +/// Return the remaining input. /// /// ```rust /// # use nom::error::ErrorKind; @@ -30,13 +30,13 @@ mod macros; #[inline] pub fn rest<T, E: ParseError<T>>(input: T) -> IResult<T, T, E> where - T: Slice<Range<usize>> + Slice<RangeFrom<usize>> + Slice<RangeTo<usize>>, + T: Slice<RangeFrom<usize>>, T: InputLength, { Ok((input.slice(input.input_len()..), input)) } -/// Return the length of the remaining input +/// Return the length of the remaining input. /// /// ```rust /// # use nom::error::ErrorKind; @@ -47,52 +47,51 @@ where #[inline] pub fn rest_len<T, E: ParseError<T>>(input: T) -> IResult<T, usize, E> where - T: Slice<Range<usize>> + Slice<RangeFrom<usize>> + Slice<RangeTo<usize>>, T: InputLength, { let len = input.input_len(); Ok((input, len)) } -/// maps a function on the result of a parser +/// Maps a function on the result of a parser. /// /// ```rust /// # #[macro_use] extern crate nom; -/// # use nom::{Err,error::ErrorKind, IResult}; +/// use nom::{Err,error::ErrorKind, IResult,Parser}; /// use nom::character::complete::digit1; /// use nom::combinator::map; /// # fn main() { /// -/// let parse = map(digit1, |s: &str| s.len()); +/// let mut parser = map(digit1, |s: &str| s.len()); /// /// // the parser will count how many characters were returned by digit1 -/// assert_eq!(parse("123456"), Ok(("", 6))); +/// assert_eq!(parser.parse("123456"), Ok(("", 6))); /// /// // this will fail if digit1 fails -/// assert_eq!(parse("abc"), Err(Err::Error(("abc", ErrorKind::Digit)))); +/// assert_eq!(parser.parse("abc"), Err(Err::Error(("abc", ErrorKind::Digit)))); /// # } /// ``` -pub fn map<I, O1, O2, E: ParseError<I>, F, G>(first: F, second: G) -> impl Fn(I) -> IResult<I, O2, E> +pub fn map<I, O1, O2, E, F, G>(mut first: F, mut second: G) -> impl FnMut(I) -> IResult<I, O2, E> where - F: Fn(I) -> IResult<I, O1, E>, - G: Fn(O1) -> O2, + F: Parser<I, O1, E>, + G: FnMut(O1) -> O2, { move |input: I| { - let (input, o1) = first(input)?; + let (input, o1) = first.parse(input)?; Ok((input, second(o1))) } } #[doc(hidden)] -pub fn mapc<I, O1, O2, E: ParseError<I>, F, G>(input: I, first: F, second: G) -> IResult<I, O2, E> +pub fn mapc<I, O1, O2, E, F, G>(input: I, first: F, second: G) -> IResult<I, O2, E> where F: Fn(I) -> IResult<I, O1, E>, G: Fn(O1) -> O2, { - map(first, second)(input) + map(first, second).parse(input) } -/// applies a function returning a Result over the result of a parser +/// Applies a function returning a `Result` over the result of a parser. /// /// ```rust /// # #[macro_use] extern crate nom; @@ -101,7 +100,7 @@ where /// use nom::combinator::map_res; /// # fn main() { /// -/// let parse = map_res(digit1, |s: &str| s.parse::<u8>()); +/// let mut parse = map_res(digit1, |s: &str| s.parse::<u8>()); /// /// // the parser will convert the result of digit1 to a number /// assert_eq!(parse("123"), Ok(("", 123))); @@ -113,23 +112,30 @@ where /// assert_eq!(parse("123456"), Err(Err::Error(("123456", ErrorKind::MapRes)))); /// # } /// ``` -pub fn map_res<I: Clone, O1, O2, E: ParseError<I>, E2, F, G>(first: F, second: G) -> impl Fn(I) -> IResult<I, O2, E> +pub fn map_res<I: Clone, O1, O2, E: FromExternalError<I, E2>, E2, F, G>( + mut first: F, + mut second: G, +) -> impl FnMut(I) -> IResult<I, O2, E> where - F: Fn(I) -> IResult<I, O1, E>, - G: Fn(O1) -> Result<O2, E2>, + F: Parser<I, O1, E>, + G: FnMut(O1) -> Result<O2, E2>, { move |input: I| { let i = input.clone(); - let (input, o1) = first(input)?; + let (input, o1) = first.parse(input)?; match second(o1) { Ok(o2) => Ok((input, o2)), - Err(_) => Err(Err::Error(E::from_error_kind(i, ErrorKind::MapRes))), + Err(e) => Err(Err::Error(E::from_external_error(i, ErrorKind::MapRes, e))), } } } #[doc(hidden)] -pub fn map_resc<I: Clone, O1, O2, E: ParseError<I>, E2, F, G>(input: I, first: F, second: G) -> IResult<I, O2, E> +pub fn map_resc<I: Clone, O1, O2, E: FromExternalError<I, E2>, E2, F, G>( + input: I, + first: F, + second: G, +) -> IResult<I, O2, E> where F: Fn(I) -> IResult<I, O1, E>, G: Fn(O1) -> Result<O2, E2>, @@ -137,7 +143,7 @@ where map_res(first, second)(input) } -/// applies a function returning an Option over the result of a parser +/// Applies a function returning an `Option` over the result of a parser. /// /// ```rust /// # #[macro_use] extern crate nom; @@ -146,7 +152,7 @@ where /// use nom::combinator::map_opt; /// # fn main() { /// -/// let parse = map_opt(digit1, |s: &str| s.parse::<u8>().ok()); +/// let mut parse = map_opt(digit1, |s: &str| s.parse::<u8>().ok()); /// /// // the parser will convert the result of digit1 to a number /// assert_eq!(parse("123"), Ok(("", 123))); @@ -158,14 +164,17 @@ where /// assert_eq!(parse("123456"), Err(Err::Error(("123456", ErrorKind::MapOpt)))); /// # } /// ``` -pub fn map_opt<I: Clone, O1, O2, E: ParseError<I>, F, G>(first: F, second: G) -> impl Fn(I) -> IResult<I, O2, E> +pub fn map_opt<I: Clone, O1, O2, E: ParseError<I>, F, G>( + mut first: F, + mut second: G, +) -> impl FnMut(I) -> IResult<I, O2, E> where - F: Fn(I) -> IResult<I, O1, E>, - G: Fn(O1) -> Option<O2>, + F: Parser<I, O1, E>, + G: FnMut(O1) -> Option<O2>, { move |input: I| { let i = input.clone(); - let (input, o1) = first(input)?; + let (input, o1) = first.parse(input)?; match second(o1) { Some(o2) => Ok((input, o2)), None => Err(Err::Error(E::from_error_kind(i, ErrorKind::MapOpt))), @@ -174,7 +183,11 @@ where } #[doc(hidden)] -pub fn map_optc<I: Clone, O1, O2, E: ParseError<I>, F, G>(input: I, first: F, second: G) -> IResult<I, O2, E> +pub fn map_optc<I: Clone, O1, O2, E: ParseError<I>, F, G>( + input: I, + first: F, + second: G, +) -> IResult<I, O2, E> where F: Fn(I) -> IResult<I, O1, E>, G: Fn(O1) -> Option<O2>, @@ -182,7 +195,7 @@ where map_opt(first, second)(input) } -/// applies a parser over the result of another one +/// Applies a parser over the result of another one. /// /// ```rust /// # #[macro_use] extern crate nom; @@ -192,65 +205,73 @@ where /// use nom::combinator::map_parser; /// # fn main() { /// -/// let parse = map_parser(take(5u8), digit1); +/// let mut parse = map_parser(take(5u8), digit1); /// /// assert_eq!(parse("12345"), Ok(("", "12345"))); /// assert_eq!(parse("123ab"), Ok(("", "123"))); /// assert_eq!(parse("123"), Err(Err::Error(("123", ErrorKind::Eof)))); /// # } /// ``` -pub fn map_parser<I: Clone, O1, O2, E: ParseError<I>, F, G>(first: F, second: G) -> impl Fn(I) -> IResult<I, O2, E> +pub fn map_parser<I, O1, O2, E: ParseError<I>, F, G>( + mut first: F, + mut second: G, +) -> impl FnMut(I) -> IResult<I, O2, E> where - F: Fn(I) -> IResult<I, O1, E>, - G: Fn(O1) -> IResult<O1, O2, E>, - O1: InputLength, + F: Parser<I, O1, E>, + G: Parser<O1, O2, E>, { move |input: I| { - let (input, o1) = first(input)?; - let (_, o2) = second(o1)?; + let (input, o1) = first.parse(input)?; + let (_, o2) = second.parse(o1)?; Ok((input, o2)) } } #[doc(hidden)] -pub fn map_parserc<I: Clone, O1, O2, E: ParseError<I>, F, G>(input: I, first: F, second: G) -> IResult<I, O2, E> +pub fn map_parserc<I, O1, O2, E: ParseError<I>, F, G>( + input: I, + first: F, + second: G, +) -> IResult<I, O2, E> where F: Fn(I) -> IResult<I, O1, E>, G: Fn(O1) -> IResult<O1, O2, E>, - O1: InputLength, { map_parser(first, second)(input) } -/// creates a new parser from the output of the first parser, then apply that parser over the rest of the input +/// Creates a new parser from the output of the first parser, then apply that parser over the rest of the input. /// /// ```rust /// # #[macro_use] extern crate nom; /// # use nom::{Err,error::ErrorKind, IResult}; /// use nom::bytes::complete::take; -/// use nom::number::complete::be_u8; +/// use nom::number::complete::u8; /// use nom::combinator::flat_map; /// # fn main() { /// -/// let parse = flat_map(be_u8, take); +/// let mut parse = flat_map(u8, take); /// /// assert_eq!(parse(&[2, 0, 1, 2][..]), Ok((&[2][..], &[0, 1][..]))); /// assert_eq!(parse(&[4, 0, 1, 2][..]), Err(Err::Error((&[0, 1, 2][..], ErrorKind::Eof)))); /// # } /// ``` -pub fn flat_map<I, O1, O2, E: ParseError<I>, F, G, H>(first: F, second: G) -> impl Fn(I) -> IResult<I, O2, E> +pub fn flat_map<I, O1, O2, E: ParseError<I>, F, G, H>( + mut first: F, + second: G, +) -> impl FnMut(I) -> IResult<I, O2, E> where - F: Fn(I) -> IResult<I, O1, E>, + F: Parser<I, O1, E>, G: Fn(O1) -> H, - H: Fn(I) -> IResult<I, O2, E> + H: Parser<I, O2, E>, { move |input: I| { - let (input, o1) = first(input)?; - second(o1)(input) + let (input, o1) = first.parse(input)?; + second(o1).parse(input) } } -/// optional parser: will return None if not successful +/// Optional parser: Will return `None` if not successful. /// /// ```rust /// # #[macro_use] extern crate nom; @@ -267,13 +288,13 @@ where /// assert_eq!(parser("123;"), Ok(("123;", None))); /// # } /// ``` -pub fn opt<I:Clone, O, E: ParseError<I>, F>(f: F) -> impl Fn(I) -> IResult<I, Option<O>, E> +pub fn opt<I: Clone, O, E: ParseError<I>, F>(mut f: F) -> impl FnMut(I) -> IResult<I, Option<O>, E> where - F: Fn(I) -> IResult<I, O, E>, + F: Parser<I, O, E>, { move |input: I| { let i = input.clone(); - match f(input) { + match f.parse(input) { Ok((i, o)) => Ok((i, Some(o))), Err(Err::Error(_)) => Ok((i, None)), Err(e) => Err(e), @@ -282,18 +303,18 @@ where } #[doc(hidden)] -pub fn optc<I:Clone, O, E: ParseError<I>, F>(input: I, f: F) -> IResult<I, Option<O>, E> +pub fn optc<I: Clone, O, E: ParseError<I>, F>(input: I, f: F) -> IResult<I, Option<O>, E> where F: Fn(I) -> IResult<I, O, E>, { opt(f)(input) } -/// calls the parser if the condition is met +/// Calls the parser if the condition is met. /// /// ```rust /// # #[macro_use] extern crate nom; -/// # use nom::{Err,error::ErrorKind, IResult}; +/// # use nom::{Err, error::{Error, ErrorKind}, IResult}; /// use nom::combinator::cond; /// use nom::character::complete::alpha1; /// # fn main() { @@ -304,17 +325,20 @@ where /// /// assert_eq!(parser(true, "abcd;"), Ok((";", Some("abcd")))); /// assert_eq!(parser(false, "abcd;"), Ok(("abcd;", None))); -/// assert_eq!(parser(true, "123;"), Err(Err::Error(("123;", ErrorKind::Alpha)))); +/// assert_eq!(parser(true, "123;"), Err(Err::Error(Error::new("123;", ErrorKind::Alpha)))); /// assert_eq!(parser(false, "123;"), Ok(("123;", None))); /// # } /// ``` -pub fn cond<I:Clone, O, E: ParseError<I>, F>(b: bool, f: F) -> impl Fn(I) -> IResult<I, Option<O>, E> +pub fn cond<I, O, E: ParseError<I>, F>( + b: bool, + mut f: F, +) -> impl FnMut(I) -> IResult<I, Option<O>, E> where - F: Fn(I) -> IResult<I, O, E>, + F: Parser<I, O, E>, { move |input: I| { if b { - match f(input) { + match f.parse(input) { Ok((i, o)) => Ok((i, Some(o))), Err(e) => Err(e), } @@ -325,14 +349,14 @@ where } #[doc(hidden)] -pub fn condc<I:Clone, O, E: ParseError<I>, F>(input: I, b: bool, f: F) -> IResult<I, Option<O>, E> +pub fn condc<I, O, E: ParseError<I>, F>(input: I, b: bool, f: F) -> IResult<I, Option<O>, E> where F: Fn(I) -> IResult<I, O, E>, { cond(b, f)(input) } -/// tries to apply its parser without consuming the input +/// Tries to apply its parser without consuming the input. /// /// ```rust /// # #[macro_use] extern crate nom; @@ -341,19 +365,19 @@ where /// use nom::character::complete::alpha1; /// # fn main() { /// -/// let parser = peek(alpha1); +/// let mut parser = peek(alpha1); /// /// assert_eq!(parser("abcd;"), Ok(("abcd;", "abcd"))); /// assert_eq!(parser("123;"), Err(Err::Error(("123;", ErrorKind::Alpha)))); /// # } /// ``` -pub fn peek<I:Clone, O, E: ParseError<I>, F>(f: F) -> impl Fn(I) -> IResult<I, O, E> +pub fn peek<I: Clone, O, E: ParseError<I>, F>(mut f: F) -> impl FnMut(I) -> IResult<I, O, E> where - F: Fn(I) -> IResult<I, O, E>, + F: Parser<I, O, E>, { move |input: I| { let i = input.clone(); - match f(input) { + match f.parse(input) { Ok((_, o)) => Ok((i, o)), Err(e) => Err(e), } @@ -361,14 +385,40 @@ where } #[doc(hidden)] -pub fn peekc<I:Clone, O, E: ParseError<I>, F>(input: I, f: F) -> IResult<I, O, E> +pub fn peekc<I: Clone, O, E: ParseError<I>, F>(input: I, f: F) -> IResult<I, O, E> where F: Fn(I) -> IResult<I, O, E>, { peek(f)(input) } -/// transforms Incomplete into Error +/// returns its input if it is at the end of input data +/// +/// When we're at the end of the data, this combinator +/// will succeed +/// +/// ``` +/// # #[macro_use] extern crate nom; +/// # use std::str; +/// # use nom::{Err, error::ErrorKind, IResult}; +/// # use nom::combinator::eof; +/// +/// # fn main() { +/// let parser = eof; +/// assert_eq!(parser("abc"), Err(Err::Error(("abc", ErrorKind::Eof)))); +/// assert_eq!(parser(""), Ok(("", ""))); +/// # } +/// ``` +pub fn eof<I: InputLength + Clone, E: ParseError<I>>(input: I) -> IResult<I, I, E> { + if input.input_len() == 0 { + let clone = input.clone(); + Ok((input, clone)) + } else { + Err(Err::Error(E::from_error_kind(input, ErrorKind::Eof))) + } +} + +/// Transforms Incomplete into `Error`. /// /// ```rust /// # #[macro_use] extern crate nom; @@ -377,23 +427,21 @@ where /// use nom::combinator::complete; /// # fn main() { /// -/// let parser = complete(take(5u8)); +/// let mut parser = complete(take(5u8)); /// /// assert_eq!(parser("abcdefg"), Ok(("fg", "abcde"))); /// assert_eq!(parser("abcd"), Err(Err::Error(("abcd", ErrorKind::Complete)))); /// # } /// ``` -pub fn complete<I: Clone, O, E: ParseError<I>, F>(f: F) -> impl Fn(I) -> IResult<I, O, E> +pub fn complete<I: Clone, O, E: ParseError<I>, F>(mut f: F) -> impl FnMut(I) -> IResult<I, O, E> where - F: Fn(I) -> IResult<I, O, E>, + F: Parser<I, O, E>, { move |input: I| { let i = input.clone(); - match f(input) { - Err(Err::Incomplete(_)) => { - Err(Err::Error(E::from_error_kind(i, ErrorKind::Complete))) - }, - rest => rest + match f.parse(input) { + Err(Err::Incomplete(_)) => Err(Err::Error(E::from_error_kind(i, ErrorKind::Complete))), + rest => rest, } } } @@ -403,10 +451,10 @@ pub fn completec<I: Clone, O, E: ParseError<I>, F>(input: I, f: F) -> IResult<I, where F: Fn(I) -> IResult<I, O, E>, { - complete(f)(input) + complete(f)(input) } -/// succeeds if all the input has been consumed by its child parser +/// Succeeds if all the input has been consumed by its child parser. /// /// ```rust /// # #[macro_use] extern crate nom; @@ -415,20 +463,20 @@ where /// use nom::character::complete::alpha1; /// # fn main() { /// -/// let parser = all_consuming(alpha1); +/// let mut parser = all_consuming(alpha1); /// /// assert_eq!(parser("abcd"), Ok(("", "abcd"))); /// assert_eq!(parser("abcd;"),Err(Err::Error((";", ErrorKind::Eof)))); /// assert_eq!(parser("123abcd;"),Err(Err::Error(("123abcd;", ErrorKind::Alpha)))); /// # } /// ``` -pub fn all_consuming<I, O, E: ParseError<I>, F>(f: F) -> impl Fn(I) -> IResult<I, O, E> +pub fn all_consuming<I, O, E: ParseError<I>, F>(mut f: F) -> impl FnMut(I) -> IResult<I, O, E> where I: InputLength, - F: Fn(I) -> IResult<I, O, E>, + F: Parser<I, O, E>, { move |input: I| { - let (input, res) = f(input)?; + let (input, res) = f.parse(input)?; if input.input_len() == 0 { Ok((input, res)) } else { @@ -437,10 +485,10 @@ where } } -/// returns the result of the child parser if it satisfies a verification function +/// Returns the result of the child parser if it satisfies a verification function. /// -/// the verification function takes as argument a reference to the output of the -/// parser +/// The verification function takes as argument a reference to the output of the +/// parser. /// /// ```rust /// # #[macro_use] extern crate nom; @@ -449,23 +497,26 @@ where /// use nom::character::complete::alpha1; /// # fn main() { /// -/// let parser = verify(alpha1, |s: &str| s.len() == 4); +/// let mut parser = verify(alpha1, |s: &str| s.len() == 4); /// /// assert_eq!(parser("abcd"), Ok(("", "abcd"))); /// assert_eq!(parser("abcde"), Err(Err::Error(("abcde", ErrorKind::Verify)))); /// assert_eq!(parser("123abcd;"),Err(Err::Error(("123abcd;", ErrorKind::Alpha)))); /// # } /// ``` -pub fn verify<I: Clone, O1, O2, E: ParseError<I>, F, G>(first: F, second: G) -> impl Fn(I) -> IResult<I, O1, E> +pub fn verify<I: Clone, O1, O2, E: ParseError<I>, F, G>( + mut first: F, + second: G, +) -> impl FnMut(I) -> IResult<I, O1, E> where - F: Fn(I) -> IResult<I, O1, E>, + F: Parser<I, O1, E>, G: Fn(&O2) -> bool, O1: Borrow<O2>, O2: ?Sized, { move |input: I| { let i = input.clone(); - let (input, o) = first(input)?; + let (input, o) = first.parse(input)?; if second(o.borrow()) { Ok((input, o)) @@ -476,7 +527,11 @@ where } #[doc(hidden)] -pub fn verifyc<I: Clone, O1, O2, E: ParseError<I>, F, G>(input: I, first: F, second: G) -> IResult<I, O1, E> +pub fn verifyc<I: Clone, O1, O2, E: ParseError<I>, F, G>( + input: I, + first: F, + second: G, +) -> IResult<I, O1, E> where F: Fn(I) -> IResult<I, O1, E>, G: Fn(&O2) -> bool, @@ -486,7 +541,7 @@ where verify(first, second)(input) } -/// returns the provided value if the child parser succeeds +/// Returns the provided value if the child parser succeeds. /// /// ```rust /// # #[macro_use] extern crate nom; @@ -495,30 +550,35 @@ where /// use nom::character::complete::alpha1; /// # fn main() { /// -/// let parser = value(1234, alpha1); +/// let mut parser = value(1234, alpha1); /// /// assert_eq!(parser("abcd"), Ok(("", 1234))); /// assert_eq!(parser("123abcd;"), Err(Err::Error(("123abcd;", ErrorKind::Alpha)))); /// # } /// ``` -pub fn value<I, O1: Clone, O2, E: ParseError<I>, F>(val: O1, parser: F) -> impl Fn(I) -> IResult<I, O1, E> +pub fn value<I, O1: Clone, O2, E: ParseError<I>, F>( + val: O1, + mut parser: F, +) -> impl FnMut(I) -> IResult<I, O1, E> where - F: Fn(I) -> IResult<I, O2, E>, + F: Parser<I, O2, E>, { - move |input: I| { - parser(input).map(|(i, _)| (i, val.clone())) - } + move |input: I| parser.parse(input).map(|(i, _)| (i, val.clone())) } #[doc(hidden)] -pub fn valuec<I, O1: Clone, O2, E: ParseError<I>, F>(input: I, val: O1, parser: F) -> IResult<I, O1, E> +pub fn valuec<I, O1: Clone, O2, E: ParseError<I>, F>( + input: I, + val: O1, + parser: F, +) -> IResult<I, O1, E> where F: Fn(I) -> IResult<I, O2, E>, { value(val, parser)(input) } -/// succeeds if the child parser returns an error +/// Succeeds if the child parser returns an error. /// /// ```rust /// # #[macro_use] extern crate nom; @@ -527,19 +587,19 @@ where /// use nom::character::complete::alpha1; /// # fn main() { /// -/// let parser = not(alpha1); +/// let mut parser = not(alpha1); /// /// assert_eq!(parser("123"), Ok(("123", ()))); /// assert_eq!(parser("abcd"), Err(Err::Error(("abcd", ErrorKind::Not)))); /// # } /// ``` -pub fn not<I: Clone, O, E: ParseError<I>, F>(parser: F) -> impl Fn(I) -> IResult<I, (), E> +pub fn not<I: Clone, O, E: ParseError<I>, F>(mut parser: F) -> impl FnMut(I) -> IResult<I, (), E> where - F: Fn(I) -> IResult<I, O, E>, + F: Parser<I, O, E>, { move |input: I| { let i = input.clone(); - match parser(input) { + match parser.parse(input) { Ok(_) => Err(Err::Error(E::from_error_kind(i, ErrorKind::Not))), Err(Err::Error(_)) => Ok((i, ())), Err(e) => Err(e), @@ -555,7 +615,7 @@ where not(parser)(input) } -/// if the child parser was successful, return the consumed input as produced value +/// If the child parser was successful, return the consumed input as produced value. /// /// ```rust /// # #[macro_use] extern crate nom; @@ -565,36 +625,108 @@ where /// use nom::sequence::separated_pair; /// # fn main() { /// -/// let parser = recognize(separated_pair(alpha1, char(','), alpha1)); +/// let mut parser = recognize(separated_pair(alpha1, char(','), alpha1)); /// /// assert_eq!(parser("abcd,efgh"), Ok(("", "abcd,efgh"))); /// assert_eq!(parser("abcd;"),Err(Err::Error((";", ErrorKind::Char)))); /// # } /// ``` -pub fn recognize<I: Clone + Offset + Slice<RangeTo<usize>>, O, E: ParseError<I>, F>(parser: F) -> impl Fn(I) -> IResult<I, I, E> +pub fn recognize<I: Clone + Offset + Slice<RangeTo<usize>>, O, E: ParseError<I>, F>( + mut parser: F, +) -> impl FnMut(I) -> IResult<I, I, E> where - F: Fn(I) -> IResult<I, O, E>, + F: Parser<I, O, E>, { move |input: I| { let i = input.clone(); - match parser(i) { + match parser.parse(i) { Ok((i, _)) => { let index = input.offset(&i); Ok((i, input.slice(..index))) - }, + } Err(e) => Err(e), } } } #[doc(hidden)] -pub fn recognizec<I: Clone + Offset + Slice<RangeTo<usize>>, O, E: ParseError<I>, F>(input: I, parser: F) -> IResult<I, I, E> +pub fn recognizec<I: Clone + Offset + Slice<RangeTo<usize>>, O, E: ParseError<I>, F>( + input: I, + parser: F, +) -> IResult<I, I, E> where F: Fn(I) -> IResult<I, O, E>, { recognize(parser)(input) } +/// if the child parser was successful, return the consumed input with the output +/// as a tuple. Functions similarly to [recognize](fn.recognize.html) except it +/// returns the parser output as well. +/// +/// This can be useful especially in cases where the output is not the same type +/// as the input, or the input is a user defined type. +/// +/// Returned tuple is of the format `(consumed input, produced output)`. +/// +/// ```rust +/// # #[macro_use] extern crate nom; +/// # use nom::{Err,error::ErrorKind, IResult}; +/// use nom::combinator::{consumed, value, recognize, map}; +/// use nom::character::complete::{char, alpha1}; +/// use nom::bytes::complete::tag; +/// use nom::sequence::separated_pair; +/// +/// fn inner_parser(input: &str) -> IResult<&str, bool> { +/// value(true, tag("1234"))(input) +/// } +/// +/// # fn main() { +/// +/// let mut consumed_parser = consumed(value(true, separated_pair(alpha1, char(','), alpha1))); +/// +/// assert_eq!(consumed_parser("abcd,efgh1"), Ok(("1", ("abcd,efgh", true)))); +/// assert_eq!(consumed_parser("abcd;"),Err(Err::Error((";", ErrorKind::Char)))); +/// +/// +/// // the first output (representing the consumed input) +/// // should be the same as that of the `recognize` parser. +/// let mut recognize_parser = recognize(inner_parser); +/// let mut consumed_parser = map(consumed(inner_parser), |(consumed, output)| consumed); +/// +/// assert_eq!(recognize_parser("1234"), consumed_parser("1234")); +/// assert_eq!(recognize_parser("abcd"), consumed_parser("abcd")); +/// # } +/// ``` +pub fn consumed<I, O, F, E>(mut parser: F) -> impl FnMut(I) -> IResult<I, (I, O), E> +where + I: Clone + Offset + Slice<RangeTo<usize>>, + E: ParseError<I>, + F: Parser<I, O, E>, +{ + move |input: I| { + let i = input.clone(); + match parser.parse(i) { + Ok((remaining, result)) => { + let index = input.offset(&remaining); + let consumed = input.slice(..index); + Ok((remaining, (consumed, result))) + } + Err(e) => Err(e), + } + } +} + +#[doc(hidden)] +pub fn consumedc<I, O, E: ParseError<I>, F>(input: I, parser: F) -> IResult<I, (I, O), E> +where + I: Clone + Offset + Slice<RangeTo<usize>>, + E: ParseError<E>, + F: Fn(I) -> IResult<I, O, E>, +{ + consumed(parser)(input) +} + /// transforms an error to failure /// /// ```rust @@ -604,37 +736,86 @@ where /// use nom::character::complete::alpha1; /// # fn main() { /// -/// let parser = cut(alpha1); +/// let mut parser = cut(alpha1); /// /// assert_eq!(parser("abcd;"), Ok((";", "abcd"))); /// assert_eq!(parser("123;"), Err(Err::Failure(("123;", ErrorKind::Alpha)))); /// # } /// ``` -pub fn cut<I: Clone + Slice<RangeTo<usize>>, O, E: ParseError<I>, F>(parser: F) -> impl Fn(I) -> IResult<I, O, E> +pub fn cut<I, O, E: ParseError<I>, F>(mut parser: F) -> impl FnMut(I) -> IResult<I, O, E> where - F: Fn(I) -> IResult<I, O, E>, + F: Parser<I, O, E>, { - move |input: I| { - let i = input.clone(); - match parser(i) { - Err(Err::Error(e)) => Err(Err::Failure(e)), - rest => rest, - } + move |input: I| match parser.parse(input) { + Err(Err::Error(e)) => Err(Err::Failure(e)), + rest => rest, } } #[doc(hidden)] -pub fn cutc<I: Clone + Slice<RangeTo<usize>>, O, E: ParseError<I>, F>(input: I, parser: F) -> IResult<I, O, E> +pub fn cutc<I, O, E: ParseError<I>, F>(input: I, parser: F) -> IResult<I, O, E> where F: Fn(I) -> IResult<I, O, E>, { cut(parser)(input) } -/// creates an iterator from input data and a parser +/// automatically converts the child parser's result to another type /// -/// call the iterator's [finish] method to get the remaining input if successful, -/// or the error value if we encountered an error +/// it will be able to convert the output value and the error value +/// as long as the `Into` implementations are available +/// +/// ```rust +/// # #[macro_use] extern crate nom; +/// # use nom::IResult; +/// use nom::combinator::into; +/// use nom::character::complete::alpha1; +/// # fn main() { +/// +/// fn parser1(i: &str) -> IResult<&str, &str> { +/// alpha1(i) +/// } +/// +/// let mut parser2 = into(parser1); +/// +/// // the parser converts the &str output of the child parser into a Vec<u8> +/// let bytes: IResult<&str, Vec<u8>> = parser2("abcd"); +/// assert_eq!(bytes, Ok(("", vec![97, 98, 99, 100]))); +/// # } +/// ``` +pub fn into<I, O1, O2, E1, E2, F>(mut parser: F) -> impl FnMut(I) -> IResult<I, O2, E2> +where + O1: Into<O2>, + E1: Into<E2>, + E1: ParseError<I>, + E2: ParseError<I>, + F: Parser<I, O1, E1>, +{ + //map(parser, Into::into) + move |input: I| match parser.parse(input) { + Ok((i, o)) => Ok((i, o.into())), + Err(Err::Error(e)) => Err(Err::Error(e.into())), + Err(Err::Failure(e)) => Err(Err::Failure(e.into())), + Err(Err::Incomplete(e)) => Err(Err::Incomplete(e)), + } +} + +#[doc(hidden)] +pub fn intoc<I, O1, O2, E1, E2, F>(input: I, parser: F) -> IResult<I, O2, E2> +where + O1: Into<O2>, + E1: Into<E2>, + E1: ParseError<I>, + E2: ParseError<I>, + F: Parser<I, O1, E1>, +{ + into(parser)(input) +} + +/// Creates an iterator from input data and a parser. +/// +/// Call the iterator's [ParserIterator::finish] method to get the remaining input if successful, +/// or the error value if we encountered an error. /// /// ```rust /// use nom::{combinator::iterator, IResult, bytes::complete::tag, character::complete::alpha1, sequence::terminated}; @@ -651,62 +832,63 @@ where /// ``` pub fn iterator<Input, Output, Error, F>(input: Input, f: F) -> ParserIterator<Input, Error, F> where - F: Fn(Input) -> IResult<Input, Output, Error>, - Error: ParseError<Input> { - - ParserIterator { - iterator: f, - input, - state: State::Running, - } + F: Parser<Input, Output, Error>, + Error: ParseError<Input>, +{ + ParserIterator { + iterator: f, + input, + state: Some(State::Running), + } } -/// main structure associated to the [iterator] function +/// Main structure associated to the [iterator] function. pub struct ParserIterator<I, E, F> { iterator: F, input: I, - state: State<E>, + state: Option<State<E>>, } -impl<I: Clone, E: Clone, F> ParserIterator<I, E, F> { - /// returns the remaining input if parsing was successful, or the error if we encountered an error - pub fn finish(self) -> IResult<I, (), E> { - match &self.state { - State::Running | State::Done => Ok((self.input.clone(), ())), - State::Failure(e) => Err(Err::Failure(e.clone())), - State::Incomplete(i) => Err(Err::Incomplete(i.clone())), +impl<I: Clone, E, F> ParserIterator<I, E, F> { + /// Returns the remaining input if parsing was successful, or the error if we encountered an error. + pub fn finish(mut self) -> IResult<I, (), E> { + match self.state.take().unwrap() { + State::Running | State::Done => Ok((self.input, ())), + State::Failure(e) => Err(Err::Failure(e)), + State::Incomplete(i) => Err(Err::Incomplete(i)), } } } -impl<'a, Input ,Output ,Error, F> core::iter::Iterator for &'a mut ParserIterator<Input, Error, F> - where - F: Fn(Input) -> IResult<Input, Output, Error>, - Input: Clone +impl<'a, Input, Output, Error, F> core::iter::Iterator for &'a mut ParserIterator<Input, Error, F> +where + F: FnMut(Input) -> IResult<Input, Output, Error>, + Input: Clone, { type Item = Output; fn next(&mut self) -> Option<Self::Item> { - if let State::Running = self.state { + if let State::Running = self.state.take().unwrap() { let input = self.input.clone(); match (self.iterator)(input) { Ok((i, o)) => { self.input = i; + self.state = Some(State::Running); Some(o) - }, + } Err(Err::Error(_)) => { - self.state = State::Done; + self.state = Some(State::Done); None - }, + } Err(Err::Failure(e)) => { - self.state = State::Failure(e); + self.state = Some(State::Failure(e)); None - }, + } Err(Err::Incomplete(i)) => { - self.state = State::Incomplete(i); + self.state = Some(State::Incomplete(i)); None - }, + } } } else { None @@ -721,14 +903,39 @@ enum State<E> { Incomplete(Needed), } +/// a parser which always succeeds with given value without consuming any input. +/// +/// It can be used for example as the last alternative in `alt` to +/// specify the default case. +/// +/// ```rust +/// # #[macro_use] extern crate nom; +/// # use nom::{Err,error::ErrorKind, IResult}; +/// use nom::branch::alt; +/// use nom::combinator::{success, value}; +/// use nom::character::complete::char; +/// # fn main() { +/// +/// let mut parser = success::<_,_,(_,ErrorKind)>(10); +/// assert_eq!(parser("xyz"), Ok(("xyz", 10))); +/// +/// let mut sign = alt((value(-1, char('-')), value(1, char('+')), success::<_,_,(_,ErrorKind)>(1))); +/// assert_eq!(sign("+10"), Ok(("10", 1))); +/// assert_eq!(sign("-10"), Ok(("10", -1))); +/// assert_eq!(sign("10"), Ok(("10", 1))); +/// # } +/// ``` +pub fn success<I, O: Clone, E: ParseError<I>>(val: O) -> impl Fn(I) -> IResult<I, O, E> { + move |input: I| Ok((input, val.clone())) +} #[cfg(test)] mod tests { use super::*; - use crate::internal::{Err, IResult, Needed}; - use crate::error::ParseError; use crate::bytes::complete::take; - use crate::number::complete::be_u8; + use crate::error::ParseError; + use crate::internal::{Err, IResult, Needed}; + use crate::number::complete::u8; macro_rules! assert_parse( ($left: expr, $right: expr) => { @@ -746,21 +953,50 @@ mod tests { assert_eq!(res, Ok((&v2[..], ()))); }*/ + #[test] + fn eof_on_slices() { + let not_over: &[u8] = &b"Hello, world!"[..]; + let is_over: &[u8] = &b""[..]; + + let res_not_over = eof(not_over); + assert_parse!( + res_not_over, + Err(Err::Error(error_position!(not_over, ErrorKind::Eof))) + ); + + let res_over = eof(is_over); + assert_parse!(res_over, Ok((is_over, is_over))); + } + + #[test] + fn eof_on_strs() { + let not_over: &str = "Hello, world!"; + let is_over: &str = ""; + + let res_not_over = eof(not_over); + assert_parse!( + res_not_over, + Err(Err::Error(error_position!(not_over, ErrorKind::Eof))) + ); + + let res_over = eof(is_over); + assert_parse!(res_over, Ok((is_over, is_over))); + } /* - #[test] - fn end_of_input() { - let not_over = &b"Hello, world!"[..]; - let is_over = &b""[..]; - named!(eof_test, eof!()); + #[test] + fn end_of_input() { + let not_over = &b"Hello, world!"[..]; + let is_over = &b""[..]; + named!(eof_test, eof!()); - let res_not_over = eof_test(not_over); - assert_eq!(res_not_over, Err(Err::Error(error_position!(not_over, ErrorKind::Eof)))); + let res_not_over = eof_test(not_over); + assert_eq!(res_not_over, Err(Err::Error(error_position!(not_over, ErrorKind::Eof)))); - let res_over = eof_test(is_over); - assert_eq!(res_over, Ok((is_over, is_over))); - } - */ + let res_over = eof_test(is_over); + assert_eq!(res_over, Ok((is_over, is_over))); + } + */ #[test] fn rest_on_slices() { @@ -808,28 +1044,46 @@ mod tests { #[test] fn test_flat_map() { - let input: &[u8] = &[3, 100, 101, 102, 103, 104][..]; - assert_parse!(flat_map(be_u8, take)(input), Ok((&[103, 104][..], &[100, 101, 102][..]))); + let input: &[u8] = &[3, 100, 101, 102, 103, 104][..]; + assert_parse!( + flat_map(u8, take)(input), + Ok((&[103, 104][..], &[100, 101, 102][..])) + ); } #[test] fn test_map_opt() { - let input: &[u8] = &[50][..]; - assert_parse!(map_opt(be_u8, |u| if u < 20 {Some(u)} else {None})(input), Err(Err::Error((&[50][..], ErrorKind::MapOpt)))); - assert_parse!(map_opt(be_u8, |u| if u > 20 {Some(u)} else {None})(input), Ok((&[][..], 50))); + let input: &[u8] = &[50][..]; + assert_parse!( + map_opt(u8, |u| if u < 20 { Some(u) } else { None })(input), + Err(Err::Error((&[50][..], ErrorKind::MapOpt))) + ); + assert_parse!( + map_opt(u8, |u| if u > 20 { Some(u) } else { None })(input), + Ok((&[][..], 50)) + ); } #[test] fn test_map_parser() { - let input: &[u8] = &[100, 101, 102, 103, 104][..]; - assert_parse!(map_parser(take(4usize), take(2usize))(input), Ok((&[104][..], &[100, 101][..]))); + let input: &[u8] = &[100, 101, 102, 103, 104][..]; + assert_parse!( + map_parser(take(4usize), take(2usize))(input), + Ok((&[104][..], &[100, 101][..])) + ); } #[test] fn test_all_consuming() { - let input: &[u8] = &[100, 101, 102][..]; - assert_parse!(all_consuming(take(2usize))(input), Err(Err::Error((&[102][..], ErrorKind::Eof)))); - assert_parse!(all_consuming(take(3usize))(input), Ok((&[][..], &[100, 101, 102][..]))); + let input: &[u8] = &[100, 101, 102][..]; + assert_parse!( + all_consuming(take(2usize))(input), + Err(Err::Error((&[102][..], ErrorKind::Eof))) + ); + assert_parse!( + all_consuming(take(3usize))(input), + Ok((&[][..], &[100, 101, 102][..])) + ); } #[test] @@ -837,10 +1091,13 @@ mod tests { fn test_verify_ref() { use crate::bytes::complete::take; - let parser1 = verify(take(3u8), |s: &[u8]| s == &b"abc"[..]); + let mut parser1 = verify(take(3u8), |s: &[u8]| s == &b"abc"[..]); assert_eq!(parser1(&b"abcd"[..]), Ok((&b"d"[..], &b"abc"[..]))); - assert_eq!(parser1(&b"defg"[..]), Err(Err::Error((&b"defg"[..], ErrorKind::Verify)))); + assert_eq!( + parser1(&b"defg"[..]), + Err(Err::Error((&b"defg"[..], ErrorKind::Verify))) + ); fn parser2(i: &[u8]) -> IResult<&[u8], u32> { verify(crate::number::streaming::be_u32, |val: &u32| *val < 3)(i) @@ -851,9 +1108,29 @@ mod tests { #[cfg(feature = "alloc")] fn test_verify_alloc() { use crate::bytes::complete::take; - let parser1 = verify(map(take(3u8), |s: &[u8]| s.to_vec()), |s: &[u8]| s == &b"abc"[..]); + let mut parser1 = verify(map(take(3u8), |s: &[u8]| s.to_vec()), |s: &[u8]| { + s == &b"abc"[..] + }); assert_eq!(parser1(&b"abcd"[..]), Ok((&b"d"[..], (&b"abc").to_vec()))); - assert_eq!(parser1(&b"defg"[..]), Err(Err::Error((&b"defg"[..], ErrorKind::Verify)))); + assert_eq!( + parser1(&b"defg"[..]), + Err(Err::Error((&b"defg"[..], ErrorKind::Verify))) + ); + } + + #[test] + #[cfg(feature = "std")] + fn test_into() { + use crate::bytes::complete::take; + use crate::{ + error::{Error, ParseError}, + Err, + }; + + let mut parser = into(take::<_, _, Error<_>>(3u8)); + let result: IResult<&[u8], Vec<u8>> = parser(&b"abcdefg"[..]); + + assert_eq!(result, Ok((&b"defg"[..], vec![97, 98, 99]))); } } diff --git a/src/error.rs b/src/error.rs index 1505161..a5aa678 100644 --- a/src/error.rs +++ b/src/error.rs @@ -3,34 +3,41 @@ //! Parsers are generic over their error type, requiring that it implements //! the `error::ParseError<Input>` trait. -/// this trait must be implemented by the error type of a nom parser +use crate::internal::Parser; +use crate::lib::std::fmt; + +/// This trait must be implemented by the error type of a nom parser. /// /// There are already implementations of it for `(Input, ErrorKind)` /// and `VerboseError<Input>`. /// /// It provides methods to create an error from some combinators, -/// and combine existing errors in combinators like `alt` +/// and combine existing errors in combinators like `alt`. pub trait ParseError<I>: Sized { - /// creates an error from the input position and an [ErrorKind] + /// Creates an error from the input position and an [ErrorKind] fn from_error_kind(input: I, kind: ErrorKind) -> Self; - /// combines an existing error with a new one created from the input + /// Combines an existing error with a new one created from the input /// position and an [ErrorKind]. This is useful when backtracking /// through a parse tree, accumulating error context on the way fn append(input: I, kind: ErrorKind, other: Self) -> Self; - /// creates an error from an input position and an expected character + /// Creates an error from an input position and an expected character fn from_char(input: I, _: char) -> Self { Self::from_error_kind(input, ErrorKind::Char) } - /// combines two existing error. This function is used to compare errors + /// Combines two existing errors. This function is used to compare errors /// generated in various branches of [alt] fn or(self, other: Self) -> Self { other } +} - /// create a new error from an input position, a static string and an existing error. +/// This trait is required by the `context` combinator to add a static string +/// to an existing error +pub trait ContextError<I>: Sized { + /// Creates a new error from an input position, a static string and an existing error. /// This is used mainly in the [context] combinator, to add user friendly information /// to errors when backtracking through a parse tree fn add_context(_input: I, _ctx: &'static str, other: Self) -> Self { @@ -38,6 +45,61 @@ pub trait ParseError<I>: Sized { } } +/// This trait is required by the [map_res] combinator to integrate +/// error types from external functions, like [std::str::FromStr] +pub trait FromExternalError<I, E> { + /// Creates a new error from an input position, an [ErrorKind] indicating the + /// wrapping parser, and an external error + fn from_external_error(input: I, kind: ErrorKind, e: E) -> Self; +} + +/// default error type, only contains the error' location and code +#[derive(Debug, PartialEq)] +pub struct Error<I> { + /// position of the error in the input data + pub input: I, + /// nom error code + pub code: ErrorKind, +} + +impl<I> Error<I> { + /// creates a new basic error + pub fn new(input: I, code: ErrorKind) -> Error<I> { + Error { input, code } + } +} + +impl<I> ParseError<I> for Error<I> { + fn from_error_kind(input: I, kind: ErrorKind) -> Self { + Error { input, code: kind } + } + + fn append(_: I, _: ErrorKind, other: Self) -> Self { + other + } +} + +impl<I> ContextError<I> for Error<I> {} + +impl<I, E> FromExternalError<I, E> for Error<I> { + /// Create a new error from an input position and an external error + fn from_external_error(input: I, kind: ErrorKind, _e: E) -> Self { + Error { input, code: kind } + } +} + +/// The Display implementation allows the std::error::Error implementation +impl<I: fmt::Display> fmt::Display for Error<I> { + fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result { + write!(f, "error {:?} at: {}", self.code, self.input) + } +} + +#[cfg(feature = "std")] +impl<I: fmt::Debug + fmt::Display> std::error::Error for Error<I> {} + +// for backward compatibility, keep those trait implementations +// for the previously used error type impl<I> ParseError<I> for (I, ErrorKind) { fn from_error_kind(input: I, kind: ErrorKind) -> Self { (input, kind) @@ -48,48 +110,65 @@ impl<I> ParseError<I> for (I, ErrorKind) { } } +impl<I> ContextError<I> for (I, ErrorKind) {} + +impl<I, E> FromExternalError<I, E> for (I, ErrorKind) { + fn from_external_error(input: I, kind: ErrorKind, _e: E) -> Self { + (input, kind) + } +} + impl<I> ParseError<I> for () { fn from_error_kind(_: I, _: ErrorKind) -> Self {} fn append(_: I, _: ErrorKind, _: Self) -> Self {} } -/// creates an error from the input position and an [ErrorKind] +impl<I> ContextError<I> for () {} + +impl<I, E> FromExternalError<I, E> for () { + fn from_external_error(_input: I, _kind: ErrorKind, _e: E) -> Self {} +} + +/// Creates an error from the input position and an [ErrorKind] pub fn make_error<I, E: ParseError<I>>(input: I, kind: ErrorKind) -> E { E::from_error_kind(input, kind) } -/// combines an existing error with a new one created from the input +/// Combines an existing error with a new one created from the input /// position and an [ErrorKind]. This is useful when backtracking /// through a parse tree, accumulating error context on the way pub fn append_error<I, E: ParseError<I>>(input: I, kind: ErrorKind, other: E) -> E { E::append(input, kind, other) } -/// this error type accumulates errors and their position when backtracking +/// This error type accumulates errors and their position when backtracking /// through a parse tree. With some post processing (cf `examples/json.rs`), /// it can be used to display user friendly error messages #[cfg(feature = "alloc")] +#[cfg_attr(feature = "docsrs", doc(cfg(feature = "alloc")))] #[derive(Clone, Debug, PartialEq)] pub struct VerboseError<I> { - /// list of errors accumulated by `VerboseError`, containing the affected + /// List of errors accumulated by `VerboseError`, containing the affected /// part of input data, and some context pub errors: crate::lib::std::vec::Vec<(I, VerboseErrorKind)>, } #[cfg(feature = "alloc")] +#[cfg_attr(feature = "docsrs", doc(cfg(feature = "alloc")))] #[derive(Clone, Debug, PartialEq)] -/// error context for `VerboseError` +/// Error context for `VerboseError` pub enum VerboseErrorKind { - /// static string added by the `context` function + /// Static string added by the `context` function Context(&'static str), - /// indicates which character was expected by the `char` function + /// Indicates which character was expected by the `char` function Char(char), - /// error kind given by various nom parsers + /// Error kind given by various nom parsers Nom(ErrorKind), } #[cfg(feature = "alloc")] +#[cfg_attr(feature = "docsrs", doc(cfg(feature = "alloc")))] impl<I> ParseError<I> for VerboseError<I> { fn from_error_kind(input: I, kind: ErrorKind) -> Self { VerboseError { @@ -107,23 +186,58 @@ impl<I> ParseError<I> for VerboseError<I> { errors: vec![(input, VerboseErrorKind::Char(c))], } } +} +#[cfg(feature = "alloc")] +#[cfg_attr(feature = "docsrs", doc(cfg(feature = "alloc")))] +impl<I> ContextError<I> for VerboseError<I> { fn add_context(input: I, ctx: &'static str, mut other: Self) -> Self { other.errors.push((input, VerboseErrorKind::Context(ctx))); other } } +#[cfg(feature = "alloc")] +#[cfg_attr(feature = "docsrs", doc(cfg(feature = "alloc")))] +impl<I, E> FromExternalError<I, E> for VerboseError<I> { + /// Create a new error from an input position and an external error + fn from_external_error(input: I, kind: ErrorKind, _e: E) -> Self { + Self::from_error_kind(input, kind) + } +} + +#[cfg(feature = "alloc")] +impl<I: fmt::Display> fmt::Display for VerboseError<I> { + fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result { + writeln!(f, "Parse error:")?; + for (input, error) in &self.errors { + match error { + VerboseErrorKind::Nom(e) => writeln!(f, "{:?} at: {}", e, input)?, + VerboseErrorKind::Char(c) => writeln!(f, "expected '{}' at: {}", c, input)?, + VerboseErrorKind::Context(s) => writeln!(f, "in section '{}', at: {}", s, input)?, + } + } + + Ok(()) + } +} + +#[cfg(feature = "std")] +impl<I: fmt::Debug + fmt::Display> std::error::Error for VerboseError<I> {} + use crate::internal::{Err, IResult}; -/// create a new error from an input position, a static string and an existing error. +/// Create a new error from an input position, a static string and an existing error. /// This is used mainly in the [context] combinator, to add user friendly information /// to errors when backtracking through a parse tree -pub fn context<I: Clone, E: ParseError<I>, F, O>(context: &'static str, f: F) -> impl Fn(I) -> IResult<I, O, E> +pub fn context<I: Clone, E: ContextError<I>, F, O>( + context: &'static str, + mut f: F, +) -> impl FnMut(I) -> IResult<I, O, E> where - F: Fn(I) -> IResult<I, O, E>, + F: Parser<I, O, E>, { - move |i: I| match f(i.clone()) { + move |i: I| match f.parse(i.clone()) { Ok(o) => Ok(o), Err(Err::Incomplete(i)) => Err(Err::Incomplete(i)), Err(Err::Error(e)) => Err(Err::Error(E::add_context(i, context, e))), @@ -131,9 +245,13 @@ where } } -/// transforms a `VerboseError` into a trace with input position information +/// Transforms a `VerboseError` into a trace with input position information #[cfg(feature = "alloc")] -pub fn convert_error(input: &str, e: VerboseError<&str>) -> crate::lib::std::string::String { +#[cfg_attr(feature = "docsrs", doc(cfg(feature = "alloc")))] +pub fn convert_error<I: core::ops::Deref<Target = str>>( + input: I, + e: VerboseError<I>, +) -> crate::lib::std::string::String { use crate::lib::std::fmt::Write; use crate::traits::Offset; @@ -144,7 +262,9 @@ pub fn convert_error(input: &str, e: VerboseError<&str>) -> crate::lib::std::str if input.is_empty() { match kind { - VerboseErrorKind::Char(c) => write!(&mut result, "{}: expected '{}', got empty input\n\n", i, c), + VerboseErrorKind::Char(c) => { + write!(&mut result, "{}: expected '{}', got empty input\n\n", i, c) + } VerboseErrorKind::Context(s) => write!(&mut result, "{}: in {}, got empty input\n\n", i, s), VerboseErrorKind::Nom(e) => write!(&mut result, "{}: in {:?}, got empty input\n\n", i, e), } @@ -156,45 +276,56 @@ pub fn convert_error(input: &str, e: VerboseError<&str>) -> crate::lib::std::str // Find the line that includes the subslice: // Find the *last* newline before the substring starts - let line_begin = prefix.iter().rev().position(|&b| b == b'\n').map(|pos| offset - pos).unwrap_or(0); + let line_begin = prefix + .iter() + .rev() + .position(|&b| b == b'\n') + .map(|pos| offset - pos) + .unwrap_or(0); // Find the full line after that newline - let line = input[line_begin..].lines().next().unwrap_or(&input[line_begin..]).trim_end(); + let line = input[line_begin..] + .lines() + .next() + .unwrap_or(&input[line_begin..]) + .trim_end(); // The (1-indexed) column number is the offset of our substring into that line let column_number = line.offset(substring) + 1; match kind { - VerboseErrorKind::Char(c) => if let Some(actual) = substring.chars().next() { - write!( - &mut result, - "{i}: at line {line_number}:\n\ + VerboseErrorKind::Char(c) => { + if let Some(actual) = substring.chars().next() { + write!( + &mut result, + "{i}: at line {line_number}:\n\ {line}\n\ {caret:>column$}\n\ expected '{expected}', found {actual}\n\n", - i = i, - line_number = line_number, - line = line, - caret = '^', - column = column_number, - expected = c, - actual = actual, - ) - } else { - write!( - &mut result, - "{i}: at line {line_number}:\n\ + i = i, + line_number = line_number, + line = line, + caret = '^', + column = column_number, + expected = c, + actual = actual, + ) + } else { + write!( + &mut result, + "{i}: at line {line_number}:\n\ {line}\n\ {caret:>column$}\n\ expected '{expected}', got end of input\n\n", - i = i, - line_number = line_number, - line = line, - caret = '^', - column = column_number, - expected = c, - ) - }, + i = i, + line_number = line_number, + line = line, + caret = '^', + column = column_number, + expected = c, + ) + } + } VerboseErrorKind::Context(s) => write!( &mut result, "{i}: at line {line_number}, in {context}:\n\ @@ -228,8 +359,8 @@ pub fn convert_error(input: &str, e: VerboseError<&str>) -> crate::lib::std::str result } -/// indicates which parser returned an error -#[cfg_attr(rustfmt, rustfmt_skip)] +/// Indicates which parser returned an error +#[rustfmt::skip] #[derive(Debug,PartialEq,Eq,Hash,Clone,Copy)] #[allow(deprecated,missing_docs)] pub enum ErrorKind { @@ -285,11 +416,12 @@ pub enum ErrorKind { Many0Count, Many1Count, Float, + Satisfy, } -#[cfg_attr(rustfmt, rustfmt_skip)] +#[rustfmt::skip] #[allow(deprecated)] -/// converts an ErrorKind to a number +/// Converts an ErrorKind to a number pub fn error_to_u32(e: &ErrorKind) -> u32 { match *e { ErrorKind::Tag => 1, @@ -344,13 +476,14 @@ pub fn error_to_u32(e: &ErrorKind) -> u32 { ErrorKind::Many0Count => 72, ErrorKind::Many1Count => 73, ErrorKind::Float => 74, + ErrorKind::Satisfy => 75, } } impl ErrorKind { - #[cfg_attr(rustfmt, rustfmt_skip)] + #[rustfmt::skip] #[allow(deprecated)] - /// converts an ErrorKind to a text description + /// Converts an ErrorKind to a text description pub fn description(&self) -> &str { match *self { ErrorKind::Tag => "Tag", @@ -405,11 +538,12 @@ impl ErrorKind { ErrorKind::Many0Count => "Count occurrence of >=0 patterns", ErrorKind::Many1Count => "Count occurrence of >=1 patterns", ErrorKind::Float => "Float", + ErrorKind::Satisfy => "Satisfy", } } } -/// creates a parse error from a `nom::ErrorKind` +/// Creates a parse error from a `nom::ErrorKind` /// and the position in the input #[allow(unused_variables)] #[macro_export(local_inner_macros)] @@ -419,9 +553,9 @@ macro_rules! error_position( }); ); -/// creates a parse error from a `nom::ErrorKind`, +/// Creates a parse error from a `nom::ErrorKind`, /// the position in the input and the next error in -/// the parsing tree. +/// the parsing tree #[allow(unused_variables)] #[macro_export(local_inner_macros)] macro_rules! error_node_position( @@ -430,27 +564,6 @@ macro_rules! error_node_position( }); ); -/* - -#[cfg(feature = "std")] -use $crate::lib::std::any::Any; -#[cfg(feature = "std")] -use $crate::lib::std::{error,fmt}; -#[cfg(feature = "std")] -impl<E: fmt::Debug+Any> error::Error for Err<E> { - fn description(&self) -> &str { - self.description() - } -} - -#[cfg(feature = "std")] -impl<E: fmt::Debug> fmt::Display for Err<E> { - fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { - write!(f, "{}", self.description()) - } -} -*/ - //FIXME: error rewrite /// translate parser result from IResult<I,O,u32> to IResult<I,O,E> with a custom type /// @@ -516,19 +629,19 @@ macro_rules! fix_error ( /// `flat_map!(R -> IResult<R,S>, S -> IResult<S,T>) => R -> IResult<R, T>` /// -/// combines a parser R -> IResult<R,S> and -/// a parser S -> IResult<S,T> to return another -/// parser R -> IResult<R,T> +/// Combines a parser `R -> IResult<R,S>` and +/// a parser `S -> IResult<S,T>` to return another +/// parser `R -> IResult<R,T>` /// /// ```rust /// # #[macro_use] extern crate nom; -/// # use nom::{Err, error::ErrorKind}; +/// # use nom::{Err, error::{Error, ErrorKind}}; /// use nom::number::complete::recognize_float; /// /// named!(parser<&str, f64>, flat_map!(recognize_float, parse_to!(f64))); /// /// assert_eq!(parser("123.45;"), Ok((";", 123.45))); -/// assert_eq!(parser("abc"), Err(Err::Error(("abc", ErrorKind::Char)))); +/// assert_eq!(parser("abc"), Err(Err::Error(Error::new("abc", ErrorKind::Char)))); /// ``` #[macro_export(local_inner_macros)] macro_rules! flat_map( @@ -559,7 +672,7 @@ mod tests { fn convert_error_panic() { let input = ""; - let result: IResult<_, _, VerboseError<&str>> = char('x')(input); + let _result: IResult<_, _, VerboseError<&str>> = char('x')(input); } } diff --git a/src/internal.rs b/src/internal.rs index 401f98a..fd683c0 100644 --- a/src/internal.rs +++ b/src/internal.rs @@ -1,38 +1,79 @@ //! Basic types to build the parsers use self::Needed::*; -use crate::error::ErrorKind; +use crate::error::{self, ErrorKind}; +use crate::lib::std::fmt; +use core::num::NonZeroUsize; /// Holds the result of parsing functions /// -/// It depends on I, the input type, O, the output type, and E, the error type (by default u32) +/// It depends on the input type `I`, the output type `O`, and the error type `E` +/// (by default `(I, nom::ErrorKind)`) /// /// The `Ok` side is a pair containing the remainder of the input (the part of the data that /// was not parsed) and the produced value. The `Err` side contains an instance of `nom::Err`. /// -pub type IResult<I, O, E=(I,ErrorKind)> = Result<(I, O), Err<E>>; +/// Outside of the parsing code, you can use the [Finish::finish] method to convert +/// it to a more common result type +pub type IResult<I, O, E = error::Error<I>> = Result<(I, O), Err<E>>; + +/// Helper trait to convert a parser's result to a more manageable type +pub trait Finish<I, O, E> { + /// converts the parser's result to a type that is more consumable by error + /// management libraries. It keeps the same `Ok` branch, and merges `Err::Error` + /// and `Err::Failure` into the `Err` side. + /// + /// *warning*: if the result is `Err(Err::Incomplete(_))`, this method will panic. + /// - "complete" parsers: It will not be an issue, `Incomplete` is never used + /// - "streaming" parsers: `Incomplete` will be returned if there's not enough data + /// for the parser to decide, and you should gather more data before parsing again. + /// Once the parser returns either `Ok(_)`, `Err(Err::Error(_))` or `Err(Err::Failure(_))`, + /// you can get out of the parsing loop and call `finish()` on the parser's result + fn finish(self) -> Result<(I, O), E>; +} + +impl<I, O, E> Finish<I, O, E> for IResult<I, O, E> { + fn finish(self) -> Result<(I, O), E> { + match self { + Ok(res) => Ok(res), + Err(Err::Error(e)) | Err(Err::Failure(e)) => Err(e), + Err(Err::Incomplete(_)) => { + panic!("Cannot call `finish()` on `Err(Err::Incomplete(_))`: this result means that the parser does not have enough data to decide, you should gather more data and try to reapply the parser instead") + } + } + } +} /// Contains information on needed data if a parser returned `Incomplete` #[derive(Debug, PartialEq, Eq, Clone, Copy)] +#[allow(missing_doc_code_examples)] pub enum Needed { - /// needs more data, but we do not know how much + /// Needs more data, but we do not know how much Unknown, - /// contains the required data size - Size(usize), + /// Contains the required data size in bytes + Size(NonZeroUsize), } impl Needed { - /// indicates if we know how many bytes we need + /// Creates `Needed` instance, returns `Needed::Unknown` if the argument is zero + pub fn new(s: usize) -> Self { + match NonZeroUsize::new(s) { + Some(sz) => Needed::Size(sz), + None => Needed::Unknown, + } + } + + /// Indicates if we know how many bytes we need pub fn is_known(&self) -> bool { *self != Unknown } /// Maps a `Needed` to `Needed` by applying a function to a contained `Size` value. #[inline] - pub fn map<F: Fn(usize) -> usize>(self, f: F) -> Needed { + pub fn map<F: Fn(NonZeroUsize) -> usize>(self, f: F) -> Needed { match self { Unknown => Unknown, - Size(n) => Size(f(n)), + Size(n) => Needed::new(f(n)), } } } @@ -52,6 +93,7 @@ impl Needed { /// to try other parsers, you were already in the right branch, so the data is invalid /// #[derive(Debug, Clone, PartialEq)] +#[allow(missing_doc_code_examples)] pub enum Err<E> { /// There was not enough data Incomplete(Needed), @@ -64,7 +106,7 @@ pub enum Err<E> { } impl<E> Err<E> { - /// tests if the result is Incomplete + /// Tests if the result is Incomplete pub fn is_incomplete(&self) -> bool { if let Err::Incomplete(_) = self { true @@ -75,7 +117,8 @@ impl<E> Err<E> { /// Applies the given function to the inner error pub fn map<E2, F>(self, f: F) -> Err<E2> - where F: FnOnce(E) -> E2 + where + F: FnOnce(E) -> E2, { match self { Err::Incomplete(n) => Err::Incomplete(n), @@ -84,18 +127,21 @@ impl<E> Err<E> { } } - /// automatically converts between errors if the underlying type supports it + /// Automatically converts between errors if the underlying type supports it pub fn convert<F>(e: Err<F>) -> Self - where E: From<F> + where + E: From<F>, { - e.map(Into::into) + e.map(crate::lib::std::convert::Into::into) } } impl<T> Err<(T, ErrorKind)> { - /// maps `Err<(T, ErrorKind)>` to `Err<(U, ErrorKind)>` with the given F: T -> U + /// Maps `Err<(T, ErrorKind)>` to `Err<(U, ErrorKind)>` with the given `F: T -> U` pub fn map_input<U, F>(self, f: F) -> Err<(U, ErrorKind)> - where F: FnOnce(T) -> U { + where + F: FnOnce(T) -> U, + { match self { Err::Incomplete(n) => Err::Incomplete(n), Err::Failure((input, k)) => Err::Failure((f(input), k)), @@ -104,17 +150,21 @@ impl<T> Err<(T, ErrorKind)> { } } -#[cfg(feature = "std")] +#[cfg(feature = "alloc")] +use crate::lib::std::{borrow::ToOwned, string::String, vec::Vec}; +#[cfg(feature = "alloc")] impl Err<(&[u8], ErrorKind)> { /// Obtaining ownership + #[cfg_attr(feature = "docsrs", doc(cfg(feature = "alloc")))] pub fn to_owned(self) -> Err<(Vec<u8>, ErrorKind)> { self.map_input(ToOwned::to_owned) } } -#[cfg(feature = "std")] +#[cfg(feature = "alloc")] impl Err<(&str, ErrorKind)> { - /// automatically converts between errors if the underlying type supports it + /// Automatically converts between errors if the underlying type supports it + #[cfg_attr(feature = "docsrs", doc(cfg(feature = "alloc")))] pub fn to_owned(self) -> Err<(String, ErrorKind)> { self.map_input(ToOwned::to_owned) } @@ -122,10 +172,6 @@ impl Err<(&str, ErrorKind)> { impl<E: Eq> Eq for Err<E> {} -#[cfg(feature = "std")] -use std::fmt; - -#[cfg(feature = "std")] impl<E> fmt::Display for Err<E> where E: fmt::Debug, @@ -153,6 +199,225 @@ where } } +/// All nom parsers implement this trait +pub trait Parser<I, O, E> { + /// A parser takes in input type, and returns a `Result` containing + /// either the remaining input and the output value, or an error + fn parse(&mut self, input: I) -> IResult<I, O, E>; + + /// Maps a function over the result of a parser + fn map<G, O2>(self, g: G) -> Map<Self, G, O> + where + G: Fn(O) -> O2, + Self: core::marker::Sized, + { + Map { + f: self, + g, + phantom: core::marker::PhantomData, + } + } + + /// Creates a second parser from the output of the first one, then apply over the rest of the input + fn flat_map<G, H, O2>(self, g: G) -> FlatMap<Self, G, O> + where + G: Fn(O) -> H, + H: Parser<I, O2, E>, + Self: core::marker::Sized, + { + FlatMap { + f: self, + g, + phantom: core::marker::PhantomData, + } + } + + /// Applies a second parser over the output of the first one + fn and_then<G, O2>(self, g: G) -> AndThen<Self, G, O> + where + G: Parser<O, O2, E>, + Self: core::marker::Sized, + { + AndThen { + f: self, + g, + phantom: core::marker::PhantomData, + } + } + + /// Applies a second parser after the first one, return their results as a tuple + fn and<G, O2>(self, g: G) -> And<Self, G> + where + G: Parser<I, O2, E>, + Self: core::marker::Sized, + { + And { f: self, g } + } + + /// Applies a second parser over the input if the first one failed + fn or<G>(self, g: G) -> Or<Self, G> + where + G: Parser<I, O, E>, + Self: core::marker::Sized, + { + Or { f: self, g } + } + + /// automatically converts the parser's output and error values to another type, as long as they + /// implement the `From` trait + fn into<O2: From<O>, E2: From<E>>(self) -> Into<Self, O, O2, E, E2> + where + Self: core::marker::Sized, + { + Into { + f: self, + phantom_out1: core::marker::PhantomData, + phantom_err1: core::marker::PhantomData, + phantom_out2: core::marker::PhantomData, + phantom_err2: core::marker::PhantomData, + } + } +} + +impl<'a, I, O, E, F> Parser<I, O, E> for F +where + F: FnMut(I) -> IResult<I, O, E> + 'a, +{ + fn parse(&mut self, i: I) -> IResult<I, O, E> { + self(i) + } +} + +#[cfg(feature = "alloc")] +use alloc::boxed::Box; + +#[cfg(feature = "alloc")] +impl<'a, I, O, E> Parser<I, O, E> for Box<dyn Parser<I, O, E> + 'a> { + fn parse(&mut self, input: I) -> IResult<I, O, E> { + (**self).parse(input) + } +} + +/// Implementation of `Parser::map` +#[allow(missing_doc_code_examples)] +pub struct Map<F, G, O1> { + f: F, + g: G, + phantom: core::marker::PhantomData<O1>, +} + +impl<'a, I, O1, O2, E, F: Parser<I, O1, E>, G: Fn(O1) -> O2> Parser<I, O2, E> for Map<F, G, O1> { + fn parse(&mut self, i: I) -> IResult<I, O2, E> { + match self.f.parse(i) { + Err(e) => Err(e), + Ok((i, o)) => Ok((i, (self.g)(o))), + } + } +} + +/// Implementation of `Parser::flat_map` +#[allow(missing_doc_code_examples)] +pub struct FlatMap<F, G, O1> { + f: F, + g: G, + phantom: core::marker::PhantomData<O1>, +} + +impl<'a, I, O1, O2, E, F: Parser<I, O1, E>, G: Fn(O1) -> H, H: Parser<I, O2, E>> Parser<I, O2, E> + for FlatMap<F, G, O1> +{ + fn parse(&mut self, i: I) -> IResult<I, O2, E> { + let (i, o1) = self.f.parse(i)?; + (self.g)(o1).parse(i) + } +} + +/// Implementation of `Parser::and_then` +#[allow(missing_doc_code_examples)] +pub struct AndThen<F, G, O1> { + f: F, + g: G, + phantom: core::marker::PhantomData<O1>, +} + +impl<'a, I, O1, O2, E, F: Parser<I, O1, E>, G: Parser<O1, O2, E>> Parser<I, O2, E> + for AndThen<F, G, O1> +{ + fn parse(&mut self, i: I) -> IResult<I, O2, E> { + let (i, o1) = self.f.parse(i)?; + let (_, o2) = self.g.parse(o1)?; + Ok((i, o2)) + } +} + +/// Implementation of `Parser::and` +#[allow(missing_doc_code_examples)] +pub struct And<F, G> { + f: F, + g: G, +} + +impl<'a, I, O1, O2, E, F: Parser<I, O1, E>, G: Parser<I, O2, E>> Parser<I, (O1, O2), E> + for And<F, G> +{ + fn parse(&mut self, i: I) -> IResult<I, (O1, O2), E> { + let (i, o1) = self.f.parse(i)?; + let (i, o2) = self.g.parse(i)?; + Ok((i, (o1, o2))) + } +} + +/// Implementation of `Parser::or` +#[allow(missing_doc_code_examples)] +pub struct Or<F, G> { + f: F, + g: G, +} + +impl<'a, I: Clone, O, E: crate::error::ParseError<I>, F: Parser<I, O, E>, G: Parser<I, O, E>> + Parser<I, O, E> for Or<F, G> +{ + fn parse(&mut self, i: I) -> IResult<I, O, E> { + match self.f.parse(i.clone()) { + Err(Err::Error(e1)) => match self.g.parse(i) { + Err(Err::Error(e2)) => Err(Err::Error(e1.or(e2))), + res => res, + }, + res => res, + } + } +} + +/// Implementation of `Parser::into` +#[allow(missing_doc_code_examples)] +pub struct Into<F, O1, O2: From<O1>, E1, E2: From<E1>> { + f: F, + phantom_out1: core::marker::PhantomData<O1>, + phantom_err1: core::marker::PhantomData<E1>, + phantom_out2: core::marker::PhantomData<O2>, + phantom_err2: core::marker::PhantomData<E2>, +} + +impl< + 'a, + I: Clone, + O1, + O2: From<O1>, + E1, + E2: crate::error::ParseError<I> + From<E1>, + F: Parser<I, O1, E1>, + > Parser<I, O2, E2> for Into<F, O1, O2, E1, E2> +{ + fn parse(&mut self, i: I) -> IResult<I, O2, E2> { + match self.f.parse(i) { + Ok((i, o)) => Ok((i, o.into())), + Err(Err::Error(e)) => Err(Err::Error(e.into())), + Err(Err::Failure(e)) => Err(Err::Failure(e.into())), + Err(Err::Incomplete(e)) => Err(Err::Incomplete(e)), + } + } +} + #[cfg(test)] mod tests { use super::*; @@ -171,8 +436,8 @@ mod tests { fn size_test() { assert_size!(IResult<&[u8], &[u8], (&[u8], u32)>, 40); assert_size!(IResult<&str, &str, u32>, 40); - assert_size!(Needed, 16); - assert_size!(Err<u32>, 24); + assert_size!(Needed, 8); + assert_size!(Err<u32>, 16); assert_size!(ErrorKind, 1); } @@ -181,5 +446,4 @@ mod tests { let e = Err::Error(1); assert_eq!(e.map(|v| v + 1), Err::Error(2)); } - } @@ -58,6 +58,7 @@ //! about [the design of nom macros](https://github.com/Geal/nom/blob/master/doc/how_nom_macros_work.md), //! [how to write parsers](https://github.com/Geal/nom/blob/master/doc/making_a_new_parser_from_scratch.md), //! or the [error management system](https://github.com/Geal/nom/blob/master/doc/error_management.md). +//! You can also check out the [recipes] module that contains examples of common patterns. //! //! **Looking for a specific combinator? Read the //! ["choose a combinator" guide](https://github.com/Geal/nom/blob/master/doc/choosing_a_combinator.md)** @@ -81,11 +82,11 @@ //! //! This gives us a few advantages: //! -//! - the parsers are small and easy to write -//! - the parsers components are easy to reuse (if they're general enough, please add them to nom!) -//! - the parsers components are easy to test separately (unit tests and property-based tests) -//! - the parser combination code looks close to the grammar you would have written -//! - you can build partial parsers, specific to the data you need at the moment, and ignore the rest +//! - The parsers are small and easy to write +//! - The parsers components are easy to reuse (if they're general enough, please add them to nom!) +//! - The parsers components are easy to test separately (unit tests and property-based tests) +//! - The parser combination code looks close to the grammar you would have written +//! - You can build partial parsers, specific to the data you need at the moment, and ignore the rest //! //! Here is an example of one such parser, to recognize text between parentheses: //! @@ -118,7 +119,7 @@ //! # fn main() { //! fn take4(i: &[u8]) -> IResult<&[u8], &[u8]>{ //! if i.len() < 4 { -//! Err(Err::Incomplete(Needed::Size(4))) +//! Err(Err::Incomplete(Needed::new(4))) //! } else { //! Ok((&i[4..], &i[0..4])) //! } @@ -165,7 +166,7 @@ //! fn parser(input: I) -> IResult<I, O, E>; //! ``` //! -//! Or like this, if you don't want to specify a custom error type (it will be `u32` by default): +//! Or like this, if you don't want to specify a custom error type (it will be `(I, ErrorKind)` by default): //! //! ```rust,ignore //! fn parser(input: I) -> IResult<I, O>; @@ -174,9 +175,9 @@ //! `IResult` is an alias for the `Result` type: //! //! ```rust -//! use nom::{Needed, error::ErrorKind}; +//! use nom::{Needed, error::Error}; //! -//! type IResult<I, O, E = (I,ErrorKind)> = Result<(I, O), Err<E>>; +//! type IResult<I, O, E = Error<I>> = Result<(I, O), Err<E>>; //! //! enum Err<E> { //! Incomplete(Needed), @@ -187,14 +188,13 @@ //! //! It can have the following values: //! -//! - a correct result `Ok((I,O))` with the first element being the remaining of the input (not parsed yet), and the second the output value; -//! - an error `Err(Err::Error(c))` with `c` an error that can be built from the input position and a parser specific error -//! - an error `Err(Err::Incomplete(Needed))` indicating that more input is necessary. `Needed` can indicate how much data is needed -//! - an error `Err(Err::Failure(c))`. It works like the `Error` case, except it indicates an unrecoverable error: we cannot backtrack and test another parser +//! - A correct result `Ok((I,O))` with the first element being the remaining of the input (not parsed yet), and the second the output value; +//! - An error `Err(Err::Error(c))` with `c` an error that can be built from the input position and a parser specific error +//! - An error `Err(Err::Incomplete(Needed))` indicating that more input is necessary. `Needed` can indicate how much data is needed +//! - An error `Err(Err::Failure(c))`. It works like the `Error` case, except it indicates an unrecoverable error: We cannot backtrack and test another parser //! //! Please refer to the ["choose a combinator" guide](https://github.com/Geal/nom/blob/master/doc/choosing_a_combinator.md) for an exhaustive list of parsers. //! See also the rest of the documentation [here](https://github.com/Geal/nom/blob/master/doc). -//! . //! //! ## Making new parsers with function combinators //! @@ -230,7 +230,7 @@ //! use nom::branch::alt; //! use nom::bytes::complete::tag; //! -//! let alt_tags = alt((tag("abcd"), tag("efgh"))); +//! let mut alt_tags = alt((tag("abcd"), tag("efgh"))); //! //! assert_eq!(alt_tags(&b"abcdxxx"[..]), Ok((&b"xxx"[..], &b"abcd"[..]))); //! assert_eq!(alt_tags(&b"efghxxx"[..]), Ok((&b"xxx"[..], &b"efgh"[..]))); @@ -276,9 +276,9 @@ //! //! Here are some basic combining macros available: //! -//! - **`opt`**: will make the parser optional (if it returns the `O` type, the new parser returns `Option<O>`) -//! - **`many0`**: will apply the parser 0 or more times (if it returns the `O` type, the new parser returns `Vec<O>`) -//! - **`many1`**: will apply the parser 1 or more times +//! - **`opt`**: Will make the parser optional (if it returns the `O` type, the new parser returns `Option<O>`) +//! - **`many0`**: Will apply the parser 0 or more times (if it returns the `O` type, the new parser returns `Vec<O>`) +//! - **`many1`**: Will apply the parser 1 or more times //! //! There are more complex (and more useful) parsers like `tuple!`, which is //! used to apply a series of parsers then assemble their results. @@ -293,7 +293,7 @@ //! bytes::streaming::{tag, take}, //! sequence::tuple}; //! -//! let tpl = tuple((be_u16, take(3u8), tag("fg"))); +//! let mut tpl = tuple((be_u16, take(3u8), tag("fg"))); //! //! assert_eq!( //! tpl(&b"abcdefgh"[..]), @@ -302,7 +302,7 @@ //! (0x6162u16, &b"cde"[..], &b"fg"[..]) //! )) //! ); -//! assert_eq!(tpl(&b"abcde"[..]), Err(nom::Err::Incomplete(Needed::Size(2)))); +//! assert_eq!(tpl(&b"abcde"[..]), Err(nom::Err::Incomplete(Needed::new(2)))); //! let input = &b"abcdejk"[..]; //! assert_eq!(tpl(input), Err(nom::Err::Error((&input[5..], ErrorKind::Tag)))); //! # } @@ -356,7 +356,7 @@ //! Here is how it works in practice: //! //! ```rust -//! use nom::{IResult, Err, Needed, error::ErrorKind, bytes, character}; +//! use nom::{IResult, Err, Needed, error::{Error, ErrorKind}, bytes, character}; //! //! fn take_streaming(i: &[u8]) -> IResult<&[u8], &[u8]> { //! bytes::streaming::take(4u8)(i) @@ -372,10 +372,10 @@ //! //! // if the input is smaller than 4 bytes, the streaming parser //! // will return `Incomplete` to indicate that we need more data -//! assert_eq!(take_streaming(&b"abc"[..]), Err(Err::Incomplete(Needed::Size(4)))); +//! assert_eq!(take_streaming(&b"abc"[..]), Err(Err::Incomplete(Needed::new(1)))); //! //! // but the complete parser will return an error -//! assert_eq!(take_complete(&b"abc"[..]), Err(Err::Error((&b"abc"[..], ErrorKind::Eof)))); +//! assert_eq!(take_complete(&b"abc"[..]), Err(Err::Error(Error::new(&b"abc"[..], ErrorKind::Eof)))); //! //! // the alpha0 function recognizes 0 or more alphabetic characters //! fn alpha0_streaming(i: &str) -> IResult<&str, &str> { @@ -393,65 +393,79 @@ //! // but when there's no limit, the streaming version returns `Incomplete`, because it cannot //! // know if more input data should be recognized. The whole input could be "abcd;", or //! // "abcde;" -//! assert_eq!(alpha0_streaming("abcd"), Err(Err::Incomplete(Needed::Size(1)))); +//! assert_eq!(alpha0_streaming("abcd"), Err(Err::Incomplete(Needed::new(1)))); //! //! // while the complete version knows that all of the data is there //! assert_eq!(alpha0_complete("abcd"), Ok(("", "abcd"))); //! ``` -//! **Going further:** read the [guides](https://github.com/Geal/nom/tree/master/doc)! -#![cfg_attr(all(not(feature = "std"), feature = "alloc"), feature(alloc))] +//! **Going further:** Read the [guides](https://github.com/Geal/nom/tree/master/doc), +//! check out the [recipes]! #![cfg_attr(not(feature = "std"), no_std)] -#![cfg_attr(feature = "cargo-clippy", allow(doc_markdown))] +#![cfg_attr(feature = "cargo-clippy", allow(clippy::doc_markdown))] #![cfg_attr(nightly, feature(test))] +#![cfg_attr(feature = "docsrs", feature(doc_cfg))] +#![cfg_attr(feature = "docsrs", feature(external_doc))] #![deny(missing_docs)] #![warn(missing_doc_code_examples)] -#[cfg(all(not(feature = "std"), feature = "alloc"))] +#[cfg(feature = "alloc")] #[macro_use] extern crate alloc; -#[cfg(feature = "regexp_macros")] -#[macro_use] -extern crate lazy_static; +#[cfg(feature = "bitvec")] +pub extern crate bitvec; +#[cfg(doctest)] +extern crate doc_comment; +#[cfg(feature = "lexical")] +extern crate lexical_core; extern crate memchr; #[cfg(feature = "regexp")] pub extern crate regex; -#[cfg(feature = "lexical")] -extern crate lexical_core; #[cfg(nightly)] extern crate test; -#[cfg(test)] -extern crate doc_comment; -//FIXME: reactivate doctest once https://github.com/rust-lang/rust/issues/62210 is done -//#[cfg(doctest)] -//doc_comment::doctest!("../README.md"); +#[cfg(doctest)] +doc_comment::doctest!("../README.md"); /// Lib module to re-export everything needed from `std` or `core`/`alloc`. This is how `serde` does /// it, albeit there it is not public. +#[allow(missing_doc_code_examples)] pub mod lib { /// `std` facade allowing `std`/`core` to be interchangeable. Reexports `alloc` crate optionally, /// as well as `core` or `std` #[cfg(not(feature = "std"))] + #[allow(missing_doc_code_examples)] /// internal std exports for no_std compatibility pub mod std { + #[doc(hidden)] + #[cfg(not(feature = "alloc"))] + pub use core::borrow; + #[cfg(feature = "alloc")] - #[cfg_attr(feature = "alloc", macro_use)] - pub use alloc::{boxed, string, vec}; + #[doc(hidden)] + pub use alloc::{borrow, boxed, string, vec}; - pub use core::{cmp, convert, fmt, iter, mem, ops, option, result, slice, str, borrow}; + #[doc(hidden)] + pub use core::{cmp, convert, fmt, iter, mem, ops, option, result, slice, str}; /// internal reproduction of std prelude + #[doc(hidden)] pub mod prelude { pub use core::prelude as v1; } } #[cfg(feature = "std")] + #[allow(missing_doc_code_examples)] /// internal std exports for no_std compatibility pub mod std { - pub use std::{alloc, boxed, cmp, collections, convert, fmt, hash, iter, mem, ops, option, result, slice, str, string, vec, borrow}; + #[doc(hidden)] + pub use std::{ + alloc, borrow, boxed, cmp, collections, convert, fmt, hash, iter, mem, ops, option, result, + slice, str, string, vec, + }; /// internal reproduction of std prelude + #[doc(hidden)] pub mod prelude { pub use std::prelude as v1; } @@ -461,12 +475,10 @@ pub mod lib { pub use regex; } +pub use self::bits::*; +pub use self::internal::*; pub use self::traits::*; pub use self::util::*; -pub use self::internal::*; -pub use self::methods::*; -pub use self::bits::*; -pub use self::whitespace::*; #[cfg(feature = "regexp")] pub use self::regexp::*; @@ -489,8 +501,6 @@ pub mod branch; pub mod sequence; #[macro_use] pub mod multi; -#[macro_use] -pub mod methods; #[macro_use] pub mod bytes; @@ -500,14 +510,16 @@ pub mod bits; #[macro_use] pub mod character; -#[macro_use] -pub mod whitespace; - #[cfg(feature = "regexp")] +#[cfg_attr(feature = "docsrs", doc(cfg(feature = "regexp")))] #[macro_use] -mod regexp; +pub mod regexp; mod str; #[macro_use] pub mod number; + +#[cfg(feature = "docsrs")] +#[cfg_attr(feature = "docsrs", doc(include = "../doc/nom_recipes.md"))] +pub mod recipes {} diff --git a/src/methods.rs b/src/methods.rs deleted file mode 100644 index 520f1e8..0000000 --- a/src/methods.rs +++ /dev/null @@ -1,20 +0,0 @@ -//! method combinators - -/// do not use: method combinators moved to the nom-methods crate -#[macro_export] -macro_rules! method ( - ($($args:tt)*) => (compile_error!("method combinators moved to the nom-methods crate");); -); - -/// do not use: method combinators moved to the nom-methods crate -#[macro_export] -macro_rules! call_m ( - ($($args:tt)*) => (compile_error!("method combinators moved to the nom-methods crate");); -); - -/// do not use: method combinators moved to the nom-methods crate -#[macro_export] -macro_rules! apply_m ( - ($($args:tt)*) => (compile_error!("method combinators moved to the nom-methods crate");); -); - diff --git a/src/multi/macros.rs b/src/multi/macros.rs index 10b66b8..c03de3a 100644 --- a/src/multi/macros.rs +++ b/src/multi/macros.rs @@ -1,16 +1,16 @@ //! Parsers for applying parsers multiple times -/// `separated_list!(I -> IResult<I,T>, I -> IResult<I,O>) => I -> IResult<I, Vec<O>>` -/// separated_list(sep, X) returns a Vec<X> +/// `separated_list0!(I -> IResult<I,T>, I -> IResult<I,O>) => I -> IResult<I, Vec<O>>` +/// `separated_list0(sep, X)` returns a `Vec<X>`. /// /// ```rust /// # #[macro_use] extern crate nom; /// # use nom::{Err, error::ErrorKind, Needed, IResult}; -/// use nom::multi::separated_list; +/// use nom::multi::separated_list0; /// use nom::bytes::complete::tag; /// /// # fn main() { -/// named!(parser<&str, Vec<&str>>, separated_list!(tag("|"), tag("abc"))); +/// named!(parser<&str, Vec<&str>>, separated_list0!(tag("|"), tag("abc"))); /// /// assert_eq!(parser("abc|abc|abc"), Ok(("", vec!["abc", "abc", "abc"]))); /// assert_eq!(parser("abc123abc"), Ok(("123abc", vec!["abc"]))); @@ -20,69 +20,71 @@ /// # } /// ``` #[cfg(feature = "alloc")] +#[cfg_attr(feature = "docsrs", doc(cfg(feature = "alloc")))] #[macro_export(local_inner_macros)] -macro_rules! separated_list( +macro_rules! separated_list0( ($i:expr, $submac:ident!( $($args:tt)* ), $submac2:ident!( $($args2:tt)* )) => ( - separated_list!($i, |i| $submac!(i, $($args)*), |i| $submac2!(i, $($args2)*)) + separated_list0!($i, |i| $submac!(i, $($args)*), |i| $submac2!(i, $($args2)*)) ); ($i:expr, $submac:ident!( $($args:tt)* ), $g:expr) => ( - separated_list!($i, |i| $submac!(i, $($args)*), $g); + separated_list0!($i, |i| $submac!(i, $($args)*), $g); ); ($i:expr, $f:expr, $submac:ident!( $($args:tt)* )) => ( - separated_list!($i, $f, |i| $submac!(i, $($args)*)); + separated_list0!($i, $f, |i| $submac!(i, $($args)*)); ); ($i:expr, $f:expr, $g:expr) => ( - $crate::multi::separated_listc($i, $f, $g) + $crate::multi::separated_list0c($i, $f, $g) ); ); -/// `separated_nonempty_list!(I -> IResult<I,T>, I -> IResult<I,O>) => I -> IResult<I, Vec<O>>` -/// separated_nonempty_list(sep, X) returns a Vec<X> +/// `separated_list1!(I -> IResult<I,T>, I -> IResult<I,O>) => I -> IResult<I, Vec<O>>` +/// `separated_list1(sep, X)` returns a `Vec<X>`. /// -/// it will return an error if there is no element in the list +/// It will return an error if there is no element in the list. /// ```rust /// # #[macro_use] extern crate nom; -/// # use nom::{Err, error::ErrorKind, Needed, IResult}; -/// use nom::multi::separated_nonempty_list; +/// # use nom::{Err, error::{Error, ErrorKind}, Needed, IResult}; +/// use nom::multi::separated_list1; /// use nom::bytes::complete::tag; /// /// # fn main() { -/// named!(parser<&str, Vec<&str>>, separated_nonempty_list!(tag("|"), tag("abc"))); +/// named!(parser<&str, Vec<&str>>, separated_list1!(tag("|"), tag("abc"))); /// /// assert_eq!(parser("abc|abc|abc"), Ok(("", vec!["abc", "abc", "abc"]))); /// assert_eq!(parser("abc123abc"), Ok(("123abc", vec!["abc"]))); /// assert_eq!(parser("abc|def"), Ok(("|def", vec!["abc"]))); -/// assert_eq!(parser(""), Err(Err::Error(("", ErrorKind::Tag)))); -/// assert_eq!(parser("def|abc"), Err(Err::Error(("def|abc", ErrorKind::Tag)))); +/// assert_eq!(parser(""), Err(Err::Error(Error::new("", ErrorKind::Tag)))); +/// assert_eq!(parser("def|abc"), Err(Err::Error(Error::new("def|abc", ErrorKind::Tag)))); /// # } /// ``` #[cfg(feature = "alloc")] +#[cfg_attr(feature = "docsrs", doc(cfg(feature = "alloc")))] #[macro_export(local_inner_macros)] -macro_rules! separated_nonempty_list( +macro_rules! separated_list1( ($i:expr, $submac:ident!( $($args:tt)* ), $submac2:ident!( $($args2:tt)* )) => ( - separated_nonempty_list!($i, |i| $submac!(i, $($args)*), |i| $submac2!(i, $($args2)*)) + separated_list1!($i, |i| $submac!(i, $($args)*), |i| $submac2!(i, $($args2)*)) ); ($i:expr, $submac:ident!( $($args:tt)* ), $g:expr) => ( - separated_nonempty_list!($i, |i| $submac!(i, $($args)*), $g); + separated_list1!($i, |i| $submac!(i, $($args)*), $g); ); ($i:expr, $f:expr, $submac:ident!( $($args:tt)* )) => ( - separated_nonempty_list!($i, $f, |i| $submac!(i, $($args)*)); + separated_list1!($i, $f, |i| $submac!(i, $($args)*)); ); ($i:expr, $f:expr, $g:expr) => ( - $crate::multi::separated_nonempty_listc($i, $f, $g) + $crate::multi::separated_list1c($i, $f, $g) ); ); /// `many0!(I -> IResult<I,O>) => I -> IResult<I, Vec<O>>` -/// Applies the parser 0 or more times and returns the list of results in a Vec. +/// Applies the parser 0 or more times and returns the list of results in a `Vec`. /// -/// The embedded parser may return Incomplete. +/// The embedded parser may return `Incomplete`. /// /// `many0` will only return `Error` if the embedded parser does not consume any input /// (to avoid infinite loops). @@ -102,6 +104,7 @@ macro_rules! separated_nonempty_list( /// ``` /// #[cfg(feature = "alloc")] +#[cfg_attr(feature = "docsrs", doc(cfg(feature = "alloc")))] #[macro_export(local_inner_macros)] macro_rules! many0( ($i:expr, $submac:ident!( $($args:tt)* )) => ( @@ -113,9 +116,9 @@ macro_rules! many0( ); /// `many1!(I -> IResult<I,O>) => I -> IResult<I, Vec<O>>` -/// Applies the parser 1 or more times and returns the list of results in a Vec +/// Applies the parser 1 or more times and returns the list of results in a `Vec`. /// -/// the embedded parser may return Incomplete +/// The embedded parser may return `Incomplete`. /// /// ``` /// # #[macro_use] extern crate nom; @@ -133,6 +136,7 @@ macro_rules! many0( /// # } /// ``` #[cfg(feature = "alloc")] +#[cfg_attr(feature = "docsrs", doc(cfg(feature = "alloc")))] #[macro_export(local_inner_macros)] macro_rules! many1( ($i:expr, $submac:ident!( $($args:tt)* )) => ( @@ -147,7 +151,7 @@ macro_rules! many1( /// Applies the first parser until the second applies. Returns a tuple containing the list /// of results from the first in a Vec and the result of the second. /// -/// The first embedded parser may return Incomplete +/// The first embedded parser may return `Incomplete`. /// /// ``` /// # #[macro_use] extern crate nom; @@ -169,6 +173,7 @@ macro_rules! many1( /// # } /// ``` #[cfg(feature = "alloc")] +#[cfg_attr(feature = "docsrs", doc(cfg(feature = "alloc")))] #[macro_export(local_inner_macros)] macro_rules! many_till( ($i:expr, $submac:ident!( $($args:tt)* ), $submac2:ident!( $($args2:tt)* )) => ( @@ -190,7 +195,7 @@ macro_rules! many_till( /// `many_m_n!(usize, usize, I -> IResult<I,O>) => I -> IResult<I, Vec<O>>` /// Applies the parser between m and n times (n included) and returns the list of -/// results in a Vec +/// results in a `Vec`. /// /// the embedded parser may return Incomplete /// @@ -213,6 +218,7 @@ macro_rules! many_till( /// # } /// ``` #[cfg(feature = "alloc")] +#[cfg_attr(feature = "docsrs", doc(cfg(feature = "alloc")))] #[macro_export(local_inner_macros)] macro_rules! many_m_n( ($i:expr, $m:expr, $n: expr, $submac:ident!( $($args:tt)* )) => ( @@ -277,7 +283,7 @@ macro_rules! many1_count { } /// `count!(I -> IResult<I,O>, nb) => I -> IResult<I, Vec<O>>` -/// Applies the child parser a specified number of times +/// Applies the child parser a specified number of times. /// /// ``` /// # #[macro_use] extern crate nom; @@ -296,6 +302,7 @@ macro_rules! many1_count { /// ``` /// #[cfg(feature = "alloc")] +#[cfg_attr(feature = "docsrs", doc(cfg(feature = "alloc")))] #[macro_export(local_inner_macros)] macro_rules! count( ($i:expr, $submac:ident!( $($args:tt)* ), $count: expr) => ( @@ -307,7 +314,7 @@ macro_rules! count( ); /// `length_count!(I -> IResult<I, nb>, I -> IResult<I,O>) => I -> IResult<I, Vec<O>>` -/// gets a number from the first parser, then applies the second parser that many times +/// Gets a number from the first parser, then applies the second parser that many times. /// /// ```rust /// # #[macro_use] extern crate nom; @@ -318,11 +325,12 @@ macro_rules! count( /// named!(parser<Vec<&[u8]>>, length_count!(be_u8, tag!("abc"))); /// /// assert_eq!(parser(&b"\x02abcabcabc"[..]), Ok(((&b"abc"[..], vec![&b"abc"[..], &b"abc"[..]])))); -/// assert_eq!(parser(&b"\x04abcabcabc"[..]), Err(Err::Incomplete(Needed::Size(3)))); +/// assert_eq!(parser(&b"\x04abcabcabc"[..]), Err(Err::Incomplete(Needed::new(3)))); /// # } /// ``` #[macro_export(local_inner_macros)] #[cfg(feature = "alloc")] +#[cfg_attr(feature = "docsrs", doc(cfg(feature = "alloc")))] macro_rules! length_count( ($i:expr, $submac:ident!( $($args:tt)* ), $submac2:ident!( $($args2:tt)* )) => ( { @@ -357,7 +365,7 @@ macro_rules! length_count( /// `length_data!(I -> IResult<I, nb>) => O` /// /// `length_data` gets a number from the first parser, then takes a subslice of the input -/// of that size and returns that subslice +/// of that size and returns that subslice. /// /// ```rust /// # #[macro_use] extern crate nom; @@ -368,7 +376,7 @@ macro_rules! length_count( /// named!(parser, length_data!(be_u8)); /// /// assert_eq!(parser(&b"\x06abcabcabc"[..]), Ok((&b"abc"[..], &b"abcabc"[..]))); -/// assert_eq!(parser(&b"\x06abc"[..]), Err(Err::Incomplete(Needed::Size(6)))); +/// assert_eq!(parser(&b"\x06abc"[..]), Err(Err::Incomplete(Needed::new(3)))); /// # } /// ``` #[macro_export(local_inner_macros)] @@ -386,7 +394,7 @@ macro_rules! length_data( /// /// Gets a number from the first parser, takes a subslice of the input of that size, /// then applies the second parser on that subslice. If the second parser returns -/// `Incomplete`, `length_value` will return an error +/// `Incomplete`, `length_value` will return an error. /// /// ```rust /// # #[macro_use] extern crate nom; @@ -399,7 +407,7 @@ macro_rules! length_data( /// named!(parser, length_value!(be_u8, alpha0)); /// /// assert_eq!(parser(&b"\x06abcabcabc"[..]), Ok((&b"abc"[..], &b"abcabc"[..]))); -/// assert_eq!(parser(&b"\x06abc"[..]), Err(Err::Incomplete(Needed::Size(6)))); +/// assert_eq!(parser(&b"\x06abc"[..]), Err(Err::Incomplete(Needed::new(3)))); /// # } /// ``` #[macro_export(local_inner_macros)] @@ -422,9 +430,9 @@ macro_rules! length_value( ); /// `fold_many0!(I -> IResult<I,O>, R, Fn(R, O) -> R) => I -> IResult<I, R>` -/// Applies the parser 0 or more times and folds the list of return values +/// Applies the parser 0 or more times and folds the list of return values. /// -/// the embedded parser may return Incomplete +/// The embedded parser may return `Incomplete`. /// /// ``` /// # #[macro_use] extern crate nom; @@ -450,14 +458,14 @@ macro_rules! fold_many0( fold_many0!($i, |i| $submac!(i, $($args)*), $init, $fold_f) ); ($i:expr, $f:expr, $init:expr, $fold_f:expr) => ( - $crate::multi::fold_many0($f, $init, $fold_f)($i) + $crate::multi::fold_many0c($i, $f, $init, $fold_f) ); ); /// `fold_many1!(I -> IResult<I,O>, R, Fn(R, O) -> R) => I -> IResult<I, R>` -/// Applies the parser 1 or more times and folds the list of return values +/// Applies the parser 1 or more times and folds the list of return values. /// -/// the embedded parser may return Incomplete +/// The embedded parser may return `Incomplete`. /// /// ``` /// # #[macro_use] extern crate nom; @@ -492,9 +500,9 @@ macro_rules! fold_many1( ); /// `fold_many_m_n!(usize, usize, I -> IResult<I,O>, R, Fn(R, O) -> R) => I -> IResult<I, R>` -/// Applies the parser between m and n times (n included) and folds the list of return value +/// Applies the parser between m and n times (n included) and folds the list of return value. /// -/// the embedded parser may return Incomplete +/// The embedded parser may return `Incomplete`. /// /// ``` /// # #[macro_use] extern crate nom; @@ -511,7 +519,7 @@ macro_rules! fold_many1( /// let b = b"abcdabcdefgh"; /// let c = b"abcdabcdabcdabcdabcdefgh"; /// -/// assert_eq!(multi(&a[..]), Err(Err::Error(error_position!(&a[..], ErrorKind::ManyMN)))); +/// assert_eq!(multi(&a[..]), Err(Err::Error(error_position!(&b"efgh"[..], ErrorKind::Tag)))); /// let res = vec![&b"abcd"[..], &b"abcd"[..]]; /// assert_eq!(multi(&b[..]),Ok((&b"efgh"[..], res))); /// let res2 = vec![&b"abcd"[..], &b"abcd"[..], &b"abcd"[..], &b"abcd"[..]]; @@ -530,14 +538,14 @@ macro_rules! fold_many_m_n( #[cfg(test)] mod tests { - use crate::internal::{Err, IResult, Needed}; + use crate::character::streaming::digit1 as digit; + use crate::error::ErrorKind; use crate::error::ParseError; + use crate::internal::{Err, IResult, Needed}; use crate::lib::std::str::{self, FromStr}; #[cfg(feature = "alloc")] use crate::lib::std::vec::Vec; - use crate::character::streaming::digit1 as digit; use crate::number::streaming::{be_u16, be_u8}; - use crate::error::ErrorKind; // reproduce the tag and take macros, because of module import order macro_rules! tag ( @@ -569,7 +577,7 @@ mod tests { let res: IResult<_,_,_> = if reduced != b { Err($crate::Err::Error($crate::error::make_error($i, $crate::error::ErrorKind::Tag))) } else if m < blen { - Err($crate::Err::Incomplete(Needed::Size(blen))) + Err($crate::Err::Incomplete(Needed::new(blen))) } else { Ok((&$i[blen..], reduced)) }; @@ -580,10 +588,11 @@ mod tests { #[test] #[cfg(feature = "alloc")] - fn separated_list() { - named!(multi<&[u8],Vec<&[u8]> >, separated_list!(tag!(","), tag!("abcd"))); - named!(multi_empty<&[u8],Vec<&[u8]> >, separated_list!(tag!(","), tag!(""))); - named!(multi_longsep<&[u8],Vec<&[u8]> >, separated_list!(tag!(".."), tag!("abcd"))); + fn separated_list0() { + named!(multi<&[u8],Vec<&[u8]> >, separated_list0!(tag!(","), tag!("abcd"))); + named!(multi_empty<&[u8],Vec<&[u8]> >, separated_list0!(tag!(","), tag!(""))); + named!(empty_sep<&[u8],Vec<&[u8]> >, separated_list0!(tag!(""), tag!("abc"))); + named!(multi_longsep<&[u8],Vec<&[u8]> >, separated_list0!(tag!(".."), tag!("abcd"))); let a = &b"abcdef"[..]; let b = &b"abcd,abcdef"[..]; @@ -593,28 +602,36 @@ mod tests { let f = &b"abc"[..]; let g = &b"abcd."[..]; let h = &b"abcd,abc"[..]; + let i = &b"abcabc"[..]; let res1 = vec![&b"abcd"[..]]; assert_eq!(multi(a), Ok((&b"ef"[..], res1))); let res2 = vec![&b"abcd"[..], &b"abcd"[..]]; assert_eq!(multi(b), Ok((&b"ef"[..], res2))); assert_eq!(multi(c), Ok((&b"azerty"[..], Vec::new()))); - assert_eq!(multi_empty(d), Err(Err::Error(error_position!(d, ErrorKind::SeparatedList)))); - //let res3 = vec![&b""[..], &b""[..], &b""[..]]; - //assert_eq!(multi_empty(d),Ok((&b"abc"[..], res3))); + let res3 = vec![&b""[..], &b""[..], &b""[..]]; + assert_eq!(multi_empty(d), Ok((&b"abc"[..], res3))); + let i_err_pos = &i[3..]; + assert_eq!( + empty_sep(i), + Err(Err::Error(error_position!( + i_err_pos, + ErrorKind::SeparatedList + ))) + ); let res4 = vec![&b"abcd"[..], &b"abcd"[..]]; assert_eq!(multi(e), Ok((&b",ef"[..], res4))); - assert_eq!(multi(f), Err(Err::Incomplete(Needed::Size(4)))); - assert_eq!(multi_longsep(g), Err(Err::Incomplete(Needed::Size(2)))); - assert_eq!(multi(h), Err(Err::Incomplete(Needed::Size(4)))); + assert_eq!(multi(f), Err(Err::Incomplete(Needed::new(4)))); + assert_eq!(multi_longsep(g), Err(Err::Incomplete(Needed::new(2)))); + assert_eq!(multi(h), Err(Err::Incomplete(Needed::new(4)))); } #[test] #[cfg(feature = "alloc")] - fn separated_nonempty_list() { - named!(multi<&[u8],Vec<&[u8]> >, separated_nonempty_list!(tag!(","), tag!("abcd"))); - named!(multi_longsep<&[u8],Vec<&[u8]> >, separated_nonempty_list!(tag!(".."), tag!("abcd"))); + fn separated_list1() { + named!(multi<&[u8],Vec<&[u8]> >, separated_list1!(tag!(","), tag!("abcd"))); + named!(multi_longsep<&[u8],Vec<&[u8]> >, separated_list1!(tag!(".."), tag!("abcd"))); let a = &b"abcdef"[..]; let b = &b"abcd,abcdef"[..]; @@ -629,13 +646,16 @@ mod tests { assert_eq!(multi(a), Ok((&b"ef"[..], res1))); let res2 = vec![&b"abcd"[..], &b"abcd"[..]]; assert_eq!(multi(b), Ok((&b"ef"[..], res2))); - assert_eq!(multi(c), Err(Err::Error(error_position!(c, ErrorKind::Tag)))); + assert_eq!( + multi(c), + Err(Err::Error(error_position!(c, ErrorKind::Tag))) + ); let res3 = vec![&b"abcd"[..], &b"abcd"[..]]; assert_eq!(multi(d), Ok((&b",ef"[..], res3))); - assert_eq!(multi(f), Err(Err::Incomplete(Needed::Size(4)))); - assert_eq!(multi_longsep(g), Err(Err::Incomplete(Needed::Size(2)))); - assert_eq!(multi(h), Err(Err::Incomplete(Needed::Size(4)))); + assert_eq!(multi(f), Err(Err::Incomplete(Needed::new(4)))); + assert_eq!(multi_longsep(g), Err(Err::Incomplete(Needed::new(2)))); + assert_eq!(multi(h), Err(Err::Incomplete(Needed::new(4)))); } #[test] @@ -647,14 +667,20 @@ mod tests { named!( multi_empty<&[u8],Vec<&[u8]> >, many0!(tag_empty) ); assert_eq!(multi(&b"abcdef"[..]), Ok((&b"ef"[..], vec![&b"abcd"[..]]))); - assert_eq!(multi(&b"abcdabcdefgh"[..]), Ok((&b"efgh"[..], vec![&b"abcd"[..], &b"abcd"[..]]))); + assert_eq!( + multi(&b"abcdabcdefgh"[..]), + Ok((&b"efgh"[..], vec![&b"abcd"[..], &b"abcd"[..]])) + ); assert_eq!(multi(&b"azerty"[..]), Ok((&b"azerty"[..], Vec::new()))); - assert_eq!(multi(&b"abcdab"[..]), Err(Err::Incomplete(Needed::Size(4)))); - assert_eq!(multi(&b"abcd"[..]), Err(Err::Incomplete(Needed::Size(4)))); - assert_eq!(multi(&b""[..]), Err(Err::Incomplete(Needed::Size(4)))); + assert_eq!(multi(&b"abcdab"[..]), Err(Err::Incomplete(Needed::new(4)))); + assert_eq!(multi(&b"abcd"[..]), Err(Err::Incomplete(Needed::new(4)))); + assert_eq!(multi(&b""[..]), Err(Err::Incomplete(Needed::new(4)))); assert_eq!( multi_empty(&b"abcdef"[..]), - Err(Err::Error(error_position!(&b"abcdef"[..], ErrorKind::Many0))) + Err(Err::Error(error_position!( + &b"abcdef"[..], + ErrorKind::Many0 + ))) ); } @@ -682,8 +708,11 @@ mod tests { assert_eq!(multi(a), Ok((&b"ef"[..], res1))); let res2 = vec![&b"abcd"[..], &b"abcd"[..]]; assert_eq!(multi(b), Ok((&b"efgh"[..], res2))); - assert_eq!(multi(c), Err(Err::Error(error_position!(c, ErrorKind::Tag)))); - assert_eq!(multi(d), Err(Err::Incomplete(Needed::Size(4)))); + assert_eq!( + multi(c), + Err(Err::Error(error_position!(c, ErrorKind::Tag))) + ); + assert_eq!(multi(d), Err(Err::Incomplete(Needed::new(4)))); } #[test] @@ -724,7 +753,10 @@ mod tests { named!(multi1<&[u8],Vec<&[u8]> >, many1!(tst)); let a = &b"abcdef"[..]; - assert_eq!(multi1(a), Err(Err::Error(error_position!(a, ErrorKind::Tag)))); + assert_eq!( + multi1(a), + Err(Err::Error(error_position!(a, ErrorKind::Tag))) + ); } #[test] @@ -738,14 +770,17 @@ mod tests { let d = &b"AbcdAbcdAbcdAbcdAbcdefgh"[..]; let e = &b"AbcdAb"[..]; - assert_eq!(multi(a), Err(Err::Error(error_position!(&b"ef"[..], ErrorKind::Tag)))); + assert_eq!( + multi(a), + Err(Err::Error(error_position!(&b"ef"[..], ErrorKind::Tag))) + ); let res1 = vec![&b"Abcd"[..], &b"Abcd"[..]]; assert_eq!(multi(b), Ok((&b"efgh"[..], res1))); let res2 = vec![&b"Abcd"[..], &b"Abcd"[..], &b"Abcd"[..], &b"Abcd"[..]]; assert_eq!(multi(c), Ok((&b"efgh"[..], res2))); let res3 = vec![&b"Abcd"[..], &b"Abcd"[..], &b"Abcd"[..], &b"Abcd"[..]]; assert_eq!(multi(d), Ok((&b"Abcdefgh"[..], res3))); - assert_eq!(multi(e), Err(Err::Incomplete(Needed::Size(4)))); + assert_eq!(multi(e), Err(Err::Incomplete(Needed::new(4)))); } #[test] @@ -755,17 +790,29 @@ mod tests { named!(tag_abc, tag!("abc")); named!( cnt_2<&[u8], Vec<&[u8]> >, count!(tag_abc, TIMES ) ); - assert_eq!(cnt_2(&b"abcabcabcdef"[..]), Ok((&b"abcdef"[..], vec![&b"abc"[..], &b"abc"[..]]))); - assert_eq!(cnt_2(&b"ab"[..]), Err(Err::Incomplete(Needed::Size(3)))); - assert_eq!(cnt_2(&b"abcab"[..]), Err(Err::Incomplete(Needed::Size(3)))); - assert_eq!(cnt_2(&b"xxx"[..]), Err(Err::Error(error_position!(&b"xxx"[..], ErrorKind::Tag)))); + assert_eq!( + cnt_2(&b"abcabcabcdef"[..]), + Ok((&b"abcdef"[..], vec![&b"abc"[..], &b"abc"[..]])) + ); + assert_eq!(cnt_2(&b"ab"[..]), Err(Err::Incomplete(Needed::new(3)))); + assert_eq!(cnt_2(&b"abcab"[..]), Err(Err::Incomplete(Needed::new(3)))); + assert_eq!( + cnt_2(&b"xxx"[..]), + Err(Err::Error(error_position!(&b"xxx"[..], ErrorKind::Tag))) + ); assert_eq!( cnt_2(&b"xxxabcabcdef"[..]), - Err(Err::Error(error_position!(&b"xxxabcabcdef"[..], ErrorKind::Tag))) + Err(Err::Error(error_position!( + &b"xxxabcabcdef"[..], + ErrorKind::Tag + ))) ); assert_eq!( cnt_2(&b"abcxxxabcdef"[..]), - Err(Err::Error(error_position!(&b"xxxabcdef"[..], ErrorKind::Tag))) + Err(Err::Error(error_position!( + &b"xxxabcdef"[..], + ErrorKind::Tag + ))) ); } @@ -794,8 +841,14 @@ mod tests { let error_2_remain = &b"abcxxxabcdef"[..]; assert_eq!(counter_2(done), Ok((rest, parsed_done))); - assert_eq!(counter_2(incomplete_1), Ok((incomplete_1, parsed_incompl_1))); - assert_eq!(counter_2(incomplete_2), Ok((incomplete_2, parsed_incompl_2))); + assert_eq!( + counter_2(incomplete_1), + Ok((incomplete_1, parsed_incompl_1)) + ); + assert_eq!( + counter_2(incomplete_2), + Ok((incomplete_2, parsed_incompl_2)) + ); assert_eq!(counter_2(error), Ok((error_remain, parsed_err))); assert_eq!(counter_2(error_1), Ok((error_1_remain, parsed_err_1))); assert_eq!(counter_2(error_2), Ok((error_2_remain, parsed_err_2))); @@ -804,7 +857,7 @@ mod tests { #[derive(Debug, Clone, PartialEq)] pub struct NilError; - impl<I> From<(I,ErrorKind)> for NilError { + impl<I> From<(I, ErrorKind)> for NilError { fn from(_: (I, ErrorKind)) -> Self { NilError } @@ -833,10 +886,16 @@ mod tests { named!(tag_abc, tag!(&b"abc"[..])); named!( cnt<&[u8], Vec<&[u8]> >, length_count!(number, tag_abc) ); - assert_eq!(cnt(&b"2abcabcabcdef"[..]), Ok((&b"abcdef"[..], vec![&b"abc"[..], &b"abc"[..]]))); - assert_eq!(cnt(&b"2ab"[..]), Err(Err::Incomplete(Needed::Size(3)))); - assert_eq!(cnt(&b"3abcab"[..]), Err(Err::Incomplete(Needed::Size(3)))); - assert_eq!(cnt(&b"xxx"[..]), Err(Err::Error(error_position!(&b"xxx"[..], ErrorKind::Digit)))); + assert_eq!( + cnt(&b"2abcabcabcdef"[..]), + Ok((&b"abcdef"[..], vec![&b"abc"[..], &b"abc"[..]])) + ); + assert_eq!(cnt(&b"2ab"[..]), Err(Err::Incomplete(Needed::new(3)))); + assert_eq!(cnt(&b"3abcab"[..]), Err(Err::Incomplete(Needed::new(3)))); + assert_eq!( + cnt(&b"xxx"[..]), + Err(Err::Error(error_position!(&b"xxx"[..], ErrorKind::Digit))) + ); assert_eq!( cnt(&b"2abcxxx"[..]), Err(Err::Error(error_position!(&b"xxx"[..], ErrorKind::Tag))) @@ -847,9 +906,15 @@ mod tests { fn length_data() { named!( take<&[u8], &[u8]>, length_data!(number) ); - assert_eq!(take(&b"6abcabcabcdef"[..]), Ok((&b"abcdef"[..], &b"abcabc"[..]))); - assert_eq!(take(&b"3ab"[..]), Err(Err::Incomplete(Needed::Size(3)))); - assert_eq!(take(&b"xxx"[..]), Err(Err::Error(error_position!(&b"xxx"[..], ErrorKind::Digit)))); + assert_eq!( + take(&b"6abcabcabcdef"[..]), + Ok((&b"abcdef"[..], &b"abcabc"[..])) + ); + assert_eq!(take(&b"3ab"[..]), Err(Err::Incomplete(Needed::new(1)))); + assert_eq!( + take(&b"xxx"[..]), + Err(Err::Error(error_position!(&b"xxx"[..], ErrorKind::Digit))) + ); assert_eq!(take(&b"2abcxxx"[..]), Ok((&b"cxxx"[..], &b"ab"[..]))); } @@ -859,8 +924,14 @@ mod tests { named!(length_value_2<&[u8], (u8, u8) >, length_value!(be_u8, tuple!(be_u8, be_u8))); let i1 = [0, 5, 6]; - assert_eq!(length_value_1(&i1), Err(Err::Error(error_position!(&b""[..], ErrorKind::Complete)))); - assert_eq!(length_value_2(&i1), Err(Err::Error(error_position!(&b""[..], ErrorKind::Complete)))); + assert_eq!( + length_value_1(&i1), + Err(Err::Error(error_position!(&b""[..], ErrorKind::Complete))) + ); + assert_eq!( + length_value_2(&i1), + Err(Err::Error(error_position!(&b""[..], ErrorKind::Complete))) + ); let i2 = [1, 5, 6, 3]; assert_eq!( @@ -887,21 +958,27 @@ mod tests { fn fold_into_vec<T>(mut acc: Vec<T>, item: T) -> Vec<T> { acc.push(item); acc - }; + } named!(tag_abcd, tag!("abcd")); named!(tag_empty, tag!("")); named!( multi<&[u8],Vec<&[u8]> >, fold_many0!(tag_abcd, Vec::new(), fold_into_vec) ); named!( multi_empty<&[u8],Vec<&[u8]> >, fold_many0!(tag_empty, Vec::new(), fold_into_vec) ); assert_eq!(multi(&b"abcdef"[..]), Ok((&b"ef"[..], vec![&b"abcd"[..]]))); - assert_eq!(multi(&b"abcdabcdefgh"[..]), Ok((&b"efgh"[..], vec![&b"abcd"[..], &b"abcd"[..]]))); + assert_eq!( + multi(&b"abcdabcdefgh"[..]), + Ok((&b"efgh"[..], vec![&b"abcd"[..], &b"abcd"[..]])) + ); assert_eq!(multi(&b"azerty"[..]), Ok((&b"azerty"[..], Vec::new()))); - assert_eq!(multi(&b"abcdab"[..]), Err(Err::Incomplete(Needed::Size(4)))); - assert_eq!(multi(&b"abcd"[..]), Err(Err::Incomplete(Needed::Size(4)))); - assert_eq!(multi(&b""[..]), Err(Err::Incomplete(Needed::Size(4)))); + assert_eq!(multi(&b"abcdab"[..]), Err(Err::Incomplete(Needed::new(4)))); + assert_eq!(multi(&b"abcd"[..]), Err(Err::Incomplete(Needed::new(4)))); + assert_eq!(multi(&b""[..]), Err(Err::Incomplete(Needed::new(4)))); assert_eq!( multi_empty(&b"abcdef"[..]), - Err(Err::Error(error_position!(&b"abcdef"[..], ErrorKind::Many0))) + Err(Err::Error(error_position!( + &b"abcdef"[..], + ErrorKind::Many0 + ))) ); } @@ -911,7 +988,7 @@ mod tests { fn fold_into_vec<T>(mut acc: Vec<T>, item: T) -> Vec<T> { acc.push(item); acc - }; + } named!(multi<&[u8],Vec<&[u8]> >, fold_many1!(tag!("abcd"), Vec::new(), fold_into_vec)); let a = &b"abcdef"[..]; @@ -923,8 +1000,11 @@ mod tests { assert_eq!(multi(a), Ok((&b"ef"[..], res1))); let res2 = vec![&b"abcd"[..], &b"abcd"[..]]; assert_eq!(multi(b), Ok((&b"efgh"[..], res2))); - assert_eq!(multi(c), Err(Err::Error(error_position!(c, ErrorKind::Many1)))); - assert_eq!(multi(d), Err(Err::Incomplete(Needed::Size(4)))); + assert_eq!( + multi(c), + Err(Err::Error(error_position!(c, ErrorKind::Many1))) + ); + assert_eq!(multi(d), Err(Err::Incomplete(Needed::new(4)))); } #[test] @@ -933,7 +1013,7 @@ mod tests { fn fold_into_vec<T>(mut acc: Vec<T>, item: T) -> Vec<T> { acc.push(item); acc - }; + } named!(multi<&[u8],Vec<&[u8]> >, fold_many_m_n!(2, 4, tag!("Abcd"), Vec::new(), fold_into_vec)); let a = &b"Abcdef"[..]; @@ -942,14 +1022,17 @@ mod tests { let d = &b"AbcdAbcdAbcdAbcdAbcdefgh"[..]; let e = &b"AbcdAb"[..]; - assert_eq!(multi(a), Err(Err::Error(error_position!(a, ErrorKind::ManyMN)))); + assert_eq!( + multi(a), + Err(Err::Error(error_position!(&b"ef"[..], ErrorKind::Tag))) + ); let res1 = vec![&b"Abcd"[..], &b"Abcd"[..]]; assert_eq!(multi(b), Ok((&b"efgh"[..], res1))); let res2 = vec![&b"Abcd"[..], &b"Abcd"[..], &b"Abcd"[..], &b"Abcd"[..]]; assert_eq!(multi(c), Ok((&b"efgh"[..], res2))); let res3 = vec![&b"Abcd"[..], &b"Abcd"[..], &b"Abcd"[..], &b"Abcd"[..]]; assert_eq!(multi(d), Ok((&b"Abcdefgh"[..], res3))); - assert_eq!(multi(e), Err(Err::Incomplete(Needed::Size(4)))); + assert_eq!(multi(e), Err(Err::Incomplete(Needed::new(4)))); } #[test] @@ -963,7 +1046,10 @@ mod tests { assert_eq!(count0_nums(&b"123,45,junk"[..]), Ok((&b"junk"[..], 2))); - assert_eq!(count0_nums(&b"1,2,3,4,5,6,7,8,9,0,junk"[..]), Ok((&b"junk"[..], 10))); + assert_eq!( + count0_nums(&b"1,2,3,4,5,6,7,8,9,0,junk"[..]), + Ok((&b"junk"[..], 10)) + ); assert_eq!(count0_nums(&b"hello"[..]), Ok((&b"hello"[..], 0))); } @@ -977,12 +1063,17 @@ mod tests { assert_eq!(count1_nums(&b"123,45,junk"[..]), Ok((&b"junk"[..], 2))); - assert_eq!(count1_nums(&b"1,2,3,4,5,6,7,8,9,0,junk"[..]), Ok((&b"junk"[..], 10))); + assert_eq!( + count1_nums(&b"1,2,3,4,5,6,7,8,9,0,junk"[..]), + Ok((&b"junk"[..], 10)) + ); assert_eq!( count1_nums(&b"hello"[..]), - Err(Err::Error(error_position!(&b"hello"[..], ErrorKind::Many1Count))) + Err(Err::Error(error_position!( + &b"hello"[..], + ErrorKind::Many1Count + ))) ); } - } diff --git a/src/multi/mod.rs b/src/multi/mod.rs index e8f26dd..46a59ed 100644 --- a/src/multi/mod.rs +++ b/src/multi/mod.rs @@ -1,14 +1,15 @@ -//! combinators applying their child parser multiple times +//! Combinators applying their child parser multiple times #[macro_use] mod macros; -use crate::internal::{Err, IResult, Needed}; +use crate::error::ErrorKind; use crate::error::ParseError; -use crate::traits::{InputLength, InputTake, ToUsize}; +use crate::internal::{Err, IResult, Needed, Parser}; #[cfg(feature = "alloc")] use crate::lib::std::vec::Vec; -use crate::error::ErrorKind; +use crate::traits::{InputLength, InputTake, ToUsize}; +use core::num::NonZeroUsize; /// Repeats the embedded parser until it fails /// and returns the results in a `Vec`. @@ -35,17 +36,17 @@ use crate::error::ErrorKind; /// assert_eq!(parser(""), Ok(("", vec![]))); /// ``` #[cfg(feature = "alloc")] -pub fn many0<I, O, E, F>(f: F) -> impl Fn(I) -> IResult<I, Vec<O>, E> +#[cfg_attr(feature = "docsrs", doc(cfg(feature = "alloc")))] +pub fn many0<I, O, E, F>(mut f: F) -> impl FnMut(I) -> IResult<I, Vec<O>, E> where I: Clone + PartialEq, - F: Fn(I) -> IResult<I, O, E>, + F: Parser<I, O, E>, E: ParseError<I>, { - move |i: I| { + move |mut i: I| { let mut acc = crate::lib::std::vec::Vec::with_capacity(4); - let mut i = i.clone(); loop { - match f(i.clone()) { + match f.parse(i.clone()) { Err(Err::Error(_)) => return Ok((i, acc)), Err(e) => return Err(e), Ok((i1, o)) => { @@ -63,6 +64,7 @@ where // this implementation is used for type inference issues in macros #[doc(hidden)] #[cfg(feature = "alloc")] +#[cfg_attr(feature = "docsrs", doc(cfg(feature = "alloc")))] pub fn many0c<I, O, E, F>(input: I, f: F) -> IResult<I, Vec<O>, E> where I: Clone + PartialEq, @@ -80,12 +82,12 @@ where /// # Arguments /// * `f` The parser to apply. /// -/// *Note*: if the parser passed to `many1` accepts empty inputs +/// *Note*: If the parser passed to `many1` accepts empty inputs /// (like `alpha0` or `digit0`), `many1` will return an error, -/// to prevent going into an infinite loop +/// to prevent going into an infinite loop. /// /// ```rust -/// # use nom::{Err, error::ErrorKind, Needed, IResult}; +/// # use nom::{Err, error::{Error, ErrorKind}, Needed, IResult}; /// use nom::multi::many1; /// use nom::bytes::complete::tag; /// @@ -95,38 +97,36 @@ where /// /// assert_eq!(parser("abcabc"), Ok(("", vec!["abc", "abc"]))); /// assert_eq!(parser("abc123"), Ok(("123", vec!["abc"]))); -/// assert_eq!(parser("123123"), Err(Err::Error(("123123", ErrorKind::Tag)))); -/// assert_eq!(parser(""), Err(Err::Error(("", ErrorKind::Tag)))); +/// assert_eq!(parser("123123"), Err(Err::Error(Error::new("123123", ErrorKind::Tag)))); +/// assert_eq!(parser(""), Err(Err::Error(Error::new("", ErrorKind::Tag)))); /// ``` #[cfg(feature = "alloc")] -pub fn many1<I, O, E, F>(f: F) -> impl Fn(I) -> IResult<I, Vec<O>, E> +#[cfg_attr(feature = "docsrs", doc(cfg(feature = "alloc")))] +pub fn many1<I, O, E, F>(mut f: F) -> impl FnMut(I) -> IResult<I, Vec<O>, E> where I: Clone + PartialEq, - F: Fn(I) -> IResult<I, O, E>, + F: Parser<I, O, E>, E: ParseError<I>, { - move |i: I| { - let mut i = i.clone(); - match f(i.clone()) { - Err(Err::Error(err)) => return Err(Err::Error(E::append(i, ErrorKind::Many1, err))), - Err(e) => return Err(e), - Ok((i1, o)) => { - let mut acc = crate::lib::std::vec::Vec::with_capacity(4); - acc.push(o); - i = i1; - - loop { - match f(i.clone()) { - Err(Err::Error(_)) => return Ok((i, acc)), - Err(e) => return Err(e), - Ok((i1, o)) => { - if i1 == i { - return Err(Err::Error(E::from_error_kind(i, ErrorKind::Many1))); - } - - i = i1; - acc.push(o); + move |mut i: I| match f.parse(i.clone()) { + Err(Err::Error(err)) => Err(Err::Error(E::append(i, ErrorKind::Many1, err))), + Err(e) => Err(e), + Ok((i1, o)) => { + let mut acc = crate::lib::std::vec::Vec::with_capacity(4); + acc.push(o); + i = i1; + + loop { + match f.parse(i.clone()) { + Err(Err::Error(_)) => return Ok((i, acc)), + Err(e) => return Err(e), + Ok((i1, o)) => { + if i1 == i { + return Err(Err::Error(E::from_error_kind(i, ErrorKind::Many1))); } + + i = i1; + acc.push(o); } } } @@ -137,9 +137,10 @@ where // this implementation is used for type inference issues in macros #[doc(hidden)] #[cfg(feature = "alloc")] +#[cfg_attr(feature = "docsrs", doc(cfg(feature = "alloc")))] pub fn many1c<I, O, E, F>(input: I, f: F) -> IResult<I, Vec<O>, E> where - I: Clone + Copy + PartialEq, + I: Clone + PartialEq, F: Fn(I) -> IResult<I, O, E>, E: ParseError<I>, { @@ -150,7 +151,7 @@ where /// a result. Returns a pair consisting of the results of /// `f` in a `Vec` and the result of `g`. /// ```rust -/// # use nom::{Err, error::ErrorKind, Needed, IResult}; +/// # use nom::{Err, error::{Error, ErrorKind}, Needed, IResult}; /// use nom::multi::many_till; /// use nom::bytes::complete::tag; /// @@ -159,29 +160,31 @@ where /// }; /// /// assert_eq!(parser("abcabcend"), Ok(("", (vec!["abc", "abc"], "end")))); -/// assert_eq!(parser("abc123end"), Err(Err::Error(("123end", ErrorKind::Tag)))); -/// assert_eq!(parser("123123end"), Err(Err::Error(("123123end", ErrorKind::Tag)))); -/// assert_eq!(parser(""), Err(Err::Error(("", ErrorKind::Tag)))); +/// assert_eq!(parser("abc123end"), Err(Err::Error(Error::new("123end", ErrorKind::Tag)))); +/// assert_eq!(parser("123123end"), Err(Err::Error(Error::new("123123end", ErrorKind::Tag)))); +/// assert_eq!(parser(""), Err(Err::Error(Error::new("", ErrorKind::Tag)))); /// assert_eq!(parser("abcendefg"), Ok(("efg", (vec!["abc"], "end")))); /// ``` #[cfg(feature = "alloc")] -pub fn many_till<I, O, P, E, F, G>(f: F, g: G) -> impl Fn(I) -> IResult<I, (Vec<O>, P), E> +#[cfg_attr(feature = "docsrs", doc(cfg(feature = "alloc")))] +pub fn many_till<I, O, P, E, F, G>( + mut f: F, + mut g: G, +) -> impl FnMut(I) -> IResult<I, (Vec<O>, P), E> where I: Clone + PartialEq, - F: Fn(I) -> IResult<I, O, E>, - G: Fn(I) -> IResult<I, P, E>, + F: Parser<I, O, E>, + G: Parser<I, P, E>, E: ParseError<I>, { - move |i: I| { + move |mut i: I| { let mut res = crate::lib::std::vec::Vec::new(); - let mut i = i.clone(); loop { - match g(i.clone()) { + match g.parse(i.clone()) { Ok((i1, o)) => return Ok((i1, (res, o))), Err(Err::Error(_)) => { - match f(i.clone()) { - Err(Err::Error(err)) => - return Err(Err::Error(E::append(i, ErrorKind::ManyTill, err))), + match f.parse(i.clone()) { + Err(Err::Error(err)) => return Err(Err::Error(E::append(i, ErrorKind::ManyTill, err))), Err(e) => return Err(e), Ok((i1, o)) => { // loop trip must always consume (otherwise infinite loops) @@ -193,7 +196,7 @@ where i = i1; } } - }, + } Err(e) => return Err(e), } } @@ -203,6 +206,7 @@ where // this implementation is used for type inference issues in macros #[doc(hidden)] #[cfg(feature = "alloc")] +#[cfg_attr(feature = "docsrs", doc(cfg(feature = "alloc")))] pub fn many_tillc<I, O, P, E, F, G>(i: I, f: F, g: G) -> IResult<I, (Vec<O>, P), E> where I: Clone + PartialEq, @@ -221,11 +225,11 @@ where /// /// ```rust /// # use nom::{Err, error::ErrorKind, Needed, IResult}; -/// use nom::multi::separated_list; +/// use nom::multi::separated_list0; /// use nom::bytes::complete::tag; /// /// fn parser(s: &str) -> IResult<&str, Vec<&str>> { -/// separated_list(tag("|"), tag("abc"))(s) +/// separated_list0(tag("|"), tag("abc"))(s) /// } /// /// assert_eq!(parser("abc|abc|abc"), Ok(("", vec!["abc", "abc", "abc"]))); @@ -235,32 +239,31 @@ where /// assert_eq!(parser("def|abc"), Ok(("def|abc", vec![]))); /// ``` #[cfg(feature = "alloc")] -pub fn separated_list<I, O, O2, E, F, G>(sep: G, f: F) -> impl Fn(I) -> IResult<I, Vec<O>, E> +#[cfg_attr(feature = "docsrs", doc(cfg(feature = "alloc")))] +pub fn separated_list0<I, O, O2, E, F, G>( + mut sep: G, + mut f: F, +) -> impl FnMut(I) -> IResult<I, Vec<O>, E> where I: Clone + PartialEq, - F: Fn(I) -> IResult<I, O, E>, - G: Fn(I) -> IResult<I, O2, E>, + F: Parser<I, O, E>, + G: Parser<I, O2, E>, E: ParseError<I>, { - move |i: I| { + move |mut i: I| { let mut res = Vec::new(); - let mut i = i.clone(); - match f(i.clone()) { + match f.parse(i.clone()) { Err(Err::Error(_)) => return Ok((i, res)), Err(e) => return Err(e), Ok((i1, o)) => { - if i1 == i { - return Err(Err::Error(E::from_error_kind(i1, ErrorKind::SeparatedList))); - } - res.push(o); i = i1; } } loop { - match sep(i.clone()) { + match sep.parse(i.clone()) { Err(Err::Error(_)) => return Ok((i, res)), Err(e) => return Err(e), Ok((i1, _)) => { @@ -268,14 +271,10 @@ where return Err(Err::Error(E::from_error_kind(i1, ErrorKind::SeparatedList))); } - match f(i1.clone()) { + match f.parse(i1.clone()) { Err(Err::Error(_)) => return Ok((i, res)), Err(e) => return Err(e), Ok((i2, o)) => { - if i2 == i { - return Err(Err::Error(E::from_error_kind(i2, ErrorKind::SeparatedList))); - } - res.push(o); i = i2; } @@ -289,14 +288,15 @@ where // this implementation is used for type inference issues in macros #[doc(hidden)] #[cfg(feature = "alloc")] -pub fn separated_listc<I, O, O2, E, F, G>(i: I, sep: G, f: F) -> IResult<I, Vec<O>, E> +#[cfg_attr(feature = "docsrs", doc(cfg(feature = "alloc")))] +pub fn separated_list0c<I, O, O2, E, F, G>(i: I, sep: G, f: F) -> IResult<I, Vec<O>, E> where I: Clone + PartialEq, F: Fn(I) -> IResult<I, O, E>, G: Fn(I) -> IResult<I, O2, E>, E: ParseError<I>, { - separated_list(sep, f)(i) + separated_list0(sep, f)(i) } /// Alternates between two parsers to produce @@ -307,47 +307,46 @@ where /// * `f` Parses the elements of the list. /// ```rust /// # #[macro_use] extern crate nom; -/// # use nom::{Err, error::ErrorKind, Needed, IResult}; -/// use nom::multi::separated_nonempty_list; +/// # use nom::{Err, error::{Error, ErrorKind}, Needed, IResult}; +/// use nom::multi::separated_list1; /// use nom::bytes::complete::tag; /// /// fn parser(s: &str) -> IResult<&str, Vec<&str>> { -/// separated_nonempty_list(tag("|"), tag("abc"))(s) +/// separated_list1(tag("|"), tag("abc"))(s) /// } /// /// assert_eq!(parser("abc|abc|abc"), Ok(("", vec!["abc", "abc", "abc"]))); /// assert_eq!(parser("abc123abc"), Ok(("123abc", vec!["abc"]))); /// assert_eq!(parser("abc|def"), Ok(("|def", vec!["abc"]))); -/// assert_eq!(parser(""), Err(Err::Error(("", ErrorKind::Tag)))); -/// assert_eq!(parser("def|abc"), Err(Err::Error(("def|abc", ErrorKind::Tag)))); +/// assert_eq!(parser(""), Err(Err::Error(Error::new("", ErrorKind::Tag)))); +/// assert_eq!(parser("def|abc"), Err(Err::Error(Error::new("def|abc", ErrorKind::Tag)))); /// ``` #[cfg(feature = "alloc")] -pub fn separated_nonempty_list<I, O, O2, E, F, G>(sep: G, f: F) -> impl Fn(I) -> IResult<I, Vec<O>, E> +#[cfg_attr(feature = "docsrs", doc(cfg(feature = "alloc")))] +pub fn separated_list1<I, O, O2, E, F, G>( + mut sep: G, + mut f: F, +) -> impl FnMut(I) -> IResult<I, Vec<O>, E> where I: Clone + PartialEq, - F: Fn(I) -> IResult<I, O, E>, - G: Fn(I) -> IResult<I, O2, E>, + F: Parser<I, O, E>, + G: Parser<I, O2, E>, E: ParseError<I>, { - move |i: I| { + move |mut i: I| { let mut res = Vec::new(); - let mut i = i.clone(); // Parse the first element - match f(i.clone()) { - Err(e)=> return Err(e), + match f.parse(i.clone()) { + Err(e) => return Err(e), Ok((i1, o)) => { - if i1 == i { - return Err(Err::Error(E::from_error_kind(i1, ErrorKind::SeparatedList))); - } - res.push(o); i = i1; } } loop { - match sep(i.clone()) { + match sep.parse(i.clone()) { Err(Err::Error(_)) => return Ok((i, res)), Err(e) => return Err(e), Ok((i1, _)) => { @@ -355,14 +354,10 @@ where return Err(Err::Error(E::from_error_kind(i1, ErrorKind::SeparatedList))); } - match f(i1.clone()) { + match f.parse(i1.clone()) { Err(Err::Error(_)) => return Ok((i, res)), Err(e) => return Err(e), Ok((i2, o)) => { - if i2 == i { - return Err(Err::Error(E::from_error_kind(i2, ErrorKind::SeparatedList))); - } - res.push(o); i = i2; } @@ -376,14 +371,15 @@ where // this implementation is used for type inference issues in macros #[doc(hidden)] #[cfg(feature = "alloc")] -pub fn separated_nonempty_listc<I, O, O2, E, F, G>(i: I, sep: G, f: F) -> IResult<I, Vec<O>, E> +#[cfg_attr(feature = "docsrs", doc(cfg(feature = "alloc")))] +pub fn separated_list1c<I, O, O2, E, F, G>(i: I, sep: G, f: F) -> IResult<I, Vec<O>, E> where I: Clone + PartialEq, F: Fn(I) -> IResult<I, O, E>, G: Fn(I) -> IResult<I, O2, E>, E: ParseError<I>, { - separated_nonempty_list(sep, f)(i) + separated_list1(sep, f)(i) } /// Repeats the embedded parser `n` times or until it fails @@ -410,40 +406,33 @@ where /// assert_eq!(parser("abcabcabc"), Ok(("abc", vec!["abc", "abc"]))); /// ``` #[cfg(feature = "alloc")] -pub fn many_m_n<I, O, E, F>(m: usize, n: usize, f: F) -> impl Fn(I) -> IResult<I, Vec<O>, E> +#[cfg_attr(feature = "docsrs", doc(cfg(feature = "alloc")))] +pub fn many_m_n<I, O, E, F>( + min: usize, + max: usize, + mut parse: F, +) -> impl FnMut(I) -> IResult<I, Vec<O>, E> where I: Clone + PartialEq, - F: Fn(I) -> IResult<I, O, E>, + F: Parser<I, O, E>, E: ParseError<I>, { - move |i: I| { - let mut res = crate::lib::std::vec::Vec::with_capacity(m); - let mut input = i.clone(); - let mut count: usize = 0; + move |mut input: I| { + let mut res = crate::lib::std::vec::Vec::with_capacity(min); - if n == 0 { - return Ok((i, vec!())) - } - - loop { - let _i = input.clone(); - match f(_i) { - Ok((i, o)) => { + for count in 0..max { + match parse.parse(input.clone()) { + Ok((tail, value)) => { // do not allow parsers that do not consume input (causes infinite loops) - if i == input { + if tail == input { return Err(Err::Error(E::from_error_kind(input, ErrorKind::ManyMN))); } - res.push(o); - input = i; - count += 1; - - if count == n { - return Ok((input, res)); - } + res.push(value); + input = tail; } Err(Err::Error(e)) => { - if count < m { + if count < min { return Err(Err::Error(E::append(input, ErrorKind::ManyMN, e))); } else { return Ok((input, res)); @@ -454,6 +443,8 @@ where } } } + + Ok((input, res)) } } @@ -488,19 +479,19 @@ where /// assert_eq!(parser("123123"), Ok(("123123", 0))); /// assert_eq!(parser(""), Ok(("", 0))); /// ``` -pub fn many0_count<I, O, E, F>(f: F) -> impl Fn(I) -> IResult<I, usize, E> +pub fn many0_count<I, O, E, F>(mut f: F) -> impl FnMut(I) -> IResult<I, usize, E> where I: Clone + PartialEq, - F: Fn(I) -> IResult<I, O, E>, + F: Parser<I, O, E>, E: ParseError<I>, { move |i: I| { - let mut input = i.clone(); + let mut input = i; let mut count = 0; loop { let input_ = input.clone(); - match f(input_) { + match f.parse(input_) { Ok((i, _)) => { // loop trip must always consume (otherwise infinite loops) if i == input { @@ -537,7 +528,7 @@ where /// * `f` The parser to apply. /// ```rust /// # #[macro_use] extern crate nom; -/// # use nom::{Err, error::ErrorKind, Needed, IResult}; +/// # use nom::{Err, error::{Error, ErrorKind}, Needed, IResult}; /// use nom::multi::many1_count; /// use nom::bytes::complete::tag; /// @@ -547,18 +538,18 @@ where /// /// assert_eq!(parser("abcabc"), Ok(("", 2))); /// assert_eq!(parser("abc123"), Ok(("123", 1))); -/// assert_eq!(parser("123123"), Err(Err::Error(("123123", ErrorKind::Many1Count)))); -/// assert_eq!(parser(""), Err(Err::Error(("", ErrorKind::Many1Count)))); +/// assert_eq!(parser("123123"), Err(Err::Error(Error::new("123123", ErrorKind::Many1Count)))); +/// assert_eq!(parser(""), Err(Err::Error(Error::new("", ErrorKind::Many1Count)))); /// ``` -pub fn many1_count<I, O, E, F>(f: F) -> impl Fn(I) -> IResult<I, usize, E> +pub fn many1_count<I, O, E, F>(mut f: F) -> impl FnMut(I) -> IResult<I, usize, E> where I: Clone + PartialEq, - F: Fn(I) -> IResult<I, O, E>, + F: Parser<I, O, E>, E: ParseError<I>, { move |i: I| { let i_ = i.clone(); - match f(i_) { + match f.parse(i_) { Err(Err::Error(_)) => Err(Err::Error(E::from_error_kind(i, ErrorKind::Many1Count))), Err(i) => Err(i), Ok((i1, _)) => { @@ -567,7 +558,7 @@ where loop { let input_ = input.clone(); - match f(input_) { + match f.parse(input_) { Err(Err::Error(_)) => return Ok((input, count)), Err(e) => return Err(e), Ok((i, _)) => { @@ -602,7 +593,7 @@ where /// * `count` How often to apply the parser. /// ```rust /// # #[macro_use] extern crate nom; -/// # use nom::{Err, error::ErrorKind, Needed, IResult}; +/// # use nom::{Err, error::{Error, ErrorKind}, Needed, IResult}; /// use nom::multi::count; /// use nom::bytes::complete::tag; /// @@ -611,25 +602,26 @@ where /// } /// /// assert_eq!(parser("abcabc"), Ok(("", vec!["abc", "abc"]))); -/// assert_eq!(parser("abc123"), Err(Err::Error(("123", ErrorKind::Tag)))); -/// assert_eq!(parser("123123"), Err(Err::Error(("123123", ErrorKind::Tag)))); -/// assert_eq!(parser(""), Err(Err::Error(("", ErrorKind::Tag)))); +/// assert_eq!(parser("abc123"), Err(Err::Error(Error::new("123", ErrorKind::Tag)))); +/// assert_eq!(parser("123123"), Err(Err::Error(Error::new("123123", ErrorKind::Tag)))); +/// assert_eq!(parser(""), Err(Err::Error(Error::new("", ErrorKind::Tag)))); /// assert_eq!(parser("abcabcabc"), Ok(("abc", vec!["abc", "abc"]))); /// ``` #[cfg(feature = "alloc")] -pub fn count<I, O, E, F>(f: F, count: usize) -> impl Fn(I) -> IResult<I, Vec<O>, E> +#[cfg_attr(feature = "docsrs", doc(cfg(feature = "alloc")))] +pub fn count<I, O, E, F>(mut f: F, count: usize) -> impl FnMut(I) -> IResult<I, Vec<O>, E> where I: Clone + PartialEq, - F: Fn(I) -> IResult<I, O, E>, + F: Parser<I, O, E>, E: ParseError<I>, { - move |i: I | { + move |i: I| { let mut input = i.clone(); - let mut res = crate::lib::std::vec::Vec::new(); + let mut res = crate::lib::std::vec::Vec::with_capacity(count); for _ in 0..count { let input_ = input.clone(); - match f(input_) { + match f.parse(input_) { Ok((i, o)) => { res.push(o); input = i; @@ -647,6 +639,58 @@ where } } +/// Runs the embedded parser repeatedly, filling the given slice with results. This parser fails if +/// the input runs out before the given slice is full. +/// # Arguments +/// * `f` The parser to apply. +/// * `buf` The slice to fill +/// ```rust +/// # #[macro_use] extern crate nom; +/// # use nom::{Err, error::{Error, ErrorKind}, Needed, IResult}; +/// use nom::multi::fill; +/// use nom::bytes::complete::tag; +/// +/// fn parser(s: &str) -> IResult<&str, [&str; 2]> { +/// let mut buf = ["", ""]; +/// let (rest, ()) = fill(tag("abc"), &mut buf)(s)?; +/// Ok((rest, buf)) +/// } +/// +/// assert_eq!(parser("abcabc"), Ok(("", ["abc", "abc"]))); +/// assert_eq!(parser("abc123"), Err(Err::Error(Error::new("123", ErrorKind::Tag)))); +/// assert_eq!(parser("123123"), Err(Err::Error(Error::new("123123", ErrorKind::Tag)))); +/// assert_eq!(parser(""), Err(Err::Error(Error::new("", ErrorKind::Tag)))); +/// assert_eq!(parser("abcabcabc"), Ok(("abc", ["abc", "abc"]))); +/// ``` +pub fn fill<'a, I, O, E, F>(f: F, buf: &'a mut [O]) -> impl FnMut(I) -> IResult<I, (), E> + 'a +where + I: Clone + PartialEq, + F: Fn(I) -> IResult<I, O, E> + 'a, + E: ParseError<I>, +{ + move |i: I| { + let mut input = i.clone(); + + for elem in buf.iter_mut() { + let input_ = input.clone(); + match f(input_) { + Ok((i, o)) => { + *elem = o; + input = i; + } + Err(Err::Error(e)) => { + return Err(Err::Error(E::append(i, ErrorKind::Count, e))); + } + Err(e) => { + return Err(e); + } + } + } + + Ok((input, ())) + } +} + /// Applies a parser until it fails and accumulates /// the results using a given function and initial value. /// # Arguments @@ -676,21 +720,25 @@ where /// assert_eq!(parser("123123"), Ok(("123123", vec![]))); /// assert_eq!(parser(""), Ok(("", vec![]))); /// ``` -pub fn fold_many0<I, O, E, F, G, R>(f: F, init: R, g: G) -> impl Fn(I) -> IResult<I, R, E> +pub fn fold_many0<I, O, E, F, G, R>( + mut f: F, + init: R, + mut g: G, +) -> impl FnMut(I) -> IResult<I, R, E> where I: Clone + PartialEq, - F: Fn(I) -> IResult<I, O, E>, - G: Fn(R, O) -> R, + F: Parser<I, O, E>, + G: FnMut(R, O) -> R, E: ParseError<I>, R: Clone, { move |i: I| { let mut res = init.clone(); - let mut input = i.clone(); + let mut input = i; loop { let i_ = input.clone(); - match f(i_) { + match f.parse(i_) { Ok((i, o)) => { // loop trip must always consume (otherwise infinite loops) if i == input { @@ -716,7 +764,7 @@ pub fn fold_many0c<I, O, E, F, G, R>(i: I, f: F, init: R, g: G) -> IResult<I, R, where I: Clone + PartialEq, F: Fn(I) -> IResult<I, O, E>, - G: Fn(R, O) -> R, + G: FnMut(R, O) -> R, E: ParseError<I>, R: Clone, { @@ -734,7 +782,7 @@ where /// the current accumulator. /// ```rust /// # #[macro_use] extern crate nom; -/// # use nom::{Err, error::ErrorKind, Needed, IResult}; +/// # use nom::{Err, error::{Error, ErrorKind}, Needed, IResult}; /// use nom::multi::fold_many1; /// use nom::bytes::complete::tag; /// @@ -751,30 +799,34 @@ where /// /// assert_eq!(parser("abcabc"), Ok(("", vec!["abc", "abc"]))); /// assert_eq!(parser("abc123"), Ok(("123", vec!["abc"]))); -/// assert_eq!(parser("123123"), Err(Err::Error(("123123", ErrorKind::Many1)))); -/// assert_eq!(parser(""), Err(Err::Error(("", ErrorKind::Many1)))); +/// assert_eq!(parser("123123"), Err(Err::Error(Error::new("123123", ErrorKind::Many1)))); +/// assert_eq!(parser(""), Err(Err::Error(Error::new("", ErrorKind::Many1)))); /// ``` -pub fn fold_many1<I, O, E, F, G, R>(f: F, init: R, g: G) -> impl Fn(I) -> IResult<I, R, E> +pub fn fold_many1<I, O, E, F, G, R>( + mut f: F, + init: R, + mut g: G, +) -> impl FnMut(I) -> IResult<I, R, E> where I: Clone + PartialEq, - F: Fn(I) -> IResult<I, O, E>, - G: Fn(R, O) -> R, + F: Parser<I, O, E>, + G: FnMut(R, O) -> R, E: ParseError<I>, R: Clone, { move |i: I| { let _i = i.clone(); let init = init.clone(); - match f(_i) { + match f.parse(_i) { Err(Err::Error(_)) => Err(Err::Error(E::from_error_kind(i, ErrorKind::Many1))), - Err(e) => return Err(e), + Err(e) => Err(e), Ok((i1, o1)) => { let mut acc = g(init, o1); let mut input = i1; loop { let _input = input.clone(); - match f(_input) { + match f.parse(_input) { Err(Err::Error(_)) => { break; } @@ -801,7 +853,7 @@ pub fn fold_many1c<I, O, E, F, G, R>(i: I, f: F, init: R, g: G) -> IResult<I, R, where I: Clone + PartialEq, F: Fn(I) -> IResult<I, O, E>, - G: Fn(R, O) -> R, + G: FnMut(R, O) -> R, E: ParseError<I>, R: Clone, { @@ -844,34 +896,40 @@ where /// assert_eq!(parser(""), Ok(("", vec![]))); /// assert_eq!(parser("abcabcabc"), Ok(("abc", vec!["abc", "abc"]))); /// ``` -pub fn fold_many_m_n<I, O, E, F, G, R>(m: usize, n: usize, f: F, init: R, g: G) -> impl Fn(I) ->IResult<I, R, E> +pub fn fold_many_m_n<I, O, E, F, G, R>( + min: usize, + max: usize, + mut parse: F, + init: R, + mut fold: G, +) -> impl FnMut(I) -> IResult<I, R, E> where I: Clone + PartialEq, - F: Fn(I) -> IResult<I, O, E>, - G: Fn(R, O) -> R, + F: Parser<I, O, E>, + G: FnMut(R, O) -> R, E: ParseError<I>, R: Clone, { - move |i: I| { + move |mut input: I| { let mut acc = init.clone(); - let mut input = i.clone(); - for count in 0..n { - let _input = input.clone(); - match f(_input) { - Ok((i, o)) => { + for count in 0..max { + match parse.parse(input.clone()) { + Ok((tail, value)) => { // do not allow parsers that do not consume input (causes infinite loops) - if i == input { - return Err(Err::Error(E::from_error_kind(i, ErrorKind::ManyMN))); + if tail == input { + return Err(Err::Error(E::from_error_kind(tail, ErrorKind::ManyMN))); } - acc = g(acc, o); - input = i; + acc = fold(acc, value); + input = tail; } //FInputXMError: handle failure properly - Err(Err::Error(_)) => if count < m { - return Err(Err::Error(E::from_error_kind(i, ErrorKind::ManyMN))); - } else { - break; + Err(Err::Error(err)) => { + if count < min { + return Err(Err::Error(E::append(input, ErrorKind::ManyMN, err))); + } else { + break; + } } Err(e) => return Err(e), } @@ -882,7 +940,14 @@ where } #[doc(hidden)] -pub fn fold_many_m_nc<I, O, E, F, G, R>(i: I, m: usize, n: usize, f: F, init: R, g: G) -> IResult<I, R, E> +pub fn fold_many_m_nc<I, O, E, F, G, R>( + input: I, + min: usize, + max: usize, + parse: F, + init: R, + fold: G, +) -> IResult<I, R, E> where I: Clone + PartialEq, F: Fn(I) -> IResult<I, O, E>, @@ -890,19 +955,18 @@ where E: ParseError<I>, R: Clone, { - fold_many_m_n(m, n, f, init, g)(i) + fold_many_m_n(min, max, parse, init, fold)(input) } /// Gets a number from the parser and returns a /// subslice of the input of that size. -/// If the parser returns Incomplete, -/// length_data will return an error. +/// If the parser returns `Incomplete`, +/// `length_data` will return an error. /// # Arguments /// * `f` The parser to apply. /// ```rust /// # #[macro_use] extern crate nom; /// # use nom::{Err, error::ErrorKind, Needed, IResult}; -/// # use nom::Needed::Size; /// use nom::number::complete::be_u16; /// use nom::multi::length_data; /// use nom::bytes::complete::tag; @@ -912,22 +976,25 @@ where /// } /// /// assert_eq!(parser(b"\x00\x03abcefg"), Ok((&b"efg"[..], &b"abc"[..]))); -/// assert_eq!(parser(b"\x00\x03"), Err(Err::Incomplete(Size(3)))); +/// assert_eq!(parser(b"\x00\x03a"), Err(Err::Incomplete(Needed::new(2)))); /// ``` -pub fn length_data<I, N, E, F>(f: F) -> impl Fn(I) -> IResult<I, I, E> +pub fn length_data<I, N, E, F>(mut f: F) -> impl FnMut(I) -> IResult<I, I, E> where - I: Clone + InputLength + InputTake, - N: Copy + ToUsize, - F: Fn(I) -> IResult<I, N, E>, + I: InputLength + InputTake, + N: ToUsize, + F: Parser<I, N, E>, E: ParseError<I>, { move |i: I| { - let (i, length) = f(i)?; + let (i, length) = f.parse(i)?; let length: usize = length.to_usize(); - if i.input_len() < length { - Err(Err::Incomplete(Needed::Size(length))) + if let Some(needed) = length + .checked_sub(i.input_len()) + .and_then(NonZeroUsize::new) + { + Err(Err::Incomplete(Needed::Size(needed))) } else { Ok(i.take_split(length)) } @@ -937,14 +1004,13 @@ where /// Gets a number from the first parser, /// takes a subslice of the input of that size, /// then applies the second parser on that subslice. -/// If the second parser returns Incomplete, -/// length_value will return an error. +/// If the second parser returns `Incomplete`, +/// `length_value` will return an error. /// # Arguments /// * `f` The parser to apply. /// ```rust /// # #[macro_use] extern crate nom; -/// # use nom::{Err, error::ErrorKind, Needed, IResult}; -/// # use nom::Needed::Size; +/// # use nom::{Err, error::{Error, ErrorKind}, Needed, IResult}; /// use nom::number::complete::be_u16; /// use nom::multi::length_value; /// use nom::bytes::complete::tag; @@ -954,31 +1020,33 @@ where /// } /// /// assert_eq!(parser(b"\x00\x03abcefg"), Ok((&b"efg"[..], &b"abc"[..]))); -/// assert_eq!(parser(b"\x00\x03123123"), Err(Err::Error((&b"123"[..], ErrorKind::Tag)))); -/// assert_eq!(parser(b"\x00\x03"), Err(Err::Incomplete(Size(3)))); +/// assert_eq!(parser(b"\x00\x03123123"), Err(Err::Error(Error::new(&b"123"[..], ErrorKind::Tag)))); +/// assert_eq!(parser(b"\x00\x03a"), Err(Err::Incomplete(Needed::new(2)))); /// ``` -pub fn length_value<I, O, N, E, F, G>(f: F, g: G) -> impl Fn(I) -> IResult<I, O, E> +pub fn length_value<I, O, N, E, F, G>(mut f: F, mut g: G) -> impl FnMut(I) -> IResult<I, O, E> where I: Clone + InputLength + InputTake, - N: Copy + ToUsize, - F: Fn(I) -> IResult<I, N, E>, - G: Fn(I) -> IResult<I, O, E>, + N: ToUsize, + F: Parser<I, N, E>, + G: Parser<I, O, E>, E: ParseError<I>, { move |i: I| { - let (i, length) = f(i)?; + let (i, length) = f.parse(i)?; let length: usize = length.to_usize(); - if i.input_len() < length { - Err(Err::Incomplete(Needed::Size(length))) + if let Some(needed) = length + .checked_sub(i.input_len()) + .and_then(NonZeroUsize::new) + { + Err(Err::Incomplete(Needed::Size(needed))) } else { let (rest, i) = i.take_split(length); - match g(i.clone()) { - Err(Err::Incomplete(_)) => - Err(Err::Error(E::from_error_kind(i, ErrorKind::Complete))), + match g.parse(i.clone()) { + Err(Err::Incomplete(_)) => Err(Err::Error(E::from_error_kind(i, ErrorKind::Complete))), Err(e) => Err(e), - Ok((_, o)) => Ok((rest,o)), + Ok((_, o)) => Ok((rest, o)), } } } @@ -988,10 +1056,67 @@ where pub fn length_valuec<I, O, N, E, F, G>(i: I, f: F, g: G) -> IResult<I, O, E> where I: Clone + InputLength + InputTake, - N: Copy + ToUsize, + N: ToUsize, F: Fn(I) -> IResult<I, N, E>, G: Fn(I) -> IResult<I, O, E>, E: ParseError<I>, { length_value(f, g)(i) } + +/// Gets a number from the first parser, +/// then applies the second parser that many times. +/// Arguments +/// * `f` The parser to apply to obtain the count. +/// * `g` The parser to apply repeatedly. +/// ```rust +/// # #[macro_use] extern crate nom; +/// # use nom::{Err, error::{Error, ErrorKind}, Needed, IResult}; +/// use nom::number::complete::u8; +/// use nom::multi::length_count; +/// use nom::bytes::complete::tag; +/// use nom::combinator::map; +/// +/// fn parser(s: &[u8]) -> IResult<&[u8], Vec<&[u8]>> { +/// length_count(map(u8, |i| { +/// println!("got number: {}", i); +/// i +/// }), tag("abc"))(s) +/// } +/// +/// assert_eq!(parser(&b"\x02abcabcabc"[..]), Ok(((&b"abc"[..], vec![&b"abc"[..], &b"abc"[..]])))); +/// assert_eq!(parser(b"\x03123123123"), Err(Err::Error(Error::new(&b"123123123"[..], ErrorKind::Tag)))); +/// ``` +#[cfg(feature = "alloc")] +pub fn length_count<I, O, N, E, F, G>(mut f: F, mut g: G) -> impl FnMut(I) -> IResult<I, Vec<O>, E> +where + I: Clone, + N: ToUsize, + F: Parser<I, N, E>, + G: Parser<I, O, E>, + E: ParseError<I>, +{ + move |i: I| { + let (i, count) = f.parse(i)?; + let mut input = i.clone(); + let mut res = Vec::new(); + + for _ in 0..count.to_usize() { + let input_ = input.clone(); + match g.parse(input_) { + Ok((i, o)) => { + res.push(o); + input = i; + } + Err(Err::Error(e)) => { + return Err(Err::Error(E::append(i, ErrorKind::Count, e))); + } + Err(e) => { + return Err(e); + } + } + } + + Ok((input, res)) + } +} diff --git a/src/number/complete.rs b/src/number/complete.rs index b8d4232..ffebf23 100644 --- a/src/number/complete.rs +++ b/src/number/complete.rs @@ -1,19 +1,19 @@ -//! parsers recognizing numbers, complete input version +//! Parsers recognizing numbers, complete input version -use crate::internal::*; +use crate::branch::alt; +use crate::character::complete::{char, digit1}; +use crate::combinator::{cut, map, opt, recognize}; use crate::error::ParseError; -use crate::traits::{AsChar, InputIter, InputLength, InputTakeAtPosition}; +use crate::error::{make_error, ErrorKind}; +use crate::internal::*; use crate::lib::std::ops::{RangeFrom, RangeTo}; +use crate::sequence::{pair, tuple}; +use crate::traits::{AsChar, InputIter, InputLength, InputTakeAtPosition}; use crate::traits::{Offset, Slice}; -use crate::error::{ErrorKind, make_error}; -use crate::character::complete::{char, digit1}; -use crate::combinator::{opt, cut, map, recognize}; -use crate::branch::alt; -use crate::sequence::{tuple, pair}; -/// Recognizes an unsigned 1 byte integer +/// Recognizes an unsigned 1 byte integer. /// -/// *complete version*: returns an error if there is not enough input data +/// *Complete version*: Returns an error if there is not enough input data. /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; /// # use nom::Needed::Size; @@ -23,21 +23,27 @@ use crate::sequence::{tuple, pair}; /// be_u8(s) /// }; /// -/// assert_eq!(parser(b"\x00\x03abcefg"), Ok((&b"\x03abcefg"[..], 0x00))); -/// assert_eq!(parser(b""), Err(Err::Error((&[][..], ErrorKind::Eof)))); +/// assert_eq!(parser(&b"\x00\x03abcefg"[..]), Ok((&b"\x03abcefg"[..], 0x00))); +/// assert_eq!(parser(&b""[..]), Err(Err::Error((&[][..], ErrorKind::Eof)))); /// ``` #[inline] -pub fn be_u8<'a, E: ParseError<&'a[u8]>>(i: &'a[u8]) -> IResult<&'a[u8], u8, E> { - if i.len() < 1 { - Err(Err::Error(make_error(i, ErrorKind::Eof))) +pub fn be_u8<I, E: ParseError<I>>(input: I) -> IResult<I, u8, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + let bound: usize = 1; + if input.input_len() < bound { + Err(Err::Error(make_error(input, ErrorKind::Eof))) } else { - Ok((&i[1..], i[0])) + let res = input.iter_elements().next().unwrap(); + + Ok((input.slice(bound..), res)) } } -/// Recognizes a big endian unsigned 2 bytes integer +/// Recognizes a big endian unsigned 2 bytes integer. /// -/// *complete version*: returns an error if there is not enough input data +/// *Complete version*: Returns an error if there is not enough input data. /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; /// # use nom::Needed::Size; @@ -47,22 +53,30 @@ pub fn be_u8<'a, E: ParseError<&'a[u8]>>(i: &'a[u8]) -> IResult<&'a[u8], u8, E> /// be_u16(s) /// }; /// -/// assert_eq!(parser(b"\x00\x03abcefg"), Ok((&b"abcefg"[..], 0x0003))); -/// assert_eq!(parser(b"\x01"), Err(Err::Error((&[0x01][..], ErrorKind::Eof)))); +/// assert_eq!(parser(&b"\x00\x03abcefg"[..]), Ok((&b"abcefg"[..], 0x0003))); +/// assert_eq!(parser(&b"\x01"[..]), Err(Err::Error((&[0x01][..], ErrorKind::Eof)))); /// ``` #[inline] -pub fn be_u16<'a, E: ParseError<&'a[u8]>>(i: &'a[u8]) -> IResult<&'a[u8], u16, E> { - if i.len() < 2 { - Err(Err::Error(make_error(i, ErrorKind::Eof))) +pub fn be_u16<I, E: ParseError<I>>(input: I) -> IResult<I, u16, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + let bound: usize = 2; + if input.input_len() < bound { + Err(Err::Error(make_error(input, ErrorKind::Eof))) } else { - let res = ((i[0] as u16) << 8) + i[1] as u16; - Ok((&i[2..], res)) + let mut res = 0u16; + for byte in input.iter_elements().take(bound) { + res = (res << 8) + byte as u16; + } + + Ok((input.slice(bound..), res)) } } -/// Recognizes a big endian unsigned 3 byte integer +/// Recognizes a big endian unsigned 3 byte integer. /// -/// *complete version*: returns an error if there is not enough input data +/// *Complete version*: Returns an error if there is not enough input data. /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; /// # use nom::Needed::Size; @@ -72,22 +86,30 @@ pub fn be_u16<'a, E: ParseError<&'a[u8]>>(i: &'a[u8]) -> IResult<&'a[u8], u16, E /// be_u24(s) /// }; /// -/// assert_eq!(parser(b"\x00\x03\x05abcefg"), Ok((&b"abcefg"[..], 0x000305))); -/// assert_eq!(parser(b"\x01"), Err(Err::Error((&[0x01][..], ErrorKind::Eof)))); +/// assert_eq!(parser(&b"\x00\x03\x05abcefg"[..]), Ok((&b"abcefg"[..], 0x000305))); +/// assert_eq!(parser(&b"\x01"[..]), Err(Err::Error((&[0x01][..], ErrorKind::Eof)))); /// ``` #[inline] -pub fn be_u24<'a, E: ParseError<&'a[u8]>>(i: &'a[u8]) -> IResult<&'a[u8], u32, E> { - if i.len() < 3 { - Err(Err::Error(make_error(i, ErrorKind::Eof))) +pub fn be_u24<I, E: ParseError<I>>(input: I) -> IResult<I, u32, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + let bound: usize = 3; + if input.input_len() < bound { + Err(Err::Error(make_error(input, ErrorKind::Eof))) } else { - let res = ((i[0] as u32) << 16) + ((i[1] as u32) << 8) + (i[2] as u32); - Ok((&i[3..], res)) + let mut res = 0u32; + for byte in input.iter_elements().take(bound) { + res = (res << 8) + byte as u32; + } + + Ok((input.slice(bound..), res)) } } -/// Recognizes a big endian unsigned 4 bytes integer +/// Recognizes a big endian unsigned 4 bytes integer. /// -/// *complete version*: returns an error if there is not enough input data +/// *Complete version*: Returns an error if there is not enough input data. /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; /// # use nom::Needed::Size; @@ -97,22 +119,30 @@ pub fn be_u24<'a, E: ParseError<&'a[u8]>>(i: &'a[u8]) -> IResult<&'a[u8], u32, E /// be_u32(s) /// }; /// -/// assert_eq!(parser(b"\x00\x03\x05\x07abcefg"), Ok((&b"abcefg"[..], 0x00030507))); -/// assert_eq!(parser(b"\x01"), Err(Err::Error((&[0x01][..], ErrorKind::Eof)))); +/// assert_eq!(parser(&b"\x00\x03\x05\x07abcefg"[..]), Ok((&b"abcefg"[..], 0x00030507))); +/// assert_eq!(parser(&b"\x01"[..]), Err(Err::Error((&[0x01][..], ErrorKind::Eof)))); /// ``` #[inline] -pub fn be_u32<'a, E: ParseError<&'a[u8]>>(i: &'a[u8]) -> IResult<&'a[u8], u32, E> { - if i.len() < 4 { - Err(Err::Error(make_error(i, ErrorKind::Eof))) +pub fn be_u32<I, E: ParseError<I>>(input: I) -> IResult<I, u32, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + let bound: usize = 4; + if input.input_len() < bound { + Err(Err::Error(make_error(input, ErrorKind::Eof))) } else { - let res = ((i[0] as u32) << 24) + ((i[1] as u32) << 16) + ((i[2] as u32) << 8) + i[3] as u32; - Ok((&i[4..], res)) + let mut res = 0u32; + for byte in input.iter_elements().take(bound) { + res = (res << 8) + byte as u32; + } + + Ok((input.slice(bound..), res)) } } -/// Recognizes a big endian unsigned 8 bytes integer +/// Recognizes a big endian unsigned 8 bytes integer. /// -/// *complete version*: returns an error if there is not enough input data +/// *Complete version*: Returns an error if there is not enough input data. /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; /// # use nom::Needed::Size; @@ -122,23 +152,30 @@ pub fn be_u32<'a, E: ParseError<&'a[u8]>>(i: &'a[u8]) -> IResult<&'a[u8], u32, E /// be_u64(s) /// }; /// -/// assert_eq!(parser(b"\x00\x01\x02\x03\x04\x05\x06\x07abcefg"), Ok((&b"abcefg"[..], 0x0001020304050607))); -/// assert_eq!(parser(b"\x01"), Err(Err::Error((&[0x01][..], ErrorKind::Eof)))); +/// assert_eq!(parser(&b"\x00\x01\x02\x03\x04\x05\x06\x07abcefg"[..]), Ok((&b"abcefg"[..], 0x0001020304050607))); +/// assert_eq!(parser(&b"\x01"[..]), Err(Err::Error((&[0x01][..], ErrorKind::Eof)))); /// ``` #[inline] -pub fn be_u64<'a, E: ParseError<&'a[u8]>>(i: &'a[u8]) -> IResult<&'a[u8], u64, E> { - if i.len() < 8 { - Err(Err::Error(make_error(i, ErrorKind::Eof))) +pub fn be_u64<I, E: ParseError<I>>(input: I) -> IResult<I, u64, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + let bound: usize = 8; + if input.input_len() < bound { + Err(Err::Error(make_error(input, ErrorKind::Eof))) } else { - let res = ((i[0] as u64) << 56) + ((i[1] as u64) << 48) + ((i[2] as u64) << 40) + ((i[3] as u64) << 32) + ((i[4] as u64) << 24) - + ((i[5] as u64) << 16) + ((i[6] as u64) << 8) + i[7] as u64; - Ok((&i[8..], res)) + let mut res = 0u64; + for byte in input.iter_elements().take(bound) { + res = (res << 8) + byte as u64; + } + + Ok((input.slice(bound..), res)) } } -/// Recognizes a big endian unsigned 16 bytes integer +/// Recognizes a big endian unsigned 16 bytes integer. /// -/// *complete version*: returns an error if there is not enough input data +/// *Complete version*: Returns an error if there is not enough input data. /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; /// # use nom::Needed::Size; @@ -148,38 +185,31 @@ pub fn be_u64<'a, E: ParseError<&'a[u8]>>(i: &'a[u8]) -> IResult<&'a[u8], u64, E /// be_u128(s) /// }; /// -/// assert_eq!(parser(b"\x00\x01\x02\x03\x04\x05\x06\x07\x00\x01\x02\x03\x04\x05\x06\x07abcefg"), Ok((&b"abcefg"[..], 0x00010203040506070001020304050607))); -/// assert_eq!(parser(b"\x01"), Err(Err::Error((&[0x01][..], ErrorKind::Eof)))); +/// assert_eq!(parser(&b"\x00\x01\x02\x03\x04\x05\x06\x07\x00\x01\x02\x03\x04\x05\x06\x07abcefg"[..]), Ok((&b"abcefg"[..], 0x00010203040506070001020304050607))); +/// assert_eq!(parser(&b"\x01"[..]), Err(Err::Error((&[0x01][..], ErrorKind::Eof)))); /// ``` #[inline] #[cfg(stable_i128)] -pub fn be_u128<'a, E: ParseError<&'a[u8]>>(i: &'a[u8]) -> IResult<&'a[u8], u128, E> { - if i.len() < 16 { - Err(Err::Error(make_error(i, ErrorKind::Eof))) +pub fn be_u128<I, E: ParseError<I>>(input: I) -> IResult<I, u128, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + let bound: usize = 16; + if input.input_len() < bound { + Err(Err::Error(make_error(input, ErrorKind::Eof))) } else { - let res = ((i[0] as u128) << 120) - + ((i[1] as u128) << 112) - + ((i[2] as u128) << 104) - + ((i[3] as u128) << 96) - + ((i[4] as u128) << 88) - + ((i[5] as u128) << 80) - + ((i[6] as u128) << 72) - + ((i[7] as u128) << 64) - + ((i[8] as u128) << 56) - + ((i[9] as u128) << 48) - + ((i[10] as u128) << 40) - + ((i[11] as u128) << 32) - + ((i[12] as u128) << 24) - + ((i[13] as u128) << 16) - + ((i[14] as u128) << 8) - + i[15] as u128; - Ok((&i[16..], res)) + let mut res = 0u128; + for byte in input.iter_elements().take(bound) { + res = (res << 8) + byte as u128; + } + + Ok((input.slice(bound..), res)) } } -/// Recognizes a signed 1 byte integer +/// Recognizes a signed 1 byte integer. /// -/// *complete version*: returns an error if there is not enough input data +/// *Complete version*: Returns an error if there is not enough input data. /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; /// # use nom::Needed::Size; @@ -189,17 +219,20 @@ pub fn be_u128<'a, E: ParseError<&'a[u8]>>(i: &'a[u8]) -> IResult<&'a[u8], u128, /// be_i8(s) /// }; /// -/// assert_eq!(parser(b"\x00\x03abcefg"), Ok((&b"\x03abcefg"[..], 0x00))); -/// assert_eq!(parser(b""), Err(Err::Error((&[][..], ErrorKind::Eof)))); +/// assert_eq!(parser(&b"\x00\x03abcefg"[..]), Ok((&b"\x03abcefg"[..], 0x00))); +/// assert_eq!(parser(&b""[..]), Err(Err::Error((&[][..], ErrorKind::Eof)))); /// ``` #[inline] -pub fn be_i8<'a, E: ParseError<&'a[u8]>>(i: &'a[u8]) -> IResult<&'a[u8], i8, E> { - map!(i, be_u8, |x| x as i8) +pub fn be_i8<I, E: ParseError<I>>(input: I) -> IResult<I, i8, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + map!(input, be_u8, |x| x as i8) } -/// Recognizes a big endian signed 2 bytes integer +/// Recognizes a big endian signed 2 bytes integer. /// -/// *complete version*: returns an error if there is not enough input data +/// *Complete version*: Returns an error if there is not enough input data. /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; /// # use nom::Needed::Size; @@ -209,17 +242,20 @@ pub fn be_i8<'a, E: ParseError<&'a[u8]>>(i: &'a[u8]) -> IResult<&'a[u8], i8, E> /// be_i16(s) /// }; /// -/// assert_eq!(parser(b"\x00\x03abcefg"), Ok((&b"abcefg"[..], 0x0003))); -/// assert_eq!(parser(b"\x01"), Err(Err::Error((&[0x01][..], ErrorKind::Eof)))); +/// assert_eq!(parser(&b"\x00\x03abcefg"[..]), Ok((&b"abcefg"[..], 0x0003))); +/// assert_eq!(parser(&b"\x01"[..]), Err(Err::Error((&[0x01][..], ErrorKind::Eof)))); /// ``` #[inline] -pub fn be_i16<'a, E: ParseError<&'a [u8]>>(i: &'a[u8]) -> IResult<&'a[u8], i16, E> { - map!(i, be_u16, |x| x as i16) +pub fn be_i16<I, E: ParseError<I>>(input: I) -> IResult<I, i16, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + map!(input, be_u16, |x| x as i16) } -/// Recognizes a big endian signed 3 bytes integer +/// Recognizes a big endian signed 3 bytes integer. /// -/// *complete version*: returns an error if there is not enough input data +/// *Complete version*: Returns an error if there is not enough input data. /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; /// # use nom::Needed::Size; @@ -229,22 +265,25 @@ pub fn be_i16<'a, E: ParseError<&'a [u8]>>(i: &'a[u8]) -> IResult<&'a[u8], i16, /// be_i24(s) /// }; /// -/// assert_eq!(parser(b"\x00\x03\x05abcefg"), Ok((&b"abcefg"[..], 0x000305))); -/// assert_eq!(parser(b"\x01"), Err(Err::Error((&[0x01][..], ErrorKind::Eof)))); +/// assert_eq!(parser(&b"\x00\x03\x05abcefg"[..]), Ok((&b"abcefg"[..], 0x000305))); +/// assert_eq!(parser(&b"\x01"[..]), Err(Err::Error((&[0x01][..], ErrorKind::Eof)))); /// ``` #[inline] -pub fn be_i24<'a, E: ParseError<&'a[u8]>>(i: &'a[u8]) -> IResult<&'a[u8], i32, E> { +pub fn be_i24<I, E: ParseError<I>>(input: I) -> IResult<I, i32, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ // Same as the unsigned version but we need to sign-extend manually here - map!(i, be_u24, |x| if x & 0x80_00_00 != 0 { + map!(input, be_u24, |x| if x & 0x80_00_00 != 0 { (x | 0xff_00_00_00) as i32 } else { x as i32 }) } -/// Recognizes a big endian signed 4 bytes integer +/// Recognizes a big endian signed 4 bytes integer. /// -/// *complete version*: returns an error if there is not enough input data +/// *Complete version*: Teturns an error if there is not enough input data. /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; /// # use nom::Needed::Size; @@ -254,17 +293,20 @@ pub fn be_i24<'a, E: ParseError<&'a[u8]>>(i: &'a[u8]) -> IResult<&'a[u8], i32, E /// be_i32(s) /// }; /// -/// assert_eq!(parser(b"\x00\x03\x05\x07abcefg"), Ok((&b"abcefg"[..], 0x00030507))); -/// assert_eq!(parser(b"\x01"), Err(Err::Error((&[0x01][..], ErrorKind::Eof)))); +/// assert_eq!(parser(&b"\x00\x03\x05\x07abcefg"[..]), Ok((&b"abcefg"[..], 0x00030507))); +/// assert_eq!(parser(&b"\x01"[..]), Err(Err::Error((&[0x01][..], ErrorKind::Eof)))); /// ``` #[inline] -pub fn be_i32<'a, E: ParseError<&'a[u8]>>(i: &'a[u8]) -> IResult<&'a[u8], i32, E> { - map!(i, be_u32, |x| x as i32) +pub fn be_i32<I, E: ParseError<I>>(input: I) -> IResult<I, i32, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + map!(input, be_u32, |x| x as i32) } -/// Recognizes a big endian signed 8 bytes integer +/// Recognizes a big endian signed 8 bytes integer. /// -/// *complete version*: returns an error if there is not enough input data +/// *Complete version*: Returns an error if there is not enough input data. /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; /// # use nom::Needed::Size; @@ -274,17 +316,20 @@ pub fn be_i32<'a, E: ParseError<&'a[u8]>>(i: &'a[u8]) -> IResult<&'a[u8], i32, E /// be_i64(s) /// }; /// -/// assert_eq!(parser(b"\x00\x01\x02\x03\x04\x05\x06\x07abcefg"), Ok((&b"abcefg"[..], 0x0001020304050607))); -/// assert_eq!(parser(b"\x01"), Err(Err::Error((&[0x01][..], ErrorKind::Eof)))); +/// assert_eq!(parser(&b"\x00\x01\x02\x03\x04\x05\x06\x07abcefg"[..]), Ok((&b"abcefg"[..], 0x0001020304050607))); +/// assert_eq!(parser(&b"\x01"[..]), Err(Err::Error((&[0x01][..], ErrorKind::Eof)))); /// ``` #[inline] -pub fn be_i64<'a, E: ParseError<&'a[u8]>>(i: &'a[u8]) -> IResult<&'a[u8], i64, E> { - map!(i, be_u64, |x| x as i64) +pub fn be_i64<I, E: ParseError<I>>(input: I) -> IResult<I, i64, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + map!(input, be_u64, |x| x as i64) } -/// Recognizes a big endian signed 16 bytes integer +/// Recognizes a big endian signed 16 bytes integer. /// -/// *complete version*: returns an error if there is not enough input data +/// *Complete version*: Returns an error if there is not enough input data. /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; /// # use nom::Needed::Size; @@ -294,18 +339,21 @@ pub fn be_i64<'a, E: ParseError<&'a[u8]>>(i: &'a[u8]) -> IResult<&'a[u8], i64, E /// be_i128(s) /// }; /// -/// assert_eq!(parser(b"\x00\x01\x02\x03\x04\x05\x06\x07\x00\x01\x02\x03\x04\x05\x06\x07abcefg"), Ok((&b"abcefg"[..], 0x00010203040506070001020304050607))); -/// assert_eq!(parser(b"\x01"), Err(Err::Error((&[0x01][..], ErrorKind::Eof)))); +/// assert_eq!(parser(&b"\x00\x01\x02\x03\x04\x05\x06\x07\x00\x01\x02\x03\x04\x05\x06\x07abcefg"[..]), Ok((&b"abcefg"[..], 0x00010203040506070001020304050607))); +/// assert_eq!(parser(&b"\x01"[..]), Err(Err::Error((&[0x01][..], ErrorKind::Eof)))); /// ``` #[inline] #[cfg(stable_i128)] -pub fn be_i128<'a, E: ParseError<&'a [u8]>>(i: &'a[u8]) -> IResult<&'a[u8], i128, E> { - map!(i, be_u128, |x| x as i128) +pub fn be_i128<I, E: ParseError<I>>(input: I) -> IResult<I, i128, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + map!(input, be_u128, |x| x as i128) } -/// Recognizes an unsigned 1 byte integer +/// Recognizes an unsigned 1 byte integer. /// -/// *complete version*: returns an error if there is not enough input data +/// *Complete version*: Returns an error if there is not enough input data. /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; /// # use nom::Needed::Size; @@ -315,21 +363,27 @@ pub fn be_i128<'a, E: ParseError<&'a [u8]>>(i: &'a[u8]) -> IResult<&'a[u8], i128 /// le_u8(s) /// }; /// -/// assert_eq!(parser(b"\x00\x03abcefg"), Ok((&b"\x03abcefg"[..], 0x00))); -/// assert_eq!(parser(b""), Err(Err::Error((&[][..], ErrorKind::Eof)))); +/// assert_eq!(parser(&b"\x00\x03abcefg"[..]), Ok((&b"\x03abcefg"[..], 0x00))); +/// assert_eq!(parser(&b""[..]), Err(Err::Error((&[][..], ErrorKind::Eof)))); /// ``` #[inline] -pub fn le_u8<'a, E: ParseError<&'a [u8]>>(i: &'a[u8]) -> IResult<&'a[u8], u8, E> { - if i.len() < 1 { - Err(Err::Error(make_error(i, ErrorKind::Eof))) +pub fn le_u8<I, E: ParseError<I>>(input: I) -> IResult<I, u8, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + let bound: usize = 1; + if input.input_len() < bound { + Err(Err::Error(make_error(input, ErrorKind::Eof))) } else { - Ok((&i[1..], i[0])) + let res = input.iter_elements().next().unwrap(); + + Ok((input.slice(bound..), res)) } } -/// Recognizes a little endian unsigned 2 bytes integer +/// Recognizes a little endian unsigned 2 bytes integer. /// -/// *complete version*: returns an error if there is not enough input data +/// *Complete version*: Returns an error if there is not enough input data. /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; /// # use nom::Needed::Size; @@ -339,22 +393,30 @@ pub fn le_u8<'a, E: ParseError<&'a [u8]>>(i: &'a[u8]) -> IResult<&'a[u8], u8, E> /// le_u16(s) /// }; /// -/// assert_eq!(parser(b"\x00\x03abcefg"), Ok((&b"abcefg"[..], 0x0300))); -/// assert_eq!(parser(b"\x01"), Err(Err::Error((&[0x01][..], ErrorKind::Eof)))); +/// assert_eq!(parser(&b"\x00\x03abcefg"[..]), Ok((&b"abcefg"[..], 0x0300))); +/// assert_eq!(parser(&b"\x01"[..]), Err(Err::Error((&[0x01][..], ErrorKind::Eof)))); /// ``` #[inline] -pub fn le_u16<'a, E: ParseError<&'a [u8]>>(i: &'a[u8]) -> IResult<&'a[u8], u16, E> { - if i.len() < 2 { - Err(Err::Error(make_error(i, ErrorKind::Eof))) +pub fn le_u16<I, E: ParseError<I>>(input: I) -> IResult<I, u16, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + let bound: usize = 2; + if input.input_len() < bound { + Err(Err::Error(make_error(input, ErrorKind::Eof))) } else { - let res = ((i[1] as u16) << 8) + i[0] as u16; - Ok((&i[2..], res)) + let mut res = 0u16; + for (index, byte) in input.iter_indices().take(bound) { + res += (byte as u16) << (8 * index); + } + + Ok((input.slice(bound..), res)) } } -/// Recognizes a little endian unsigned 3 byte integer +/// Recognizes a little endian unsigned 3 byte integer. /// -/// *complete version*: returns an error if there is not enough input data +/// *Complete version*: Returns an error if there is not enough input data. /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; /// # use nom::Needed::Size; @@ -364,22 +426,30 @@ pub fn le_u16<'a, E: ParseError<&'a [u8]>>(i: &'a[u8]) -> IResult<&'a[u8], u16, /// le_u24(s) /// }; /// -/// assert_eq!(parser(b"\x00\x03\x05abcefg"), Ok((&b"abcefg"[..], 0x050300))); -/// assert_eq!(parser(b"\x01"), Err(Err::Error((&[0x01][..], ErrorKind::Eof)))); +/// assert_eq!(parser(&b"\x00\x03\x05abcefg"[..]), Ok((&b"abcefg"[..], 0x050300))); +/// assert_eq!(parser(&b"\x01"[..]), Err(Err::Error((&[0x01][..], ErrorKind::Eof)))); /// ``` #[inline] -pub fn le_u24<'a, E: ParseError<&'a [u8]>>(i: &'a[u8]) -> IResult<&'a[u8], u32, E> { - if i.len() < 3 { - Err(Err::Error(make_error(i, ErrorKind::Eof))) +pub fn le_u24<I, E: ParseError<I>>(input: I) -> IResult<I, u32, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + let bound: usize = 3; + if input.input_len() < bound { + Err(Err::Error(make_error(input, ErrorKind::Eof))) } else { - let res = (i[0] as u32) + ((i[1] as u32) << 8) + ((i[2] as u32) << 16); - Ok((&i[3..], res)) + let mut res = 0u32; + for (index, byte) in input.iter_indices().take(bound) { + res += (byte as u32) << (8 * index); + } + + Ok((input.slice(bound..), res)) } } -/// Recognizes a little endian unsigned 4 bytes integer +/// Recognizes a little endian unsigned 4 bytes integer. /// -/// *complete version*: returns an error if there is not enough input data +/// *Complete version*: Returns an error if there is not enough input data. /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; /// # use nom::Needed::Size; @@ -389,22 +459,30 @@ pub fn le_u24<'a, E: ParseError<&'a [u8]>>(i: &'a[u8]) -> IResult<&'a[u8], u32, /// le_u32(s) /// }; /// -/// assert_eq!(parser(b"\x00\x03\x05\x07abcefg"), Ok((&b"abcefg"[..], 0x07050300))); -/// assert_eq!(parser(b"\x01"), Err(Err::Error((&[0x01][..], ErrorKind::Eof)))); +/// assert_eq!(parser(&b"\x00\x03\x05\x07abcefg"[..]), Ok((&b"abcefg"[..], 0x07050300))); +/// assert_eq!(parser(&b"\x01"[..]), Err(Err::Error((&[0x01][..], ErrorKind::Eof)))); /// ``` #[inline] -pub fn le_u32<'a, E: ParseError<&'a [u8]>>(i: &'a[u8]) -> IResult<&'a[u8], u32, E> { - if i.len() < 4 { - Err(Err::Error(make_error(i, ErrorKind::Eof))) +pub fn le_u32<I, E: ParseError<I>>(input: I) -> IResult<I, u32, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + let bound: usize = 4; + if input.input_len() < bound { + Err(Err::Error(make_error(input, ErrorKind::Eof))) } else { - let res = ((i[3] as u32) << 24) + ((i[2] as u32) << 16) + ((i[1] as u32) << 8) + i[0] as u32; - Ok((&i[4..], res)) + let mut res = 0u32; + for (index, byte) in input.iter_indices().take(bound) { + res += (byte as u32) << (8 * index); + } + + Ok((input.slice(bound..), res)) } } -/// Recognizes a little endian unsigned 8 bytes integer +/// Recognizes a little endian unsigned 8 bytes integer. /// -/// *complete version*: returns an error if there is not enough input data +/// *Complete version*: Returns an error if there is not enough input data. /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; /// # use nom::Needed::Size; @@ -414,23 +492,30 @@ pub fn le_u32<'a, E: ParseError<&'a [u8]>>(i: &'a[u8]) -> IResult<&'a[u8], u32, /// le_u64(s) /// }; /// -/// assert_eq!(parser(b"\x00\x01\x02\x03\x04\x05\x06\x07abcefg"), Ok((&b"abcefg"[..], 0x0706050403020100))); -/// assert_eq!(parser(b"\x01"), Err(Err::Error((&[0x01][..], ErrorKind::Eof)))); +/// assert_eq!(parser(&b"\x00\x01\x02\x03\x04\x05\x06\x07abcefg"[..]), Ok((&b"abcefg"[..], 0x0706050403020100))); +/// assert_eq!(parser(&b"\x01"[..]), Err(Err::Error((&[0x01][..], ErrorKind::Eof)))); /// ``` #[inline] -pub fn le_u64<'a, E: ParseError<&'a [u8]>>(i: &'a[u8]) -> IResult<&'a[u8], u64, E> { - if i.len() < 8 { - Err(Err::Error(make_error(i, ErrorKind::Eof))) +pub fn le_u64<I, E: ParseError<I>>(input: I) -> IResult<I, u64, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + let bound: usize = 8; + if input.input_len() < bound { + Err(Err::Error(make_error(input, ErrorKind::Eof))) } else { - let res = ((i[7] as u64) << 56) + ((i[6] as u64) << 48) + ((i[5] as u64) << 40) + ((i[4] as u64) << 32) + ((i[3] as u64) << 24) - + ((i[2] as u64) << 16) + ((i[1] as u64) << 8) + i[0] as u64; - Ok((&i[8..], res)) + let mut res = 0u64; + for (index, byte) in input.iter_indices().take(bound) { + res += (byte as u64) << (8 * index); + } + + Ok((input.slice(bound..), res)) } } -/// Recognizes a little endian unsigned 16 bytes integer +/// Recognizes a little endian unsigned 16 bytes integer. /// -/// *complete version*: returns an error if there is not enough input data +/// *Complete version*: Returns an error if there is not enough input data. /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; /// # use nom::Needed::Size; @@ -440,38 +525,31 @@ pub fn le_u64<'a, E: ParseError<&'a [u8]>>(i: &'a[u8]) -> IResult<&'a[u8], u64, /// le_u128(s) /// }; /// -/// assert_eq!(parser(b"\x00\x01\x02\x03\x04\x05\x06\x07\x00\x01\x02\x03\x04\x05\x06\x07abcefg"), Ok((&b"abcefg"[..], 0x07060504030201000706050403020100))); -/// assert_eq!(parser(b"\x01"), Err(Err::Error((&[0x01][..], ErrorKind::Eof)))); +/// assert_eq!(parser(&b"\x00\x01\x02\x03\x04\x05\x06\x07\x00\x01\x02\x03\x04\x05\x06\x07abcefg"[..]), Ok((&b"abcefg"[..], 0x07060504030201000706050403020100))); +/// assert_eq!(parser(&b"\x01"[..]), Err(Err::Error((&[0x01][..], ErrorKind::Eof)))); /// ``` #[inline] #[cfg(stable_i128)] -pub fn le_u128<'a, E: ParseError<&'a [u8]>>(i: &'a[u8]) -> IResult<&'a[u8], u128, E> { - if i.len() < 16 { - Err(Err::Error(make_error(i, ErrorKind::Eof))) +pub fn le_u128<I, E: ParseError<I>>(input: I) -> IResult<I, u128, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + let bound: usize = 16; + if input.input_len() < bound { + Err(Err::Error(make_error(input, ErrorKind::Eof))) } else { - let res = ((i[15] as u128) << 120) - + ((i[14] as u128) << 112) - + ((i[13] as u128) << 104) - + ((i[12] as u128) << 96) - + ((i[11] as u128) << 88) - + ((i[10] as u128) << 80) - + ((i[9] as u128) << 72) - + ((i[8] as u128) << 64) - + ((i[7] as u128) << 56) - + ((i[6] as u128) << 48) - + ((i[5] as u128) << 40) - + ((i[4] as u128) << 32) - + ((i[3] as u128) << 24) - + ((i[2] as u128) << 16) - + ((i[1] as u128) << 8) - + i[0] as u128; - Ok((&i[16..], res)) + let mut res = 0u128; + for (index, byte) in input.iter_indices().take(bound) { + res += (byte as u128) << (8 * index); + } + + Ok((input.slice(bound..), res)) } } -/// Recognizes a signed 1 byte integer +/// Recognizes a signed 1 byte integer. /// -/// *complete version*: returns an error if there is not enough input data +/// *Complete version*: Returns an error if there is not enough input data. /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; /// # use nom::Needed::Size; @@ -481,17 +559,20 @@ pub fn le_u128<'a, E: ParseError<&'a [u8]>>(i: &'a[u8]) -> IResult<&'a[u8], u128 /// le_i8(s) /// }; /// -/// assert_eq!(parser(b"\x00\x03abcefg"), Ok((&b"\x03abcefg"[..], 0x00))); -/// assert_eq!(parser(b""), Err(Err::Error((&[][..], ErrorKind::Eof)))); +/// assert_eq!(parser(&b"\x00\x03abcefg"[..]), Ok((&b"\x03abcefg"[..], 0x00))); +/// assert_eq!(parser(&b""[..]), Err(Err::Error((&[][..], ErrorKind::Eof)))); /// ``` #[inline] -pub fn le_i8<'a, E: ParseError<&'a [u8]>>(i: &'a[u8]) -> IResult<&'a[u8], i8, E> { - map!(i, le_u8, |x| x as i8) +pub fn le_i8<I, E: ParseError<I>>(input: I) -> IResult<I, i8, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + map!(input, be_u8, |x| x as i8) } -/// Recognizes a little endian signed 2 bytes integer +/// Recognizes a little endian signed 2 bytes integer. /// -/// *complete version*: returns an error if there is not enough input data +/// *Complete version*: Returns an error if there is not enough input data. /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; /// # use nom::Needed::Size; @@ -501,17 +582,20 @@ pub fn le_i8<'a, E: ParseError<&'a [u8]>>(i: &'a[u8]) -> IResult<&'a[u8], i8, E> /// le_i16(s) /// }; /// -/// assert_eq!(parser(b"\x00\x03abcefg"), Ok((&b"abcefg"[..], 0x0300))); -/// assert_eq!(parser(b"\x01"), Err(Err::Error((&[0x01][..], ErrorKind::Eof)))); +/// assert_eq!(parser(&b"\x00\x03abcefg"[..]), Ok((&b"abcefg"[..], 0x0300))); +/// assert_eq!(parser(&b"\x01"[..]), Err(Err::Error((&[0x01][..], ErrorKind::Eof)))); /// ``` #[inline] -pub fn le_i16<'a, E: ParseError<&'a [u8]>>(i: &'a[u8]) -> IResult<&'a[u8], i16, E> { - map!(i, le_u16, |x| x as i16) +pub fn le_i16<I, E: ParseError<I>>(input: I) -> IResult<I, i16, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + map!(input, le_u16, |x| x as i16) } -/// Recognizes a little endian signed 3 bytes integer +/// Recognizes a little endian signed 3 bytes integer. /// -/// *complete version*: returns an error if there is not enough input data +/// *Complete version*: Returns an error if there is not enough input data. /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; /// # use nom::Needed::Size; @@ -521,22 +605,25 @@ pub fn le_i16<'a, E: ParseError<&'a [u8]>>(i: &'a[u8]) -> IResult<&'a[u8], i16, /// le_i24(s) /// }; /// -/// assert_eq!(parser(b"\x00\x03\x05abcefg"), Ok((&b"abcefg"[..], 0x050300))); -/// assert_eq!(parser(b"\x01"), Err(Err::Error((&[0x01][..], ErrorKind::Eof)))); +/// assert_eq!(parser(&b"\x00\x03\x05abcefg"[..]), Ok((&b"abcefg"[..], 0x050300))); +/// assert_eq!(parser(&b"\x01"[..]), Err(Err::Error((&[0x01][..], ErrorKind::Eof)))); /// ``` #[inline] -pub fn le_i24<'a, E: ParseError<&'a [u8]>>(i: &'a[u8]) -> IResult<&'a[u8], i32, E> { +pub fn le_i24<I, E: ParseError<I>>(input: I) -> IResult<I, i32, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ // Same as the unsigned version but we need to sign-extend manually here - map!(i, le_u24, |x| if x & 0x80_00_00 != 0 { + map!(input, le_u24, |x| if x & 0x80_00_00 != 0 { (x | 0xff_00_00_00) as i32 } else { x as i32 }) } -/// Recognizes a little endian signed 4 bytes integer +/// Recognizes a little endian signed 4 bytes integer. /// -/// *complete version*: returns an error if there is not enough input data +/// *Complete version*: Returns an error if there is not enough input data. /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; /// # use nom::Needed::Size; @@ -546,17 +633,20 @@ pub fn le_i24<'a, E: ParseError<&'a [u8]>>(i: &'a[u8]) -> IResult<&'a[u8], i32, /// le_i32(s) /// }; /// -/// assert_eq!(parser(b"\x00\x03\x05\x07abcefg"), Ok((&b"abcefg"[..], 0x07050300))); -/// assert_eq!(parser(b"\x01"), Err(Err::Error((&[0x01][..], ErrorKind::Eof)))); +/// assert_eq!(parser(&b"\x00\x03\x05\x07abcefg"[..]), Ok((&b"abcefg"[..], 0x07050300))); +/// assert_eq!(parser(&b"\x01"[..]), Err(Err::Error((&[0x01][..], ErrorKind::Eof)))); /// ``` #[inline] -pub fn le_i32<'a, E: ParseError<&'a [u8]>>(i: &'a[u8]) -> IResult<&'a[u8], i32, E> { - map!(i, le_u32, |x| x as i32) +pub fn le_i32<I, E: ParseError<I>>(input: I) -> IResult<I, i32, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + map!(input, le_u32, |x| x as i32) } -/// Recognizes a little endian signed 8 bytes integer +/// Recognizes a little endian signed 8 bytes integer. /// -/// *complete version*: returns an error if there is not enough input data +/// *Complete version*: Returns an error if there is not enough input data. /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; /// # use nom::Needed::Size; @@ -566,17 +656,20 @@ pub fn le_i32<'a, E: ParseError<&'a [u8]>>(i: &'a[u8]) -> IResult<&'a[u8], i32, /// le_i64(s) /// }; /// -/// assert_eq!(parser(b"\x00\x01\x02\x03\x04\x05\x06\x07abcefg"), Ok((&b"abcefg"[..], 0x0706050403020100))); -/// assert_eq!(parser(b"\x01"), Err(Err::Error((&[0x01][..], ErrorKind::Eof)))); +/// assert_eq!(parser(&b"\x00\x01\x02\x03\x04\x05\x06\x07abcefg"[..]), Ok((&b"abcefg"[..], 0x0706050403020100))); +/// assert_eq!(parser(&b"\x01"[..]), Err(Err::Error((&[0x01][..], ErrorKind::Eof)))); /// ``` #[inline] -pub fn le_i64<'a, E: ParseError<&'a [u8]>>(i: &'a[u8]) -> IResult<&'a[u8], i64, E> { - map!(i, le_u64, |x| x as i64) +pub fn le_i64<I, E: ParseError<I>>(input: I) -> IResult<I, i64, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + map!(input, le_u64, |x| x as i64) } -/// Recognizes a little endian signed 16 bytes integer +/// Recognizes a little endian signed 16 bytes integer. /// -/// *complete version*: returns an error if there is not enough input data +/// *Complete version*: Returns an error if there is not enough input data. /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; /// # use nom::Needed::Size; @@ -586,21 +679,472 @@ pub fn le_i64<'a, E: ParseError<&'a [u8]>>(i: &'a[u8]) -> IResult<&'a[u8], i64, /// le_i128(s) /// }; /// -/// assert_eq!(parser(b"\x00\x01\x02\x03\x04\x05\x06\x07\x00\x01\x02\x03\x04\x05\x06\x07abcefg"), Ok((&b"abcefg"[..], 0x07060504030201000706050403020100))); -/// assert_eq!(parser(b"\x01"), Err(Err::Error((&[0x01][..], ErrorKind::Eof)))); +/// assert_eq!(parser(&b"\x00\x01\x02\x03\x04\x05\x06\x07\x00\x01\x02\x03\x04\x05\x06\x07abcefg"[..]), Ok((&b"abcefg"[..], 0x07060504030201000706050403020100))); +/// assert_eq!(parser(&b"\x01"[..]), Err(Err::Error((&[0x01][..], ErrorKind::Eof)))); +/// ``` +#[inline] +#[cfg(stable_i128)] +pub fn le_i128<I, E: ParseError<I>>(input: I) -> IResult<I, i128, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + map!(input, le_u128, |x| x as i128) +} + +/// Recognizes an unsigned 1 byte integer +/// +/// Note that endianness does not apply to 1 byte numbers. +/// *complete version*: returns an error if there is not enough input data +/// ```rust +/// # use nom::{Err, error::ErrorKind, Needed}; +/// # use nom::Needed::Size; +/// use nom::number::complete::u8; +/// +/// let parser = |s| { +/// u8(s) +/// }; +/// +/// assert_eq!(parser(&b"\x00\x03abcefg"[..]), Ok((&b"\x03abcefg"[..], 0x00))); +/// assert_eq!(parser(&b""[..]), Err(Err::Error((&[][..], ErrorKind::Eof)))); +/// ``` +#[inline] +pub fn u8<I, E: ParseError<I>>(input: I) -> IResult<I, u8, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + let bound: usize = 1; + if input.input_len() < bound { + Err(Err::Error(make_error(input, ErrorKind::Eof))) + } else { + let res = input.iter_elements().next().unwrap(); + + Ok((input.slice(bound..), res)) + } +} + +/// Recognizes an unsigned 2 bytes integer +/// +/// If the parameter is `nom::number::Endianness::Big`, parse a big endian u16 integer, +/// otherwise if `nom::number::Endianness::Little` parse a little endian u16 integer. +/// *complete version*: returns an error if there is not enough input data +/// +/// ```rust +/// # use nom::{Err, error::ErrorKind, Needed}; +/// # use nom::Needed::Size; +/// use nom::number::complete::u16; +/// +/// let be_u16 = |s| { +/// u16(nom::number::Endianness::Big)(s) +/// }; +/// +/// assert_eq!(be_u16(&b"\x00\x03abcefg"[..]), Ok((&b"abcefg"[..], 0x0003))); +/// assert_eq!(be_u16(&b"\x01"[..]), Err(Err::Error((&[0x01][..], ErrorKind::Eof)))); +/// +/// let le_u16 = |s| { +/// u16(nom::number::Endianness::Little)(s) +/// }; +/// +/// assert_eq!(le_u16(&b"\x00\x03abcefg"[..]), Ok((&b"abcefg"[..], 0x0300))); +/// assert_eq!(le_u16(&b"\x01"[..]), Err(Err::Error((&[0x01][..], ErrorKind::Eof)))); +/// ``` +#[inline] +pub fn u16<I, E: ParseError<I>>(endian: crate::number::Endianness) -> fn(I) -> IResult<I, u16, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + match endian { + crate::number::Endianness::Big => be_u16, + crate::number::Endianness::Little => le_u16, + #[cfg(target_endian = "big")] + crate::number::Endianness::Native => be_u16, + #[cfg(target_endian = "little")] + crate::number::Endianness::Native => le_u16, + } +} + +/// Recognizes an unsigned 3 byte integer +/// +/// If the parameter is `nom::number::Endianness::Big`, parse a big endian u24 integer, +/// otherwise if `nom::number::Endianness::Little` parse a little endian u24 integer. +/// *complete version*: returns an error if there is not enough input data +/// ```rust +/// # use nom::{Err, error::ErrorKind, Needed}; +/// # use nom::Needed::Size; +/// use nom::number::complete::u24; +/// +/// let be_u24 = |s| { +/// u24(nom::number::Endianness::Big)(s) +/// }; +/// +/// assert_eq!(be_u24(&b"\x00\x03\x05abcefg"[..]), Ok((&b"abcefg"[..], 0x000305))); +/// assert_eq!(be_u24(&b"\x01"[..]), Err(Err::Error((&[0x01][..], ErrorKind::Eof)))); +/// +/// let le_u24 = |s| { +/// u24(nom::number::Endianness::Little)(s) +/// }; +/// +/// assert_eq!(le_u24(&b"\x00\x03\x05abcefg"[..]), Ok((&b"abcefg"[..], 0x050300))); +/// assert_eq!(le_u24(&b"\x01"[..]), Err(Err::Error((&[0x01][..], ErrorKind::Eof)))); +/// ``` +#[inline] +pub fn u24<I, E: ParseError<I>>(endian: crate::number::Endianness) -> fn(I) -> IResult<I, u32, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + match endian { + crate::number::Endianness::Big => be_u24, + crate::number::Endianness::Little => le_u24, + #[cfg(target_endian = "big")] + crate::number::Endianness::Native => be_u24, + #[cfg(target_endian = "little")] + crate::number::Endianness::Native => le_u24, + } +} + +/// Recognizes an unsigned 4 byte integer +/// +/// If the parameter is `nom::number::Endianness::Big`, parse a big endian u32 integer, +/// otherwise if `nom::number::Endianness::Little` parse a little endian u32 integer. +/// *complete version*: returns an error if there is not enough input data +/// ```rust +/// # use nom::{Err, error::ErrorKind, Needed}; +/// # use nom::Needed::Size; +/// use nom::number::complete::u32; +/// +/// let be_u32 = |s| { +/// u32(nom::number::Endianness::Big)(s) +/// }; +/// +/// assert_eq!(be_u32(&b"\x00\x03\x05\x07abcefg"[..]), Ok((&b"abcefg"[..], 0x00030507))); +/// assert_eq!(be_u32(&b"\x01"[..]), Err(Err::Error((&[0x01][..], ErrorKind::Eof)))); +/// +/// let le_u32 = |s| { +/// u32(nom::number::Endianness::Little)(s) +/// }; +/// +/// assert_eq!(le_u32(&b"\x00\x03\x05\x07abcefg"[..]), Ok((&b"abcefg"[..], 0x07050300))); +/// assert_eq!(le_u32(&b"\x01"[..]), Err(Err::Error((&[0x01][..], ErrorKind::Eof)))); +/// ``` +#[inline] +pub fn u32<I, E: ParseError<I>>(endian: crate::number::Endianness) -> fn(I) -> IResult<I, u32, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + match endian { + crate::number::Endianness::Big => be_u32, + crate::number::Endianness::Little => le_u32, + #[cfg(target_endian = "big")] + crate::number::Endianness::Native => be_u32, + #[cfg(target_endian = "little")] + crate::number::Endianness::Native => le_u32, + } +} + +/// Recognizes an unsigned 8 byte integer +/// +/// If the parameter is `nom::number::Endianness::Big`, parse a big endian u64 integer, +/// otherwise if `nom::number::Endianness::Little` parse a little endian u64 integer. +/// *complete version*: returns an error if there is not enough input data +/// ```rust +/// # use nom::{Err, error::ErrorKind, Needed}; +/// # use nom::Needed::Size; +/// use nom::number::complete::u64; +/// +/// let be_u64 = |s| { +/// u64(nom::number::Endianness::Big)(s) +/// }; +/// +/// assert_eq!(be_u64(&b"\x00\x01\x02\x03\x04\x05\x06\x07abcefg"[..]), Ok((&b"abcefg"[..], 0x0001020304050607))); +/// assert_eq!(be_u64(&b"\x01"[..]), Err(Err::Error((&[0x01][..], ErrorKind::Eof)))); +/// +/// let le_u64 = |s| { +/// u64(nom::number::Endianness::Little)(s) +/// }; +/// +/// assert_eq!(le_u64(&b"\x00\x01\x02\x03\x04\x05\x06\x07abcefg"[..]), Ok((&b"abcefg"[..], 0x0706050403020100))); +/// assert_eq!(le_u64(&b"\x01"[..]), Err(Err::Error((&[0x01][..], ErrorKind::Eof)))); +/// ``` +#[inline] +pub fn u64<I, E: ParseError<I>>(endian: crate::number::Endianness) -> fn(I) -> IResult<I, u64, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + match endian { + crate::number::Endianness::Big => be_u64, + crate::number::Endianness::Little => le_u64, + #[cfg(target_endian = "big")] + crate::number::Endianness::Native => be_u64, + #[cfg(target_endian = "little")] + crate::number::Endianness::Native => le_u64, + } +} + +/// Recognizes an unsigned 16 byte integer +/// +/// If the parameter is `nom::number::Endianness::Big`, parse a big endian u128 integer, +/// otherwise if `nom::number::Endianness::Little` parse a little endian u128 integer. +/// *complete version*: returns an error if there is not enough input data +/// ```rust +/// # use nom::{Err, error::ErrorKind, Needed}; +/// # use nom::Needed::Size; +/// use nom::number::complete::u128; +/// +/// let be_u128 = |s| { +/// u128(nom::number::Endianness::Big)(s) +/// }; +/// +/// assert_eq!(be_u128(&b"\x00\x01\x02\x03\x04\x05\x06\x07\x00\x01\x02\x03\x04\x05\x06\x07abcefg"[..]), Ok((&b"abcefg"[..], 0x00010203040506070001020304050607))); +/// assert_eq!(be_u128(&b"\x01"[..]), Err(Err::Error((&[0x01][..], ErrorKind::Eof)))); +/// +/// let le_u128 = |s| { +/// u128(nom::number::Endianness::Little)(s) +/// }; +/// +/// assert_eq!(le_u128(&b"\x00\x01\x02\x03\x04\x05\x06\x07\x00\x01\x02\x03\x04\x05\x06\x07abcefg"[..]), Ok((&b"abcefg"[..], 0x07060504030201000706050403020100))); +/// assert_eq!(le_u128(&b"\x01"[..]), Err(Err::Error((&[0x01][..], ErrorKind::Eof)))); /// ``` #[inline] #[cfg(stable_i128)] -pub fn le_i128<'a, E: ParseError<&'a [u8]>>(i: &'a[u8]) -> IResult<&'a[u8], i128, E> { - map!(i, le_u128, |x| x as i128) +pub fn u128<I, E: ParseError<I>>(endian: crate::number::Endianness) -> fn(I) -> IResult<I, u128, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + match endian { + crate::number::Endianness::Big => be_u128, + crate::number::Endianness::Little => le_u128, + #[cfg(target_endian = "big")] + crate::number::Endianness::Native => be_u128, + #[cfg(target_endian = "little")] + crate::number::Endianness::Native => le_u128, + } +} + +/// Recognizes a signed 1 byte integer +/// +/// Note that endianness does not apply to 1 byte numbers. +/// *complete version*: returns an error if there is not enough input data +/// ```rust +/// # use nom::{Err, error::ErrorKind, Needed}; +/// # use nom::Needed::Size; +/// use nom::number::complete::i8; +/// +/// let parser = |s| { +/// i8(s) +/// }; +/// +/// assert_eq!(parser(&b"\x00\x03abcefg"[..]), Ok((&b"\x03abcefg"[..], 0x00))); +/// assert_eq!(parser(&b""[..]), Err(Err::Error((&[][..], ErrorKind::Eof)))); +/// ``` +#[inline] +pub fn i8<I, E: ParseError<I>>(i: I) -> IResult<I, i8, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + map!(i, u8, |x| x as i8) +} + +/// Recognizes a signed 2 byte integer +/// +/// If the parameter is `nom::number::Endianness::Big`, parse a big endian i16 integer, +/// otherwise if `nom::number::Endianness::Little` parse a little endian i16 integer. +/// *complete version*: returns an error if there is not enough input data +/// ```rust +/// # use nom::{Err, error::ErrorKind, Needed}; +/// # use nom::Needed::Size; +/// use nom::number::complete::i16; +/// +/// let be_i16 = |s| { +/// i16(nom::number::Endianness::Big)(s) +/// }; +/// +/// assert_eq!(be_i16(&b"\x00\x03abcefg"[..]), Ok((&b"abcefg"[..], 0x0003))); +/// assert_eq!(be_i16(&b"\x01"[..]), Err(Err::Error((&[0x01][..], ErrorKind::Eof)))); +/// +/// let le_i16 = |s| { +/// i16(nom::number::Endianness::Little)(s) +/// }; +/// +/// assert_eq!(le_i16(&b"\x00\x03abcefg"[..]), Ok((&b"abcefg"[..], 0x0300))); +/// assert_eq!(le_i16(&b"\x01"[..]), Err(Err::Error((&[0x01][..], ErrorKind::Eof)))); +/// ``` +#[inline] +pub fn i16<I, E: ParseError<I>>(endian: crate::number::Endianness) -> fn(I) -> IResult<I, i16, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + match endian { + crate::number::Endianness::Big => be_i16, + crate::number::Endianness::Little => le_i16, + #[cfg(target_endian = "big")] + crate::number::Endianness::Native => be_i16, + #[cfg(target_endian = "little")] + crate::number::Endianness::Native => le_i16, + } +} + +/// Recognizes a signed 3 byte integer +/// +/// If the parameter is `nom::number::Endianness::Big`, parse a big endian i24 integer, +/// otherwise if `nom::number::Endianness::Little` parse a little endian i24 integer. +/// *complete version*: returns an error if there is not enough input data +/// ```rust +/// # use nom::{Err, error::ErrorKind, Needed}; +/// # use nom::Needed::Size; +/// use nom::number::complete::i24; +/// +/// let be_i24 = |s| { +/// i24(nom::number::Endianness::Big)(s) +/// }; +/// +/// assert_eq!(be_i24(&b"\x00\x03\x05abcefg"[..]), Ok((&b"abcefg"[..], 0x000305))); +/// assert_eq!(be_i24(&b"\x01"[..]), Err(Err::Error((&[0x01][..], ErrorKind::Eof)))); +/// +/// let le_i24 = |s| { +/// i24(nom::number::Endianness::Little)(s) +/// }; +/// +/// assert_eq!(le_i24(&b"\x00\x03\x05abcefg"[..]), Ok((&b"abcefg"[..], 0x050300))); +/// assert_eq!(le_i24(&b"\x01"[..]), Err(Err::Error((&[0x01][..], ErrorKind::Eof)))); +/// ``` +#[inline] +pub fn i24<I, E: ParseError<I>>(endian: crate::number::Endianness) -> fn(I) -> IResult<I, i32, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + match endian { + crate::number::Endianness::Big => be_i24, + crate::number::Endianness::Little => le_i24, + #[cfg(target_endian = "big")] + crate::number::Endianness::Native => be_i24, + #[cfg(target_endian = "little")] + crate::number::Endianness::Native => le_i24, + } +} + +/// Recognizes a signed 4 byte integer +/// +/// If the parameter is `nom::number::Endianness::Big`, parse a big endian i32 integer, +/// otherwise if `nom::number::Endianness::Little` parse a little endian i32 integer. +/// *complete version*: returns an error if there is not enough input data +/// ```rust +/// # use nom::{Err, error::ErrorKind, Needed}; +/// # use nom::Needed::Size; +/// use nom::number::complete::i32; +/// +/// let be_i32 = |s| { +/// i32(nom::number::Endianness::Big)(s) +/// }; +/// +/// assert_eq!(be_i32(&b"\x00\x03\x05\x07abcefg"[..]), Ok((&b"abcefg"[..], 0x00030507))); +/// assert_eq!(be_i32(&b"\x01"[..]), Err(Err::Error((&[0x01][..], ErrorKind::Eof)))); +/// +/// let le_i32 = |s| { +/// i32(nom::number::Endianness::Little)(s) +/// }; +/// +/// assert_eq!(le_i32(&b"\x00\x03\x05\x07abcefg"[..]), Ok((&b"abcefg"[..], 0x07050300))); +/// assert_eq!(le_i32(&b"\x01"[..]), Err(Err::Error((&[0x01][..], ErrorKind::Eof)))); +/// ``` +#[inline] +pub fn i32<I, E: ParseError<I>>(endian: crate::number::Endianness) -> fn(I) -> IResult<I, i32, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + match endian { + crate::number::Endianness::Big => be_i32, + crate::number::Endianness::Little => le_i32, + #[cfg(target_endian = "big")] + crate::number::Endianness::Native => be_i32, + #[cfg(target_endian = "little")] + crate::number::Endianness::Native => le_i32, + } +} + +/// Recognizes a signed 8 byte integer +/// +/// If the parameter is `nom::number::Endianness::Big`, parse a big endian i64 integer, +/// otherwise if `nom::number::Endianness::Little` parse a little endian i64 integer. +/// *complete version*: returns an error if there is not enough input data +/// ```rust +/// # use nom::{Err, error::ErrorKind, Needed}; +/// # use nom::Needed::Size; +/// use nom::number::complete::i64; +/// +/// let be_i64 = |s| { +/// i64(nom::number::Endianness::Big)(s) +/// }; +/// +/// assert_eq!(be_i64(&b"\x00\x01\x02\x03\x04\x05\x06\x07abcefg"[..]), Ok((&b"abcefg"[..], 0x0001020304050607))); +/// assert_eq!(be_i64(&b"\x01"[..]), Err(Err::Error((&[0x01][..], ErrorKind::Eof)))); +/// +/// let le_i64 = |s| { +/// i64(nom::number::Endianness::Little)(s) +/// }; +/// +/// assert_eq!(le_i64(&b"\x00\x01\x02\x03\x04\x05\x06\x07abcefg"[..]), Ok((&b"abcefg"[..], 0x0706050403020100))); +/// assert_eq!(le_i64(&b"\x01"[..]), Err(Err::Error((&[0x01][..], ErrorKind::Eof)))); +/// ``` +#[inline] +pub fn i64<I, E: ParseError<I>>(endian: crate::number::Endianness) -> fn(I) -> IResult<I, i64, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + match endian { + crate::number::Endianness::Big => be_i64, + crate::number::Endianness::Little => le_i64, + #[cfg(target_endian = "big")] + crate::number::Endianness::Native => be_i64, + #[cfg(target_endian = "little")] + crate::number::Endianness::Native => le_i64, + } } -/// Recognizes a big endian 4 bytes floating point number +/// Recognizes a signed 16 byte integer /// +/// If the parameter is `nom::number::Endianness::Big`, parse a big endian i128 integer, +/// otherwise if `nom::number::Endianness::Little` parse a little endian i128 integer. /// *complete version*: returns an error if there is not enough input data /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; /// # use nom::Needed::Size; +/// use nom::number::complete::i128; +/// +/// let be_i128 = |s| { +/// i128(nom::number::Endianness::Big)(s) +/// }; +/// +/// assert_eq!(be_i128(&b"\x00\x01\x02\x03\x04\x05\x06\x07\x00\x01\x02\x03\x04\x05\x06\x07abcefg"[..]), Ok((&b"abcefg"[..], 0x00010203040506070001020304050607))); +/// assert_eq!(be_i128(&b"\x01"[..]), Err(Err::Error((&[0x01][..], ErrorKind::Eof)))); +/// +/// let le_i128 = |s| { +/// i128(nom::number::Endianness::Little)(s) +/// }; +/// +/// assert_eq!(le_i128(&b"\x00\x01\x02\x03\x04\x05\x06\x07\x00\x01\x02\x03\x04\x05\x06\x07abcefg"[..]), Ok((&b"abcefg"[..], 0x07060504030201000706050403020100))); +/// assert_eq!(le_i128(&b"\x01"[..]), Err(Err::Error((&[0x01][..], ErrorKind::Eof)))); +/// ``` +#[inline] +#[cfg(stable_i128)] +pub fn i128<I, E: ParseError<I>>(endian: crate::number::Endianness) -> fn(I) -> IResult<I, i128, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + match endian { + crate::number::Endianness::Big => be_i128, + crate::number::Endianness::Little => le_i128, + #[cfg(target_endian = "big")] + crate::number::Endianness::Native => be_i128, + #[cfg(target_endian = "little")] + crate::number::Endianness::Native => le_i128, + } +} + +/// Recognizes a big endian 4 bytes floating point number. +/// +/// *Complete version*: Returns an error if there is not enough input data. +/// ```rust +/// # use nom::{Err, error::ErrorKind, Needed}; +/// # use nom::Needed::Size; /// use nom::number::complete::be_f32; /// /// let parser = |s| { @@ -608,19 +1152,22 @@ pub fn le_i128<'a, E: ParseError<&'a [u8]>>(i: &'a[u8]) -> IResult<&'a[u8], i128 /// }; /// /// assert_eq!(parser(&[0x41, 0x48, 0x00, 0x00][..]), Ok((&b""[..], 12.5))); -/// assert_eq!(parser(b"abc"), Err(Err::Error((&b"abc"[..], ErrorKind::Eof)))); +/// assert_eq!(parser(&b"abc"[..]), Err(Err::Error((&b"abc"[..], ErrorKind::Eof)))); /// ``` #[inline] -pub fn be_f32<'a, E: ParseError<&'a [u8]>>(input: &'a[u8]) -> IResult<&'a[u8], f32, E> { +pub fn be_f32<I, E: ParseError<I>>(input: I) -> IResult<I, f32, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ match be_u32(input) { Err(e) => Err(e), Ok((i, o)) => Ok((i, f32::from_bits(o))), } } -/// Recognizes a big endian 8 bytes floating point number +/// Recognizes a big endian 8 bytes floating point number. /// -/// *complete version*: returns an error if there is not enough input data +/// *Complete version*: Returns an error if there is not enough input data. /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; /// # use nom::Needed::Size; @@ -631,19 +1178,22 @@ pub fn be_f32<'a, E: ParseError<&'a [u8]>>(input: &'a[u8]) -> IResult<&'a[u8], f /// }; /// /// assert_eq!(parser(&[0x40, 0x29, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00][..]), Ok((&b""[..], 12.5))); -/// assert_eq!(parser(b"abc"), Err(Err::Error((&b"abc"[..], ErrorKind::Eof)))); +/// assert_eq!(parser(&b"abc"[..]), Err(Err::Error((&b"abc"[..], ErrorKind::Eof)))); /// ``` #[inline] -pub fn be_f64<'a, E: ParseError<&'a [u8]>>(input: &'a[u8]) -> IResult<&'a[u8], f64, E> { +pub fn be_f64<I, E: ParseError<I>>(input: I) -> IResult<I, f64, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ match be_u64(input) { Err(e) => Err(e), Ok((i, o)) => Ok((i, f64::from_bits(o))), } } -/// Recognizes a little endian 4 bytes floating point number +/// Recognizes a little endian 4 bytes floating point number. /// -/// *complete version*: returns an error if there is not enough input data +/// *Complete version*: Returns an error if there is not enough input data. /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; /// # use nom::Needed::Size; @@ -654,19 +1204,22 @@ pub fn be_f64<'a, E: ParseError<&'a [u8]>>(input: &'a[u8]) -> IResult<&'a[u8], f /// }; /// /// assert_eq!(parser(&[0x00, 0x00, 0x48, 0x41][..]), Ok((&b""[..], 12.5))); -/// assert_eq!(parser(b"abc"), Err(Err::Error((&b"abc"[..], ErrorKind::Eof)))); +/// assert_eq!(parser(&b"abc"[..]), Err(Err::Error((&b"abc"[..], ErrorKind::Eof)))); /// ``` #[inline] -pub fn le_f32<'a, E: ParseError<&'a [u8]>>(input: &'a[u8]) -> IResult<&'a[u8], f32, E> { +pub fn le_f32<I, E: ParseError<I>>(input: I) -> IResult<I, f32, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ match le_u32(input) { Err(e) => Err(e), Ok((i, o)) => Ok((i, f32::from_bits(o))), } } -/// Recognizes a little endian 8 bytes floating point number +/// Recognizes a little endian 8 bytes floating point number. /// -/// *complete version*: returns an error if there is not enough input data +/// *Complete version*: Returns an error if there is not enough input data. /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; /// # use nom::Needed::Size; @@ -677,19 +1230,100 @@ pub fn le_f32<'a, E: ParseError<&'a [u8]>>(input: &'a[u8]) -> IResult<&'a[u8], f /// }; /// /// assert_eq!(parser(&[0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x29, 0x40][..]), Ok((&b""[..], 12.5))); -/// assert_eq!(parser(b"abc"), Err(Err::Error((&b"abc"[..], ErrorKind::Eof)))); +/// assert_eq!(parser(&b"abc"[..]), Err(Err::Error((&b"abc"[..], ErrorKind::Eof)))); /// ``` #[inline] -pub fn le_f64<'a, E: ParseError<&'a [u8]>>(input: &'a[u8]) -> IResult<&'a[u8], f64, E> { +pub fn le_f64<I, E: ParseError<I>>(input: I) -> IResult<I, f64, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ match le_u64(input) { Err(e) => Err(e), Ok((i, o)) => Ok((i, f64::from_bits(o))), } } -/// Recognizes a hex-encoded integer +/// Recognizes a 4 byte floating point number /// -/// *complete version*: will parse until the end of input if it has less than 8 bytes +/// If the parameter is `nom::number::Endianness::Big`, parse a big endian f32 float, +/// otherwise if `nom::number::Endianness::Little` parse a little endian f32 float. +/// *complete version*: returns an error if there is not enough input data +/// ```rust +/// # use nom::{Err, error::ErrorKind, Needed}; +/// # use nom::Needed::Size; +/// use nom::number::complete::f32; +/// +/// let be_f32 = |s| { +/// f32(nom::number::Endianness::Big)(s) +/// }; +/// +/// assert_eq!(be_f32(&[0x41, 0x48, 0x00, 0x00][..]), Ok((&b""[..], 12.5))); +/// assert_eq!(be_f32(&b"abc"[..]), Err(Err::Error((&b"abc"[..], ErrorKind::Eof)))); +/// +/// let le_f32 = |s| { +/// f32(nom::number::Endianness::Little)(s) +/// }; +/// +/// assert_eq!(le_f32(&[0x00, 0x00, 0x48, 0x41][..]), Ok((&b""[..], 12.5))); +/// assert_eq!(le_f32(&b"abc"[..]), Err(Err::Error((&b"abc"[..], ErrorKind::Eof)))); +/// ``` +#[inline] +pub fn f32<I, E: ParseError<I>>(endian: crate::number::Endianness) -> fn(I) -> IResult<I, f32, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + match endian { + crate::number::Endianness::Big => be_f32, + crate::number::Endianness::Little => le_f32, + #[cfg(target_endian = "big")] + crate::number::Endianness::Native => be_f32, + #[cfg(target_endian = "little")] + crate::number::Endianness::Native => le_f32, + } +} + +/// Recognizes an 8 byte floating point number +/// +/// If the parameter is `nom::number::Endianness::Big`, parse a big endian f64 float, +/// otherwise if `nom::number::Endianness::Little` parse a little endian f64 float. +/// *complete version*: returns an error if there is not enough input data +/// ```rust +/// # use nom::{Err, error::ErrorKind, Needed}; +/// # use nom::Needed::Size; +/// use nom::number::complete::f64; +/// +/// let be_f64 = |s| { +/// f64(nom::number::Endianness::Big)(s) +/// }; +/// +/// assert_eq!(be_f64(&[0x40, 0x29, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00][..]), Ok((&b""[..], 12.5))); +/// assert_eq!(be_f64(&b"abc"[..]), Err(Err::Error((&b"abc"[..], ErrorKind::Eof)))); +/// +/// let le_f64 = |s| { +/// f64(nom::number::Endianness::Little)(s) +/// }; +/// +/// assert_eq!(le_f64(&[0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x29, 0x40][..]), Ok((&b""[..], 12.5))); +/// assert_eq!(le_f64(&b"abc"[..]), Err(Err::Error((&b"abc"[..], ErrorKind::Eof)))); +/// ``` +#[inline] +pub fn f64<I, E: ParseError<I>>(endian: crate::number::Endianness) -> fn(I) -> IResult<I, f64, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + match endian { + crate::number::Endianness::Big => be_f64, + crate::number::Endianness::Little => le_f64, + #[cfg(target_endian = "big")] + crate::number::Endianness::Native => be_f64, + #[cfg(target_endian = "little")] + crate::number::Endianness::Native => le_f64, + } +} + +/// Recognizes a hex-encoded integer. +/// +/// *Complete version*: Will parse until the end of input if it has less than 8 bytes. /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; /// # use nom::Needed::Size; @@ -699,12 +1333,12 @@ pub fn le_f64<'a, E: ParseError<&'a [u8]>>(input: &'a[u8]) -> IResult<&'a[u8], f /// hex_u32(s) /// }; /// -/// assert_eq!(parser(b"01AE"), Ok((&b""[..], 0x01AE))); -/// assert_eq!(parser(b"abc"), Ok((&b""[..], 0x0ABC))); -/// assert_eq!(parser(b"ggg"), Err(Err::Error((&b"ggg"[..], ErrorKind::IsA)))); +/// assert_eq!(parser(&b"01AE"[..]), Ok((&b""[..], 0x01AE))); +/// assert_eq!(parser(&b"abc"[..]), Ok((&b""[..], 0x0ABC))); +/// assert_eq!(parser(&b"ggg"[..]), Err(Err::Error((&b"ggg"[..], ErrorKind::IsA)))); /// ``` #[inline] -pub fn hex_u32<'a, E: ParseError<&'a [u8]>>(input: &'a[u8]) -> IResult<&'a[u8], u32, E> { +pub fn hex_u32<'a, E: ParseError<&'a [u8]>>(input: &'a [u8]) -> IResult<&'a [u8], u32, E> { let (i, o) = crate::bytes::complete::is_a(&b"0123456789abcdefABCDEF"[..])(input)?; // Do not parse more than 8 characters for a u32 let (parsed, remaining) = if o.len() <= 8 { @@ -726,9 +1360,9 @@ pub fn hex_u32<'a, E: ParseError<&'a [u8]>>(input: &'a[u8]) -> IResult<&'a[u8], Ok((remaining, res)) } -/// Recognizes floating point number in a byte string and returns the corresponding slice +/// Recognizes floating point number in a byte string and returns the corresponding slice. /// -/// *complete version*: can parse until the end of input +/// *Complete version*: Can parse until the end of input. /// /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; @@ -745,7 +1379,7 @@ pub fn hex_u32<'a, E: ParseError<&'a [u8]>>(input: &'a[u8]) -> IResult<&'a[u8], /// assert_eq!(parser("abc"), Err(Err::Error(("abc", ErrorKind::Char)))); /// ``` #[allow(unused_imports)] -#[cfg_attr(rustfmt, rustfmt_skip)] +#[rustfmt::skip] pub fn recognize_float<T, E:ParseError<T>>(input: T) -> IResult<T, T, E> where T: Slice<RangeFrom<usize>> + Slice<RangeTo<usize>>, @@ -771,9 +1405,9 @@ where )(input) } -/// Recognizes floating point number in a byte string and returns a f32 +/// Recognizes floating point number in a byte string and returns a f32. /// -/// *complete version*: can parse until the end of input +/// *Complete version*: Can parse until the end of input. /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; /// # use nom::Needed::Size; @@ -789,30 +1423,30 @@ where /// assert_eq!(parser("abc"), Err(Err::Error(("abc", ErrorKind::Char)))); /// ``` #[cfg(not(feature = "lexical"))] -pub fn float<T, E:ParseError<T>>(input: T) -> IResult<T, f32, E> +pub fn float<T, E: ParseError<T>>(input: T) -> IResult<T, f32, E> where T: Slice<RangeFrom<usize>> + Slice<RangeTo<usize>>, T: Clone + Offset, T: InputIter + InputLength + crate::traits::ParseTo<f32>, <T as InputIter>::Item: AsChar, T: InputTakeAtPosition, - <T as InputTakeAtPosition>::Item: AsChar + <T as InputTakeAtPosition>::Item: AsChar, { match recognize_float(input) { Err(e) => Err(e), Ok((i, s)) => match s.parse_to() { Some(n) => Ok((i, n)), - None => Err(Err::Error(E::from_error_kind(i, ErrorKind::Float))) - } + None => Err(Err::Error(E::from_error_kind(i, ErrorKind::Float))), + }, } } -/// Recognizes floating point number in a byte string and returns a f32 +/// Recognizes floating point number in a byte string and returns a f32. /// -/// *complete version*: can parse until the end of input +/// *Complete version*: Can parse until the end of input. /// -/// this function uses the lexical-core crate for float parsing by default, you -/// can deactivate it by removing the "lexical" feature +/// This function uses the `lexical-core` crate for float parsing by default, you +/// can deactivate it by removing the "lexical" feature. /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; /// # use nom::Needed::Size; @@ -828,19 +1462,19 @@ where /// assert_eq!(parser("abc"), Err(Err::Error(("abc", ErrorKind::Float)))); /// ``` #[cfg(feature = "lexical")] -pub fn float<T, E:ParseError<T>>(input: T) -> IResult<T, f32, E> +pub fn float<T, E: ParseError<T>>(input: T) -> IResult<T, f32, E> where T: crate::traits::AsBytes + InputLength + Slice<RangeFrom<usize>>, { match ::lexical_core::parse_partial(input.as_bytes()) { Ok((value, processed)) => Ok((input.slice(processed..), value)), - Err(_) => Err(Err::Error(E::from_error_kind(input, ErrorKind::Float))) + Err(_) => Err(Err::Error(E::from_error_kind(input, ErrorKind::Float))), } } -/// Recognizes floating point number in a byte string and returns a f64 +/// Recognizes floating point number in a byte string and returns a f64. /// -/// *complete version*: can parse until the end of input +/// *Complete version*: Can parse until the end of input. /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; /// # use nom::Needed::Size; @@ -856,30 +1490,30 @@ where /// assert_eq!(parser("abc"), Err(Err::Error(("abc", ErrorKind::Char)))); /// ``` #[cfg(not(feature = "lexical"))] -pub fn double<T, E:ParseError<T>>(input: T) -> IResult<T, f64, E> +pub fn double<T, E: ParseError<T>>(input: T) -> IResult<T, f64, E> where T: Slice<RangeFrom<usize>> + Slice<RangeTo<usize>>, T: Clone + Offset, T: InputIter + InputLength + crate::traits::ParseTo<f64>, <T as InputIter>::Item: AsChar, T: InputTakeAtPosition, - <T as InputTakeAtPosition>::Item: AsChar + <T as InputTakeAtPosition>::Item: AsChar, { match recognize_float(input) { Err(e) => Err(e), Ok((i, s)) => match s.parse_to() { Some(n) => Ok((i, n)), - None => Err(Err::Error(E::from_error_kind(i, ErrorKind::Float))) - } + None => Err(Err::Error(E::from_error_kind(i, ErrorKind::Float))), + }, } } -/// Recognizes floating point number in a byte string and returns a f64 +/// Recognizes floating point number in a byte string and returns a f64. /// -/// *complete version*: can parse until the end of input +/// *Complete version*: Can parse until the end of input. /// -/// this function uses the lexical-core crate for float parsing by default, you -/// can deactivate it by removing the "lexical" feature +/// This function uses the `lexical-core` crate for float parsing by default, you +/// can deactivate it by removing the "lexical" feature. /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; /// # use nom::Needed::Size; @@ -895,21 +1529,21 @@ where /// assert_eq!(parser("abc"), Err(Err::Error(("abc", ErrorKind::Float)))); /// ``` #[cfg(feature = "lexical")] -pub fn double<T, E:ParseError<T>>(input: T) -> IResult<T, f64, E> +pub fn double<T, E: ParseError<T>>(input: T) -> IResult<T, f64, E> where T: crate::traits::AsBytes + InputLength + Slice<RangeFrom<usize>>, { match ::lexical_core::parse_partial(input.as_bytes()) { Ok((value, processed)) => Ok((input.slice(processed..), value)), - Err(_) => Err(Err::Error(E::from_error_kind(input, ErrorKind::Float))) + Err(_) => Err(Err::Error(E::from_error_kind(input, ErrorKind::Float))), } } #[cfg(test)] mod tests { use super::*; - use crate::internal::Err; use crate::error::ErrorKind; + use crate::internal::Err; macro_rules! assert_parse( ($left: expr, $right: expr) => { @@ -920,129 +1554,175 @@ mod tests { #[test] fn i8_tests() { - assert_parse!(be_i8(&[0x00]), Ok((&b""[..], 0))); - assert_parse!(be_i8(&[0x7f]), Ok((&b""[..], 127))); - assert_parse!(be_i8(&[0xff]), Ok((&b""[..], -1))); - assert_parse!(be_i8(&[0x80]), Ok((&b""[..], -128))); + assert_parse!(i8(&[0x00][..]), Ok((&b""[..], 0))); + assert_parse!(i8(&[0x7f][..]), Ok((&b""[..], 127))); + assert_parse!(i8(&[0xff][..]), Ok((&b""[..], -1))); + assert_parse!(i8(&[0x80][..]), Ok((&b""[..], -128))); } #[test] - fn i16_tests() { - assert_parse!(be_i16(&[0x00, 0x00]), Ok((&b""[..], 0))); - assert_parse!(be_i16(&[0x7f, 0xff]), Ok((&b""[..], 32_767_i16))); - assert_parse!(be_i16(&[0xff, 0xff]), Ok((&b""[..], -1))); - assert_parse!(be_i16(&[0x80, 0x00]), Ok((&b""[..], -32_768_i16))); + fn be_i8_tests() { + assert_parse!(be_i8(&[0x00][..]), Ok((&b""[..], 0))); + assert_parse!(be_i8(&[0x7f][..]), Ok((&b""[..], 127))); + assert_parse!(be_i8(&[0xff][..]), Ok((&b""[..], -1))); + assert_parse!(be_i8(&[0x80][..]), Ok((&b""[..], -128))); } #[test] - fn u24_tests() { - assert_parse!(be_u24(&[0x00, 0x00, 0x00]), Ok((&b""[..], 0))); - assert_parse!(be_u24(&[0x00, 0xFF, 0xFF]), Ok((&b""[..], 65_535_u32))); - assert_parse!(be_u24(&[0x12, 0x34, 0x56]), Ok((&b""[..], 1_193_046_u32))); + fn be_i16_tests() { + assert_parse!(be_i16(&[0x00, 0x00][..]), Ok((&b""[..], 0))); + assert_parse!(be_i16(&[0x7f, 0xff][..]), Ok((&b""[..], 32_767_i16))); + assert_parse!(be_i16(&[0xff, 0xff][..]), Ok((&b""[..], -1))); + assert_parse!(be_i16(&[0x80, 0x00][..]), Ok((&b""[..], -32_768_i16))); } #[test] - fn i24_tests() { - assert_parse!(be_i24(&[0xFF, 0xFF, 0xFF]), Ok((&b""[..], -1_i32))); - assert_parse!(be_i24(&[0xFF, 0x00, 0x00]), Ok((&b""[..], -65_536_i32))); - assert_parse!(be_i24(&[0xED, 0xCB, 0xAA]), Ok((&b""[..], -1_193_046_i32))); + fn be_u24_tests() { + assert_parse!(be_u24(&[0x00, 0x00, 0x00][..]), Ok((&b""[..], 0))); + assert_parse!(be_u24(&[0x00, 0xFF, 0xFF][..]), Ok((&b""[..], 65_535_u32))); + assert_parse!( + be_u24(&[0x12, 0x34, 0x56][..]), + Ok((&b""[..], 1_193_046_u32)) + ); + } + + #[test] + fn be_i24_tests() { + assert_parse!(be_i24(&[0xFF, 0xFF, 0xFF][..]), Ok((&b""[..], -1_i32))); + assert_parse!(be_i24(&[0xFF, 0x00, 0x00][..]), Ok((&b""[..], -65_536_i32))); + assert_parse!( + be_i24(&[0xED, 0xCB, 0xAA][..]), + Ok((&b""[..], -1_193_046_i32)) + ); } #[test] - fn i32_tests() { - assert_parse!(be_i32(&[0x00, 0x00, 0x00, 0x00]), Ok((&b""[..], 0))); + fn be_i32_tests() { + assert_parse!(be_i32(&[0x00, 0x00, 0x00, 0x00][..]), Ok((&b""[..], 0))); assert_parse!( - be_i32(&[0x7f, 0xff, 0xff, 0xff]), + be_i32(&[0x7f, 0xff, 0xff, 0xff][..]), Ok((&b""[..], 2_147_483_647_i32)) ); - assert_parse!(be_i32(&[0xff, 0xff, 0xff, 0xff]), Ok((&b""[..], -1))); + assert_parse!(be_i32(&[0xff, 0xff, 0xff, 0xff][..]), Ok((&b""[..], -1))); assert_parse!( - be_i32(&[0x80, 0x00, 0x00, 0x00]), + be_i32(&[0x80, 0x00, 0x00, 0x00][..]), Ok((&b""[..], -2_147_483_648_i32)) ); } #[test] - fn i64_tests() { + fn be_i64_tests() { assert_parse!( - be_i64(&[0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00]), + be_i64(&[0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00][..]), Ok((&b""[..], 0)) ); assert_parse!( - be_i64(&[0x7f, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff]), + be_i64(&[0x7f, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff][..]), Ok((&b""[..], 9_223_372_036_854_775_807_i64)) ); assert_parse!( - be_i64(&[0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff]), + be_i64(&[0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff][..]), Ok((&b""[..], -1)) ); assert_parse!( - be_i64(&[0x80, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00]), + be_i64(&[0x80, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00][..]), Ok((&b""[..], -9_223_372_036_854_775_808_i64)) ); } #[test] #[cfg(stable_i128)] - fn i128_tests() { + fn be_i128_tests() { assert_parse!( - be_i128(&[0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00]), + be_i128( + &[ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00 + ][..] + ), Ok((&b""[..], 0)) ); assert_parse!( - be_i128(&[0x7f, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff]), - Ok((&b""[..], 170_141_183_460_469_231_731_687_303_715_884_105_727_i128)) + be_i128( + &[ + 0x7f, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, + 0xff + ][..] + ), + Ok(( + &b""[..], + 170_141_183_460_469_231_731_687_303_715_884_105_727_i128 + )) ); assert_parse!( - be_i128(&[0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff]), + be_i128( + &[ + 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, + 0xff + ][..] + ), Ok((&b""[..], -1)) ); assert_parse!( - be_i128(&[0x80, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00]), - Ok((&b""[..], -170_141_183_460_469_231_731_687_303_715_884_105_728_i128)) + be_i128( + &[ + 0x80, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00 + ][..] + ), + Ok(( + &b""[..], + -170_141_183_460_469_231_731_687_303_715_884_105_728_i128 + )) ); } #[test] fn le_i8_tests() { - assert_parse!(le_i8(&[0x00]), Ok((&b""[..], 0))); - assert_parse!(le_i8(&[0x7f]), Ok((&b""[..], 127))); - assert_parse!(le_i8(&[0xff]), Ok((&b""[..], -1))); - assert_parse!(le_i8(&[0x80]), Ok((&b""[..], -128))); + assert_parse!(le_i8(&[0x00][..]), Ok((&b""[..], 0))); + assert_parse!(le_i8(&[0x7f][..]), Ok((&b""[..], 127))); + assert_parse!(le_i8(&[0xff][..]), Ok((&b""[..], -1))); + assert_parse!(le_i8(&[0x80][..]), Ok((&b""[..], -128))); } #[test] fn le_i16_tests() { - assert_parse!(le_i16(&[0x00, 0x00]), Ok((&b""[..], 0))); - assert_parse!(le_i16(&[0xff, 0x7f]), Ok((&b""[..], 32_767_i16))); - assert_parse!(le_i16(&[0xff, 0xff]), Ok((&b""[..], -1))); - assert_parse!(le_i16(&[0x00, 0x80]), Ok((&b""[..], -32_768_i16))); + assert_parse!(le_i16(&[0x00, 0x00][..]), Ok((&b""[..], 0))); + assert_parse!(le_i16(&[0xff, 0x7f][..]), Ok((&b""[..], 32_767_i16))); + assert_parse!(le_i16(&[0xff, 0xff][..]), Ok((&b""[..], -1))); + assert_parse!(le_i16(&[0x00, 0x80][..]), Ok((&b""[..], -32_768_i16))); } #[test] fn le_u24_tests() { - assert_parse!(le_u24(&[0x00, 0x00, 0x00]), Ok((&b""[..], 0))); - assert_parse!(le_u24(&[0xFF, 0xFF, 0x00]), Ok((&b""[..], 65_535_u32))); - assert_parse!(le_u24(&[0x56, 0x34, 0x12]), Ok((&b""[..], 1_193_046_u32))); + assert_parse!(le_u24(&[0x00, 0x00, 0x00][..]), Ok((&b""[..], 0))); + assert_parse!(le_u24(&[0xFF, 0xFF, 0x00][..]), Ok((&b""[..], 65_535_u32))); + assert_parse!( + le_u24(&[0x56, 0x34, 0x12][..]), + Ok((&b""[..], 1_193_046_u32)) + ); } #[test] fn le_i24_tests() { - assert_parse!(le_i24(&[0xFF, 0xFF, 0xFF]), Ok((&b""[..], -1_i32))); - assert_parse!(le_i24(&[0x00, 0x00, 0xFF]), Ok((&b""[..], -65_536_i32))); - assert_parse!(le_i24(&[0xAA, 0xCB, 0xED]), Ok((&b""[..], -1_193_046_i32))); + assert_parse!(le_i24(&[0xFF, 0xFF, 0xFF][..]), Ok((&b""[..], -1_i32))); + assert_parse!(le_i24(&[0x00, 0x00, 0xFF][..]), Ok((&b""[..], -65_536_i32))); + assert_parse!( + le_i24(&[0xAA, 0xCB, 0xED][..]), + Ok((&b""[..], -1_193_046_i32)) + ); } #[test] fn le_i32_tests() { - assert_parse!(le_i32(&[0x00, 0x00, 0x00, 0x00]), Ok((&b""[..], 0))); + assert_parse!(le_i32(&[0x00, 0x00, 0x00, 0x00][..]), Ok((&b""[..], 0))); assert_parse!( - le_i32(&[0xff, 0xff, 0xff, 0x7f]), + le_i32(&[0xff, 0xff, 0xff, 0x7f][..]), Ok((&b""[..], 2_147_483_647_i32)) ); - assert_parse!(le_i32(&[0xff, 0xff, 0xff, 0xff]), Ok((&b""[..], -1))); + assert_parse!(le_i32(&[0xff, 0xff, 0xff, 0xff][..]), Ok((&b""[..], -1))); assert_parse!( - le_i32(&[0x00, 0x00, 0x00, 0x80]), + le_i32(&[0x00, 0x00, 0x00, 0x80][..]), Ok((&b""[..], -2_147_483_648_i32)) ); } @@ -1050,19 +1730,19 @@ mod tests { #[test] fn le_i64_tests() { assert_parse!( - le_i64(&[0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00]), + le_i64(&[0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00][..]), Ok((&b""[..], 0)) ); assert_parse!( - le_i64(&[0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x7f]), + le_i64(&[0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x7f][..]), Ok((&b""[..], 9_223_372_036_854_775_807_i64)) ); assert_parse!( - le_i64(&[0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff]), + le_i64(&[0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff][..]), Ok((&b""[..], -1)) ); assert_parse!( - le_i64(&[0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x80]), + le_i64(&[0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x80][..]), Ok((&b""[..], -9_223_372_036_854_775_808_i64)) ); } @@ -1071,28 +1751,54 @@ mod tests { #[cfg(stable_i128)] fn le_i128_tests() { assert_parse!( - le_i128(&[0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00]), + le_i128( + &[ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00 + ][..] + ), Ok((&b""[..], 0)) ); assert_parse!( - le_i128(&[0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x7f]), - Ok((&b""[..], 170_141_183_460_469_231_731_687_303_715_884_105_727_i128)) + le_i128( + &[ + 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, + 0x7f + ][..] + ), + Ok(( + &b""[..], + 170_141_183_460_469_231_731_687_303_715_884_105_727_i128 + )) ); assert_parse!( - le_i128(&[0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff]), + le_i128( + &[ + 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, + 0xff + ][..] + ), Ok((&b""[..], -1)) ); assert_parse!( - le_i128(&[0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x80]), - Ok((&b""[..], -170_141_183_460_469_231_731_687_303_715_884_105_728_i128)) + le_i128( + &[ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x80 + ][..] + ), + Ok(( + &b""[..], + -170_141_183_460_469_231_731_687_303_715_884_105_728_i128 + )) ); } #[test] fn be_f32_tests() { - assert_parse!(be_f32(&[0x00, 0x00, 0x00, 0x00]), Ok((&b""[..], 0_f32))); + assert_parse!(be_f32(&[0x00, 0x00, 0x00, 0x00][..]), Ok((&b""[..], 0_f32))); assert_parse!( - be_f32(&[0x4d, 0x31, 0x1f, 0xd8]), + be_f32(&[0x4d, 0x31, 0x1f, 0xd8][..]), Ok((&b""[..], 185_728_392_f32)) ); } @@ -1100,20 +1806,20 @@ mod tests { #[test] fn be_f64_tests() { assert_parse!( - be_f64(&[0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00]), + be_f64(&[0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00][..]), Ok((&b""[..], 0_f64)) ); assert_parse!( - be_f64(&[0x41, 0xa6, 0x23, 0xfb, 0x10, 0x00, 0x00, 0x00]), + be_f64(&[0x41, 0xa6, 0x23, 0xfb, 0x10, 0x00, 0x00, 0x00][..]), Ok((&b""[..], 185_728_392_f64)) ); } #[test] fn le_f32_tests() { - assert_parse!(le_f32(&[0x00, 0x00, 0x00, 0x00]), Ok((&b""[..], 0_f32))); + assert_parse!(le_f32(&[0x00, 0x00, 0x00, 0x00][..]), Ok((&b""[..], 0_f32))); assert_parse!( - le_f32(&[0xd8, 0x1f, 0x31, 0x4d]), + le_f32(&[0xd8, 0x1f, 0x31, 0x4d][..]), Ok((&b""[..], 185_728_392_f32)) ); } @@ -1121,11 +1827,11 @@ mod tests { #[test] fn le_f64_tests() { assert_parse!( - le_f64(&[0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00]), + le_f64(&[0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00][..]), Ok((&b""[..], 0_f64)) ); assert_parse!( - le_f64(&[0x00, 0x00, 0x00, 0x10, 0xfb, 0x23, 0xa6, 0x41]), + le_f64(&[0x00, 0x00, 0x00, 0x10, 0xfb, 0x23, 0xa6, 0x41][..]), Ok((&b""[..], 185_728_392_f64)) ); } @@ -1194,5 +1900,4 @@ mod tests { Err(Err::Failure(("", ErrorKind::Digit))) ); } - } diff --git a/src/number/macros.rs b/src/number/macros.rs index cb8c2a9..27fdcce 100644 --- a/src/number/macros.rs +++ b/src/number/macros.rs @@ -1,7 +1,7 @@ -//! parsers recognizing numbers +//! Parsers recognizing numbers -/// if the parameter is nom::Endianness::Big, parse a big endian u16 integer, -/// otherwise a little endian u16 integer +/// If the parameter is `nom::number::Endianness::Big`, parse a big endian u16 integer, +/// otherwise a little endian u16 integer. /// /// ```rust /// # #[macro_use] extern crate nom; @@ -12,19 +12,19 @@ /// named!(be<u16>, u16!(Endianness::Big)); /// /// assert_eq!(be(b"\x00\x01abcd"), Ok((&b"abcd"[..], 0x0001))); -/// assert_eq!(be(b"\x01"), Err(Err::Incomplete(Needed::Size(2)))); +/// assert_eq!(be(b"\x01"), Err(Err::Incomplete(Needed::new(1)))); /// /// named!(le<u16>, u16!(Endianness::Little)); /// /// assert_eq!(le(b"\x00\x01abcd"), Ok((&b"abcd"[..], 0x0100))); -/// assert_eq!(le(b"\x01"), Err(Err::Incomplete(Needed::Size(2)))); +/// assert_eq!(le(b"\x01"), Err(Err::Incomplete(Needed::new(1)))); /// # } /// ``` #[macro_export(local_inner_macros)] macro_rules! u16 ( ($i:expr, $e:expr) => ( {if $crate::number::Endianness::Big == $e { $crate::number::streaming::be_u16($i) } else { $crate::number::streaming::le_u16($i) } } );); -/// if the parameter is nom::Endianness::Big, parse a big endian u32 integer, -/// otherwise a little endian u32 integer +/// If the parameter is `nom::number::Endianness::Big`, parse a big endian u32 integer, +/// otherwise a little endian u32 integer. /// /// ```rust /// # #[macro_use] extern crate nom; @@ -35,19 +35,19 @@ macro_rules! u16 ( ($i:expr, $e:expr) => ( {if $crate::number::Endianness::Big = /// named!(be<u32>, u32!(Endianness::Big)); /// /// assert_eq!(be(b"\x00\x01\x02\x03abcd"), Ok((&b"abcd"[..], 0x00010203))); -/// assert_eq!(be(b"\x01"), Err(Err::Incomplete(Needed::Size(4)))); +/// assert_eq!(be(b"\x01"), Err(Err::Incomplete(Needed::new(3)))); /// /// named!(le<u32>, u32!(Endianness::Little)); /// /// assert_eq!(le(b"\x00\x01\x02\x03abcd"), Ok((&b"abcd"[..], 0x03020100))); -/// assert_eq!(le(b"\x01"), Err(Err::Incomplete(Needed::Size(4)))); +/// assert_eq!(le(b"\x01"), Err(Err::Incomplete(Needed::new(3)))); /// # } /// ``` #[macro_export(local_inner_macros)] macro_rules! u32 ( ($i:expr, $e:expr) => ( {if $crate::number::Endianness::Big == $e { $crate::number::streaming::be_u32($i) } else { $crate::number::streaming::le_u32($i) } } );); -/// if the parameter is nom::Endianness::Big, parse a big endian u64 integer, -/// otherwise a little endian u64 integer +/// If the parameter is `nom::number::Endianness::Big`, parse a big endian u64 integer, +/// otherwise a little endian u64 integer. /// /// ```rust /// # #[macro_use] extern crate nom; @@ -58,19 +58,19 @@ macro_rules! u32 ( ($i:expr, $e:expr) => ( {if $crate::number::Endianness::Big = /// named!(be<u64>, u64!(Endianness::Big)); /// /// assert_eq!(be(b"\x00\x01\x02\x03\x04\x05\x06\x07abcd"), Ok((&b"abcd"[..], 0x0001020304050607))); -/// assert_eq!(be(b"\x01"), Err(Err::Incomplete(Needed::Size(8)))); +/// assert_eq!(be(b"\x01"), Err(Err::Incomplete(Needed::new(7)))); /// /// named!(le<u64>, u64!(Endianness::Little)); /// /// assert_eq!(le(b"\x00\x01\x02\x03\x04\x05\x06\x07abcd"), Ok((&b"abcd"[..], 0x0706050403020100))); -/// assert_eq!(le(b"\x01"), Err(Err::Incomplete(Needed::Size(8)))); +/// assert_eq!(le(b"\x01"), Err(Err::Incomplete(Needed::new(7)))); /// # } /// ``` #[macro_export(local_inner_macros)] macro_rules! u64 ( ($i:expr, $e:expr) => ( {if $crate::number::Endianness::Big == $e { $crate::number::streaming::be_u64($i) } else { $crate::number::streaming::le_u64($i) } } );); -/// if the parameter is nom::Endianness::Big, parse a big endian u128 integer, -/// otherwise a little endian u128 integer +/// If the parameter is `nom::number::Endianness::Big`, parse a big endian u128 integer, +/// otherwise a little endian u128 integer. /// /// ```rust /// # #[macro_use] extern crate nom; @@ -81,20 +81,20 @@ macro_rules! u64 ( ($i:expr, $e:expr) => ( {if $crate::number::Endianness::Big = /// named!(be<u128>, u128!(Endianness::Big)); /// /// assert_eq!(be(b"\x00\x01\x02\x03\x04\x05\x06\x07\x08\x09\x10\x11\x12\x13\x14\x15abcd"), Ok((&b"abcd"[..], 0x00010203040506070809101112131415))); -/// assert_eq!(be(b"\x01"), Err(Err::Incomplete(Needed::Size(16)))); +/// assert_eq!(be(b"\x01"), Err(Err::Incomplete(Needed::new(15)))); /// /// named!(le<u128>, u128!(Endianness::Little)); /// /// assert_eq!(le(b"\x00\x01\x02\x03\x04\x05\x06\x07\x08\x09\x10\x11\x12\x13\x14\x15abcd"), Ok((&b"abcd"[..], 0x15141312111009080706050403020100))); -/// assert_eq!(le(b"\x01"), Err(Err::Incomplete(Needed::Size(16)))); +/// assert_eq!(le(b"\x01"), Err(Err::Incomplete(Needed::new(15)))); /// # } /// ``` #[macro_export(local_inner_macros)] #[cfg(stable_i128)] macro_rules! u128 ( ($i:expr, $e:expr) => ( {if $crate::number::Endianness::Big == $e { $crate::number::streaming::be_u128($i) } else { $crate::number::streaming::le_u128($i) } } );); -/// if the parameter is nom::Endianness::Big, parse a big endian i16 integer, -/// otherwise a little endian i16 integer +/// If the parameter is `nom::number::Endianness::Big`, parse a big endian i16 integer, +/// otherwise a little endian i16 integer. /// /// ```rust /// # #[macro_use] extern crate nom; @@ -105,19 +105,19 @@ macro_rules! u128 ( ($i:expr, $e:expr) => ( {if $crate::number::Endianness::Big /// named!(be<i16>, i16!(Endianness::Big)); /// /// assert_eq!(be(b"\x00\x01abcd"), Ok((&b"abcd"[..], 0x0001))); -/// assert_eq!(be(b"\x01"), Err(Err::Incomplete(Needed::Size(2)))); +/// assert_eq!(be(b"\x01"), Err(Err::Incomplete(Needed::new(1)))); /// /// named!(le<i16>, i16!(Endianness::Little)); /// /// assert_eq!(le(b"\x00\x01abcd"), Ok((&b"abcd"[..], 0x0100))); -/// assert_eq!(le(b"\x01"), Err(Err::Incomplete(Needed::Size(2)))); +/// assert_eq!(le(b"\x01"), Err(Err::Incomplete(Needed::new(1)))); /// # } /// ``` #[macro_export(local_inner_macros)] macro_rules! i16 ( ($i:expr, $e:expr) => ( {if $crate::number::Endianness::Big == $e { $crate::number::streaming::be_i16($i) } else { $crate::number::streaming::le_i16($i) } } );); -/// if the parameter is nom::Endianness::Big, parse a big endian i32 integer, -/// otherwise a little endian i32 integer +/// If the parameter is `nom::number::Endianness::Big`, parse a big endian i32 integer, +/// otherwise a little endian i32 integer. /// /// ```rust /// # #[macro_use] extern crate nom; @@ -128,19 +128,19 @@ macro_rules! i16 ( ($i:expr, $e:expr) => ( {if $crate::number::Endianness::Big = /// named!(be<i32>, i32!(Endianness::Big)); /// /// assert_eq!(be(b"\x00\x01\x02\x03abcd"), Ok((&b"abcd"[..], 0x00010203))); -/// assert_eq!(be(b"\x01"), Err(Err::Incomplete(Needed::Size(4)))); +/// assert_eq!(be(b"\x01"), Err(Err::Incomplete(Needed::new(3)))); /// /// named!(le<i32>, i32!(Endianness::Little)); /// /// assert_eq!(le(b"\x00\x01\x02\x03abcd"), Ok((&b"abcd"[..], 0x03020100))); -/// assert_eq!(le(b"\x01"), Err(Err::Incomplete(Needed::Size(4)))); +/// assert_eq!(le(b"\x01"), Err(Err::Incomplete(Needed::new(3)))); /// # } /// ``` #[macro_export(local_inner_macros)] macro_rules! i32 ( ($i:expr, $e:expr) => ( {if $crate::number::Endianness::Big == $e { $crate::number::streaming::be_i32($i) } else { $crate::number::streaming::le_i32($i) } } );); -/// if the parameter is nom::Endianness::Big, parse a big endian i64 integer, -/// otherwise a little endian i64 integer +/// If the parameter is `nom::number::Endianness::Big`, parse a big endian i64 integer, +/// otherwise a little endian i64 integer. /// /// ```rust /// # #[macro_use] extern crate nom; @@ -151,19 +151,19 @@ macro_rules! i32 ( ($i:expr, $e:expr) => ( {if $crate::number::Endianness::Big = /// named!(be<i64>, i64!(Endianness::Big)); /// /// assert_eq!(be(b"\x00\x01\x02\x03\x04\x05\x06\x07abcd"), Ok((&b"abcd"[..], 0x0001020304050607))); -/// assert_eq!(be(b"\x01"), Err(Err::Incomplete(Needed::Size(8)))); +/// assert_eq!(be(b"\x01"), Err(Err::Incomplete(Needed::new(7)))); /// /// named!(le<i64>, i64!(Endianness::Little)); /// /// assert_eq!(le(b"\x00\x01\x02\x03\x04\x05\x06\x07abcd"), Ok((&b"abcd"[..], 0x0706050403020100))); -/// assert_eq!(le(b"\x01"), Err(Err::Incomplete(Needed::Size(8)))); +/// assert_eq!(le(b"\x01"), Err(Err::Incomplete(Needed::new(7)))); /// # } /// ``` #[macro_export(local_inner_macros)] macro_rules! i64 ( ($i:expr, $e:expr) => ( {if $crate::number::Endianness::Big == $e { $crate::number::streaming::be_i64($i) } else { $crate::number::streaming::le_i64($i) } } );); -/// if the parameter is nom::Endianness::Big, parse a big endian i64 integer, -/// otherwise a little endian i64 integer +/// If the parameter is `nom::number::Endianness::Big`, parse a big endian i64 integer, +/// otherwise a little endian i64 integer. /// /// ```rust /// # #[macro_use] extern crate nom; @@ -174,12 +174,12 @@ macro_rules! i64 ( ($i:expr, $e:expr) => ( {if $crate::number::Endianness::Big = /// named!(be<i128>, i128!(Endianness::Big)); /// /// assert_eq!(be(b"\x00\x01\x02\x03\x04\x05\x06\x07\x08\x09\x10\x11\x12\x13\x14\x15abcd"), Ok((&b"abcd"[..], 0x00010203040506070809101112131415))); -/// assert_eq!(be(b"\x01"), Err(Err::Incomplete(Needed::Size(16)))); +/// assert_eq!(be(b"\x01"), Err(Err::Incomplete(Needed::new(15)))); /// /// named!(le<i128>, i128!(Endianness::Little)); /// /// assert_eq!(le(b"\x00\x01\x02\x03\x04\x05\x06\x07\x08\x09\x10\x11\x12\x13\x14\x15abcd"), Ok((&b"abcd"[..], 0x15141312111009080706050403020100))); -/// assert_eq!(le(b"\x01"), Err(Err::Incomplete(Needed::Size(16)))); +/// assert_eq!(le(b"\x01"), Err(Err::Incomplete(Needed::new(15)))); /// # } /// ``` #[macro_export(local_inner_macros)] diff --git a/src/number/mod.rs b/src/number/mod.rs index 9be4c38..6ab2cd5 100644 --- a/src/number/mod.rs +++ b/src/number/mod.rs @@ -1,17 +1,18 @@ -//! parsers recognizing numbers +//! Parsers recognizing numbers #[macro_use] mod macros; -pub mod streaming; pub mod complete; +pub mod streaming; /// Configurable endianness #[derive(Debug, PartialEq, Eq, Clone, Copy)] pub enum Endianness { - /// big endian + /// Big endian Big, - /// little endian + /// Little endian Little, + /// Will match the host's endianness + Native, } - diff --git a/src/number/streaming.rs b/src/number/streaming.rs index 15a89a4..f502519 100644 --- a/src/number/streaming.rs +++ b/src/number/streaming.rs @@ -1,694 +1,1301 @@ -//! parsers recognizing numbers, streaming version +//! Parsers recognizing numbers, streaming version -use crate::internal::*; +use crate::branch::alt; +use crate::character::streaming::{char, digit1}; +use crate::combinator::{cut, map, opt, recognize}; use crate::error::{ErrorKind, ParseError}; -use crate::traits::{AsChar, InputIter, InputLength, InputTakeAtPosition}; +use crate::internal::*; use crate::lib::std::ops::{RangeFrom, RangeTo}; -use crate::traits::{Offset, Slice}; -use crate::character::streaming::{char, digit1}; use crate::sequence::{pair, tuple}; -use crate::combinator::{cut, map, opt, recognize}; -use crate::branch::alt; +use crate::traits::{AsChar, InputIter, InputLength, InputTakeAtPosition}; +use crate::traits::{Offset, Slice}; -/// Recognizes an unsigned 1 byte integer +/// Recognizes an unsigned 1 byte integer. /// -/// *streaming version*: will return `Err(nom::Err::Incomplete(_))` if there is not enough data +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there is not enough data. /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; -/// # use nom::Needed::Size; /// use nom::number::streaming::be_u8; /// -/// let parser = be_u8::<(_, ErrorKind)>; +/// let parser = |s| { +/// be_u8::<_, (_, ErrorKind)>(s) +/// }; /// -/// assert_eq!(parser(b"\x00\x01abcd"), Ok((&b"\x01abcd"[..], 0x00))); -/// assert_eq!(parser(b""), Err(Err::Incomplete(Needed::Size(1)))); +/// assert_eq!(parser(&b"\x00\x01abcd"[..]), Ok((&b"\x01abcd"[..], 0x00))); +/// assert_eq!(parser(&b""[..]), Err(Err::Incomplete(Needed::new(1)))); /// ``` #[inline] -pub fn be_u8<'a, E: ParseError<&'a[u8]>>(i: &'a[u8]) -> IResult<&'a[u8], u8, E> { - if i.len() < 1 { - Err(Err::Incomplete(Needed::Size(1))) +pub fn be_u8<I, E: ParseError<I>>(input: I) -> IResult<I, u8, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + let bound: usize = 1; + if input.input_len() < bound { + Err(Err::Incomplete(Needed::new(1))) } else { - Ok((&i[1..], i[0])) + let res = input.iter_elements().next().unwrap(); + + Ok((input.slice(bound..), res)) } } -/// Recognizes a big endian unsigned 2 bytes integer +/// Recognizes a big endian unsigned 2 bytes integer. /// -/// *streaming version*: will return `Err(nom::Err::Incomplete(_))` if there is not enough data +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there is not enough data. /// /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; -/// # use nom::Needed::Size; /// use nom::number::streaming::be_u16; /// /// let parser = |s| { -/// be_u16::<(_, ErrorKind)>(s) +/// be_u16::<_, (_, ErrorKind)>(s) /// }; /// -/// assert_eq!(parser(b"\x00\x01abcd"), Ok((&b"abcd"[..], 0x0001))); -/// assert_eq!(parser(b"\x01"), Err(Err::Incomplete(Needed::Size(2)))); +/// assert_eq!(parser(&b"\x00\x01abcd"[..]), Ok((&b"abcd"[..], 0x0001))); +/// assert_eq!(parser(&b"\x01"[..]), Err(Err::Incomplete(Needed::new(1)))); /// ``` #[inline] -pub fn be_u16<'a, E: ParseError<&'a[u8]>>(i: &'a[u8]) -> IResult<&'a[u8], u16, E> { - if i.len() < 2 { - Err(Err::Incomplete(Needed::Size(2))) +pub fn be_u16<I, E: ParseError<I>>(input: I) -> IResult<I, u16, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + let bound: usize = 2; + if input.input_len() < bound { + Err(Err::Incomplete(Needed::new(bound - input.input_len()))) } else { - let res = ((i[0] as u16) << 8) + i[1] as u16; - Ok((&i[2..], res)) + let mut res = 0u16; + for byte in input.iter_elements().take(bound) { + res = (res << 8) + byte as u16; + } + + Ok((input.slice(bound..), res)) } } -/// Recognizes a big endian unsigned 3 byte integer +/// Recognizes a big endian unsigned 3 byte integer. /// -/// *streaming version*: will return `Err(nom::Err::Incomplete(_))` if there is not enough data +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there is not enough data. /// /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; -/// # use nom::Needed::Size; /// use nom::number::streaming::be_u24; /// /// let parser = |s| { -/// be_u24::<(_, ErrorKind)>(s) +/// be_u24::<_, (_, ErrorKind)>(s) /// }; /// -/// assert_eq!(parser(b"\x00\x01\x02abcd"), Ok((&b"abcd"[..], 0x000102))); -/// assert_eq!(parser(b"\x01"), Err(Err::Incomplete(Needed::Size(3)))); +/// assert_eq!(parser(&b"\x00\x01\x02abcd"[..]), Ok((&b"abcd"[..], 0x000102))); +/// assert_eq!(parser(&b"\x01"[..]), Err(Err::Incomplete(Needed::new(2)))); /// ``` #[inline] -pub fn be_u24<'a, E: ParseError<&'a[u8]>>(i: &'a[u8]) -> IResult<&'a[u8], u32, E> { - if i.len() < 3 { - Err(Err::Incomplete(Needed::Size(3))) +pub fn be_u24<I, E: ParseError<I>>(input: I) -> IResult<I, u32, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + let bound: usize = 3; + if input.input_len() < bound { + Err(Err::Incomplete(Needed::new(bound - input.input_len()))) } else { - let res = ((i[0] as u32) << 16) + ((i[1] as u32) << 8) + (i[2] as u32); - Ok((&i[3..], res)) + let mut res = 0u32; + for byte in input.iter_elements().take(bound) { + res = (res << 8) + byte as u32; + } + + Ok((input.slice(bound..), res)) } } -/// Recognizes a big endian unsigned 4 bytes integer +/// Recognizes a big endian unsigned 4 bytes integer. /// -/// *streaming version*: will return `Err(nom::Err::Incomplete(_))` if there is not enough data +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there is not enough data. /// /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; -/// # use nom::Needed::Size; /// use nom::number::streaming::be_u32; /// /// let parser = |s| { -/// be_u32::<(_, ErrorKind)>(s) +/// be_u32::<_, (_, ErrorKind)>(s) /// }; /// -/// assert_eq!(parser(b"\x00\x01\x02\x03abcd"), Ok((&b"abcd"[..], 0x00010203))); -/// assert_eq!(parser(b"\x01"), Err(Err::Incomplete(Needed::Size(4)))); +/// assert_eq!(parser(&b"\x00\x01\x02\x03abcd"[..]), Ok((&b"abcd"[..], 0x00010203))); +/// assert_eq!(parser(&b"\x01"[..]), Err(Err::Incomplete(Needed::new(3)))); /// ``` #[inline] -pub fn be_u32<'a, E: ParseError<&'a[u8]>>(i: &'a[u8]) -> IResult<&'a[u8], u32, E> { - if i.len() < 4 { - Err(Err::Incomplete(Needed::Size(4))) +pub fn be_u32<I, E: ParseError<I>>(input: I) -> IResult<I, u32, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + let bound: usize = 4; + if input.input_len() < bound { + Err(Err::Incomplete(Needed::new(bound - input.input_len()))) } else { - let res = ((i[0] as u32) << 24) + ((i[1] as u32) << 16) + ((i[2] as u32) << 8) + i[3] as u32; - Ok((&i[4..], res)) + let mut res = 0u32; + for byte in input.iter_elements().take(bound) { + res = (res << 8) + byte as u32; + } + + Ok((input.slice(bound..), res)) } } -/// Recognizes a big endian unsigned 8 bytes integer +/// Recognizes a big endian unsigned 8 bytes integer. /// -/// *streaming version*: will return `Err(nom::Err::Incomplete(_))` if there is not enough data +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there is not enough data. /// /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; -/// # use nom::Needed::Size; /// use nom::number::streaming::be_u64; /// /// let parser = |s| { -/// be_u64::<(_, ErrorKind)>(s) +/// be_u64::<_, (_, ErrorKind)>(s) /// }; /// -/// assert_eq!(parser(b"\x00\x01\x02\x03\x04\x05\x06\x07abcd"), Ok((&b"abcd"[..], 0x0001020304050607))); -/// assert_eq!(parser(b"\x01"), Err(Err::Incomplete(Needed::Size(8)))); +/// assert_eq!(parser(&b"\x00\x01\x02\x03\x04\x05\x06\x07abcd"[..]), Ok((&b"abcd"[..], 0x0001020304050607))); +/// assert_eq!(parser(&b"\x01"[..]), Err(Err::Incomplete(Needed::new(7)))); /// ``` #[inline] -pub fn be_u64<'a, E: ParseError<&'a[u8]>>(i: &'a[u8]) -> IResult<&'a[u8], u64, E> { - if i.len() < 8 { - Err(Err::Incomplete(Needed::Size(8))) +pub fn be_u64<I, E: ParseError<I>>(input: I) -> IResult<I, u64, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + let bound: usize = 8; + if input.input_len() < bound { + Err(Err::Incomplete(Needed::new(bound - input.input_len()))) } else { - let res = ((i[0] as u64) << 56) + ((i[1] as u64) << 48) + ((i[2] as u64) << 40) + ((i[3] as u64) << 32) + ((i[4] as u64) << 24) - + ((i[5] as u64) << 16) + ((i[6] as u64) << 8) + i[7] as u64; - Ok((&i[8..], res)) + let mut res = 0u64; + for byte in input.iter_elements().take(bound) { + res = (res << 8) + byte as u64; + } + + Ok((input.slice(bound..), res)) } } -/// Recognizes a big endian unsigned 16 bytes integer +/// Recognizes a big endian unsigned 16 bytes integer. /// -/// *streaming version*: will return `Err(nom::Err::Incomplete(_))` if there is not enough data +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there is not enough data. /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; -/// # use nom::Needed::Size; /// use nom::number::streaming::be_u128; /// /// let parser = |s| { -/// be_u128::<(_, ErrorKind)>(s) +/// be_u128::<_, (_, ErrorKind)>(s) /// }; /// -/// assert_eq!(parser(b"\x00\x01\x02\x03\x04\x05\x06\x07\x08\x09\x10\x11\x12\x13\x14\x15abcd"), Ok((&b"abcd"[..], 0x00010203040506070809101112131415))); -/// assert_eq!(parser(b"\x01"), Err(Err::Incomplete(Needed::Size(16)))); +/// assert_eq!(parser(&b"\x00\x01\x02\x03\x04\x05\x06\x07\x08\x09\x10\x11\x12\x13\x14\x15abcd"[..]), Ok((&b"abcd"[..], 0x00010203040506070809101112131415))); +/// assert_eq!(parser(&b"\x01"[..]), Err(Err::Incomplete(Needed::new(15)))); /// ``` #[inline] #[cfg(stable_i128)] -pub fn be_u128<'a, E: ParseError<&'a[u8]>>(i: &'a[u8]) -> IResult<&'a[u8], u128, E> { - if i.len() < 16 { - Err(Err::Incomplete(Needed::Size(16))) +pub fn be_u128<I, E: ParseError<I>>(input: I) -> IResult<I, u128, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + let bound: usize = 16; + if input.input_len() < bound { + Err(Err::Incomplete(Needed::new(bound - input.input_len()))) } else { - let res = ((i[0] as u128) << 120) - + ((i[1] as u128) << 112) - + ((i[2] as u128) << 104) - + ((i[3] as u128) << 96) - + ((i[4] as u128) << 88) - + ((i[5] as u128) << 80) - + ((i[6] as u128) << 72) - + ((i[7] as u128) << 64) - + ((i[8] as u128) << 56) - + ((i[9] as u128) << 48) - + ((i[10] as u128) << 40) - + ((i[11] as u128) << 32) - + ((i[12] as u128) << 24) - + ((i[13] as u128) << 16) - + ((i[14] as u128) << 8) - + i[15] as u128; - Ok((&i[16..], res)) + let mut res = 0u128; + for byte in input.iter_elements().take(bound) { + res = (res << 8) + byte as u128; + } + + Ok((input.slice(bound..), res)) } } -/// Recognizes a signed 1 byte integer +/// Recognizes a signed 1 byte integer. /// -/// *streaming version*: will return `Err(nom::Err::Incomplete(_))` if there is not enough data +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there is not enough data. /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; -/// # use nom::Needed::Size; /// use nom::number::streaming::be_i8; /// -/// let parser = be_i8::<(_, ErrorKind)>; +/// let parser = be_i8::<_, (_, ErrorKind)>; /// -/// assert_eq!(parser(b"\x00\x01abcd"), Ok((&b"\x01abcd"[..], 0x00))); -/// assert_eq!(parser(b""), Err(Err::Incomplete(Needed::Size(1)))); +/// assert_eq!(parser(&b"\x00\x01abcd"[..]), Ok((&b"\x01abcd"[..], 0x00))); +/// assert_eq!(parser(&b""[..]), Err(Err::Incomplete(Needed::new(1)))); /// ``` #[inline] -pub fn be_i8<'a, E: ParseError<&'a[u8]>>(i: &'a[u8]) -> IResult<&'a[u8], i8, E> { - map!(i, be_u8, |x| x as i8) +pub fn be_i8<I, E: ParseError<I>>(input: I) -> IResult<I, i8, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + map!(input, be_u8, |x| x as i8) } -/// Recognizes a big endian signed 2 bytes integer +/// Recognizes a big endian signed 2 bytes integer. /// -/// *streaming version*: will return `Err(nom::Err::Incomplete(_))` if there is not enough data +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there is not enough data. /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; -/// # use nom::Needed::Size; /// use nom::number::streaming::be_i16; /// -/// let parser = be_i16::<(_, ErrorKind)>; +/// let parser = be_i16::<_, (_, ErrorKind)>; /// -/// assert_eq!(parser(b"\x00\x01abcd"), Ok((&b"abcd"[..], 0x0001))); -/// assert_eq!(parser(b""), Err(Err::Incomplete(Needed::Size(2)))); +/// assert_eq!(parser(&b"\x00\x01abcd"[..]), Ok((&b"abcd"[..], 0x0001))); +/// assert_eq!(parser(&b""[..]), Err(Err::Incomplete(Needed::new(2)))); /// ``` #[inline] -pub fn be_i16<'a, E: ParseError<&'a [u8]>>(i: &'a[u8]) -> IResult<&'a[u8], i16, E> { - map!(i, be_u16, |x| x as i16) +pub fn be_i16<I, E: ParseError<I>>(input: I) -> IResult<I, i16, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + map!(input, be_u16, |x| x as i16) } -/// Recognizes a big endian signed 3 bytes integer +/// Recognizes a big endian signed 3 bytes integer. /// -/// *streaming version*: will return `Err(nom::Err::Incomplete(_))` if there is not enough data +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there is not enough data. /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; -/// # use nom::Needed::Size; /// use nom::number::streaming::be_i24; /// -/// let parser = be_i24::<(_, ErrorKind)>; +/// let parser = be_i24::<_, (_, ErrorKind)>; /// -/// assert_eq!(parser(b"\x00\x01\x02abcd"), Ok((&b"abcd"[..], 0x000102))); -/// assert_eq!(parser(b""), Err(Err::Incomplete(Needed::Size(3)))); +/// assert_eq!(parser(&b"\x00\x01\x02abcd"[..]), Ok((&b"abcd"[..], 0x000102))); +/// assert_eq!(parser(&b""[..]), Err(Err::Incomplete(Needed::new(3)))); /// ``` #[inline] -pub fn be_i24<'a, E: ParseError<&'a[u8]>>(i: &'a[u8]) -> IResult<&'a[u8], i32, E> { +pub fn be_i24<I, E: ParseError<I>>(input: I) -> IResult<I, i32, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ // Same as the unsigned version but we need to sign-extend manually here - map!(i, be_u24, |x| if x & 0x80_00_00 != 0 { + map!(input, be_u24, |x| if x & 0x80_00_00 != 0 { (x | 0xff_00_00_00) as i32 } else { x as i32 }) } -/// Recognizes a big endian signed 4 bytes integer +/// Recognizes a big endian signed 4 bytes integer. /// -/// *streaming version*: will return `Err(nom::Err::Incomplete(_))` if there is not enough data +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there is not enough data. /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; -/// # use nom::Needed::Size; /// use nom::number::streaming::be_i32; /// -/// let parser = be_i32::<(_, ErrorKind)>; +/// let parser = be_i32::<_, (_, ErrorKind)>; /// -/// assert_eq!(parser(b"\x00\x01\x02\x03abcd"), Ok((&b"abcd"[..], 0x00010203))); -/// assert_eq!(parser(b""), Err(Err::Incomplete(Needed::Size(4)))); +/// assert_eq!(parser(&b"\x00\x01\x02\x03abcd"[..]), Ok((&b"abcd"[..], 0x00010203))); +/// assert_eq!(parser(&b""[..]), Err(Err::Incomplete(Needed::new(4)))); /// ``` #[inline] -pub fn be_i32<'a, E: ParseError<&'a[u8]>>(i: &'a[u8]) -> IResult<&'a[u8], i32, E> { - map!(i, be_u32, |x| x as i32) +pub fn be_i32<I, E: ParseError<I>>(input: I) -> IResult<I, i32, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + map!(input, be_u32, |x| x as i32) } -/// Recognizes a big endian signed 8 bytes integer +/// Recognizes a big endian signed 8 bytes integer. /// -/// *streaming version*: will return `Err(nom::Err::Incomplete(_))` if there is not enough data +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there is not enough data. /// /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; -/// # use nom::Needed::Size; /// use nom::number::streaming::be_i64; /// -/// let parser = be_i64::<(_, ErrorKind)>; +/// let parser = be_i64::<_, (_, ErrorKind)>; /// -/// assert_eq!(parser(b"\x00\x01\x02\x03\x04\x05\x06\x07abcd"), Ok((&b"abcd"[..], 0x0001020304050607))); -/// assert_eq!(parser(b"\x01"), Err(Err::Incomplete(Needed::Size(8)))); +/// assert_eq!(parser(&b"\x00\x01\x02\x03\x04\x05\x06\x07abcd"[..]), Ok((&b"abcd"[..], 0x0001020304050607))); +/// assert_eq!(parser(&b"\x01"[..]), Err(Err::Incomplete(Needed::new(7)))); /// ``` #[inline] -pub fn be_i64<'a, E: ParseError<&'a[u8]>>(i: &'a[u8]) -> IResult<&'a[u8], i64, E> { - map!(i, be_u64, |x| x as i64) +pub fn be_i64<I, E: ParseError<I>>(input: I) -> IResult<I, i64, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + map!(input, be_u64, |x| x as i64) } -/// Recognizes a big endian signed 16 bytes integer +/// Recognizes a big endian signed 16 bytes integer. /// -/// *streaming version*: will return `Err(nom::Err::Incomplete(_))` if there is not enough data +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there is not enough data. /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; -/// # use nom::Needed::Size; /// use nom::number::streaming::be_i128; /// -/// let parser = be_i128::<(_, ErrorKind)>; +/// let parser = be_i128::<_, (_, ErrorKind)>; /// -/// assert_eq!(parser(b"\x00\x01\x02\x03\x04\x05\x06\x07\x08\x09\x10\x11\x12\x13\x14\x15abcd"), Ok((&b"abcd"[..], 0x00010203040506070809101112131415))); -/// assert_eq!(parser(b"\x01"), Err(Err::Incomplete(Needed::Size(16)))); +/// assert_eq!(parser(&b"\x00\x01\x02\x03\x04\x05\x06\x07\x08\x09\x10\x11\x12\x13\x14\x15abcd"[..]), Ok((&b"abcd"[..], 0x00010203040506070809101112131415))); +/// assert_eq!(parser(&b"\x01"[..]), Err(Err::Incomplete(Needed::new(15)))); /// ``` #[inline] #[cfg(stable_i128)] -pub fn be_i128<'a, E: ParseError<&'a [u8]>>(i: &'a[u8]) -> IResult<&'a[u8], i128, E> { - map!(i, be_u128, |x| x as i128) +pub fn be_i128<I, E: ParseError<I>>(input: I) -> IResult<I, i128, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + map!(input, be_u128, |x| x as i128) } -/// Recognizes an unsigned 1 byte integer +/// Recognizes an unsigned 1 byte integer. /// -/// *streaming version*: will return `Err(nom::Err::Incomplete(_))` if there is not enough data +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there is not enough data. /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; -/// # use nom::Needed::Size; /// use nom::number::streaming::le_u8; /// -/// let parser = le_u8::<(_, ErrorKind)>; +/// let parser = le_u8::<_, (_, ErrorKind)>; /// -/// assert_eq!(parser(b"\x00\x01abcd"), Ok((&b"\x01abcd"[..], 0x00))); -/// assert_eq!(parser(b""), Err(Err::Incomplete(Needed::Size(1)))); +/// assert_eq!(parser(&b"\x00\x01abcd"[..]), Ok((&b"\x01abcd"[..], 0x00))); +/// assert_eq!(parser(&b""[..]), Err(Err::Incomplete(Needed::new(1)))); /// ``` #[inline] -pub fn le_u8<'a, E: ParseError<&'a [u8]>>(i: &'a[u8]) -> IResult<&'a[u8], u8, E> { - if i.len() < 1 { - Err(Err::Incomplete(Needed::Size(1))) +pub fn le_u8<I, E: ParseError<I>>(input: I) -> IResult<I, u8, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + let bound: usize = 1; + if input.input_len() < bound { + Err(Err::Incomplete(Needed::new(1))) } else { - Ok((&i[1..], i[0])) + let res = input.iter_elements().next().unwrap(); + + Ok((input.slice(bound..), res)) } } -/// Recognizes a little endian unsigned 2 bytes integer +/// Recognizes a little endian unsigned 2 bytes integer. /// -/// *streaming version*: will return `Err(nom::Err::Incomplete(_))` if there is not enough data +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there is not enough data. /// /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; -/// # use nom::Needed::Size; /// use nom::number::streaming::le_u16; /// /// let parser = |s| { -/// le_u16::<(_, ErrorKind)>(s) +/// le_u16::<_, (_, ErrorKind)>(s) /// }; /// -/// assert_eq!(parser(b"\x00\x01abcd"), Ok((&b"abcd"[..], 0x0100))); -/// assert_eq!(parser(b"\x01"), Err(Err::Incomplete(Needed::Size(2)))); +/// assert_eq!(parser(&b"\x00\x01abcd"[..]), Ok((&b"abcd"[..], 0x0100))); +/// assert_eq!(parser(&b"\x01"[..]), Err(Err::Incomplete(Needed::new(1)))); /// ``` #[inline] -pub fn le_u16<'a, E: ParseError<&'a [u8]>>(i: &'a[u8]) -> IResult<&'a[u8], u16, E> { - if i.len() < 2 { - Err(Err::Incomplete(Needed::Size(2))) +pub fn le_u16<I, E: ParseError<I>>(input: I) -> IResult<I, u16, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + let bound: usize = 2; + if input.input_len() < bound { + Err(Err::Incomplete(Needed::new(bound - input.input_len()))) } else { - let res = ((i[1] as u16) << 8) + i[0] as u16; - Ok((&i[2..], res)) + let mut res = 0u16; + for (index, byte) in input.iter_indices().take(bound) { + res += (byte as u16) << (8 * index); + } + + Ok((input.slice(bound..), res)) } } -/// Recognizes a little endian unsigned 3 bytes integer +/// Recognizes a little endian unsigned 3 bytes integer. /// -/// *streaming version*: will return `Err(nom::Err::Incomplete(_))` if there is not enough data +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there is not enough data. /// /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; -/// # use nom::Needed::Size; /// use nom::number::streaming::le_u24; /// /// let parser = |s| { -/// le_u24::<(_, ErrorKind)>(s) +/// le_u24::<_, (_, ErrorKind)>(s) /// }; /// -/// assert_eq!(parser(b"\x00\x01\x02abcd"), Ok((&b"abcd"[..], 0x020100))); -/// assert_eq!(parser(b"\x01"), Err(Err::Incomplete(Needed::Size(3)))); +/// assert_eq!(parser(&b"\x00\x01\x02abcd"[..]), Ok((&b"abcd"[..], 0x020100))); +/// assert_eq!(parser(&b"\x01"[..]), Err(Err::Incomplete(Needed::new(2)))); /// ``` #[inline] -pub fn le_u24<'a, E: ParseError<&'a [u8]>>(i: &'a[u8]) -> IResult<&'a[u8], u32, E> { - if i.len() < 3 { - Err(Err::Incomplete(Needed::Size(3))) +pub fn le_u24<I, E: ParseError<I>>(input: I) -> IResult<I, u32, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + let bound: usize = 3; + if input.input_len() < bound { + Err(Err::Incomplete(Needed::new(bound - input.input_len()))) } else { - let res = (i[0] as u32) + ((i[1] as u32) << 8) + ((i[2] as u32) << 16); - Ok((&i[3..], res)) + let mut res = 0u32; + for (index, byte) in input.iter_indices().take(bound) { + res += (byte as u32) << (8 * index); + } + + Ok((input.slice(bound..), res)) } } -/// Recognizes a little endian unsigned 4 bytes integer +/// Recognizes a little endian unsigned 4 bytes integer. /// -/// *streaming version*: will return `Err(nom::Err::Incomplete(_))` if there is not enough data +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there is not enough data. /// /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; -/// # use nom::Needed::Size; /// use nom::number::streaming::le_u32; /// /// let parser = |s| { -/// le_u32::<(_, ErrorKind)>(s) +/// le_u32::<_, (_, ErrorKind)>(s) /// }; /// -/// assert_eq!(parser(b"\x00\x01\x02\x03abcd"), Ok((&b"abcd"[..], 0x03020100))); -/// assert_eq!(parser(b"\x01"), Err(Err::Incomplete(Needed::Size(4)))); +/// assert_eq!(parser(&b"\x00\x01\x02\x03abcd"[..]), Ok((&b"abcd"[..], 0x03020100))); +/// assert_eq!(parser(&b"\x01"[..]), Err(Err::Incomplete(Needed::new(3)))); /// ``` #[inline] -pub fn le_u32<'a, E: ParseError<&'a [u8]>>(i: &'a[u8]) -> IResult<&'a[u8], u32, E> { - if i.len() < 4 { - Err(Err::Incomplete(Needed::Size(4))) +pub fn le_u32<I, E: ParseError<I>>(input: I) -> IResult<I, u32, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + let bound: usize = 4; + if input.input_len() < bound { + Err(Err::Incomplete(Needed::new(bound - input.input_len()))) } else { - let res = ((i[3] as u32) << 24) + ((i[2] as u32) << 16) + ((i[1] as u32) << 8) + i[0] as u32; - Ok((&i[4..], res)) + let mut res = 0u32; + for (index, byte) in input.iter_indices().take(bound) { + res += (byte as u32) << (8 * index); + } + + Ok((input.slice(bound..), res)) } } -/// Recognizes a little endian unsigned 8 bytes integer +/// Recognizes a little endian unsigned 8 bytes integer. /// -/// *streaming version*: will return `Err(nom::Err::Incomplete(_))` if there is not enough data +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there is not enough data. /// /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; -/// # use nom::Needed::Size; /// use nom::number::streaming::le_u64; /// /// let parser = |s| { -/// le_u64::<(_, ErrorKind)>(s) +/// le_u64::<_, (_, ErrorKind)>(s) /// }; /// -/// assert_eq!(parser(b"\x00\x01\x02\x03\x04\x05\x06\x07abcd"), Ok((&b"abcd"[..], 0x0706050403020100))); -/// assert_eq!(parser(b"\x01"), Err(Err::Incomplete(Needed::Size(8)))); +/// assert_eq!(parser(&b"\x00\x01\x02\x03\x04\x05\x06\x07abcd"[..]), Ok((&b"abcd"[..], 0x0706050403020100))); +/// assert_eq!(parser(&b"\x01"[..]), Err(Err::Incomplete(Needed::new(7)))); /// ``` #[inline] -pub fn le_u64<'a, E: ParseError<&'a [u8]>>(i: &'a[u8]) -> IResult<&'a[u8], u64, E> { - if i.len() < 8 { - Err(Err::Incomplete(Needed::Size(8))) +pub fn le_u64<I, E: ParseError<I>>(input: I) -> IResult<I, u64, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + let bound: usize = 8; + if input.input_len() < bound { + Err(Err::Incomplete(Needed::new(bound - input.input_len()))) } else { - let res = ((i[7] as u64) << 56) + ((i[6] as u64) << 48) + ((i[5] as u64) << 40) + ((i[4] as u64) << 32) + ((i[3] as u64) << 24) - + ((i[2] as u64) << 16) + ((i[1] as u64) << 8) + i[0] as u64; - Ok((&i[8..], res)) + let mut res = 0u64; + for (index, byte) in input.iter_indices().take(bound) { + res += (byte as u64) << (8 * index); + } + + Ok((input.slice(bound..), res)) } } -/// Recognizes a little endian unsigned 16 bytes integer +/// Recognizes a little endian unsigned 16 bytes integer. /// -/// *streaming version*: will return `Err(nom::Err::Incomplete(_))` if there is not enough data +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there is not enough data. /// /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; -/// # use nom::Needed::Size; /// use nom::number::streaming::le_u128; /// /// let parser = |s| { -/// le_u128::<(_, ErrorKind)>(s) +/// le_u128::<_, (_, ErrorKind)>(s) /// }; /// -/// assert_eq!(parser(b"\x00\x01\x02\x03\x04\x05\x06\x07\x08\x09\x10\x11\x12\x13\x14\x15abcd"), Ok((&b"abcd"[..], 0x15141312111009080706050403020100))); -/// assert_eq!(parser(b"\x01"), Err(Err::Incomplete(Needed::Size(16)))); +/// assert_eq!(parser(&b"\x00\x01\x02\x03\x04\x05\x06\x07\x08\x09\x10\x11\x12\x13\x14\x15abcd"[..]), Ok((&b"abcd"[..], 0x15141312111009080706050403020100))); +/// assert_eq!(parser(&b"\x01"[..]), Err(Err::Incomplete(Needed::new(15)))); /// ``` #[inline] #[cfg(stable_i128)] -pub fn le_u128<'a, E: ParseError<&'a [u8]>>(i: &'a[u8]) -> IResult<&'a[u8], u128, E> { - if i.len() < 16 { - Err(Err::Incomplete(Needed::Size(16))) +pub fn le_u128<I, E: ParseError<I>>(input: I) -> IResult<I, u128, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + let bound: usize = 16; + if input.input_len() < bound { + Err(Err::Incomplete(Needed::new(bound - input.input_len()))) } else { - let res = ((i[15] as u128) << 120) - + ((i[14] as u128) << 112) - + ((i[13] as u128) << 104) - + ((i[12] as u128) << 96) - + ((i[11] as u128) << 88) - + ((i[10] as u128) << 80) - + ((i[9] as u128) << 72) - + ((i[8] as u128) << 64) - + ((i[7] as u128) << 56) - + ((i[6] as u128) << 48) - + ((i[5] as u128) << 40) - + ((i[4] as u128) << 32) - + ((i[3] as u128) << 24) - + ((i[2] as u128) << 16) - + ((i[1] as u128) << 8) - + i[0] as u128; - Ok((&i[16..], res)) + let mut res = 0u128; + for (index, byte) in input.iter_indices().take(bound) { + res += (byte as u128) << (8 * index); + } + + Ok((input.slice(bound..), res)) } } -/// Recognizes a signed 1 byte integer +/// Recognizes a signed 1 byte integer. /// -/// *streaming version*: will return `Err(nom::Err::Incomplete(_))` if there is not enough data +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there is not enough data. /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; -/// # use nom::Needed::Size; /// use nom::number::streaming::le_i8; /// -/// let parser = le_i8::<(_, ErrorKind)>; +/// let parser = le_i8::<_, (_, ErrorKind)>; /// -/// assert_eq!(parser(b"\x00\x01abcd"), Ok((&b"\x01abcd"[..], 0x00))); -/// assert_eq!(parser(b""), Err(Err::Incomplete(Needed::Size(1)))); +/// assert_eq!(parser(&b"\x00\x01abcd"[..]), Ok((&b"\x01abcd"[..], 0x00))); +/// assert_eq!(parser(&b""[..]), Err(Err::Incomplete(Needed::new(1)))); /// ``` #[inline] -pub fn le_i8<'a, E: ParseError<&'a [u8]>>(i: &'a[u8]) -> IResult<&'a[u8], i8, E> { - map!(i, le_u8, |x| x as i8) +pub fn le_i8<I, E: ParseError<I>>(input: I) -> IResult<I, i8, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + map!(input, le_u8, |x| x as i8) } -/// Recognizes a little endian signed 2 bytes integer +/// Recognizes a little endian signed 2 bytes integer. /// -/// *streaming version*: will return `Err(nom::Err::Incomplete(_))` if there is not enough data +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there is not enough data. /// /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; -/// # use nom::Needed::Size; /// use nom::number::streaming::le_i16; /// /// let parser = |s| { -/// le_i16::<(_, ErrorKind)>(s) +/// le_i16::<_, (_, ErrorKind)>(s) /// }; /// -/// assert_eq!(parser(b"\x00\x01abcd"), Ok((&b"abcd"[..], 0x0100))); -/// assert_eq!(parser(b"\x01"), Err(Err::Incomplete(Needed::Size(2)))); +/// assert_eq!(parser(&b"\x00\x01abcd"[..]), Ok((&b"abcd"[..], 0x0100))); +/// assert_eq!(parser(&b"\x01"[..]), Err(Err::Incomplete(Needed::new(1)))); /// ``` #[inline] -pub fn le_i16<'a, E: ParseError<&'a [u8]>>(i: &'a[u8]) -> IResult<&'a[u8], i16, E> { - map!(i, le_u16, |x| x as i16) +pub fn le_i16<I, E: ParseError<I>>(input: I) -> IResult<I, i16, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + map!(input, le_u16, |x| x as i16) } -/// Recognizes a little endian signed 3 bytes integer +/// Recognizes a little endian signed 3 bytes integer. /// -/// *streaming version*: will return `Err(nom::Err::Incomplete(_))` if there is not enough data +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there is not enough data. /// /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; -/// # use nom::Needed::Size; /// use nom::number::streaming::le_i24; /// /// let parser = |s| { -/// le_i24::<(_, ErrorKind)>(s) +/// le_i24::<_, (_, ErrorKind)>(s) /// }; /// -/// assert_eq!(parser(b"\x00\x01\x02abcd"), Ok((&b"abcd"[..], 0x020100))); -/// assert_eq!(parser(b"\x01"), Err(Err::Incomplete(Needed::Size(3)))); +/// assert_eq!(parser(&b"\x00\x01\x02abcd"[..]), Ok((&b"abcd"[..], 0x020100))); +/// assert_eq!(parser(&b"\x01"[..]), Err(Err::Incomplete(Needed::new(2)))); /// ``` #[inline] -pub fn le_i24<'a, E: ParseError<&'a [u8]>>(i: &'a[u8]) -> IResult<&'a[u8], i32, E> { +pub fn le_i24<I, E: ParseError<I>>(input: I) -> IResult<I, i32, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ // Same as the unsigned version but we need to sign-extend manually here - map!(i, le_u24, |x| if x & 0x80_00_00 != 0 { + map!(input, le_u24, |x| if x & 0x80_00_00 != 0 { (x | 0xff_00_00_00) as i32 } else { x as i32 }) } -/// Recognizes a little endian signed 4 bytes integer +/// Recognizes a little endian signed 4 bytes integer. /// -/// *streaming version*: will return `Err(nom::Err::Incomplete(_))` if there is not enough data +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there is not enough data. /// /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; -/// # use nom::Needed::Size; /// use nom::number::streaming::le_i32; /// /// let parser = |s| { -/// le_i32::<(_, ErrorKind)>(s) +/// le_i32::<_, (_, ErrorKind)>(s) /// }; /// -/// assert_eq!(parser(b"\x00\x01\x02\x03abcd"), Ok((&b"abcd"[..], 0x03020100))); -/// assert_eq!(parser(b"\x01"), Err(Err::Incomplete(Needed::Size(4)))); +/// assert_eq!(parser(&b"\x00\x01\x02\x03abcd"[..]), Ok((&b"abcd"[..], 0x03020100))); +/// assert_eq!(parser(&b"\x01"[..]), Err(Err::Incomplete(Needed::new(3)))); /// ``` #[inline] -pub fn le_i32<'a, E: ParseError<&'a [u8]>>(i: &'a[u8]) -> IResult<&'a[u8], i32, E> { - map!(i, le_u32, |x| x as i32) +pub fn le_i32<I, E: ParseError<I>>(input: I) -> IResult<I, i32, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + map!(input, le_u32, |x| x as i32) } -/// Recognizes a little endian signed 8 bytes integer +/// Recognizes a little endian signed 8 bytes integer. /// -/// *streaming version*: will return `Err(nom::Err::Incomplete(_))` if there is not enough data +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there is not enough data. /// /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; -/// # use nom::Needed::Size; /// use nom::number::streaming::le_i64; /// /// let parser = |s| { -/// le_i64::<(_, ErrorKind)>(s) +/// le_i64::<_, (_, ErrorKind)>(s) /// }; /// -/// assert_eq!(parser(b"\x00\x01\x02\x03\x04\x05\x06\x07abcd"), Ok((&b"abcd"[..], 0x0706050403020100))); -/// assert_eq!(parser(b"\x01"), Err(Err::Incomplete(Needed::Size(8)))); +/// assert_eq!(parser(&b"\x00\x01\x02\x03\x04\x05\x06\x07abcd"[..]), Ok((&b"abcd"[..], 0x0706050403020100))); +/// assert_eq!(parser(&b"\x01"[..]), Err(Err::Incomplete(Needed::new(7)))); /// ``` #[inline] -pub fn le_i64<'a, E: ParseError<&'a [u8]>>(i: &'a[u8]) -> IResult<&'a[u8], i64, E> { - map!(i, le_u64, |x| x as i64) +pub fn le_i64<I, E: ParseError<I>>(input: I) -> IResult<I, i64, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + map!(input, le_u64, |x| x as i64) } -/// Recognizes a little endian signed 16 bytes integer +/// Recognizes a little endian signed 16 bytes integer. /// -/// *streaming version*: will return `Err(nom::Err::Incomplete(_))` if there is not enough data +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there is not enough data. /// /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; -/// # use nom::Needed::Size; /// use nom::number::streaming::le_i128; /// /// let parser = |s| { -/// le_i128::<(_, ErrorKind)>(s) +/// le_i128::<_, (_, ErrorKind)>(s) /// }; /// -/// assert_eq!(parser(b"\x00\x01\x02\x03\x04\x05\x06\x07\x08\x09\x10\x11\x12\x13\x14\x15abcd"), Ok((&b"abcd"[..], 0x15141312111009080706050403020100))); -/// assert_eq!(parser(b"\x01"), Err(Err::Incomplete(Needed::Size(16)))); +/// assert_eq!(parser(&b"\x00\x01\x02\x03\x04\x05\x06\x07\x08\x09\x10\x11\x12\x13\x14\x15abcd"[..]), Ok((&b"abcd"[..], 0x15141312111009080706050403020100))); +/// assert_eq!(parser(&b"\x01"[..]), Err(Err::Incomplete(Needed::new(15)))); /// ``` #[inline] #[cfg(stable_i128)] -pub fn le_i128<'a, E: ParseError<&'a [u8]>>(i: &'a[u8]) -> IResult<&'a[u8], i128, E> { - map!(i, le_u128, |x| x as i128) +pub fn le_i128<I, E: ParseError<I>>(input: I) -> IResult<I, i128, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + map!(input, le_u128, |x| x as i128) +} + +/// Recognizes an unsigned 1 byte integer +/// +/// Note that endianness does not apply to 1 byte numbers. +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there is not enough data. +/// ```rust +/// # use nom::{Err, error::ErrorKind, Needed}; +/// # use nom::Needed::Size; +/// use nom::number::streaming::u8; +/// +/// let parser = |s| { +/// u8::<_, (_, ErrorKind)>(s) +/// }; +/// +/// assert_eq!(parser(&b"\x00\x03abcefg"[..]), Ok((&b"\x03abcefg"[..], 0x00))); +/// assert_eq!(parser(&b""[..]), Err(Err::Incomplete(Needed::new(1)))); +/// ``` +#[inline] +pub fn u8<I, E: ParseError<I>>(input: I) -> IResult<I, u8, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + let bound: usize = 1; + if input.input_len() < bound { + Err(Err::Incomplete(Needed::new(1))) + } else { + let res = input.iter_elements().next().unwrap(); + + Ok((input.slice(bound..), res)) + } +} + +/// Recognizes an unsigned 2 bytes integer +/// +/// If the parameter is `nom::number::Endianness::Big`, parse a big endian u16 integer, +/// otherwise if `nom::number::Endianness::Little` parse a little endian u16 integer. +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there is not enough data. +/// +/// ```rust +/// # use nom::{Err, error::ErrorKind, Needed}; +/// # use nom::Needed::Size; +/// use nom::number::streaming::u16; +/// +/// let be_u16 = |s| { +/// u16::<_, (_, ErrorKind)>(nom::number::Endianness::Big)(s) +/// }; +/// +/// assert_eq!(be_u16(&b"\x00\x03abcefg"[..]), Ok((&b"abcefg"[..], 0x0003))); +/// assert_eq!(be_u16(&b"\x01"[..]), Err(Err::Incomplete(Needed::new(1)))); +/// +/// let le_u16 = |s| { +/// u16::<_, (_, ErrorKind)>(nom::number::Endianness::Little)(s) +/// }; +/// +/// assert_eq!(le_u16(&b"\x00\x03abcefg"[..]), Ok((&b"abcefg"[..], 0x0300))); +/// assert_eq!(le_u16(&b"\x01"[..]), Err(Err::Incomplete(Needed::new(1)))); +/// ``` +#[inline] +pub fn u16<I, E: ParseError<I>>(endian: crate::number::Endianness) -> fn(I) -> IResult<I, u16, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + match endian { + crate::number::Endianness::Big => be_u16, + crate::number::Endianness::Little => le_u16, + #[cfg(target_endian = "big")] + crate::number::Endianness::Native => be_u16, + #[cfg(target_endian = "little")] + crate::number::Endianness::Native => le_u16, + } +} + +/// Recognizes an unsigned 3 byte integer +/// +/// If the parameter is `nom::number::Endianness::Big`, parse a big endian u24 integer, +/// otherwise if `nom::number::Endianness::Little` parse a little endian u24 integer. +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there is not enough data. +/// ```rust +/// # use nom::{Err, error::ErrorKind, Needed}; +/// # use nom::Needed::Size; +/// use nom::number::streaming::u24; +/// +/// let be_u24 = |s| { +/// u24::<_,(_, ErrorKind)>(nom::number::Endianness::Big)(s) +/// }; +/// +/// assert_eq!(be_u24(&b"\x00\x03\x05abcefg"[..]), Ok((&b"abcefg"[..], 0x000305))); +/// assert_eq!(be_u24(&b"\x01"[..]), Err(Err::Incomplete(Needed::new(2)))); +/// +/// let le_u24 = |s| { +/// u24::<_, (_, ErrorKind)>(nom::number::Endianness::Little)(s) +/// }; +/// +/// assert_eq!(le_u24(&b"\x00\x03\x05abcefg"[..]), Ok((&b"abcefg"[..], 0x050300))); +/// assert_eq!(le_u24(&b"\x01"[..]), Err(Err::Incomplete(Needed::new(2)))); +/// ``` +#[inline] +pub fn u24<I, E: ParseError<I>>(endian: crate::number::Endianness) -> fn(I) -> IResult<I, u32, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + match endian { + crate::number::Endianness::Big => be_u24, + crate::number::Endianness::Little => le_u24, + #[cfg(target_endian = "big")] + crate::number::Endianness::Native => be_u24, + #[cfg(target_endian = "little")] + crate::number::Endianness::Native => le_u24, + } +} + +/// Recognizes an unsigned 4 byte integer +/// +/// If the parameter is `nom::number::Endianness::Big`, parse a big endian u32 integer, +/// otherwise if `nom::number::Endianness::Little` parse a little endian u32 integer. +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there is not enough data. +/// ```rust +/// # use nom::{Err, error::ErrorKind, Needed}; +/// # use nom::Needed::Size; +/// use nom::number::streaming::u32; +/// +/// let be_u32 = |s| { +/// u32::<_, (_, ErrorKind)>(nom::number::Endianness::Big)(s) +/// }; +/// +/// assert_eq!(be_u32(&b"\x00\x03\x05\x07abcefg"[..]), Ok((&b"abcefg"[..], 0x00030507))); +/// assert_eq!(be_u32(&b"\x01"[..]), Err(Err::Incomplete(Needed::new(3)))); +/// +/// let le_u32 = |s| { +/// u32::<_, (_, ErrorKind)>(nom::number::Endianness::Little)(s) +/// }; +/// +/// assert_eq!(le_u32(&b"\x00\x03\x05\x07abcefg"[..]), Ok((&b"abcefg"[..], 0x07050300))); +/// assert_eq!(le_u32(&b"\x01"[..]), Err(Err::Incomplete(Needed::new(3)))); +/// ``` +#[inline] +pub fn u32<I, E: ParseError<I>>(endian: crate::number::Endianness) -> fn(I) -> IResult<I, u32, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + match endian { + crate::number::Endianness::Big => be_u32, + crate::number::Endianness::Little => le_u32, + #[cfg(target_endian = "big")] + crate::number::Endianness::Native => be_u32, + #[cfg(target_endian = "little")] + crate::number::Endianness::Native => le_u32, + } +} + +/// Recognizes an unsigned 8 byte integer +/// +/// If the parameter is `nom::number::Endianness::Big`, parse a big endian u64 integer, +/// otherwise if `nom::number::Endianness::Little` parse a little endian u64 integer. +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there is not enough data. +/// ```rust +/// # use nom::{Err, error::ErrorKind, Needed}; +/// # use nom::Needed::Size; +/// use nom::number::streaming::u64; +/// +/// let be_u64 = |s| { +/// u64::<_, (_, ErrorKind)>(nom::number::Endianness::Big)(s) +/// }; +/// +/// assert_eq!(be_u64(&b"\x00\x01\x02\x03\x04\x05\x06\x07abcefg"[..]), Ok((&b"abcefg"[..], 0x0001020304050607))); +/// assert_eq!(be_u64(&b"\x01"[..]), Err(Err::Incomplete(Needed::new(7)))); +/// +/// let le_u64 = |s| { +/// u64::<_, (_, ErrorKind)>(nom::number::Endianness::Little)(s) +/// }; +/// +/// assert_eq!(le_u64(&b"\x00\x01\x02\x03\x04\x05\x06\x07abcefg"[..]), Ok((&b"abcefg"[..], 0x0706050403020100))); +/// assert_eq!(le_u64(&b"\x01"[..]), Err(Err::Incomplete(Needed::new(7)))); +/// ``` +#[inline] +pub fn u64<I, E: ParseError<I>>(endian: crate::number::Endianness) -> fn(I) -> IResult<I, u64, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + match endian { + crate::number::Endianness::Big => be_u64, + crate::number::Endianness::Little => le_u64, + #[cfg(target_endian = "big")] + crate::number::Endianness::Native => be_u64, + #[cfg(target_endian = "little")] + crate::number::Endianness::Native => le_u64, + } +} + +/// Recognizes an unsigned 16 byte integer +/// +/// If the parameter is `nom::number::Endianness::Big`, parse a big endian u128 integer, +/// otherwise if `nom::number::Endianness::Little` parse a little endian u128 integer. +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there is not enough data. +/// ```rust +/// # use nom::{Err, error::ErrorKind, Needed}; +/// # use nom::Needed::Size; +/// use nom::number::streaming::u128; +/// +/// let be_u128 = |s| { +/// u128::<_, (_, ErrorKind)>(nom::number::Endianness::Big)(s) +/// }; +/// +/// assert_eq!(be_u128(&b"\x00\x01\x02\x03\x04\x05\x06\x07\x00\x01\x02\x03\x04\x05\x06\x07abcefg"[..]), Ok((&b"abcefg"[..], 0x00010203040506070001020304050607))); +/// assert_eq!(be_u128(&b"\x01"[..]), Err(Err::Incomplete(Needed::new(15)))); +/// +/// let le_u128 = |s| { +/// u128::<_, (_, ErrorKind)>(nom::number::Endianness::Little)(s) +/// }; +/// +/// assert_eq!(le_u128(&b"\x00\x01\x02\x03\x04\x05\x06\x07\x00\x01\x02\x03\x04\x05\x06\x07abcefg"[..]), Ok((&b"abcefg"[..], 0x07060504030201000706050403020100))); +/// assert_eq!(le_u128(&b"\x01"[..]), Err(Err::Incomplete(Needed::new(15)))); +/// ``` +#[inline] +#[cfg(stable_i128)] +pub fn u128<I, E: ParseError<I>>(endian: crate::number::Endianness) -> fn(I) -> IResult<I, u128, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + match endian { + crate::number::Endianness::Big => be_u128, + crate::number::Endianness::Little => le_u128, + #[cfg(target_endian = "big")] + crate::number::Endianness::Native => be_u128, + #[cfg(target_endian = "little")] + crate::number::Endianness::Native => le_u128, + } +} + +/// Recognizes a signed 1 byte integer +/// +/// Note that endianness does not apply to 1 byte numbers. +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there is not enough data. +/// ```rust +/// # use nom::{Err, error::ErrorKind, Needed}; +/// # use nom::Needed::Size; +/// use nom::number::streaming::i8; +/// +/// let parser = |s| { +/// i8::<_, (_, ErrorKind)>(s) +/// }; +/// +/// assert_eq!(parser(&b"\x00\x03abcefg"[..]), Ok((&b"\x03abcefg"[..], 0x00))); +/// assert_eq!(parser(&b""[..]), Err(Err::Incomplete(Needed::new(1)))); +/// ``` +#[inline] +pub fn i8<I, E: ParseError<I>>(i: I) -> IResult<I, i8, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + map!(i, u8, |x| x as i8) } -/// Recognizes a big endian 4 bytes floating point number +/// Recognizes a signed 2 byte integer /// -/// *streaming version*: will return `Err(nom::Err::Incomplete(_))` if there is not enough data +/// If the parameter is `nom::number::Endianness::Big`, parse a big endian i16 integer, +/// otherwise if `nom::number::Endianness::Little` parse a little endian i16 integer. +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there is not enough data. /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; /// # use nom::Needed::Size; +/// use nom::number::streaming::i16; +/// +/// let be_i16 = |s| { +/// i16::<_, (_, ErrorKind)>(nom::number::Endianness::Big)(s) +/// }; +/// +/// assert_eq!(be_i16(&b"\x00\x03abcefg"[..]), Ok((&b"abcefg"[..], 0x0003))); +/// assert_eq!(be_i16(&b"\x01"[..]), Err(Err::Incomplete(Needed::new(1)))); +/// +/// let le_i16 = |s| { +/// i16::<_, (_, ErrorKind)>(nom::number::Endianness::Little)(s) +/// }; +/// +/// assert_eq!(le_i16(&b"\x00\x03abcefg"[..]), Ok((&b"abcefg"[..], 0x0300))); +/// assert_eq!(le_i16(&b"\x01"[..]), Err(Err::Incomplete(Needed::new(1)))); +/// ``` +#[inline] +pub fn i16<I, E: ParseError<I>>(endian: crate::number::Endianness) -> fn(I) -> IResult<I, i16, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + match endian { + crate::number::Endianness::Big => be_i16, + crate::number::Endianness::Little => le_i16, + #[cfg(target_endian = "big")] + crate::number::Endianness::Native => be_i16, + #[cfg(target_endian = "little")] + crate::number::Endianness::Native => le_i16, + } +} + +/// Recognizes a signed 3 byte integer +/// +/// If the parameter is `nom::number::Endianness::Big`, parse a big endian i24 integer, +/// otherwise if `nom::number::Endianness::Little` parse a little endian i24 integer. +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there is not enough data. +/// ```rust +/// # use nom::{Err, error::ErrorKind, Needed}; +/// # use nom::Needed::Size; +/// use nom::number::streaming::i24; +/// +/// let be_i24 = |s| { +/// i24::<_, (_, ErrorKind)>(nom::number::Endianness::Big)(s) +/// }; +/// +/// assert_eq!(be_i24(&b"\x00\x03\x05abcefg"[..]), Ok((&b"abcefg"[..], 0x000305))); +/// assert_eq!(be_i24(&b"\x01"[..]), Err(Err::Incomplete(Needed::new(2)))); +/// +/// let le_i24 = |s| { +/// i24::<_, (_, ErrorKind)>(nom::number::Endianness::Little)(s) +/// }; +/// +/// assert_eq!(le_i24(&b"\x00\x03\x05abcefg"[..]), Ok((&b"abcefg"[..], 0x050300))); +/// assert_eq!(le_i24(&b"\x01"[..]), Err(Err::Incomplete(Needed::new(2)))); +/// ``` +#[inline] +pub fn i24<I, E: ParseError<I>>(endian: crate::number::Endianness) -> fn(I) -> IResult<I, i32, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + match endian { + crate::number::Endianness::Big => be_i24, + crate::number::Endianness::Little => le_i24, + #[cfg(target_endian = "big")] + crate::number::Endianness::Native => be_i24, + #[cfg(target_endian = "little")] + crate::number::Endianness::Native => le_i24, + } +} + +/// Recognizes a signed 4 byte integer +/// +/// If the parameter is `nom::number::Endianness::Big`, parse a big endian i32 integer, +/// otherwise if `nom::number::Endianness::Little` parse a little endian i32 integer. +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there is not enough data. +/// ```rust +/// # use nom::{Err, error::ErrorKind, Needed}; +/// # use nom::Needed::Size; +/// use nom::number::streaming::i32; +/// +/// let be_i32 = |s| { +/// i32::<_, (_, ErrorKind)>(nom::number::Endianness::Big)(s) +/// }; +/// +/// assert_eq!(be_i32(&b"\x00\x03\x05\x07abcefg"[..]), Ok((&b"abcefg"[..], 0x00030507))); +/// assert_eq!(be_i32(&b"\x01"[..]), Err(Err::Incomplete(Needed::new(3)))); +/// +/// let le_i32 = |s| { +/// i32::<_, (_, ErrorKind)>(nom::number::Endianness::Little)(s) +/// }; +/// +/// assert_eq!(le_i32(&b"\x00\x03\x05\x07abcefg"[..]), Ok((&b"abcefg"[..], 0x07050300))); +/// assert_eq!(le_i32(&b"\x01"[..]), Err(Err::Incomplete(Needed::new(3)))); +/// ``` +#[inline] +pub fn i32<I, E: ParseError<I>>(endian: crate::number::Endianness) -> fn(I) -> IResult<I, i32, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + match endian { + crate::number::Endianness::Big => be_i32, + crate::number::Endianness::Little => le_i32, + #[cfg(target_endian = "big")] + crate::number::Endianness::Native => be_i32, + #[cfg(target_endian = "little")] + crate::number::Endianness::Native => le_i32, + } +} + +/// Recognizes a signed 8 byte integer +/// +/// If the parameter is `nom::number::Endianness::Big`, parse a big endian i64 integer, +/// otherwise if `nom::number::Endianness::Little` parse a little endian i64 integer. +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there is not enough data. +/// ```rust +/// # use nom::{Err, error::ErrorKind, Needed}; +/// # use nom::Needed::Size; +/// use nom::number::streaming::i64; +/// +/// let be_i64 = |s| { +/// i64::<_, (_, ErrorKind)>(nom::number::Endianness::Big)(s) +/// }; +/// +/// assert_eq!(be_i64(&b"\x00\x01\x02\x03\x04\x05\x06\x07abcefg"[..]), Ok((&b"abcefg"[..], 0x0001020304050607))); +/// assert_eq!(be_i64(&b"\x01"[..]), Err(Err::Incomplete(Needed::new(7)))); +/// +/// let le_i64 = |s| { +/// i64::<_, (_, ErrorKind)>(nom::number::Endianness::Little)(s) +/// }; +/// +/// assert_eq!(le_i64(&b"\x00\x01\x02\x03\x04\x05\x06\x07abcefg"[..]), Ok((&b"abcefg"[..], 0x0706050403020100))); +/// assert_eq!(le_i64(&b"\x01"[..]), Err(Err::Incomplete(Needed::new(7)))); +/// ``` +#[inline] +pub fn i64<I, E: ParseError<I>>(endian: crate::number::Endianness) -> fn(I) -> IResult<I, i64, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + match endian { + crate::number::Endianness::Big => be_i64, + crate::number::Endianness::Little => le_i64, + #[cfg(target_endian = "big")] + crate::number::Endianness::Native => be_i64, + #[cfg(target_endian = "little")] + crate::number::Endianness::Native => le_i64, + } +} + +/// Recognizes a signed 16 byte integer +/// +/// If the parameter is `nom::number::Endianness::Big`, parse a big endian i128 integer, +/// otherwise if `nom::number::Endianness::Little` parse a little endian i128 integer. +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there is not enough data. +/// ```rust +/// # use nom::{Err, error::ErrorKind, Needed}; +/// # use nom::Needed::Size; +/// use nom::number::streaming::i128; +/// +/// let be_i128 = |s| { +/// i128::<_, (_, ErrorKind)>(nom::number::Endianness::Big)(s) +/// }; +/// +/// assert_eq!(be_i128(&b"\x00\x01\x02\x03\x04\x05\x06\x07\x00\x01\x02\x03\x04\x05\x06\x07abcefg"[..]), Ok((&b"abcefg"[..], 0x00010203040506070001020304050607))); +/// assert_eq!(be_i128(&b"\x01"[..]), Err(Err::Incomplete(Needed::new(15)))); +/// +/// let le_i128 = |s| { +/// i128::<_, (_, ErrorKind)>(nom::number::Endianness::Little)(s) +/// }; +/// +/// assert_eq!(le_i128(&b"\x00\x01\x02\x03\x04\x05\x06\x07\x00\x01\x02\x03\x04\x05\x06\x07abcefg"[..]), Ok((&b"abcefg"[..], 0x07060504030201000706050403020100))); +/// assert_eq!(le_i128(&b"\x01"[..]), Err(Err::Incomplete(Needed::new(15)))); +/// ``` +#[inline] +#[cfg(stable_i128)] +pub fn i128<I, E: ParseError<I>>(endian: crate::number::Endianness) -> fn(I) -> IResult<I, i128, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + match endian { + crate::number::Endianness::Big => be_i128, + crate::number::Endianness::Little => le_i128, + #[cfg(target_endian = "big")] + crate::number::Endianness::Native => be_i128, + #[cfg(target_endian = "little")] + crate::number::Endianness::Native => le_i128, + } +} + +/// Recognizes a big endian 4 bytes floating point number. +/// +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there is not enough data. +/// ```rust +/// # use nom::{Err, error::ErrorKind, Needed}; /// use nom::number::streaming::be_f32; /// /// let parser = |s| { -/// be_f32::<(_, ErrorKind)>(s) +/// be_f32::<_, (_, ErrorKind)>(s) /// }; /// /// assert_eq!(parser(&[0x40, 0x29, 0x00, 0x00][..]), Ok((&b""[..], 2.640625))); -/// assert_eq!(parser(&[0x01]), Err(Err::Incomplete(Needed::Size(4)))); +/// assert_eq!(parser(&[0x01][..]), Err(Err::Incomplete(Needed::new(3)))); /// ``` #[inline] -pub fn be_f32<'a, E: ParseError<&'a [u8]>>(input: &'a[u8]) -> IResult<&'a[u8], f32, E> { +pub fn be_f32<I, E: ParseError<I>>(input: I) -> IResult<I, f32, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ match be_u32(input) { Err(e) => Err(e), Ok((i, o)) => Ok((i, f32::from_bits(o))), } } -/// Recognizes a big endian 8 bytes floating point number +/// Recognizes a big endian 8 bytes floating point number. /// -/// *streaming version*: will return `Err(nom::Err::Incomplete(_))` if there is not enough data +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there is not enough data. /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; -/// # use nom::Needed::Size; /// use nom::number::streaming::be_f64; /// /// let parser = |s| { -/// be_f64::<(_, ErrorKind)>(s) +/// be_f64::<_, (_, ErrorKind)>(s) /// }; /// /// assert_eq!(parser(&[0x40, 0x29, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00][..]), Ok((&b""[..], 12.5))); -/// assert_eq!(parser(&[0x01]), Err(Err::Incomplete(Needed::Size(8)))); +/// assert_eq!(parser(&[0x01][..]), Err(Err::Incomplete(Needed::new(7)))); /// ``` #[inline] -pub fn be_f64<'a, E: ParseError<&'a [u8]>>(input: &'a[u8]) -> IResult<&'a[u8], f64, E> { +pub fn be_f64<I, E: ParseError<I>>(input: I) -> IResult<I, f64, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ match be_u64(input) { Err(e) => Err(e), Ok((i, o)) => Ok((i, f64::from_bits(o))), } } -/// Recognizes a little endian 4 bytes floating point number +/// Recognizes a little endian 4 bytes floating point number. /// -/// *streaming version*: will return `Err(nom::Err::Incomplete(_))` if there is not enough data +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there is not enough data. /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; -/// # use nom::Needed::Size; /// use nom::number::streaming::le_f32; /// /// let parser = |s| { -/// le_f32::<(_, ErrorKind)>(s) +/// le_f32::<_, (_, ErrorKind)>(s) /// }; /// /// assert_eq!(parser(&[0x00, 0x00, 0x48, 0x41][..]), Ok((&b""[..], 12.5))); -/// assert_eq!(parser(&[0x01]), Err(Err::Incomplete(Needed::Size(4)))); +/// assert_eq!(parser(&[0x01][..]), Err(Err::Incomplete(Needed::new(3)))); /// ``` #[inline] -pub fn le_f32<'a, E: ParseError<&'a [u8]>>(input: &'a[u8]) -> IResult<&'a[u8], f32, E> { +pub fn le_f32<I, E: ParseError<I>>(input: I) -> IResult<I, f32, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ match le_u32(input) { Err(e) => Err(e), Ok((i, o)) => Ok((i, f32::from_bits(o))), } } -/// Recognizes a little endian 8 bytes floating point number +/// Recognizes a little endian 8 bytes floating point number. /// -/// *streaming version*: will return `Err(nom::Err::Incomplete(_))` if there is not enough data +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there is not enough data. /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; -/// # use nom::Needed::Size; /// use nom::number::streaming::le_f64; /// /// let parser = |s| { -/// le_f64::<(_, ErrorKind)>(s) +/// le_f64::<_, (_, ErrorKind)>(s) /// }; /// /// assert_eq!(parser(&[0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x48, 0x41][..]), Ok((&b""[..], 3145728.0))); -/// assert_eq!(parser(&[0x01]), Err(Err::Incomplete(Needed::Size(8)))); +/// assert_eq!(parser(&[0x01][..]), Err(Err::Incomplete(Needed::new(7)))); /// ``` #[inline] -pub fn le_f64<'a, E: ParseError<&'a [u8]>>(input: &'a[u8]) -> IResult<&'a[u8], f64, E> { +pub fn le_f64<I, E: ParseError<I>>(input: I) -> IResult<I, f64, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ match le_u64(input) { Err(e) => Err(e), Ok((i, o)) => Ok((i, f64::from_bits(o))), } } -/// Recognizes a hex-encoded integer +/// Recognizes a 4 byte floating point number +/// +/// If the parameter is `nom::number::Endianness::Big`, parse a big endian f32 float, +/// otherwise if `nom::number::Endianness::Little` parse a little endian f32 float. +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there is not enough data. +/// ```rust +/// # use nom::{Err, error::ErrorKind, Needed}; +/// # use nom::Needed::Size; +/// use nom::number::streaming::f32; +/// +/// let be_f32 = |s| { +/// f32::<_, (_, ErrorKind)>(nom::number::Endianness::Big)(s) +/// }; +/// +/// assert_eq!(be_f32(&[0x41, 0x48, 0x00, 0x00][..]), Ok((&b""[..], 12.5))); +/// assert_eq!(be_f32(&b"abc"[..]), Err(Err::Incomplete(Needed::new(1)))); +/// +/// let le_f32 = |s| { +/// f32::<_, (_, ErrorKind)>(nom::number::Endianness::Little)(s) +/// }; +/// +/// assert_eq!(le_f32(&[0x00, 0x00, 0x48, 0x41][..]), Ok((&b""[..], 12.5))); +/// assert_eq!(le_f32(&b"abc"[..]), Err(Err::Incomplete(Needed::new(1)))); +/// ``` +#[inline] +pub fn f32<I, E: ParseError<I>>(endian: crate::number::Endianness) -> fn(I) -> IResult<I, f32, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + match endian { + crate::number::Endianness::Big => be_f32, + crate::number::Endianness::Little => le_f32, + #[cfg(target_endian = "big")] + crate::number::Endianness::Native => be_f32, + #[cfg(target_endian = "little")] + crate::number::Endianness::Native => le_f32, + } +} + +/// Recognizes an 8 byte floating point number /// -/// *streaming version*: will return `Err(nom::Err::Incomplete(_))` if there is not enough data +/// If the parameter is `nom::number::Endianness::Big`, parse a big endian f64 float, +/// otherwise if `nom::number::Endianness::Little` parse a little endian f64 float. +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there is not enough data. /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; /// # use nom::Needed::Size; +/// use nom::number::streaming::f64; +/// +/// let be_f64 = |s| { +/// f64::<_, (_, ErrorKind)>(nom::number::Endianness::Big)(s) +/// }; +/// +/// assert_eq!(be_f64(&[0x40, 0x29, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00][..]), Ok((&b""[..], 12.5))); +/// assert_eq!(be_f64(&b"abc"[..]), Err(Err::Incomplete(Needed::new(5)))); +/// +/// let le_f64 = |s| { +/// f64::<_, (_, ErrorKind)>(nom::number::Endianness::Little)(s) +/// }; +/// +/// assert_eq!(le_f64(&[0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x29, 0x40][..]), Ok((&b""[..], 12.5))); +/// assert_eq!(le_f64(&b"abc"[..]), Err(Err::Incomplete(Needed::new(5)))); +/// ``` +#[inline] +pub fn f64<I, E: ParseError<I>>(endian: crate::number::Endianness) -> fn(I) -> IResult<I, f64, E> +where + I: Slice<RangeFrom<usize>> + InputIter<Item = u8> + InputLength, +{ + match endian { + crate::number::Endianness::Big => be_f64, + crate::number::Endianness::Little => le_f64, + #[cfg(target_endian = "big")] + crate::number::Endianness::Native => be_f64, + #[cfg(target_endian = "little")] + crate::number::Endianness::Native => le_f64, + } +} + +/// Recognizes a hex-encoded integer. +/// +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there is not enough data. +/// ```rust +/// # use nom::{Err, error::ErrorKind, Needed}; /// use nom::number::streaming::hex_u32; /// /// let parser = |s| { @@ -696,11 +1303,11 @@ pub fn le_f64<'a, E: ParseError<&'a [u8]>>(input: &'a[u8]) -> IResult<&'a[u8], f /// }; /// /// assert_eq!(parser(b"01AE;"), Ok((&b";"[..], 0x01AE))); -/// assert_eq!(parser(b"abc"), Err(Err::Incomplete(Needed::Size(1)))); +/// assert_eq!(parser(b"abc"), Err(Err::Incomplete(Needed::new(1)))); /// assert_eq!(parser(b"ggg"), Err(Err::Error((&b"ggg"[..], ErrorKind::IsA)))); /// ``` #[inline] -pub fn hex_u32<'a, E: ParseError<&'a [u8]>>(input: &'a[u8]) -> IResult<&'a[u8], u32, E> { +pub fn hex_u32<'a, E: ParseError<&'a [u8]>>(input: &'a [u8]) -> IResult<&'a [u8], u32, E> { let (i, o) = crate::bytes::streaming::is_a(&b"0123456789abcdefABCDEF"[..])(input)?; // Do not parse more than 8 characters for a u32 @@ -723,13 +1330,12 @@ pub fn hex_u32<'a, E: ParseError<&'a [u8]>>(input: &'a[u8]) -> IResult<&'a[u8], Ok((remaining, res)) } -/// Recognizes a floating point number in text format and returns the corresponding part of the input +/// Recognizes a floating point number in text format and returns the corresponding part of the input. /// -/// *streaming version*: will return `Err(nom::Err::Incomplete(_))` if it reaches the end of input +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if it reaches the end of input. /// /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; -/// # use nom::Needed::Size; /// use nom::number::streaming::recognize_float; /// /// let parser = |s| { @@ -742,14 +1348,14 @@ pub fn hex_u32<'a, E: ParseError<&'a [u8]>>(input: &'a[u8]) -> IResult<&'a[u8], /// assert_eq!(parser("abc"), Err(Err::Error(("abc", ErrorKind::Char)))); /// ``` #[allow(unused_imports)] -#[cfg_attr(rustfmt, rustfmt_skip)] +#[rustfmt::skip] pub fn recognize_float<T, E:ParseError<T>>(input: T) -> IResult<T, T, E> where T: Slice<RangeFrom<usize>> + Slice<RangeTo<usize>>, T: Clone + Offset, T: InputIter, <T as InputIter>::Item: AsChar, - T: InputTakeAtPosition, + T: InputTakeAtPosition + InputLength, <T as InputTakeAtPosition>::Item: AsChar { recognize( @@ -768,12 +1374,11 @@ where )(input) } -/// Recognizes floating point number in a byte string and returns a f32 +/// Recognizes floating point number in a byte string and returns a `f32`. /// -/// *streaming version*: will return `Err(nom::Err::Incomplete(_))` if it reaches the end of input +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if it reaches the end of input. /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; -/// # use nom::Needed::Size; /// use nom::number::streaming::float; /// /// let parser = |s| { @@ -786,30 +1391,29 @@ where /// assert_eq!(parser("abc"), Err(Err::Error(("abc", ErrorKind::Char)))); /// ``` #[cfg(not(feature = "lexical"))] -pub fn float<T, E:ParseError<T>>(input: T) -> IResult<T, f32, E> +pub fn float<T, E: ParseError<T>>(input: T) -> IResult<T, f32, E> where T: Slice<RangeFrom<usize>> + Slice<RangeTo<usize>>, T: Clone + Offset, T: InputIter + InputLength + crate::traits::ParseTo<f32>, <T as InputIter>::Item: AsChar, T: InputTakeAtPosition, - <T as InputTakeAtPosition>::Item: AsChar + <T as InputTakeAtPosition>::Item: AsChar, { match recognize_float(input) { Err(e) => Err(e), Ok((i, s)) => match s.parse_to() { Some(n) => Ok((i, n)), - None => Err(Err::Error(E::from_error_kind(i, ErrorKind::Float))) - } + None => Err(Err::Error(E::from_error_kind(i, ErrorKind::Float))), + }, } } -/// Recognizes floating point number in a byte string and returns a f32 +/// Recognizes floating point number in a byte string and returns a `f32`. /// -/// *streaming version*: will return `Err(nom::Err::Incomplete(_))` if it reaches the end of input +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if it reaches the end of input. /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; -/// # use nom::Needed::Size; /// use nom::number::streaming::float; /// /// let parser = |s| { @@ -825,7 +1429,7 @@ where /// this function uses the lexical-core crate for float parsing by default, you /// can deactivate it by removing the "lexical" feature #[cfg(feature = "lexical")] -pub fn float<T, E:ParseError<T>>(input: T) -> IResult<T, f32, E> +pub fn float<T, E: ParseError<T>>(input: T) -> IResult<T, f32, E> where T: crate::traits::AsBytes + InputLength + Slice<RangeFrom<usize>>, { @@ -836,17 +1440,16 @@ where } else { Ok((input.slice(processed..), value)) } - }, - Err(_) => Err(Err::Error(E::from_error_kind(input, ErrorKind::Float))) + } + Err(_) => Err(Err::Error(E::from_error_kind(input, ErrorKind::Float))), } } -/// Recognizes floating point number in a byte string and returns a f64 +/// Recognizes floating point number in a byte string and returns a `f64`. /// -/// *streaming version*: will return `Err(nom::Err::Incomplete(_))` if it reaches the end of input +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if it reaches the end of input. /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; -/// # use nom::Needed::Size; /// use nom::number::streaming::double; /// /// let parser = |s| { @@ -859,30 +1462,29 @@ where /// assert_eq!(parser("abc"), Err(Err::Error(("abc", ErrorKind::Char)))); /// ``` #[cfg(not(feature = "lexical"))] -pub fn double<T, E:ParseError<T>>(input: T) -> IResult<T, f64, E> +pub fn double<T, E: ParseError<T>>(input: T) -> IResult<T, f64, E> where T: Slice<RangeFrom<usize>> + Slice<RangeTo<usize>>, T: Clone + Offset, T: InputIter + InputLength + crate::traits::ParseTo<f64>, <T as InputIter>::Item: AsChar, T: InputTakeAtPosition, - <T as InputTakeAtPosition>::Item: AsChar + <T as InputTakeAtPosition>::Item: AsChar, { match recognize_float(input) { Err(e) => Err(e), Ok((i, s)) => match s.parse_to() { Some(n) => Ok((i, n)), - None => Err(Err::Error(E::from_error_kind(i, ErrorKind::Float))) - } + None => Err(Err::Error(E::from_error_kind(i, ErrorKind::Float))), + }, } } -/// Recognizes floating point number in a byte string and returns a f64 +/// Recognizes floating point number in a byte string and returns a `f64`. /// -/// *streaming version*: will return `Err(nom::Err::Incomplete(_))` if it reaches the end of input +/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if it reaches the end of input. /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; -/// # use nom::Needed::Size; /// use nom::number::streaming::double; /// /// let parser = |s| { @@ -898,7 +1500,7 @@ where /// this function uses the lexical-core crate for float parsing by default, you /// can deactivate it by removing the "lexical" feature #[cfg(feature = "lexical")] -pub fn double<T, E:ParseError<T>>(input: T) -> IResult<T, f64, E> +pub fn double<T, E: ParseError<T>>(input: T) -> IResult<T, f64, E> where T: crate::traits::AsBytes + InputLength + Slice<RangeFrom<usize>>, { @@ -909,16 +1511,16 @@ where } else { Ok((input.slice(processed..), value)) } - }, - Err(_) => Err(Err::Error(E::from_error_kind(input, ErrorKind::Float))) + } + Err(_) => Err(Err::Error(E::from_error_kind(input, ErrorKind::Float))), } } #[cfg(test)] mod tests { use super::*; - use crate::internal::{Err, Needed}; use crate::error::ErrorKind; + use crate::internal::{Err, Needed}; macro_rules! assert_parse( ($left: expr, $right: expr) => { @@ -929,129 +1531,281 @@ mod tests { #[test] fn i8_tests() { - assert_parse!(be_i8(&[0x00]), Ok((&b""[..], 0))); - assert_parse!(be_i8(&[0x7f]), Ok((&b""[..], 127))); - assert_parse!(be_i8(&[0xff]), Ok((&b""[..], -1))); - assert_parse!(be_i8(&[0x80]), Ok((&b""[..], -128))); + assert_parse!(be_i8(&[0x00][..]), Ok((&b""[..], 0))); + assert_parse!(be_i8(&[0x7f][..]), Ok((&b""[..], 127))); + assert_parse!(be_i8(&[0xff][..]), Ok((&b""[..], -1))); + assert_parse!(be_i8(&[0x80][..]), Ok((&b""[..], -128))); + assert_parse!(be_i8(&[][..]), Err(Err::Incomplete(Needed::new(1)))); } #[test] fn i16_tests() { - assert_parse!(be_i16(&[0x00, 0x00]), Ok((&b""[..], 0))); - assert_parse!(be_i16(&[0x7f, 0xff]), Ok((&b""[..], 32_767_i16))); - assert_parse!(be_i16(&[0xff, 0xff]), Ok((&b""[..], -1))); - assert_parse!(be_i16(&[0x80, 0x00]), Ok((&b""[..], -32_768_i16))); + assert_parse!(be_i16(&[0x00, 0x00][..]), Ok((&b""[..], 0))); + assert_parse!(be_i16(&[0x7f, 0xff][..]), Ok((&b""[..], 32_767_i16))); + assert_parse!(be_i16(&[0xff, 0xff][..]), Ok((&b""[..], -1))); + assert_parse!(be_i16(&[0x80, 0x00][..]), Ok((&b""[..], -32_768_i16))); + assert_parse!(be_i16(&[][..]), Err(Err::Incomplete(Needed::new(2)))); + assert_parse!(be_i16(&[0x00][..]), Err(Err::Incomplete(Needed::new(1)))); } #[test] fn u24_tests() { - assert_parse!(be_u24(&[0x00, 0x00, 0x00]), Ok((&b""[..], 0))); - assert_parse!(be_u24(&[0x00, 0xFF, 0xFF]), Ok((&b""[..], 65_535_u32))); - assert_parse!(be_u24(&[0x12, 0x34, 0x56]), Ok((&b""[..], 1_193_046_u32))); + assert_parse!(be_u24(&[0x00, 0x00, 0x00][..]), Ok((&b""[..], 0))); + assert_parse!(be_u24(&[0x00, 0xFF, 0xFF][..]), Ok((&b""[..], 65_535_u32))); + assert_parse!( + be_u24(&[0x12, 0x34, 0x56][..]), + Ok((&b""[..], 1_193_046_u32)) + ); + assert_parse!(be_u24(&[][..]), Err(Err::Incomplete(Needed::new(3)))); + assert_parse!(be_u24(&[0x00][..]), Err(Err::Incomplete(Needed::new(2)))); + assert_parse!( + be_u24(&[0x00, 0x00][..]), + Err(Err::Incomplete(Needed::new(1))) + ); } #[test] fn i24_tests() { - assert_parse!(be_i24(&[0xFF, 0xFF, 0xFF]), Ok((&b""[..], -1_i32))); - assert_parse!(be_i24(&[0xFF, 0x00, 0x00]), Ok((&b""[..], -65_536_i32))); - assert_parse!(be_i24(&[0xED, 0xCB, 0xAA]), Ok((&b""[..], -1_193_046_i32))); + assert_parse!(be_i24(&[0xFF, 0xFF, 0xFF][..]), Ok((&b""[..], -1_i32))); + assert_parse!(be_i24(&[0xFF, 0x00, 0x00][..]), Ok((&b""[..], -65_536_i32))); + assert_parse!( + be_i24(&[0xED, 0xCB, 0xAA][..]), + Ok((&b""[..], -1_193_046_i32)) + ); + assert_parse!(be_i24(&[][..]), Err(Err::Incomplete(Needed::new(3)))); + assert_parse!(be_i24(&[0x00][..]), Err(Err::Incomplete(Needed::new(2)))); + assert_parse!( + be_i24(&[0x00, 0x00][..]), + Err(Err::Incomplete(Needed::new(1))) + ); } #[test] fn i32_tests() { - assert_parse!(be_i32(&[0x00, 0x00, 0x00, 0x00]), Ok((&b""[..], 0))); + assert_parse!(be_i32(&[0x00, 0x00, 0x00, 0x00][..]), Ok((&b""[..], 0))); assert_parse!( - be_i32(&[0x7f, 0xff, 0xff, 0xff]), + be_i32(&[0x7f, 0xff, 0xff, 0xff][..]), Ok((&b""[..], 2_147_483_647_i32)) ); - assert_parse!(be_i32(&[0xff, 0xff, 0xff, 0xff]), Ok((&b""[..], -1))); + assert_parse!(be_i32(&[0xff, 0xff, 0xff, 0xff][..]), Ok((&b""[..], -1))); assert_parse!( - be_i32(&[0x80, 0x00, 0x00, 0x00]), + be_i32(&[0x80, 0x00, 0x00, 0x00][..]), Ok((&b""[..], -2_147_483_648_i32)) ); + assert_parse!(be_i32(&[][..]), Err(Err::Incomplete(Needed::new(4)))); + assert_parse!(be_i32(&[0x00][..]), Err(Err::Incomplete(Needed::new(3)))); + assert_parse!( + be_i32(&[0x00, 0x00][..]), + Err(Err::Incomplete(Needed::new(2))) + ); + assert_parse!( + be_i32(&[0x00, 0x00, 0x00][..]), + Err(Err::Incomplete(Needed::new(1))) + ); } #[test] fn i64_tests() { assert_parse!( - be_i64(&[0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00]), + be_i64(&[0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00][..]), Ok((&b""[..], 0)) ); assert_parse!( - be_i64(&[0x7f, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff]), + be_i64(&[0x7f, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff][..]), Ok((&b""[..], 9_223_372_036_854_775_807_i64)) ); assert_parse!( - be_i64(&[0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff]), + be_i64(&[0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff][..]), Ok((&b""[..], -1)) ); assert_parse!( - be_i64(&[0x80, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00]), + be_i64(&[0x80, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00][..]), Ok((&b""[..], -9_223_372_036_854_775_808_i64)) ); + assert_parse!(be_i64(&[][..]), Err(Err::Incomplete(Needed::new(8)))); + assert_parse!(be_i64(&[0x00][..]), Err(Err::Incomplete(Needed::new(7)))); + assert_parse!( + be_i64(&[0x00, 0x00][..]), + Err(Err::Incomplete(Needed::new(6))) + ); + assert_parse!( + be_i64(&[0x00, 0x00, 0x00][..]), + Err(Err::Incomplete(Needed::new(5))) + ); + assert_parse!( + be_i64(&[0x00, 0x00, 0x00, 0x00][..]), + Err(Err::Incomplete(Needed::new(4))) + ); + assert_parse!( + be_i64(&[0x00, 0x00, 0x00, 0x00, 0x00][..]), + Err(Err::Incomplete(Needed::new(3))) + ); + assert_parse!( + be_i64(&[0x00, 0x00, 0x00, 0x00, 0x00, 0x00][..]), + Err(Err::Incomplete(Needed::new(2))) + ); + assert_parse!( + be_i64(&[0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00][..]), + Err(Err::Incomplete(Needed::new(1))) + ); } #[test] #[cfg(stable_i128)] fn i128_tests() { assert_parse!( - be_i128(&[0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00]), + be_i128( + &[ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00 + ][..] + ), Ok((&b""[..], 0)) ); assert_parse!( - be_i128(&[0x7f, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff]), - Ok((&b""[..], 170_141_183_460_469_231_731_687_303_715_884_105_727_i128)) + be_i128( + &[ + 0x7f, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, + 0xff + ][..] + ), + Ok(( + &b""[..], + 170_141_183_460_469_231_731_687_303_715_884_105_727_i128 + )) ); assert_parse!( - be_i128(&[0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff]), + be_i128( + &[ + 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, + 0xff + ][..] + ), Ok((&b""[..], -1)) ); assert_parse!( - be_i128(&[0x80, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00]), - Ok((&b""[..], -170_141_183_460_469_231_731_687_303_715_884_105_728_i128)) + be_i128( + &[ + 0x80, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00 + ][..] + ), + Ok(( + &b""[..], + -170_141_183_460_469_231_731_687_303_715_884_105_728_i128 + )) + ); + assert_parse!(be_i128(&[][..]), Err(Err::Incomplete(Needed::new(16)))); + assert_parse!(be_i128(&[0x00][..]), Err(Err::Incomplete(Needed::new(15)))); + assert_parse!( + be_i128(&[0x00, 0x00][..]), + Err(Err::Incomplete(Needed::new(14))) + ); + assert_parse!( + be_i128(&[0x00, 0x00, 0x00][..]), + Err(Err::Incomplete(Needed::new(13))) + ); + assert_parse!( + be_i128(&[0x00, 0x00, 0x00, 0x00][..]), + Err(Err::Incomplete(Needed::new(12))) + ); + assert_parse!( + be_i128(&[0x00, 0x00, 0x00, 0x00, 0x00][..]), + Err(Err::Incomplete(Needed::new(11))) + ); + assert_parse!( + be_i128(&[0x00, 0x00, 0x00, 0x00, 0x00, 0x00][..]), + Err(Err::Incomplete(Needed::new(10))) + ); + assert_parse!( + be_i128(&[0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00][..]), + Err(Err::Incomplete(Needed::new(9))) + ); + assert_parse!( + be_i128(&[0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00][..]), + Err(Err::Incomplete(Needed::new(8))) + ); + assert_parse!( + be_i128(&[0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00][..]), + Err(Err::Incomplete(Needed::new(7))) + ); + assert_parse!( + be_i128(&[0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00][..]), + Err(Err::Incomplete(Needed::new(6))) + ); + assert_parse!( + be_i128(&[0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00][..]), + Err(Err::Incomplete(Needed::new(5))) + ); + assert_parse!( + be_i128(&[0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00][..]), + Err(Err::Incomplete(Needed::new(4))) + ); + assert_parse!( + be_i128(&[0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00][..]), + Err(Err::Incomplete(Needed::new(3))) + ); + assert_parse!( + be_i128( + &[0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00][..] + ), + Err(Err::Incomplete(Needed::new(2))) + ); + assert_parse!( + be_i128( + &[0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00] + [..] + ), + Err(Err::Incomplete(Needed::new(1))) ); } #[test] fn le_i8_tests() { - assert_parse!(le_i8(&[0x00]), Ok((&b""[..], 0))); - assert_parse!(le_i8(&[0x7f]), Ok((&b""[..], 127))); - assert_parse!(le_i8(&[0xff]), Ok((&b""[..], -1))); - assert_parse!(le_i8(&[0x80]), Ok((&b""[..], -128))); + assert_parse!(le_i8(&[0x00][..]), Ok((&b""[..], 0))); + assert_parse!(le_i8(&[0x7f][..]), Ok((&b""[..], 127))); + assert_parse!(le_i8(&[0xff][..]), Ok((&b""[..], -1))); + assert_parse!(le_i8(&[0x80][..]), Ok((&b""[..], -128))); } #[test] fn le_i16_tests() { - assert_parse!(le_i16(&[0x00, 0x00]), Ok((&b""[..], 0))); - assert_parse!(le_i16(&[0xff, 0x7f]), Ok((&b""[..], 32_767_i16))); - assert_parse!(le_i16(&[0xff, 0xff]), Ok((&b""[..], -1))); - assert_parse!(le_i16(&[0x00, 0x80]), Ok((&b""[..], -32_768_i16))); + assert_parse!(le_i16(&[0x00, 0x00][..]), Ok((&b""[..], 0))); + assert_parse!(le_i16(&[0xff, 0x7f][..]), Ok((&b""[..], 32_767_i16))); + assert_parse!(le_i16(&[0xff, 0xff][..]), Ok((&b""[..], -1))); + assert_parse!(le_i16(&[0x00, 0x80][..]), Ok((&b""[..], -32_768_i16))); } #[test] fn le_u24_tests() { - assert_parse!(le_u24(&[0x00, 0x00, 0x00]), Ok((&b""[..], 0))); - assert_parse!(le_u24(&[0xFF, 0xFF, 0x00]), Ok((&b""[..], 65_535_u32))); - assert_parse!(le_u24(&[0x56, 0x34, 0x12]), Ok((&b""[..], 1_193_046_u32))); + assert_parse!(le_u24(&[0x00, 0x00, 0x00][..]), Ok((&b""[..], 0))); + assert_parse!(le_u24(&[0xFF, 0xFF, 0x00][..]), Ok((&b""[..], 65_535_u32))); + assert_parse!( + le_u24(&[0x56, 0x34, 0x12][..]), + Ok((&b""[..], 1_193_046_u32)) + ); } #[test] fn le_i24_tests() { - assert_parse!(le_i24(&[0xFF, 0xFF, 0xFF]), Ok((&b""[..], -1_i32))); - assert_parse!(le_i24(&[0x00, 0x00, 0xFF]), Ok((&b""[..], -65_536_i32))); - assert_parse!(le_i24(&[0xAA, 0xCB, 0xED]), Ok((&b""[..], -1_193_046_i32))); + assert_parse!(le_i24(&[0xFF, 0xFF, 0xFF][..]), Ok((&b""[..], -1_i32))); + assert_parse!(le_i24(&[0x00, 0x00, 0xFF][..]), Ok((&b""[..], -65_536_i32))); + assert_parse!( + le_i24(&[0xAA, 0xCB, 0xED][..]), + Ok((&b""[..], -1_193_046_i32)) + ); } #[test] fn le_i32_tests() { - assert_parse!(le_i32(&[0x00, 0x00, 0x00, 0x00]), Ok((&b""[..], 0))); + assert_parse!(le_i32(&[0x00, 0x00, 0x00, 0x00][..]), Ok((&b""[..], 0))); assert_parse!( - le_i32(&[0xff, 0xff, 0xff, 0x7f]), + le_i32(&[0xff, 0xff, 0xff, 0x7f][..]), Ok((&b""[..], 2_147_483_647_i32)) ); - assert_parse!(le_i32(&[0xff, 0xff, 0xff, 0xff]), Ok((&b""[..], -1))); + assert_parse!(le_i32(&[0xff, 0xff, 0xff, 0xff][..]), Ok((&b""[..], -1))); assert_parse!( - le_i32(&[0x00, 0x00, 0x00, 0x80]), + le_i32(&[0x00, 0x00, 0x00, 0x80][..]), Ok((&b""[..], -2_147_483_648_i32)) ); } @@ -1059,19 +1813,19 @@ mod tests { #[test] fn le_i64_tests() { assert_parse!( - le_i64(&[0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00]), + le_i64(&[0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00][..]), Ok((&b""[..], 0)) ); assert_parse!( - le_i64(&[0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x7f]), + le_i64(&[0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x7f][..]), Ok((&b""[..], 9_223_372_036_854_775_807_i64)) ); assert_parse!( - le_i64(&[0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff]), + le_i64(&[0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff][..]), Ok((&b""[..], -1)) ); assert_parse!( - le_i64(&[0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x80]), + le_i64(&[0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x80][..]), Ok((&b""[..], -9_223_372_036_854_775_808_i64)) ); } @@ -1080,28 +1834,54 @@ mod tests { #[cfg(stable_i128)] fn le_i128_tests() { assert_parse!( - le_i128(&[0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00]), + le_i128( + &[ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00 + ][..] + ), Ok((&b""[..], 0)) ); assert_parse!( - le_i128(&[0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x7f]), - Ok((&b""[..], 170_141_183_460_469_231_731_687_303_715_884_105_727_i128)) + le_i128( + &[ + 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, + 0x7f + ][..] + ), + Ok(( + &b""[..], + 170_141_183_460_469_231_731_687_303_715_884_105_727_i128 + )) ); assert_parse!( - le_i128(&[0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff]), + le_i128( + &[ + 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, + 0xff + ][..] + ), Ok((&b""[..], -1)) ); assert_parse!( - le_i128(&[0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x80]), - Ok((&b""[..], -170_141_183_460_469_231_731_687_303_715_884_105_728_i128)) + le_i128( + &[ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x80 + ][..] + ), + Ok(( + &b""[..], + -170_141_183_460_469_231_731_687_303_715_884_105_728_i128 + )) ); } #[test] fn be_f32_tests() { - assert_parse!(be_f32(&[0x00, 0x00, 0x00, 0x00]), Ok((&b""[..], 0_f32))); + assert_parse!(be_f32(&[0x00, 0x00, 0x00, 0x00][..]), Ok((&b""[..], 0_f32))); assert_parse!( - be_f32(&[0x4d, 0x31, 0x1f, 0xd8]), + be_f32(&[0x4d, 0x31, 0x1f, 0xd8][..]), Ok((&b""[..], 185_728_392_f32)) ); } @@ -1109,20 +1889,20 @@ mod tests { #[test] fn be_f64_tests() { assert_parse!( - be_f64(&[0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00]), + be_f64(&[0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00][..]), Ok((&b""[..], 0_f64)) ); assert_parse!( - be_f64(&[0x41, 0xa6, 0x23, 0xfb, 0x10, 0x00, 0x00, 0x00]), + be_f64(&[0x41, 0xa6, 0x23, 0xfb, 0x10, 0x00, 0x00, 0x00][..]), Ok((&b""[..], 185_728_392_f64)) ); } #[test] fn le_f32_tests() { - assert_parse!(le_f32(&[0x00, 0x00, 0x00, 0x00]), Ok((&b""[..], 0_f32))); + assert_parse!(le_f32(&[0x00, 0x00, 0x00, 0x00][..]), Ok((&b""[..], 0_f32))); assert_parse!( - le_f32(&[0xd8, 0x1f, 0x31, 0x4d]), + le_f32(&[0xd8, 0x1f, 0x31, 0x4d][..]), Ok((&b""[..], 185_728_392_f32)) ); } @@ -1130,11 +1910,11 @@ mod tests { #[test] fn le_f64_tests() { assert_parse!( - le_f64(&[0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00]), + le_f64(&[0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00][..]), Ok((&b""[..], 0_f64)) ); assert_parse!( - le_f64(&[0x00, 0x00, 0x00, 0x10, 0xfb, 0x23, 0xa6, 0x41]), + le_f64(&[0x00, 0x00, 0x00, 0x10, 0xfb, 0x23, 0xa6, 0x41][..]), Ok((&b""[..], 185_728_392_f64)) ); } @@ -1156,7 +1936,7 @@ mod tests { ); assert_parse!(hex_u32(&b"ffffffff;"[..]), Ok((&b";"[..], 4_294_967_295))); assert_parse!(hex_u32(&b"0x1be2;"[..]), Ok((&b"x1be2;"[..], 0))); - assert_parse!(hex_u32(&b"12af"[..]), Err(Err::Incomplete(Needed::Size(1)))); + assert_parse!(hex_u32(&b"12af"[..]), Err(Err::Incomplete(Needed::new(1)))); } #[test] @@ -1200,8 +1980,7 @@ mod tests { let remaining_exponent = "-1.234E-"; assert_parse!( recognize_float(remaining_exponent), - Err(Err::Incomplete(Needed::Size(1))) + Err(Err::Incomplete(Needed::new(1))) ); } - } diff --git a/src/regexp.rs b/src/regexp.rs deleted file mode 100644 index fd165e1..0000000 --- a/src/regexp.rs +++ /dev/null @@ -1,1090 +0,0 @@ -#[doc(hidden)] -#[macro_export] -macro_rules! regex ( - ($re: ident, $s:expr) => ( - lazy_static! { - static ref $re: $crate::lib::regex::Regex = $crate::lib::regex::Regex::new($s).unwrap(); - } - ); -); - -#[doc(hidden)] -#[macro_export] -macro_rules! regex_bytes ( - ($re: ident, $s:expr) => ( - lazy_static! { - static ref $re: $crate::lib::regex::bytes::Regex = $crate::lib::regex::bytes::Regex::new($s).unwrap(); - } - ); -); - -/// `re_match!(regexp) => &[T] -> IResult<&[T], &[T]>` -/// Returns the whole input if a match is found -/// -/// requires the `regexp` feature -#[macro_export(local_inner_macros)] -macro_rules! re_match ( - ($i:expr, $re:expr) => ( - { - use $crate::lib::std::result::Result::*; - use $crate::{Err,error::ErrorKind,IResult}; - - use $crate::InputLength; - use $crate::Slice; - let re = $crate::lib::regex::Regex::new($re).unwrap(); - if re.is_match(&$i) { - Ok(($i.slice($i.input_len()..), $i)) - } else { - let res: IResult<_,_> = Err(Err::Error(error_position!($i, ErrorKind::RegexpMatch))); - res - } - } - ) -); - -#[cfg(feature = "regexp_macros")] -/// `re_match_static!(regexp) => &[T] -> IResult<&[T], &[T]>` -/// Returns the whole input if a match is found. Regular expression calculated at compile time -/// -/// requires the `regexp_macros` feature -#[macro_export(local_inner_macros)] -macro_rules! re_match_static ( - ($i:expr, $re:expr) => ( - { - use $crate::lib::std::result::Result::*; - use $crate::{Err,error::ErrorKind,IResult}; - - use $crate::InputLength; - use $crate::Slice; - regex!(RE, $re); - if RE.is_match(&$i) { - Ok(($i.slice($i.input_len()..), $i)) - } else { - let res: IResult<_,_> = Err(Err::Error(error_position!($i, ErrorKind::RegexpMatch))); - res - } - } - ) -); - -/// `re_bytes_match!(regexp) => &[T] -> IResult<&[T], &[T]>` -/// Returns the whole input if a match is found -/// -/// requires the `regexp` feature -#[macro_export(local_inner_macros)] -macro_rules! re_bytes_match ( - ($i:expr, $re:expr) => ( - { - use $crate::lib::std::result::Result::*; - use $crate::{Err,error::ErrorKind,IResult}; - - use $crate::InputLength; - use $crate::Slice; - let re = $crate::lib::regex::bytes::Regex::new($re).unwrap(); - if re.is_match(&$i) { - Ok(($i.slice($i.input_len()..), $i)) - } else { - let res: IResult<_,_> = Err(Err::Error(error_position!($i, ErrorKind::RegexpMatch))); - res - } - } - ) -); - -#[cfg(feature = "regexp_macros")] -/// `re_bytes_match_static!(regexp) => &[T] -> IResult<&[T], &[T]>` -/// Returns the whole input if a match is found. Regular expression calculated at compile time -/// -/// requires the `regexp_macros` feature -#[macro_export(local_inner_macros)] -macro_rules! re_bytes_match_static ( - ($i:expr, $re:expr) => ( - { - use $crate::lib::std::result::Result::*; - use $crate::{Err,error::ErrorKind,IResult}; - - use $crate::InputLength; - use $crate::Slice; - regex_bytes!(RE, $re); - if RE.is_match(&$i) { - Ok(($i.slice($i.input_len()..), $i)) - } else { - let res: IResult<_,_> = Err(Err::Error(error_position!($i, ErrorKind::RegexpMatch))); - res - } - } - ) -); - -/// `re_find!(regexp) => &[T] -> IResult<&[T], &[T]>` -/// Returns the first match -/// -/// requires the `regexp` feature -#[macro_export(local_inner_macros)] -macro_rules! re_find ( - ($i:expr, $re:expr) => ( - { - use $crate::lib::std::result::Result::*; - use $crate::{Err,error::ErrorKind,IResult}; - - use $crate::Slice; - let re = $crate::lib::regex::Regex::new($re).unwrap(); - if let Some(m) = re.find(&$i) { - Ok(($i.slice(m.end()..), $i.slice(m.start()..m.end()))) - } else { - let res: IResult<_,_> = Err(Err::Error(error_position!($i, ErrorKind::RegexpFind))); - res - } - } - ) -); - -#[cfg(feature = "regexp_macros")] -/// `re_find_static!(regexp) => &[T] -> IResult<&[T], &[T]>` -/// Returns the first match. Regular expression calculated at compile time -/// -/// requires the `regexp_macros` feature -#[macro_export(local_inner_macros)] -macro_rules! re_find_static ( - ($i:expr, $re:expr) => ( - { - use $crate::lib::std::result::Result::*; - use $crate::{Err,error::ErrorKind,IResult}; - - use $crate::Slice; - regex!(RE, $re); - if let Some(m) = RE.find(&$i) { - Ok(($i.slice(m.end()..), $i.slice(m.start()..m.end()))) - } else { - let res: IResult<_,_> = Err(Err::Error(error_position!($i, ErrorKind::RegexpFind))); - res - } - } - - ) -); - -/// `re_bytes_find!(regexp) => &[T] -> IResult<&[T], &[T]>` -/// Returns the first match -/// -/// requires the `regexp` feature -#[macro_export(local_inner_macros)] -macro_rules! re_bytes_find ( - ($i:expr, $re:expr) => ( - { - use $crate::lib::std::result::Result::*; - use $crate::{Err,error::ErrorKind,IResult}; - - use $crate::Slice; - let re = $crate::lib::regex::bytes::Regex::new($re).unwrap(); - if let Some(m) = re.find(&$i) { - Ok(($i.slice(m.end()..), $i.slice(m.start()..m.end()))) - } else { - let res: IResult<_,_> = Err(Err::Error(error_position!($i, ErrorKind::RegexpFind))); - res - } - } - ) -); - -#[cfg(feature = "regexp_macros")] -/// `re_bytes_find!(regexp) => &[T] -> IResult<&[T], &[T]>` -/// Returns the first match. Regular expression calculated at compile time -/// -/// requires the `regexp_macros` feature -#[macro_export(local_inner_macros)] -macro_rules! re_bytes_find_static ( - ($i:expr, $re:expr) => ( - { - use $crate::lib::std::result::Result::*; - use $crate::{Err,error::ErrorKind,IResult}; - - use $crate::Slice; - regex_bytes!(RE, $re); - if let Some(m) = RE.find(&$i) { - Ok(($i.slice(m.end()..), $i.slice(m.start()..m.end()))) - } else { - let res: IResult<_,_> = Err(Err::Error(error_position!($i, ErrorKind::RegexpFind))); - res - } - } - - ) -); - -/// `re_matches!(regexp) => &[T] -> IResult<&[T], Vec<&[T]>>` -/// Returns all the matched parts -/// -/// requires the `regexp` feature -#[macro_export(local_inner_macros)] -macro_rules! re_matches ( - ($i:expr, $re:expr) => ( - { - use $crate::lib::std::result::Result::*; - use $crate::{Err,error::ErrorKind,IResult}; - - use $crate::Slice; - let re = $crate::lib::regex::Regex::new($re).unwrap(); - let v: Vec<_> = re.find_iter(&$i).map(|m| $i.slice(m.start()..m.end())).collect(); - if v.len() != 0 { - let offset = { - let end = v.last().unwrap(); - end.as_ptr() as usize + end.len() - $i.as_ptr() as usize - }; - Ok(($i.slice(offset..), v)) - } else { - let res: IResult<_,_> = Err(Err::Error(error_position!($i, ErrorKind::RegexpMatches))); - res - } - } - ) -); - -#[cfg(feature = "regexp_macros")] -/// `re_matches_static!(regexp) => &[T] -> IResult<&[T], Vec<&[T]>>` -/// Returns all the matched parts. Regular expression calculated at compile time -/// -/// requires the `regexp_macros` feature -#[macro_export(local_inner_macros)] -macro_rules! re_matches_static ( - ($i:expr, $re:expr) => ( - { - use $crate::lib::std::result::Result::*; - use $crate::{Err,error::ErrorKind,IResult}; - - use $crate::Slice; - regex!(RE, $re); - let v: Vec<_> = RE.find_iter(&$i).map(|m| $i.slice(m.start()..m.end())).collect(); - if v.len() != 0 { - let offset = { - let end = v.last().unwrap(); - end.as_ptr() as usize + end.len() - $i.as_ptr() as usize - }; - Ok(($i.slice(offset..), v)) - } else { - let res: IResult<_,_> = Err(Err::Error(error_position!($i, ErrorKind::RegexpMatches))); - res - } - } - ) -); - -/// `re_bytes_matches!(regexp) => &[T] -> IResult<&[T], Vec<&[T]>>` -/// Returns all the matched parts -/// -/// requires the `regexp` feature -#[macro_export(local_inner_macros)] -macro_rules! re_bytes_matches ( - ($i:expr, $re:expr) => ( - { - use $crate::lib::std::result::Result::*; - use $crate::{Err,error::ErrorKind,IResult}; - - use $crate::Slice; - let re = $crate::lib::regex::bytes::Regex::new($re).unwrap(); - let v: Vec<_> = re.find_iter(&$i).map(|m| $i.slice(m.start()..m.end())).collect(); - if v.len() != 0 { - let offset = { - let end = v.last().unwrap(); - end.as_ptr() as usize + end.len() - $i.as_ptr() as usize - }; - Ok(($i.slice(offset..), v)) - } else { - let res: IResult<_,_> = Err(Err::Error(error_position!($i, ErrorKind::RegexpMatches))); - res - } - } - ) -); - -#[cfg(feature = "regexp_macros")] -/// `re_bytes_matches_static!(regexp) => &[T] -> IResult<&[T], Vec<&[T]>>` -/// Returns all the matched parts. Regular expression calculated at compile time -/// -/// requires the `regexp_macros` feature -#[macro_export(local_inner_macros)] -macro_rules! re_bytes_matches_static ( - ($i:expr, $re:expr) => ( - { - use $crate::lib::std::result::Result::*; - use $crate::{Err,error::ErrorKind,IResult}; - - use $crate::Slice; - regex_bytes!(RE, $re); - let v: Vec<_> = RE.find_iter(&$i).map(|m| $i.slice(m.start()..m.end())).collect(); - if v.len() != 0 { - let offset = { - let end = v.last().unwrap(); - end.as_ptr() as usize + end.len() - $i.as_ptr() as usize - }; - Ok(($i.slice(offset..), v)) - } else { - let res: IResult<_,_> = Err(Err::Error(error_position!($i, ErrorKind::RegexpMatches))); - res - } - } - ) -); - -/// `re_capture!(regexp) => &[T] -> IResult<&[T], Vec<&[T]>>` -/// Returns the first capture group -/// -/// requires the `regexp` feature -#[macro_export(local_inner_macros)] -macro_rules! re_capture ( - ($i:expr, $re:expr) => ( - { - use $crate::lib::std::result::Result::*; - use $crate::{Err,error::ErrorKind,IResult}; - - use $crate::Slice; - let re = $crate::lib::regex::Regex::new($re).unwrap(); - if let Some(c) = re.captures(&$i) { - let v:Vec<_> = c.iter().filter(|el| el.is_some()).map(|el| el.unwrap()).map(|m| $i.slice(m.start()..m.end())).collect(); - let offset = { - let end = v.last().unwrap(); - end.as_ptr() as usize + end.len() - $i.as_ptr() as usize - }; - Ok(($i.slice(offset..), v)) - } else { - let res: IResult<_,_> = Err(Err::Error(error_position!($i, ErrorKind::RegexpCapture))); - res - } - } - ) -); - -#[cfg(feature = "regexp_macros")] -/// `re_capture_static!(regexp) => &[T] -> IResult<&[T], Vec<&[T]>>` -/// Returns the first capture group. Regular expression calculated at compile time -/// -/// requires the `regexp_macros` feature -#[macro_export(local_inner_macros)] -macro_rules! re_capture_static ( - ($i:expr, $re:expr) => ( - { - use $crate::lib::std::result::Result::*; - use $crate::{Err,error::ErrorKind,IResult}; - - use $crate::Slice; - regex!(RE, $re); - if let Some(c) = RE.captures(&$i) { - let v:Vec<_> = c.iter().filter(|el| el.is_some()).map(|el| el.unwrap()).map(|m| $i.slice(m.start()..m.end())).collect(); - let offset = { - let end = v.last().unwrap(); - end.as_ptr() as usize + end.len() - $i.as_ptr() as usize - }; - Ok(($i.slice(offset..), v)) - } else { - let res: IResult<_,_> = Err(Err::Error(error_position!($i, ErrorKind::RegexpCapture))); - res - } - } - ) -); - -/// `re_bytes_capture!(regexp) => &[T] -> IResult<&[T], Vec<&[T]>>` -/// Returns the first capture group -/// -/// requires the `regexp` feature -#[macro_export(local_inner_macros)] -macro_rules! re_bytes_capture ( - ($i:expr, $re:expr) => ( - { - use $crate::lib::std::result::Result::*; - use $crate::{Err,error::ErrorKind,IResult}; - - use $crate::Slice; - let re = $crate::lib::regex::bytes::Regex::new($re).unwrap(); - if let Some(c) = re.captures(&$i) { - let v:Vec<_> = c.iter().filter(|el| el.is_some()).map(|el| el.unwrap()).map(|m| $i.slice(m.start()..m.end())).collect(); - let offset = { - let end = v.last().unwrap(); - end.as_ptr() as usize + end.len() - $i.as_ptr() as usize - }; - Ok(($i.slice(offset..), v)) - } else { - let res: IResult<_,_> = Err(Err::Error(error_position!($i, ErrorKind::RegexpCapture))); - res - } - } - ) -); - -#[cfg(feature = "regexp_macros")] -/// `re_bytes_capture_static!(regexp) => &[T] -> IResult<&[T], Vec<&[T]>>` -/// Returns the first capture group. Regular expression calculated at compile time -/// -/// requires the `regexp_macros` feature -#[macro_export(local_inner_macros)] -macro_rules! re_bytes_capture_static ( - ($i:expr, $re:expr) => ( - { - use $crate::lib::std::result::Result::*; - use $crate::{Err,error::ErrorKind,IResult}; - - use $crate::Slice; - regex_bytes!(RE, $re); - if let Some(c) = RE.captures(&$i) { - let v:Vec<_> = c.iter().filter(|el| el.is_some()).map(|el| el.unwrap()).map(|m| $i.slice(m.start()..m.end())).collect(); - let offset = { - let end = v.last().unwrap(); - end.as_ptr() as usize + end.len() - $i.as_ptr() as usize - }; - Ok(($i.slice(offset..), v)) - } else { - let res: IResult<_,_> = Err(Err::Error(error_position!($i, ErrorKind::RegexpCapture))); - res - } - } - ) -); - -/// `re_captures!(regexp) => &[T] -> IResult<&[T], Vec<Vec<&[T]>>>` -/// Returns all the capture groups -/// -/// requires the `regexp` feature -#[macro_export(local_inner_macros)] -macro_rules! re_captures ( - ($i:expr, $re:expr) => ( - { - use $crate::lib::std::result::Result::*; - use $crate::{Err,error::ErrorKind,IResult}; - - use $crate::Slice; - let re = $crate::lib::regex::Regex::new($re).unwrap(); - let v:Vec<Vec<_>> = re.captures_iter(&$i) - .map(|c| c.iter().filter(|el| el.is_some()).map(|el| el.unwrap()) - .map(|m| $i.slice(m.start()..m.end())).collect()).collect(); - if v.len() != 0 { - let offset = { - let end = v.last().unwrap().last().unwrap(); - end.as_ptr() as usize + end.len() - $i.as_ptr() as usize - }; - Ok(($i.slice(offset..), v)) - } else { - let res: IResult<_,_> = Err(Err::Error(error_position!($i, ErrorKind::RegexpCapture))); - res - } - } - ) -); - -#[cfg(feature = "regexp_macros")] -/// `re_captures_static!(regexp) => &[T] -> IResult<&[T], Vec<Vec<&[T]>>>` -/// Returns all the capture groups. Regular expression calculated at compile time -/// -/// requires the `regexp_macros` feature -#[macro_export(local_inner_macros)] -macro_rules! re_captures_static ( - ($i:expr, $re:expr) => ( - { - use $crate::lib::std::result::Result::*; - use $crate::{Err,error::ErrorKind,IResult}; - - use $crate::Slice; - regex!(RE, $re); - let v:Vec<Vec<_>> = RE.captures_iter(&$i) - .map(|c| c.iter().filter(|el| el.is_some()).map(|el| el.unwrap()).map(|m| $i.slice(m.start()..m.end())).collect()).collect(); - if v.len() != 0 { - let offset = { - let end = v.last().unwrap().last().unwrap(); - end.as_ptr() as usize + end.len() - $i.as_ptr() as usize - }; - Ok(($i.slice(offset..), v)) - } else { - let res: IResult<_,_> = Err(Err::Error(error_position!($i, ErrorKind::RegexpCapture))); - res - } - } - ) -); - -/// `re_bytes_captures!(regexp) => &[T] -> IResult<&[T], Vec<Vec<&[T]>>>` -/// Returns all the capture groups -/// -/// requires the `regexp` feature -#[macro_export(local_inner_macros)] -macro_rules! re_bytes_captures ( - ($i:expr, $re:expr) => ( - { - use $crate::lib::std::result::Result::*; - use $crate::{Err,error::ErrorKind,IResult}; - - use $crate::Slice; - let re = $crate::lib::regex::bytes::Regex::new($re).unwrap(); - let v:Vec<Vec<_>> = re.captures_iter(&$i) - .map(|c| c.iter().filter(|el| el.is_some()).map(|el| el.unwrap()).map(|m| $i.slice(m.start()..m.end())).collect()).collect(); - if v.len() != 0 { - let offset = { - let end = v.last().unwrap().last().unwrap(); - end.as_ptr() as usize + end.len() - $i.as_ptr() as usize - }; - Ok(($i.slice(offset..), v)) - } else { - let res: IResult<_,_> = Err(Err::Error(error_position!($i, ErrorKind::RegexpCapture))); - res - } - } - ) -); - -#[cfg(feature = "regexp_macros")] -/// `re_bytes_captures_static!(regexp) => &[T] -> IResult<&[T], Vec<Vec<&[T]>>>` -/// Returns all the capture groups. Regular expression calculated at compile time -/// -/// requires the `regexp_macros` feature -#[macro_export(local_inner_macros)] -macro_rules! re_bytes_captures_static ( - ($i:expr, $re:expr) => ( - { - use $crate::lib::std::result::Result::*; - use $crate::{Err,error::ErrorKind,IResult}; - - use $crate::Slice; - regex_bytes!(RE, $re); - let v:Vec<Vec<_>> = RE.captures_iter(&$i) - .map(|c| c.iter().filter(|el| el.is_some()).map(|el| el.unwrap()).map(|m| $i.slice(m.start()..m.end())).collect()).collect(); - if v.len() != 0 { - let offset = { - let end = v.last().unwrap().last().unwrap(); - end.as_ptr() as usize + end.len() - $i.as_ptr() as usize - }; - Ok(($i.slice(offset..), v)) - } else { - let res: IResult<_,_> = Err(Err::Error(error_position!($i, ErrorKind::RegexpCapture))); - res - } - } - ) -); -#[cfg(test)] -mod tests { - #[cfg(feature = "alloc")] - use crate::lib::std::vec::Vec; - use crate::error::ErrorKind; - use crate::internal::Err; - - #[test] - fn re_match() { - named!(rm<&str,&str>, re_match!(r"^\d{4}-\d{2}-\d{2}")); - assert_eq!(rm("2015-09-07"), Ok(("", "2015-09-07"))); - assert_eq!( - rm("blah"), - Err(Err::Error(error_position!( - &"blah"[..], - ErrorKind::RegexpMatch - ),)) - ); - assert_eq!(rm("2015-09-07blah"), Ok(("", "2015-09-07blah"))); - } - - #[cfg(feature = "regexp_macros")] - #[test] - fn re_match_static() { - named!(rm<&str,&str>, re_match_static!(r"^\d{4}-\d{2}-\d{2}")); - assert_eq!(rm("2015-09-07"), Ok(("", "2015-09-07"))); - assert_eq!( - rm("blah"), - Err(Err::Error(error_position!( - &"blah"[..], - ErrorKind::RegexpMatch - ),)) - ); - assert_eq!(rm("2015-09-07blah"), Ok(("", "2015-09-07blah"))); - } - - #[test] - fn re_find() { - named!(rm<&str,&str>, re_find!(r"^\d{4}-\d{2}-\d{2}")); - assert_eq!(rm("2015-09-07"), Ok(("", "2015-09-07"))); - assert_eq!( - rm("blah"), - Err(Err::Error(error_position!( - &"blah"[..], - ErrorKind::RegexpFind - ),)) - ); - assert_eq!(rm("2015-09-07blah"), Ok(("blah", "2015-09-07"))); - } - - #[cfg(feature = "regexp_macros")] - #[test] - fn re_find_static() { - named!(rm<&str,&str>, re_find_static!(r"^\d{4}-\d{2}-\d{2}")); - assert_eq!(rm("2015-09-07"), Ok(("", "2015-09-07"))); - assert_eq!( - rm("blah"), - Err(Err::Error(error_position!( - &"blah"[..], - ErrorKind::RegexpFind - ),)) - ); - assert_eq!(rm("2015-09-07blah"), Ok(("blah", "2015-09-07"))); - } - - #[cfg(feature = "alloc")] - #[test] - fn re_matches() { - named!(rm< &str,Vec<&str> >, re_matches!(r"\d{4}-\d{2}-\d{2}")); - assert_eq!(rm("2015-09-07"), Ok(("", vec!["2015-09-07"]))); - assert_eq!( - rm("blah"), - Err(Err::Error(error_position!( - &"blah"[..], - ErrorKind::RegexpMatches - ))) - ); - assert_eq!( - rm("aaa2015-09-07blah2015-09-09pouet"), - Ok(("pouet", vec!["2015-09-07", "2015-09-09"])) - ); - } - - #[cfg(feature = "regexp_macros")] - #[cfg(feature = "alloc")] - #[test] - fn re_matches_static() { - named!(rm< &str,Vec<&str> >, re_matches_static!(r"\d{4}-\d{2}-\d{2}")); - assert_eq!(rm("2015-09-07"), Ok(("", vec!["2015-09-07"]))); - assert_eq!( - rm("blah"), - Err(Err::Error(error_position!( - &"blah"[..], - ErrorKind::RegexpMatches - ))) - ); - assert_eq!( - rm("aaa2015-09-07blah2015-09-09pouet"), - Ok(("pouet", vec!["2015-09-07", "2015-09-09"])) - ); - } - - #[cfg(feature = "alloc")] - #[test] - fn re_capture() { - named!(rm< &str,Vec<&str> >, re_capture!(r"([[:alpha:]]+)\s+((\d+).(\d+).(\d+))")); - assert_eq!( - rm("blah nom 0.3.11pouet"), - Ok(("pouet", vec!["nom 0.3.11", "nom", "0.3.11", "0", "3", "11"])) - ); - assert_eq!( - rm("blah"), - Err(Err::Error(error_position!( - &"blah"[..], - ErrorKind::RegexpCapture - ))) - ); - assert_eq!( - rm("hello nom 0.3.11 world regex 0.1.41"), - Ok(( - " world regex 0.1.41", - vec!["nom 0.3.11", "nom", "0.3.11", "0", "3", "11"] - )) - ); - } - - #[cfg(feature = "alloc")] - #[cfg(feature = "regexp_macros")] - #[test] - fn re_capture_static() { - named!(rm< &str,Vec<&str> >, re_capture_static!(r"([[:alpha:]]+)\s+((\d+).(\d+).(\d+))")); - assert_eq!( - rm("blah nom 0.3.11pouet"), - Ok(("pouet", vec!["nom 0.3.11", "nom", "0.3.11", "0", "3", "11"])) - ); - assert_eq!( - rm("blah"), - Err(Err::Error(error_position!( - &"blah"[..], - ErrorKind::RegexpCapture - ))) - ); - assert_eq!( - rm("hello nom 0.3.11 world regex 0.1.41"), - Ok(( - " world regex 0.1.41", - vec!["nom 0.3.11", "nom", "0.3.11", "0", "3", "11"] - )) - ); - } - - #[cfg(feature = "alloc")] - #[test] - fn re_captures() { - named!(rm< &str,Vec<Vec<&str>> >, re_captures!(r"([[:alpha:]]+)\s+((\d+).(\d+).(\d+))")); - assert_eq!( - rm("blah nom 0.3.11pouet"), - Ok(( - "pouet", - vec![vec!["nom 0.3.11", "nom", "0.3.11", "0", "3", "11"]] - )) - ); - assert_eq!( - rm("blah"), - Err(Err::Error(error_position!( - &"blah"[..], - ErrorKind::RegexpCapture - ))) - ); - assert_eq!( - rm("hello nom 0.3.11 world regex 0.1.41 aaa"), - Ok(( - " aaa", - vec![ - vec!["nom 0.3.11", "nom", "0.3.11", "0", "3", "11"], - vec!["regex 0.1.41", "regex", "0.1.41", "0", "1", "41"], - ] - )) - ); - } - - #[cfg(feature = "alloc")] - #[cfg(feature = "regexp_macros")] - #[test] - fn re_captures_static() { - named!(rm< &str,Vec<Vec<&str>> >, re_captures_static!(r"([[:alpha:]]+)\s+((\d+).(\d+).(\d+))")); - assert_eq!( - rm("blah nom 0.3.11pouet"), - Ok(( - "pouet", - vec![vec!["nom 0.3.11", "nom", "0.3.11", "0", "3", "11"]] - )) - ); - assert_eq!( - rm("blah"), - Err(Err::Error(error_position!( - &"blah"[..], - ErrorKind::RegexpCapture - ))) - ); - assert_eq!( - rm("hello nom 0.3.11 world regex 0.1.41 aaa"), - Ok(( - " aaa", - vec![ - vec!["nom 0.3.11", "nom", "0.3.11", "0", "3", "11"], - vec!["regex 0.1.41", "regex", "0.1.41", "0", "1", "41"], - ] - )) - ); - } - - #[test] - fn re_bytes_match() { - named!(rm, re_bytes_match!(r"^\d{4}-\d{2}-\d{2}")); - assert_eq!(rm(&b"2015-09-07"[..]), Ok((&b""[..], &b"2015-09-07"[..]))); - assert_eq!( - rm(&b"blah"[..]), - Err(Err::Error(error_position!( - &b"blah"[..], - ErrorKind::RegexpMatch - ))) - ); - assert_eq!( - rm(&b"2015-09-07blah"[..]), - Ok((&b""[..], &b"2015-09-07blah"[..])) - ); - } - - #[cfg(feature = "regexp_macros")] - #[test] - fn re_bytes_match_static() { - named!(rm, re_bytes_match_static!(r"^\d{4}-\d{2}-\d{2}")); - assert_eq!(rm(&b"2015-09-07"[..]), Ok((&b""[..], &b"2015-09-07"[..]))); - assert_eq!( - rm(&b"blah"[..]), - Err(Err::Error(error_position!( - &b"blah"[..], - ErrorKind::RegexpMatch - ))) - ); - assert_eq!( - rm(&b"2015-09-07blah"[..]), - Ok((&b""[..], &b"2015-09-07blah"[..])) - ); - } - - #[test] - fn re_bytes_find() { - named!(rm, re_bytes_find!(r"^\d{4}-\d{2}-\d{2}")); - assert_eq!(rm(&b"2015-09-07"[..]), Ok((&b""[..], &b"2015-09-07"[..]))); - assert_eq!( - rm(&b"blah"[..]), - Err(Err::Error(error_position!( - &b"blah"[..], - ErrorKind::RegexpFind - ))) - ); - assert_eq!( - rm(&b"2015-09-07blah"[..]), - Ok((&b"blah"[..], &b"2015-09-07"[..])) - ); - } - - #[cfg(feature = "regexp_macros")] - #[test] - fn re_bytes_find_static() { - named!(rm, re_bytes_find_static!(r"^\d{4}-\d{2}-\d{2}")); - assert_eq!(rm(&b"2015-09-07"[..]), Ok((&b""[..], &b"2015-09-07"[..]))); - assert_eq!( - rm(&b"blah"[..]), - Err(Err::Error(error_position!( - &b"blah"[..], - ErrorKind::RegexpFind - ))) - ); - assert_eq!( - rm(&b"2015-09-07blah"[..]), - Ok((&b"blah"[..], &b"2015-09-07"[..])) - ); - } - - #[cfg(feature = "alloc")] - #[test] - fn re_bytes_matches() { - named!(rm<Vec<&[u8]>>, re_bytes_matches!(r"\d{4}-\d{2}-\d{2}")); - assert_eq!( - rm(&b"2015-09-07"[..]), - Ok((&b""[..], vec![&b"2015-09-07"[..]])) - ); - assert_eq!( - rm(&b"blah"[..]), - Err(Err::Error(error_position!( - &b"blah"[..], - ErrorKind::RegexpMatches - ))) - ); - assert_eq!( - rm(&b"aaa2015-09-07blah2015-09-09pouet"[..]), - Ok((&b"pouet"[..], vec![&b"2015-09-07"[..], &b"2015-09-09"[..]])) - ); - } - - #[cfg(feature = "alloc")] - #[cfg(feature = "regexp_macros")] - #[test] - fn re_bytes_matches_static() { - named!( - rm<Vec<&[u8]>>, - re_bytes_matches_static!(r"\d{4}-\d{2}-\d{2}") - ); - assert_eq!( - rm(&b"2015-09-07"[..]), - Ok((&b""[..], vec![&b"2015-09-07"[..]])) - ); - assert_eq!( - rm(&b"blah"[..]), - Err(Err::Error(error_position!( - &b"blah"[..], - ErrorKind::RegexpMatches - ))) - ); - assert_eq!( - rm(&b"aaa2015-09-07blah2015-09-09pouet"[..]), - Ok((&b"pouet"[..], vec![&b"2015-09-07"[..], &b"2015-09-09"[..]])) - ); - } - - #[cfg(feature = "alloc")] - #[test] - fn re_bytes_capture() { - named!( - rm<Vec<&[u8]>>, - re_bytes_capture!(r"([[:alpha:]]+)\s+((\d+).(\d+).(\d+))") - ); - assert_eq!( - rm(&b"blah nom 0.3.11pouet"[..]), - Ok(( - &b"pouet"[..], - vec![ - &b"nom 0.3.11"[..], - &b"nom"[..], - &b"0.3.11"[..], - &b"0"[..], - &b"3"[..], - &b"11"[..], - ] - )) - ); - assert_eq!( - rm(&b"blah"[..]), - Err(Err::Error(error_position!( - &b"blah"[..], - ErrorKind::RegexpCapture - ))) - ); - assert_eq!( - rm(&b"hello nom 0.3.11 world regex 0.1.41"[..]), - Ok(( - &b" world regex 0.1.41"[..], - vec![ - &b"nom 0.3.11"[..], - &b"nom"[..], - &b"0.3.11"[..], - &b"0"[..], - &b"3"[..], - &b"11"[..], - ] - )) - ); - } - - #[cfg(feature = "alloc")] - #[cfg(feature = "regexp_macros")] - #[test] - fn re_bytes_capture_static() { - named!( - rm<Vec<&[u8]>>, - re_bytes_capture_static!(r"([[:alpha:]]+)\s+((\d+).(\d+).(\d+))") - ); - assert_eq!( - rm(&b"blah nom 0.3.11pouet"[..]), - Ok(( - &b"pouet"[..], - vec![ - &b"nom 0.3.11"[..], - &b"nom"[..], - &b"0.3.11"[..], - &b"0"[..], - &b"3"[..], - &b"11"[..], - ] - )) - ); - assert_eq!( - rm(&b"blah"[..]), - Err(Err::Error(error_position!( - &b"blah"[..], - ErrorKind::RegexpCapture - ))) - ); - assert_eq!( - rm(&b"hello nom 0.3.11 world regex 0.1.41"[..]), - Ok(( - &b" world regex 0.1.41"[..], - vec![ - &b"nom 0.3.11"[..], - &b"nom"[..], - &b"0.3.11"[..], - &b"0"[..], - &b"3"[..], - &b"11"[..], - ] - )) - ); - } - - #[cfg(feature = "alloc")] - #[test] - fn re_bytes_captures() { - named!( - rm<Vec<Vec<&[u8]>>>, - re_bytes_captures!(r"([[:alpha:]]+)\s+((\d+).(\d+).(\d+))") - ); - assert_eq!( - rm(&b"blah nom 0.3.11pouet"[..]), - Ok(( - &b"pouet"[..], - vec![ - vec![ - &b"nom 0.3.11"[..], - &b"nom"[..], - &b"0.3.11"[..], - &b"0"[..], - &b"3"[..], - &b"11"[..], - ], - ] - )) - ); - assert_eq!( - rm(&b"blah"[..]), - Err(Err::Error(error_position!( - &b"blah"[..], - ErrorKind::RegexpCapture - ))) - ); - assert_eq!( - rm(&b"hello nom 0.3.11 world regex 0.1.41 aaa"[..]), - Ok(( - &b" aaa"[..], - vec![ - vec![ - &b"nom 0.3.11"[..], - &b"nom"[..], - &b"0.3.11"[..], - &b"0"[..], - &b"3"[..], - &b"11"[..], - ], - vec![ - &b"regex 0.1.41"[..], - &b"regex"[..], - &b"0.1.41"[..], - &b"0"[..], - &b"1"[..], - &b"41"[..], - ], - ] - )) - ); - } - - #[cfg(feature = "alloc")] - #[cfg(feature = "regexp_macros")] - #[test] - fn re_bytes_captures_static() { - named!( - rm<Vec<Vec<&[u8]>>>, - re_bytes_captures_static!(r"([[:alpha:]]+)\s+((\d+).(\d+).(\d+))") - ); - assert_eq!( - rm(&b"blah nom 0.3.11pouet"[..]), - Ok(( - &b"pouet"[..], - vec![ - vec![ - &b"nom 0.3.11"[..], - &b"nom"[..], - &b"0.3.11"[..], - &b"0"[..], - &b"3"[..], - &b"11"[..], - ], - ] - )) - ); - assert_eq!( - rm(&b"blah"[..]), - Err(Err::Error(error_position!( - &b"blah"[..], - ErrorKind::RegexpCapture - ))) - ); - assert_eq!( - rm(&b"hello nom 0.3.11 world regex 0.1.41 aaa"[..]), - Ok(( - &b" aaa"[..], - vec![ - vec![ - &b"nom 0.3.11"[..], - &b"nom"[..], - &b"0.3.11"[..], - &b"0"[..], - &b"3"[..], - &b"11"[..], - ], - vec![ - &b"regex 0.1.41"[..], - &b"regex"[..], - &b"0.1.41"[..], - &b"0"[..], - &b"1"[..], - &b"41"[..], - ], - ] - )) - ); - } -} diff --git a/src/regexp/macros.rs b/src/regexp/macros.rs new file mode 100644 index 0000000..ff84787 --- /dev/null +++ b/src/regexp/macros.rs @@ -0,0 +1,409 @@ +/// `re_match!(regexp) => &[T] -> IResult<&[T], &[T]>` +/// Returns the whole input if a match is found. +/// +/// Requires the `regexp` feature. +#[macro_export(local_inner_macros)] +#[cfg_attr(feature = "docsrs", doc(cfg(feature = "regexp")))] +macro_rules! re_match ( + ($i:expr, $re:expr) => ( { + let r = $crate::lib::regex::Regex::new($re).unwrap(); + $crate::regexp::str::re_match(r)($i) + } ) +); + +/// `re_bytes_match!(regexp) => &[T] -> IResult<&[T], &[T]>` +/// Returns the whole input if a match is found. +/// +/// Requires the `regexp` feature. +#[macro_export(local_inner_macros)] +#[cfg_attr(feature = "docsrs", doc(cfg(feature = "regexp")))] +macro_rules! re_bytes_match ( + ($i:expr, $re:expr) => ( { + let r = $crate::lib::regex::bytes::Regex::new($re).unwrap(); + $crate::regexp::bytes::re_match(r)($i) + } ) +); + +/// `re_find!(regexp) => &[T] -> IResult<&[T], &[T]>` +/// Returns the first match. +/// +/// Requires the `regexp` feature. +#[macro_export(local_inner_macros)] +#[cfg_attr(feature = "docsrs", doc(cfg(feature = "regexp")))] +macro_rules! re_find ( + ($i:expr, $re:expr) => ( { + let r = $crate::lib::regex::Regex::new($re).unwrap(); + $crate::regexp::str::re_find(r)($i) + } ) +); + +/// `re_bytes_find!(regexp) => &[T] -> IResult<&[T], &[T]>` +/// Returns the first match. +/// +/// Requires the `regexp` feature. +#[macro_export(local_inner_macros)] +#[cfg_attr(feature = "docsrs", doc(cfg(feature = "regexp")))] +macro_rules! re_bytes_find ( + ($i:expr, $re:expr) => ( { + let r = $crate::lib::regex::bytes::Regex::new($re).unwrap(); + $crate::regexp::bytes::re_find(r)($i) + } ) +); + +/// `re_matches!(regexp) => &[T] -> IResult<&[T], Vec<&[T]>>` +/// Returns all the matched parts. +/// +/// Requires the `regexp` feature. +#[macro_export(local_inner_macros)] +#[cfg_attr( + feature = "docsrs", + doc(cfg(all(feature = "regexp", feature = "alloc"))) +)] +macro_rules! re_matches ( + ($i:expr, $re:expr) => ( { + let r = $crate::lib::regex::Regex::new($re).unwrap(); + $crate::regexp::str::re_matches(r)($i) + } ) +); + +/// `re_bytes_matches!(regexp) => &[T] -> IResult<&[T], Vec<&[T]>>` +/// Returns all the matched parts. +/// +/// Requires the `regexp` feature. +#[macro_export(local_inner_macros)] +#[cfg_attr( + feature = "docsrs", + doc(cfg(all(feature = "regexp", feature = "alloc"))) +)] +macro_rules! re_bytes_matches ( + ($i:expr, $re:expr) => ( { + let r = $crate::lib::regex::bytes::Regex::new($re).unwrap(); + $crate::regexp::bytes::re_matches(r)($i) + } ) +); + +/// `re_capture!(regexp) => &[T] -> IResult<&[T], Vec<&[T]>>` +/// Returns the capture groups of the first match. +/// +/// Requires the `regexp` feature. +#[macro_export(local_inner_macros)] +#[cfg_attr( + feature = "docsrs", + doc(cfg(all(feature = "regexp", feature = "alloc"))) +)] +macro_rules! re_capture ( + ($i:expr, $re:expr) => ( { + let r = $crate::lib::regex::Regex::new($re).unwrap(); + $crate::regexp::str::re_capture(r)($i) + } ) +); + +/// `re_bytes_capture!(regexp) => &[T] -> IResult<&[T], Vec<&[T]>>` +/// Returns the capture groups of the first match. +/// +/// Requires the `regexp` feature. +#[macro_export(local_inner_macros)] +#[cfg_attr( + feature = "docsrs", + doc(cfg(all(feature = "regexp", feature = "alloc"))) +)] +macro_rules! re_bytes_capture ( + ($i:expr, $re:expr) => ( { + let r = $crate::lib::regex::bytes::Regex::new($re).unwrap(); + $crate::regexp::bytes::re_capture(r)($i) + } + ) +); + +/// `re_captures!(regexp) => &[T] -> IResult<&[T], Vec<Vec<&[T]>>>` +/// Returns the capture groups of all matches. +/// +/// Requires the `regexp` feature. +#[macro_export(local_inner_macros)] +#[cfg_attr( + feature = "docsrs", + doc(cfg(all(feature = "regexp", feature = "alloc"))) +)] +macro_rules! re_captures ( + ($i:expr, $re:expr) => ( { + let r = $crate::lib::regex::Regex::new($re).unwrap(); + $crate::regexp::str::re_captures(r)($i) + } ) +); + +/// `re_bytes_captures!(regexp) => &[T] -> IResult<&[T], Vec<Vec<&[T]>>>` +/// Returns the capture groups of all matches. +/// +/// Requires the `regexp` feature. +#[macro_export(local_inner_macros)] +#[cfg_attr( + feature = "docsrs", + doc(cfg(all(feature = "regexp", feature = "alloc"))) +)] +macro_rules! re_bytes_captures ( + ($i:expr, $re:expr) => ( { + let r = $crate::lib::regex::bytes::Regex::new($re).unwrap(); + $crate::regexp::bytes::re_captures(r)($i) + } ) +); + +#[cfg(test)] +mod tests { + use crate::error::ErrorKind; + use crate::internal::Err; + #[cfg(feature = "alloc")] + use crate::lib::std::vec::Vec; + + #[test] + fn re_match() { + named!(rm<&str,&str>, re_match!(r"^\d{4}-\d{2}-\d{2}")); + assert_eq!(rm("2015-09-07"), Ok(("", "2015-09-07"))); + assert_eq!( + rm("blah"), + Err(Err::Error(error_position!( + &"blah"[..], + ErrorKind::RegexpMatch + ),)) + ); + assert_eq!(rm("2015-09-07blah"), Ok(("", "2015-09-07blah"))); + } + + #[test] + fn re_find() { + named!(rm<&str,&str>, re_find!(r"^\d{4}-\d{2}-\d{2}")); + assert_eq!(rm("2015-09-07"), Ok(("", "2015-09-07"))); + assert_eq!( + rm("blah"), + Err(Err::Error(error_position!( + &"blah"[..], + ErrorKind::RegexpFind + ),)) + ); + assert_eq!(rm("2015-09-07blah"), Ok(("blah", "2015-09-07"))); + } + + #[cfg(feature = "alloc")] + #[test] + fn re_matches() { + named!(rm< &str,Vec<&str> >, re_matches!(r"\d{4}-\d{2}-\d{2}")); + assert_eq!(rm("2015-09-07"), Ok(("", vec!["2015-09-07"]))); + assert_eq!( + rm("blah"), + Err(Err::Error(error_position!( + &"blah"[..], + ErrorKind::RegexpMatches + ))) + ); + assert_eq!( + rm("aaa2015-09-07blah2015-09-09pouet"), + Ok(("pouet", vec!["2015-09-07", "2015-09-09"])) + ); + } + + #[cfg(feature = "alloc")] + #[test] + fn re_capture() { + named!(rm< &str,Vec<&str> >, re_capture!(r"([[:alpha:]]+)\s+((\d+).(\d+).(\d+))")); + assert_eq!( + rm("blah nom 0.3.11pouet"), + Ok(("pouet", vec!["nom 0.3.11", "nom", "0.3.11", "0", "3", "11"])) + ); + assert_eq!( + rm("blah"), + Err(Err::Error(error_position!( + &"blah"[..], + ErrorKind::RegexpCapture + ))) + ); + assert_eq!( + rm("hello nom 0.3.11 world regex 0.1.41"), + Ok(( + " world regex 0.1.41", + vec!["nom 0.3.11", "nom", "0.3.11", "0", "3", "11"] + )) + ); + } + + #[cfg(feature = "alloc")] + #[test] + fn re_captures() { + named!(rm< &str,Vec<Vec<&str>> >, re_captures!(r"([[:alpha:]]+)\s+((\d+).(\d+).(\d+))")); + assert_eq!( + rm("blah nom 0.3.11pouet"), + Ok(( + "pouet", + vec![vec!["nom 0.3.11", "nom", "0.3.11", "0", "3", "11"]] + )) + ); + assert_eq!( + rm("blah"), + Err(Err::Error(error_position!( + &"blah"[..], + ErrorKind::RegexpCapture + ))) + ); + assert_eq!( + rm("hello nom 0.3.11 world regex 0.1.41 aaa"), + Ok(( + " aaa", + vec![ + vec!["nom 0.3.11", "nom", "0.3.11", "0", "3", "11"], + vec!["regex 0.1.41", "regex", "0.1.41", "0", "1", "41"], + ] + )) + ); + } + + #[test] + fn re_bytes_match() { + named!(rm, re_bytes_match!(r"^\d{4}-\d{2}-\d{2}")); + assert_eq!(rm(&b"2015-09-07"[..]), Ok((&b""[..], &b"2015-09-07"[..]))); + assert_eq!( + rm(&b"blah"[..]), + Err(Err::Error(error_position!( + &b"blah"[..], + ErrorKind::RegexpMatch + ))) + ); + assert_eq!( + rm(&b"2015-09-07blah"[..]), + Ok((&b""[..], &b"2015-09-07blah"[..])) + ); + } + + #[test] + fn re_bytes_find() { + named!(rm, re_bytes_find!(r"^\d{4}-\d{2}-\d{2}")); + assert_eq!(rm(&b"2015-09-07"[..]), Ok((&b""[..], &b"2015-09-07"[..]))); + assert_eq!( + rm(&b"blah"[..]), + Err(Err::Error(error_position!( + &b"blah"[..], + ErrorKind::RegexpFind + ))) + ); + assert_eq!( + rm(&b"2015-09-07blah"[..]), + Ok((&b"blah"[..], &b"2015-09-07"[..])) + ); + } + + #[cfg(feature = "alloc")] + #[test] + fn re_bytes_matches() { + named!(rm<Vec<&[u8]>>, re_bytes_matches!(r"\d{4}-\d{2}-\d{2}")); + assert_eq!( + rm(&b"2015-09-07"[..]), + Ok((&b""[..], vec![&b"2015-09-07"[..]])) + ); + assert_eq!( + rm(&b"blah"[..]), + Err(Err::Error(error_position!( + &b"blah"[..], + ErrorKind::RegexpMatches + ))) + ); + assert_eq!( + rm(&b"aaa2015-09-07blah2015-09-09pouet"[..]), + Ok((&b"pouet"[..], vec![&b"2015-09-07"[..], &b"2015-09-09"[..]])) + ); + } + + #[cfg(feature = "alloc")] + #[test] + fn re_bytes_capture() { + named!( + rm<Vec<&[u8]>>, + re_bytes_capture!(r"([[:alpha:]]+)\s+((\d+).(\d+).(\d+))") + ); + assert_eq!( + rm(&b"blah nom 0.3.11pouet"[..]), + Ok(( + &b"pouet"[..], + vec![ + &b"nom 0.3.11"[..], + &b"nom"[..], + &b"0.3.11"[..], + &b"0"[..], + &b"3"[..], + &b"11"[..], + ] + )) + ); + assert_eq!( + rm(&b"blah"[..]), + Err(Err::Error(error_position!( + &b"blah"[..], + ErrorKind::RegexpCapture + ))) + ); + assert_eq!( + rm(&b"hello nom 0.3.11 world regex 0.1.41"[..]), + Ok(( + &b" world regex 0.1.41"[..], + vec![ + &b"nom 0.3.11"[..], + &b"nom"[..], + &b"0.3.11"[..], + &b"0"[..], + &b"3"[..], + &b"11"[..], + ] + )) + ); + } + + #[cfg(feature = "alloc")] + #[test] + fn re_bytes_captures() { + named!( + rm<Vec<Vec<&[u8]>>>, + re_bytes_captures!(r"([[:alpha:]]+)\s+((\d+).(\d+).(\d+))") + ); + assert_eq!( + rm(&b"blah nom 0.3.11pouet"[..]), + Ok(( + &b"pouet"[..], + vec![vec![ + &b"nom 0.3.11"[..], + &b"nom"[..], + &b"0.3.11"[..], + &b"0"[..], + &b"3"[..], + &b"11"[..], + ],] + )) + ); + assert_eq!( + rm(&b"blah"[..]), + Err(Err::Error(error_position!( + &b"blah"[..], + ErrorKind::RegexpCapture + ))) + ); + assert_eq!( + rm(&b"hello nom 0.3.11 world regex 0.1.41 aaa"[..]), + Ok(( + &b" aaa"[..], + vec![ + vec![ + &b"nom 0.3.11"[..], + &b"nom"[..], + &b"0.3.11"[..], + &b"0"[..], + &b"3"[..], + &b"11"[..], + ], + vec![ + &b"regex 0.1.41"[..], + &b"regex"[..], + &b"0.1.41"[..], + &b"0"[..], + &b"1"[..], + &b"41"[..], + ], + ] + )) + ); + } +} diff --git a/src/regexp/mod.rs b/src/regexp/mod.rs new file mode 100644 index 0000000..b1b856c --- /dev/null +++ b/src/regexp/mod.rs @@ -0,0 +1,675 @@ +//! Parser combinators that use regular expressions. + +mod macros; + +///Regular expression parser combinators for strings. +pub mod str { + use crate::error::{ErrorKind, ParseError}; + use crate::lib::regex::Regex; + #[cfg(feature = "alloc")] + use crate::lib::std::vec::Vec; + use crate::traits::{InputLength, Slice}; + use crate::{Err, IResult}; + + /// Compares the input with a regular expression and returns the + /// whole input if a match is found. + /// + /// Requires the `regexp` feature. + /// # Example + /// + /// ``` + /// # use nom::{Err, error::ErrorKind}; + /// # use nom::regexp::str::re_match; + /// # fn main() { + /// let re = regex::Regex::new(r"^\d{4}").unwrap(); + /// let parser = re_match::<(&str, ErrorKind)>(re); + /// assert_eq!(parser("2019"), Ok(("", "2019"))); + /// assert_eq!(parser("abc"), Err(Err::Error(("abc", ErrorKind::RegexpMatch)))); + /// assert_eq!(parser("2019-10"), Ok(("", "2019-10"))); + /// # } + /// ``` + #[cfg(feature = "regexp")] + #[cfg_attr(feature = "docsrs", doc(cfg(feature = "regexp")))] + pub fn re_match<'a, E>(re: Regex) -> impl Fn(&'a str) -> IResult<&'a str, &'a str, E> + where + E: ParseError<&'a str>, + { + move |i| { + if re.is_match(i) { + Ok((i.slice(i.input_len()..), i)) + } else { + Err(Err::Error(E::from_error_kind(i, ErrorKind::RegexpMatch))) + } + } + } + + /// Compares the input with a regular expression and returns all matches in a `Vec`. + /// + /// Requires the `regexp` feature. + /// # Example + /// + /// ``` + /// # use nom::{Err, error::ErrorKind}; + /// # use nom::regexp::str::re_matches; + /// # fn main() { + /// let re = regex::Regex::new(r"a\d").unwrap(); + /// let parser = re_matches::<(&str, ErrorKind)>(re); + /// assert_eq!(parser("a1ba2"), Ok(("", vec!["a1", "a2"]))); + /// assert_eq!(parser("abc"), Err(Err::Error(("abc", ErrorKind::RegexpMatches)))); + /// # } + /// ``` + #[cfg(all(feature = "regexp", feature = "alloc"))] + #[cfg_attr( + feature = "docsrs", + doc(cfg(all(feature = "regexp", feature = "alloc"))) + )] + pub fn re_matches<'a, E>(re: Regex) -> impl Fn(&'a str) -> IResult<&'a str, Vec<&'a str>, E> + where + E: ParseError<&'a str>, + { + move |i| { + let v: Vec<_> = re + .find_iter(i) + .map(|m| i.slice(m.start()..m.end())) + .collect(); + if !v.is_empty() { + let offset = { + let end = v.last().unwrap(); + end.as_ptr() as usize + end.len() - i.as_ptr() as usize + }; + Ok((i.slice(offset..), v)) + } else { + Err(Err::Error(E::from_error_kind(i, ErrorKind::RegexpMatches))) + } + } + } + + /// Compares the input with a regular expression and returns the + /// first match. + /// + /// Requires the `regexp` feature. + /// # Example + /// + /// ``` + /// # use nom::{Err, error::ErrorKind}; + /// # use nom::regexp::str::re_find; + /// # fn main() { + /// let re = regex::Regex::new(r"\d{4}").unwrap(); + /// let parser = re_find::<(&str, ErrorKind)>(re); + /// assert_eq!(parser("abc2019"), Ok(("", "2019"))); + /// assert_eq!(parser("abc"), Err(Err::Error(("abc", ErrorKind::RegexpFind)))); + /// assert_eq!(parser("2019-10"), Ok(("-10", "2019"))); + /// # } + /// ``` + #[cfg(feature = "regexp")] + #[cfg_attr(feature = "docsrs", doc(cfg(feature = "regexp")))] + pub fn re_find<'a, E>(re: Regex) -> impl Fn(&'a str) -> IResult<&'a str, &'a str, E> + where + E: ParseError<&'a str>, + { + move |i| { + if let Some(m) = re.find(i) { + Ok((i.slice(m.end()..), i.slice(m.start()..m.end()))) + } else { + Err(Err::Error(E::from_error_kind(i, ErrorKind::RegexpFind))) + } + } + } + + /// Compares the input with a regular expression and returns + /// the capture groups of the first match in a `Vec`. + /// + /// Requires the `regexp` feature. + /// # Example + /// + /// ``` + /// # use nom::{Err, error::ErrorKind}; + /// # use nom::regexp::str::re_capture; + /// # fn main() { + /// let re = regex::Regex::new(r"(a)(\d)").unwrap(); + /// let parser = re_capture::<(&str, ErrorKind)>(re); + /// assert_eq!(parser("a1ba2"), Ok(("ba2", vec!["a1", "a", "1"]))); + /// assert_eq!(parser("abc"), Err(Err::Error(("abc", ErrorKind::RegexpCapture)))); + /// # } + /// ``` + #[cfg(all(feature = "regexp", feature = "alloc"))] + #[cfg_attr( + feature = "docsrs", + doc(cfg(all(feature = "regexp", feature = "alloc"))) + )] + pub fn re_capture<'a, E>(re: Regex) -> impl Fn(&'a str) -> IResult<&'a str, Vec<&'a str>, E> + where + E: ParseError<&'a str>, + { + move |i| { + if let Some(c) = re.captures(i) { + let v: Vec<_> = c + .iter() + .filter(|el| el.is_some()) + .map(|el| el.unwrap()) + .map(|m| i.slice(m.start()..m.end())) + .collect(); + let offset = { + let end = v.last().unwrap(); + end.as_ptr() as usize + end.len() - i.as_ptr() as usize + }; + Ok((i.slice(offset..), v)) + } else { + Err(Err::Error(E::from_error_kind(i, ErrorKind::RegexpCapture))) + } + } + } + + /// Compares the input with a regular expression and returns + /// the capture groups of all matches in a nested `Vec`. + /// + /// Requires the `regexp` feature. + /// # Example + /// + /// ``` + /// # use nom::{Err, error::ErrorKind}; + /// # use nom::regexp::str::re_captures; + /// # fn main() { + /// let re = regex::Regex::new(r"(a)(\d)").unwrap(); + /// let parser = re_captures::<(&str, ErrorKind)>(re); + /// assert_eq!(parser("a1ba2"), Ok(("", vec![vec!["a1", "a", "1"], vec!["a2", "a", "2"]]))); + /// assert_eq!(parser("abc"), Err(Err::Error(("abc", ErrorKind::RegexpCapture)))); + /// # } + /// ``` + #[cfg(all(feature = "regexp", feature = "alloc"))] + #[cfg_attr( + feature = "docsrs", + doc(cfg(all(feature = "regexp", feature = "alloc"))) + )] + pub fn re_captures<'a, E>(re: Regex) -> impl Fn(&'a str) -> IResult<&'a str, Vec<Vec<&'a str>>, E> + where + E: ParseError<&'a str>, + { + move |i| { + let v: Vec<Vec<_>> = re + .captures_iter(i) + .map(|c| { + c.iter() + .filter(|el| el.is_some()) + .map(|el| el.unwrap()) + .map(|m| i.slice(m.start()..m.end())) + .collect() + }) + .collect(); + if !v.is_empty() { + let offset = { + let end = v.last().unwrap().last().unwrap(); + end.as_ptr() as usize + end.len() - i.as_ptr() as usize + }; + Ok((i.slice(offset..), v)) + } else { + Err(Err::Error(E::from_error_kind(i, ErrorKind::RegexpCapture))) + } + } + } + + #[cfg(test)] + mod tests { + use super::*; + use crate::error::ErrorKind; + use crate::internal::Err; + use crate::lib::regex::Regex; + + macro_rules! assert_parse( + ($left: expr, $right: expr) => { + let res: $crate::IResult<_, _, (_, ErrorKind)> = $left; + assert_eq!(res, $right); + }; + ); + + #[test] + fn re_match_str() { + let re = Regex::new(r"^\d{4}-\d{2}-\d{2}").unwrap(); + let rm = re_match(re); + assert_parse!(rm("2015-09-07"), Ok(("", "2015-09-07"))); + assert_eq!( + rm("blah"), + Err(Err::Error((&"blah"[..], ErrorKind::RegexpMatch))) + ); + assert_eq!(rm("2015-09-07blah"), Ok(("", "2015-09-07blah"))); + } + + #[test] + fn re_find_str() { + let re = Regex::new(r"^\d{4}-\d{2}-\d{2}").unwrap(); + let rm = re_find(re); + assert_parse!(rm("2015-09-07"), Ok(("", "2015-09-07"))); + assert_eq!( + rm("blah"), + Err(Err::Error((&"blah"[..], ErrorKind::RegexpFind))) + ); + assert_eq!(rm("2015-09-07blah"), Ok(("blah", "2015-09-07"))); + } + + #[cfg(feature = "alloc")] + #[test] + fn re_matches_str() { + let re = Regex::new(r"\d{4}-\d{2}-\d{2}").unwrap(); + let rm = re_matches(re); + assert_parse!(rm("2015-09-07"), Ok(("", vec!["2015-09-07"]))); + assert_eq!( + rm("blah"), + Err(Err::Error((&"blah"[..], ErrorKind::RegexpMatches))) + ); + assert_eq!( + rm("aaa2015-09-07blah2015-09-09pouet"), + Ok(("pouet", vec!["2015-09-07", "2015-09-09"])) + ); + } + + #[cfg(feature = "alloc")] + #[test] + fn re_capture_str() { + let re = Regex::new(r"([[:alpha:]]+)\s+((\d+).(\d+).(\d+))").unwrap(); + let rm = re_capture(re); + assert_parse!( + rm("blah nom 0.3.11pouet"), + Ok(("pouet", vec!["nom 0.3.11", "nom", "0.3.11", "0", "3", "11"])) + ); + assert_eq!( + rm("blah"), + Err(Err::Error(("blah", ErrorKind::RegexpCapture))) + ); + assert_eq!( + rm("hello nom 0.3.11 world regex 0.1.41"), + Ok(( + " world regex 0.1.41", + vec!["nom 0.3.11", "nom", "0.3.11", "0", "3", "11"] + )) + ); + } + + #[cfg(feature = "alloc")] + #[test] + fn re_captures_str() { + let re = Regex::new(r"([[:alpha:]]+)\s+((\d+).(\d+).(\d+))").unwrap(); + let rm = re_captures(re); + assert_parse!( + rm("blah nom 0.3.11pouet"), + Ok(( + "pouet", + vec![vec!["nom 0.3.11", "nom", "0.3.11", "0", "3", "11"]] + )) + ); + assert_eq!( + rm("blah"), + Err(Err::Error((&"blah"[..], ErrorKind::RegexpCapture))) + ); + assert_eq!( + rm("hello nom 0.3.11 world regex 0.1.41 aaa"), + Ok(( + " aaa", + vec![ + vec!["nom 0.3.11", "nom", "0.3.11", "0", "3", "11"], + vec!["regex 0.1.41", "regex", "0.1.41", "0", "1", "41"], + ] + )) + ); + } + } +} + +///Regular expression parser combinators for bytes. +pub mod bytes { + use crate::error::{ErrorKind, ParseError}; + use crate::lib::regex::bytes::Regex; + #[cfg(feature = "alloc")] + use crate::lib::std::vec::Vec; + use crate::traits::{InputLength, Slice}; + use crate::{Err, IResult}; + + /// Compares the input with a regular expression and returns the + /// whole input if a match is found. + /// + /// Requires the `regexp` feature. + /// # Example + /// + /// ``` + /// # use nom::{Err, error::ErrorKind}; + /// # use nom::regexp::bytes::re_match; + /// # fn main() { + /// let re = regex::bytes::Regex::new(r"^\d{4}").unwrap(); + /// let parser = re_match::<(&[u8], ErrorKind)>(re); + /// assert_eq!(parser(&b"2019"[..]), Ok((&b""[..], &b"2019"[..]))); + /// assert_eq!(parser(&b"abc"[..]), Err(Err::Error((&b"abc"[..], ErrorKind::RegexpMatch)))); + /// assert_eq!(parser(&b"2019-10"[..]), Ok((&b""[..], &b"2019-10"[..]))); + /// # } + /// ``` + #[cfg(feature = "regexp")] + pub fn re_match<'a, E>(re: Regex) -> impl Fn(&'a [u8]) -> IResult<&'a [u8], &'a [u8], E> + where + E: ParseError<&'a [u8]>, + { + move |i| { + if re.is_match(i) { + Ok((i.slice(i.input_len()..), i)) + } else { + Err(Err::Error(E::from_error_kind(i, ErrorKind::RegexpMatch))) + } + } + } + + /// Compares the input with a regular expression and returns all matches in a `Vec`. + /// + /// Requires the `regexp` feature. + /// # Example + /// + /// ``` + /// # use nom::{Err, error::ErrorKind}; + /// # use nom::regexp::bytes::re_matches; + /// # fn main() { + /// let re = regex::bytes::Regex::new(r"a\d").unwrap(); + /// let parser = re_matches::<(&[u8], ErrorKind)>(re); + /// assert_eq!(parser(&b"a1ba2"[..]), Ok((&b""[..], vec![&b"a1"[..], &b"a2"[..]]))); + /// assert_eq!(parser(&b"abc"[..]), Err(Err::Error((&b"abc"[..], ErrorKind::RegexpMatches)))); + /// # } + /// ``` + #[cfg(all(feature = "regexp", feature = "alloc"))] + #[cfg_attr( + feature = "docsrs", + doc(cfg(all(feature = "regexp", feature = "alloc"))) + )] + pub fn re_matches<'a, E>(re: Regex) -> impl Fn(&'a [u8]) -> IResult<&'a [u8], Vec<&'a [u8]>, E> + where + E: ParseError<&'a [u8]>, + { + move |i| { + let v: Vec<_> = re + .find_iter(i) + .map(|m| i.slice(m.start()..m.end())) + .collect(); + if !v.is_empty() { + let offset = { + let end = v.last().unwrap(); + end.as_ptr() as usize + end.len() - i.as_ptr() as usize + }; + Ok((i.slice(offset..), v)) + } else { + Err(Err::Error(E::from_error_kind(i, ErrorKind::RegexpMatches))) + } + } + } + + /// Compares the input with a regular expression and returns the + /// first match. + /// + /// Requires the `regexp` feature. + /// # Example + /// + /// ``` + /// # use nom::{Err, error::ErrorKind}; + /// # use nom::regexp::bytes::re_find; + /// # fn main() { + /// let re = regex::bytes::Regex::new(r"\d{4}").unwrap(); + /// let parser = re_find::<(&[u8], ErrorKind)>(re); + /// assert_eq!(parser(&b"abc2019"[..]), Ok((&b""[..], &b"2019"[..]))); + /// assert_eq!(parser(&b"abc"[..]), Err(Err::Error((&b"abc"[..], ErrorKind::RegexpFind)))); + /// assert_eq!(parser(&b"2019-10"[..]), Ok((&b"-10"[..], &b"2019"[..]))); + /// # } + /// ``` + #[cfg(feature = "regexp")] + pub fn re_find<'a, E>(re: Regex) -> impl Fn(&'a [u8]) -> IResult<&'a [u8], &'a [u8], E> + where + E: ParseError<&'a [u8]>, + { + move |i| { + if let Some(m) = re.find(i) { + Ok((i.slice(m.end()..), i.slice(m.start()..m.end()))) + } else { + Err(Err::Error(E::from_error_kind(i, ErrorKind::RegexpFind))) + } + } + } + + /// Compares the input with a regular expression and returns + /// the capture groups of the first match in a `Vec`. + /// + /// Requires the `regexp` feature. + /// # Example + /// + /// ``` + /// # use nom::{Err, error::ErrorKind}; + /// # use nom::regexp::bytes::re_capture; + /// # fn main() { + /// let re = regex::bytes::Regex::new(r"(a)(\d)").unwrap(); + /// let parser = re_capture::<(&[u8], ErrorKind)>(re); + /// assert_eq!(parser(&b"a1ba2"[..]), Ok((&b"ba2"[..], vec![&b"a1"[..], &b"a"[..], &b"1"[..]]))); + /// assert_eq!(parser(&b"abc"[..]), Err(Err::Error((&b"abc"[..], ErrorKind::RegexpCapture)))); + /// # } + /// ``` + #[cfg(all(feature = "regexp", feature = "alloc"))] + #[cfg_attr( + feature = "docsrs", + doc(cfg(all(feature = "regexp", feature = "alloc"))) + )] + pub fn re_capture<'a, E>(re: Regex) -> impl Fn(&'a [u8]) -> IResult<&'a [u8], Vec<&'a [u8]>, E> + where + E: ParseError<&'a [u8]>, + { + move |i| { + if let Some(c) = re.captures(i) { + let v: Vec<_> = c + .iter() + .filter(|el| el.is_some()) + .map(|el| el.unwrap()) + .map(|m| i.slice(m.start()..m.end())) + .collect(); + let offset = { + let end = v.last().unwrap(); + end.as_ptr() as usize + end.len() - i.as_ptr() as usize + }; + Ok((i.slice(offset..), v)) + } else { + Err(Err::Error(E::from_error_kind(i, ErrorKind::RegexpCapture))) + } + } + } + + /// Compares the input with a regular expression and returns + /// the capture groups of all matches in a nested `Vec`. + /// + /// Requires the `regexp` feature. + /// # Example + /// + /// ``` + /// # use nom::{Err, error::ErrorKind}; + /// # use nom::regexp::bytes::re_captures; + /// # fn main() { + /// let re = regex::bytes::Regex::new(r"(a)(\d)").unwrap(); + /// let parser = re_captures::<(&[u8], ErrorKind)>(re); + /// assert_eq!(parser(&b"a1ba2"[..]), Ok((&b""[..], vec![vec![&b"a1"[..], &b"a"[..], &b"1"[..]], vec![&b"a2"[..], &b"a"[..], &b"2"[..]]]))); + /// assert_eq!(parser(&b"abc"[..]), Err(Err::Error((&b"abc"[..], ErrorKind::RegexpCapture)))); + /// # } + /// ``` + #[cfg(all(feature = "regexp", feature = "alloc"))] + #[cfg_attr( + feature = "docsrs", + doc(cfg(all(feature = "regexp", feature = "alloc"))) + )] + pub fn re_captures<'a, E>( + re: Regex, + ) -> impl Fn(&'a [u8]) -> IResult<&'a [u8], Vec<Vec<&'a [u8]>>, E> + where + E: ParseError<&'a [u8]>, + { + move |i| { + let v: Vec<Vec<_>> = re + .captures_iter(i) + .map(|c| { + c.iter() + .filter(|el| el.is_some()) + .map(|el| el.unwrap()) + .map(|m| i.slice(m.start()..m.end())) + .collect() + }) + .collect(); + if !v.is_empty() { + let offset = { + let end = v.last().unwrap().last().unwrap(); + end.as_ptr() as usize + end.len() - i.as_ptr() as usize + }; + Ok((i.slice(offset..), v)) + } else { + Err(Err::Error(E::from_error_kind(i, ErrorKind::RegexpCapture))) + } + } + } + + #[cfg(test)] + mod tests { + use super::*; + use crate::error::ErrorKind; + use crate::internal::Err; + use crate::lib::regex::bytes::Regex; + + macro_rules! assert_parse( + ($left: expr, $right: expr) => { + let res: $crate::IResult<_, _, (_, ErrorKind)> = $left; + assert_eq!(res, $right); + }; + ); + + #[test] + fn re_match_bytes() { + let re = Regex::new(r"^\d{4}-\d{2}-\d{2}").unwrap(); + let rm = re_match(re); + assert_parse!(rm(&b"2015-09-07"[..]), Ok((&b""[..], &b"2015-09-07"[..]))); + assert_eq!( + rm(&b"blah"[..]), + Err(Err::Error((&b"blah"[..], ErrorKind::RegexpMatch))) + ); + assert_eq!( + rm(&b"2015-09-07blah"[..]), + Ok((&b""[..], &b"2015-09-07blah"[..])) + ); + } + + #[test] + fn re_find_bytes() { + let re = Regex::new(r"^\d{4}-\d{2}-\d{2}").unwrap(); + let rm = re_find(re); + assert_parse!(rm(&b"2015-09-07"[..]), Ok((&b""[..], &b"2015-09-07"[..]))); + assert_eq!( + rm(&b"blah"[..]), + Err(Err::Error((&b"blah"[..], ErrorKind::RegexpFind))) + ); + assert_eq!( + rm(&b"2015-09-07blah"[..]), + Ok((&b"blah"[..], &b"2015-09-07"[..])) + ); + } + + #[cfg(feature = "alloc")] + #[test] + fn re_matches_bytes() { + let re = Regex::new(r"\d{4}-\d{2}-\d{2}").unwrap(); + let rm = re_matches(re); + assert_parse!( + rm(&b"2015-09-07"[..]), + Ok((&b""[..], vec![&b"2015-09-07"[..]])) + ); + assert_eq!( + rm(&b"blah"[..]), + Err(Err::Error((&b"blah"[..], ErrorKind::RegexpMatches))) + ); + assert_eq!( + rm(&b"aaa2015-09-07blah2015-09-09pouet"[..]), + Ok((&b"pouet"[..], vec![&b"2015-09-07"[..], &b"2015-09-09"[..]])) + ); + } + + #[cfg(feature = "alloc")] + #[test] + fn re_capture_bytes() { + let re = Regex::new(r"([[:alpha:]]+)\s+((\d+).(\d+).(\d+))").unwrap(); + let rm = re_capture(re); + assert_parse!( + rm(&b"blah nom 0.3.11pouet"[..]), + Ok(( + &b"pouet"[..], + vec![ + &b"nom 0.3.11"[..], + &b"nom"[..], + &b"0.3.11"[..], + &b"0"[..], + &b"3"[..], + &b"11"[..] + ] + )) + ); + assert_eq!( + rm(&b"blah"[..]), + Err(Err::Error((&b"blah"[..], ErrorKind::RegexpCapture))) + ); + assert_eq!( + rm(&b"hello nom 0.3.11 world regex 0.1.41"[..]), + Ok(( + &b" world regex 0.1.41"[..], + vec![ + &b"nom 0.3.11"[..], + &b"nom"[..], + &b"0.3.11"[..], + &b"0"[..], + &b"3"[..], + &b"11"[..] + ] + )) + ); + } + + #[cfg(feature = "alloc")] + #[test] + fn re_captures_bytes() { + let re = Regex::new(r"([[:alpha:]]+)\s+((\d+).(\d+).(\d+))").unwrap(); + let rm = re_captures(re); + assert_parse!( + rm(&b"blah nom 0.3.11pouet"[..]), + Ok(( + &b"pouet"[..], + vec![vec![ + &b"nom 0.3.11"[..], + &b"nom"[..], + &b"0.3.11"[..], + &b"0"[..], + &b"3"[..], + &b"11"[..] + ]] + )) + ); + + assert_eq!( + rm(&b"blah"[..]), + Err(Err::Error((&b"blah"[..], ErrorKind::RegexpCapture))) + ); + assert_eq!( + rm(&b"hello nom 0.3.11 world regex 0.1.41 aaa"[..]), + Ok(( + &b" aaa"[..], + vec![ + vec![ + &b"nom 0.3.11"[..], + &b"nom"[..], + &b"0.3.11"[..], + &b"0"[..], + &b"3"[..], + &b"11"[..] + ], + vec![ + &b"regex 0.1.41"[..], + &b"regex"[..], + &b"0.1.41"[..], + &b"0"[..], + &b"1"[..], + &b"41"[..] + ], + ] + )) + ); + } + } +} diff --git a/src/sequence/macros.rs b/src/sequence/macros.rs index a5a3351..a2bfb2c 100644 --- a/src/sequence/macros.rs +++ b/src/sequence/macros.rs @@ -4,7 +4,7 @@ /// The input type `I` must implement `nom::InputLength`. /// /// This combinator will count how much data is consumed by every child parser -/// and take it into account if there is not enough data +/// and take it into account if there is not enough data. /// /// ``` /// # #[macro_use] extern crate nom; @@ -38,7 +38,7 @@ macro_rules! tuple ( ); ); -/// Internal parser, do not use directly +/// Internal parser, do not use directly. #[doc(hidden)] #[macro_export(local_inner_macros)] macro_rules! tuple_parser ( @@ -88,19 +88,19 @@ macro_rules! tuple_parser ( ); /// `pair!(I -> IResult<I,O>, I -> IResult<I,P>) => I -> IResult<I, (O,P)>` -/// pair returns a tuple of the results of its two child parsers of both succeed +/// `pair` returns a tuple of the results of its two child parsers if both succeed. /// /// ``` /// # #[macro_use] extern crate nom; /// # use nom::Err; -/// # use nom::error::ErrorKind; +/// # use nom::error::{Error, ErrorKind}; /// # use nom::character::complete::{alpha1, digit1}; /// named!(parser<&str, (&str, &str)>, pair!(alpha1, digit1)); /// /// # fn main() { /// assert_eq!(parser("abc123"), Ok(("", ("abc", "123")))); -/// assert_eq!(parser("123abc"), Err(Err::Error(("123abc", ErrorKind::Alpha)))); -/// assert_eq!(parser("abc;123"), Err(Err::Error((";123", ErrorKind::Digit)))); +/// assert_eq!(parser("123abc"), Err(Err::Error(Error::new("123abc", ErrorKind::Alpha)))); +/// assert_eq!(parser("abc;123"), Err(Err::Error(Error::new(";123", ErrorKind::Digit)))); /// # } /// ``` #[macro_export(local_inner_macros)] @@ -123,20 +123,20 @@ macro_rules! pair( ); /// `separated_pair!(I -> IResult<I,O>, I -> IResult<I, T>, I -> IResult<I,P>) => I -> IResult<I, (O,P)>` -/// separated_pair(X,sep,Y) returns a tuple of its first and third child parsers -/// if all 3 succeed +/// `separated_pair(X,sep,Y)` returns a tuple of its first and third child parsers +/// if all 3 succeed. /// /// ``` /// # #[macro_use] extern crate nom; /// # use nom::Err; -/// # use nom::error::ErrorKind; +/// # use nom::error::{Error, ErrorKind}; /// # use nom::character::complete::{alpha1, digit1}; /// named!(parser<&str, (&str, &str)>, separated_pair!(alpha1, char!(','), digit1)); /// /// # fn main() { /// assert_eq!(parser("abc,123"), Ok(("", ("abc", "123")))); -/// assert_eq!(parser("123,abc"), Err(Err::Error(("123,abc", ErrorKind::Alpha)))); -/// assert_eq!(parser("abc;123"), Err(Err::Error((";123", ErrorKind::Char)))); +/// assert_eq!(parser("123,abc"), Err(Err::Error(Error::new("123,abc", ErrorKind::Alpha)))); +/// assert_eq!(parser("abc;123"), Err(Err::Error(Error::new(";123", ErrorKind::Char)))); /// # } /// ``` #[macro_export(local_inner_macros)] @@ -156,19 +156,19 @@ macro_rules! separated_pair( ); /// `preceded!(I -> IResult<I,T>, I -> IResult<I,O>) => I -> IResult<I, O>` -/// preceded returns the result of its second parser if both succeed +/// `preceded` returns the result of its second parser if both succeed. /// /// ``` /// # #[macro_use] extern crate nom; /// # use nom::Err; -/// # use nom::error::ErrorKind; +/// # use nom::error::{Error, ErrorKind}; /// # use nom::character::complete::{alpha1}; /// named!(parser<&str, &str>, preceded!(char!('-'), alpha1)); /// /// # fn main() { /// assert_eq!(parser("-abc"), Ok(("", "abc"))); -/// assert_eq!(parser("abc"), Err(Err::Error(("abc", ErrorKind::Char)))); -/// assert_eq!(parser("-123"), Err(Err::Error(("123", ErrorKind::Alpha)))); +/// assert_eq!(parser("abc"), Err(Err::Error(Error::new("abc", ErrorKind::Char)))); +/// assert_eq!(parser("-123"), Err(Err::Error(Error::new("123", ErrorKind::Alpha)))); /// # } /// ``` #[macro_export(local_inner_macros)] @@ -191,19 +191,19 @@ macro_rules! preceded( ); /// `terminated!(I -> IResult<I,O>, I -> IResult<I,T>) => I -> IResult<I, O>` -/// terminated returns the result of its first parser if both succeed +/// `terminated` returns the result of its first parser if both succeed. /// /// ``` /// # #[macro_use] extern crate nom; /// # use nom::Err; -/// # use nom::error::ErrorKind; +/// # use nom::error::{Error, ErrorKind}; /// # use nom::character::complete::{alpha1}; /// named!(parser<&str, &str>, terminated!(alpha1, char!(';'))); /// /// # fn main() { /// assert_eq!(parser("abc;"), Ok(("", "abc"))); -/// assert_eq!(parser("abc,"), Err(Err::Error((",", ErrorKind::Char)))); -/// assert_eq!(parser("123;"), Err(Err::Error(("123;", ErrorKind::Alpha)))); +/// assert_eq!(parser("abc,"), Err(Err::Error(Error::new(",", ErrorKind::Char)))); +/// assert_eq!(parser("123;"), Err(Err::Error(Error::new("123;", ErrorKind::Alpha)))); /// # } /// ``` #[macro_export(local_inner_macros)] @@ -226,7 +226,7 @@ macro_rules! terminated( ); /// `delimited!(I -> IResult<I,T>, I -> IResult<I,O>, I -> IResult<I,U>) => I -> IResult<I, O>` -/// delimited(opening, X, closing) returns X +/// `delimited(opening, X, closing)` returns X. /// /// ``` /// # #[macro_use] extern crate nom; @@ -261,14 +261,14 @@ macro_rules! delimited( ); /// `do_parse!(I->IResult<I,A> >> I->IResult<I,B> >> ... I->IResult<I,X> , ( O ) ) => I -> IResult<I, O>` -/// do_parse applies sub parsers in a sequence. -/// it can store intermediary results and make them available -/// for later parsers +/// `do_parse` applies sub parsers in a sequence. +/// It can store intermediary results and make them available +/// for later parsers. /// /// The input type `I` must implement `nom::InputLength`. /// /// This combinator will count how much data is consumed by every child parser -/// and take it into account if there is not enough data +/// and take it into account if there is not enough data. /// /// ``` /// # #[macro_use] extern crate nom; @@ -306,7 +306,7 @@ macro_rules! delimited( /// // so the parser will tell you that you need 7 bytes: one for the tag, /// // one for the length, then 5 bytes /// let b: Vec<u8> = vec!(42, 5, 3, 4, 5); -/// assert_eq!(tag_length_value(&b[..]), Err(Err::Incomplete(Needed::Size(5)))); +/// assert_eq!(tag_length_value(&b[..]), Err(Err::Incomplete(Needed::new(2)))); /// # } /// ``` /// @@ -473,9 +473,9 @@ macro_rules! nom_compile_error ( #[cfg(test)] mod tests { + use crate::error::ErrorKind; use crate::internal::{Err, IResult, Needed}; use crate::number::streaming::be_u16; - use crate::error::ErrorKind; // reproduce the tag and take macros, because of module import order macro_rules! tag ( @@ -508,7 +508,7 @@ mod tests { let res: IResult<_,_,_> = if reduced != b { Err($crate::Err::Error(error_position!($i, $crate::error::ErrorKind::Tag))) } else if m < blen { - Err($crate::Err::Incomplete(Needed::Size(blen))) + Err($crate::Err::Incomplete(Needed::new(blen))) } else { Ok((&$i[blen..], reduced)) }; @@ -522,7 +522,7 @@ mod tests { { let cnt = $count as usize; let res:IResult<&[u8],&[u8],_> = if $i.len() < cnt { - Err($crate::Err::Incomplete(Needed::Size(cnt))) + Err($crate::Err::Incomplete(Needed::new(cnt))) } else { Ok((&$i[cnt..],&$i[0..cnt])) }; @@ -547,7 +547,7 @@ mod tests { use util::{add_error_pattern, error_to_list, print_error}; #[cfg(feature = "std")] - #[cfg_attr(rustfmt, rustfmt_skip)] + #[rustfmt::skip] fn error_to_string<P: Clone + PartialEq>(e: &Context<P, u32>) -> &'static str { let v: Vec<(P, ErrorKind<u32>)> = error_to_list(e); // do it this way if you can use slice patterns @@ -582,7 +582,7 @@ mod tests { //} */ - #[cfg_attr(rustfmt, rustfmt_skip)] + #[rustfmt::skip] #[allow(unused_variables)] #[test] fn add_err() { @@ -622,7 +622,7 @@ mod tests { assert_eq!(res_c, Ok((&b""[..], &b"mnop"[..]))); } - #[cfg_attr(rustfmt, rustfmt_skip)] + #[rustfmt::skip] #[test] fn complete() { named!(err_test, @@ -645,9 +645,18 @@ mod tests { named!(tag_def, tag!("def")); named!( pair_abc_def<&[u8],(&[u8], &[u8])>, pair!(tag_abc, tag_def) ); - assert_eq!(pair_abc_def(&b"abcdefghijkl"[..]), Ok((&b"ghijkl"[..], (&b"abc"[..], &b"def"[..])))); - assert_eq!(pair_abc_def(&b"ab"[..]), Err(Err::Incomplete(Needed::Size(3)))); - assert_eq!(pair_abc_def(&b"abcd"[..]), Err(Err::Incomplete(Needed::Size(3)))); + assert_eq!( + pair_abc_def(&b"abcdefghijkl"[..]), + Ok((&b"ghijkl"[..], (&b"abc"[..], &b"def"[..]))) + ); + assert_eq!( + pair_abc_def(&b"ab"[..]), + Err(Err::Incomplete(Needed::new(3))) + ); + assert_eq!( + pair_abc_def(&b"abcd"[..]), + Err(Err::Incomplete(Needed::new(3))) + ); assert_eq!( pair_abc_def(&b"xxx"[..]), Err(Err::Error(error_position!(&b"xxx"[..], ErrorKind::Tag))) @@ -673,8 +682,14 @@ mod tests { sep_pair_abc_def(&b"abc,defghijkl"[..]), Ok((&b"ghijkl"[..], (&b"abc"[..], &b"def"[..]))) ); - assert_eq!(sep_pair_abc_def(&b"ab"[..]), Err(Err::Incomplete(Needed::Size(3)))); - assert_eq!(sep_pair_abc_def(&b"abc,d"[..]), Err(Err::Incomplete(Needed::Size(3)))); + assert_eq!( + sep_pair_abc_def(&b"ab"[..]), + Err(Err::Incomplete(Needed::new(3))) + ); + assert_eq!( + sep_pair_abc_def(&b"abc,d"[..]), + Err(Err::Incomplete(Needed::new(3))) + ); assert_eq!( sep_pair_abc_def(&b"xxx"[..]), Err(Err::Error(error_position!(&b"xxx"[..], ErrorKind::Tag))) @@ -695,9 +710,18 @@ mod tests { named!(tag_efgh, tag!("efgh")); named!( preceded_abcd_efgh<&[u8], &[u8]>, preceded!(tag_abcd, tag_efgh) ); - assert_eq!(preceded_abcd_efgh(&b"abcdefghijkl"[..]), Ok((&b"ijkl"[..], &b"efgh"[..]))); - assert_eq!(preceded_abcd_efgh(&b"ab"[..]), Err(Err::Incomplete(Needed::Size(4)))); - assert_eq!(preceded_abcd_efgh(&b"abcde"[..]), Err(Err::Incomplete(Needed::Size(4)))); + assert_eq!( + preceded_abcd_efgh(&b"abcdefghijkl"[..]), + Ok((&b"ijkl"[..], &b"efgh"[..])) + ); + assert_eq!( + preceded_abcd_efgh(&b"ab"[..]), + Err(Err::Incomplete(Needed::new(4))) + ); + assert_eq!( + preceded_abcd_efgh(&b"abcde"[..]), + Err(Err::Incomplete(Needed::new(4))) + ); assert_eq!( preceded_abcd_efgh(&b"xxx"[..]), Err(Err::Error(error_position!(&b"xxx"[..], ErrorKind::Tag))) @@ -718,9 +742,18 @@ mod tests { named!(tag_efgh, tag!("efgh")); named!( terminated_abcd_efgh<&[u8], &[u8]>, terminated!(tag_abcd, tag_efgh) ); - assert_eq!(terminated_abcd_efgh(&b"abcdefghijkl"[..]), Ok((&b"ijkl"[..], &b"abcd"[..]))); - assert_eq!(terminated_abcd_efgh(&b"ab"[..]), Err(Err::Incomplete(Needed::Size(4)))); - assert_eq!(terminated_abcd_efgh(&b"abcde"[..]), Err(Err::Incomplete(Needed::Size(4)))); + assert_eq!( + terminated_abcd_efgh(&b"abcdefghijkl"[..]), + Ok((&b"ijkl"[..], &b"abcd"[..])) + ); + assert_eq!( + terminated_abcd_efgh(&b"ab"[..]), + Err(Err::Incomplete(Needed::new(4))) + ); + assert_eq!( + terminated_abcd_efgh(&b"abcde"[..]), + Err(Err::Incomplete(Needed::new(4))) + ); assert_eq!( terminated_abcd_efgh(&b"xxx"[..]), Err(Err::Error(error_position!(&b"xxx"[..], ErrorKind::Tag))) @@ -742,17 +775,32 @@ mod tests { named!(tag_ghi, tag!("ghi")); named!( delimited_abc_def_ghi<&[u8], &[u8]>, delimited!(tag_abc, tag_def, tag_ghi) ); - assert_eq!(delimited_abc_def_ghi(&b"abcdefghijkl"[..]), Ok((&b"jkl"[..], &b"def"[..]))); - assert_eq!(delimited_abc_def_ghi(&b"ab"[..]), Err(Err::Incomplete(Needed::Size(3)))); - assert_eq!(delimited_abc_def_ghi(&b"abcde"[..]), Err(Err::Incomplete(Needed::Size(3)))); - assert_eq!(delimited_abc_def_ghi(&b"abcdefgh"[..]), Err(Err::Incomplete(Needed::Size(3)))); + assert_eq!( + delimited_abc_def_ghi(&b"abcdefghijkl"[..]), + Ok((&b"jkl"[..], &b"def"[..])) + ); + assert_eq!( + delimited_abc_def_ghi(&b"ab"[..]), + Err(Err::Incomplete(Needed::new(3))) + ); + assert_eq!( + delimited_abc_def_ghi(&b"abcde"[..]), + Err(Err::Incomplete(Needed::new(3))) + ); + assert_eq!( + delimited_abc_def_ghi(&b"abcdefgh"[..]), + Err(Err::Incomplete(Needed::new(3))) + ); assert_eq!( delimited_abc_def_ghi(&b"xxx"[..]), Err(Err::Error(error_position!(&b"xxx"[..], ErrorKind::Tag))) ); assert_eq!( delimited_abc_def_ghi(&b"xxxdefghi"[..]), - Err(Err::Error(error_position!(&b"xxxdefghi"[..], ErrorKind::Tag),)) + Err(Err::Error(error_position!( + &b"xxxdefghi"[..], + ErrorKind::Tag + ),)) ); assert_eq!( delimited_abc_def_ghi(&b"abcxxxghi"[..]), @@ -769,9 +817,12 @@ mod tests { named!(tuple_3<&[u8], (u16, &[u8], &[u8]) >, tuple!( be_u16 , take!(3), tag!("fg") ) ); - assert_eq!(tuple_3(&b"abcdefgh"[..]), Ok((&b"h"[..], (0x6162u16, &b"cde"[..], &b"fg"[..])))); - assert_eq!(tuple_3(&b"abcd"[..]), Err(Err::Incomplete(Needed::Size(3)))); - assert_eq!(tuple_3(&b"abcde"[..]), Err(Err::Incomplete(Needed::Size(2)))); + assert_eq!( + tuple_3(&b"abcdefgh"[..]), + Ok((&b"h"[..], (0x6162u16, &b"cde"[..], &b"fg"[..]))) + ); + assert_eq!(tuple_3(&b"abcd"[..]), Err(Err::Incomplete(Needed::new(3)))); + assert_eq!(tuple_3(&b"abcde"[..]), Err(Err::Incomplete(Needed::new(2)))); assert_eq!( tuple_3(&b"abcdejk"[..]), Err(Err::Error(error_position!(&b"jk"[..], ErrorKind::Tag))) @@ -782,10 +833,10 @@ mod tests { fn do_parse() { fn ret_int1(i: &[u8]) -> IResult<&[u8], u8> { Ok((i, 1)) - }; + } fn ret_int2(i: &[u8]) -> IResult<&[u8], u8> { Ok((i, 2)) - }; + } //trace_macros!(true); named!(do_parser<&[u8], (u8, u8)>, @@ -807,13 +858,22 @@ mod tests { //trace_macros!(false); - assert_eq!(do_parser(&b"abcdabcdefghefghX"[..]), Ok((&b"X"[..], (1, 2)))); + assert_eq!( + do_parser(&b"abcdabcdefghefghX"[..]), + Ok((&b"X"[..], (1, 2))) + ); assert_eq!(do_parser(&b"abcdefghefghX"[..]), Ok((&b"X"[..], (1, 2)))); - assert_eq!(do_parser(&b"abcdab"[..]), Err(Err::Incomplete(Needed::Size(4)))); - assert_eq!(do_parser(&b"abcdefghef"[..]), Err(Err::Incomplete(Needed::Size(4)))); + assert_eq!( + do_parser(&b"abcdab"[..]), + Err(Err::Incomplete(Needed::new(4))) + ); + assert_eq!( + do_parser(&b"abcdefghef"[..]), + Err(Err::Incomplete(Needed::new(4))) + ); } - #[cfg_attr(rustfmt, rustfmt_skip)] + #[rustfmt::skip] #[test] fn do_parse_dependency() { use crate::number::streaming::be_u8; @@ -830,7 +890,7 @@ mod tests { let res_a = [3u8, 4]; assert_eq!(length_value(&a[..]), Ok((&a[3..], &res_a[..]))); let b = [5u8, 3, 4, 5]; - assert_eq!(length_value(&b[..]), Err(Err::Incomplete(Needed::Size(5)))); + assert_eq!(length_value(&b[..]), Err(Err::Incomplete(Needed::new(5)))); } /* diff --git a/src/sequence/mod.rs b/src/sequence/mod.rs index a0eafb9..3d76725 100644 --- a/src/sequence/mod.rs +++ b/src/sequence/mod.rs @@ -1,10 +1,10 @@ -//! combinators applying parsers in sequence +//! Combinators applying parsers in sequence #[macro_use] mod macros; -use crate::internal::IResult; use crate::error::ParseError; +use crate::internal::{IResult, Parser}; /// Gets an object from the first parser, /// then gets another object from the second parser. @@ -18,27 +18,34 @@ use crate::error::ParseError; /// use nom::sequence::pair; /// use nom::bytes::complete::tag; /// -/// let parser = pair(tag("abc"), tag("efg")); +/// let mut parser = pair(tag("abc"), tag("efg")); /// /// assert_eq!(parser("abcefg"), Ok(("", ("abc", "efg")))); /// assert_eq!(parser("abcefghij"), Ok(("hij", ("abc", "efg")))); /// assert_eq!(parser(""), Err(Err::Error(("", ErrorKind::Tag)))); /// assert_eq!(parser("123"), Err(Err::Error(("123", ErrorKind::Tag)))); /// ``` -pub fn pair<I, O1, O2, E: ParseError<I>, F, G>(first: F, second: G) -> impl Fn(I) -> IResult<I, (O1, O2), E> +pub fn pair<I, O1, O2, E: ParseError<I>, F, G>( + mut first: F, + mut second: G, +) -> impl FnMut(I) -> IResult<I, (O1, O2), E> where - F: Fn(I) -> IResult<I, O1, E>, - G: Fn(I) -> IResult<I, O2, E>, + F: Parser<I, O1, E>, + G: Parser<I, O2, E>, { move |input: I| { - let (input, o1) = first(input)?; - second(input).map(|(i, o2)| (i, (o1, o2))) + let (input, o1) = first.parse(input)?; + second.parse(input).map(|(i, o2)| (i, (o1, o2))) } } // this implementation is used for type inference issues in macros #[doc(hidden)] -pub fn pairc<I, O1, O2, E: ParseError<I>, F, G>(input: I, first: F, second: G) -> IResult<I, (O1, O2), E> +pub fn pairc<I, O1, O2, E: ParseError<I>, F, G>( + input: I, + first: F, + second: G, +) -> IResult<I, (O1, O2), E> where F: Fn(I) -> IResult<I, O1, E>, G: Fn(I) -> IResult<I, O2, E>, @@ -58,27 +65,34 @@ where /// use nom::sequence::preceded; /// use nom::bytes::complete::tag; /// -/// let parser = preceded(tag("abc"), tag("efg")); +/// let mut parser = preceded(tag("abc"), tag("efg")); /// /// assert_eq!(parser("abcefg"), Ok(("", "efg"))); /// assert_eq!(parser("abcefghij"), Ok(("hij", "efg"))); /// assert_eq!(parser(""), Err(Err::Error(("", ErrorKind::Tag)))); /// assert_eq!(parser("123"), Err(Err::Error(("123", ErrorKind::Tag)))); /// ``` -pub fn preceded<I, O1, O2, E: ParseError<I>, F, G>(first: F, second: G) -> impl Fn(I) -> IResult<I, O2, E> +pub fn preceded<I, O1, O2, E: ParseError<I>, F, G>( + mut first: F, + mut second: G, +) -> impl FnMut(I) -> IResult<I, O2, E> where - F: Fn(I) -> IResult<I, O1, E>, - G: Fn(I) -> IResult<I, O2, E>, + F: Parser<I, O1, E>, + G: Parser<I, O2, E>, { move |input: I| { - let (input, _) = first(input)?; - second(input) + let (input, _) = first.parse(input)?; + second.parse(input) } } // this implementation is used for type inference issues in macros #[doc(hidden)] -pub fn precededc<I, O1, O2, E: ParseError<I>, F, G>(input: I, first: F, second: G) -> IResult<I, O2, E> +pub fn precededc<I, O1, O2, E: ParseError<I>, F, G>( + input: I, + first: F, + second: G, +) -> IResult<I, O2, E> where F: Fn(I) -> IResult<I, O1, E>, G: Fn(I) -> IResult<I, O2, E>, @@ -98,27 +112,34 @@ where /// use nom::sequence::terminated; /// use nom::bytes::complete::tag; /// -/// let parser = terminated(tag("abc"), tag("efg")); +/// let mut parser = terminated(tag("abc"), tag("efg")); /// /// assert_eq!(parser("abcefg"), Ok(("", "abc"))); /// assert_eq!(parser("abcefghij"), Ok(("hij", "abc"))); /// assert_eq!(parser(""), Err(Err::Error(("", ErrorKind::Tag)))); /// assert_eq!(parser("123"), Err(Err::Error(("123", ErrorKind::Tag)))); /// ``` -pub fn terminated<I, O1, O2, E: ParseError<I>, F, G>(first: F, second: G) -> impl Fn(I) -> IResult<I, O1, E> +pub fn terminated<I, O1, O2, E: ParseError<I>, F, G>( + mut first: F, + mut second: G, +) -> impl FnMut(I) -> IResult<I, O1, E> where - F: Fn(I) -> IResult<I, O1, E>, - G: Fn(I) -> IResult<I, O2, E>, + F: Parser<I, O1, E>, + G: Parser<I, O2, E>, { move |input: I| { - let (input, o1) = first(input)?; - second(input).map(|(i, _)| (i, o1)) + let (input, o1) = first.parse(input)?; + second.parse(input).map(|(i, _)| (i, o1)) } } // this implementation is used for type inference issues in macros #[doc(hidden)] -pub fn terminatedc<I, O1, O2, E: ParseError<I>, F, G>(input: I, first: F, second: G) -> IResult<I, O1, E> +pub fn terminatedc<I, O1, O2, E: ParseError<I>, F, G>( + input: I, + first: F, + second: G, +) -> IResult<I, O1, E> where F: Fn(I) -> IResult<I, O1, E>, G: Fn(I) -> IResult<I, O2, E>, @@ -140,29 +161,38 @@ where /// use nom::sequence::separated_pair; /// use nom::bytes::complete::tag; /// -/// let parser = separated_pair(tag("abc"), tag("|"), tag("efg")); +/// let mut parser = separated_pair(tag("abc"), tag("|"), tag("efg")); /// /// assert_eq!(parser("abc|efg"), Ok(("", ("abc", "efg")))); /// assert_eq!(parser("abc|efghij"), Ok(("hij", ("abc", "efg")))); /// assert_eq!(parser(""), Err(Err::Error(("", ErrorKind::Tag)))); /// assert_eq!(parser("123"), Err(Err::Error(("123", ErrorKind::Tag)))); /// ``` -pub fn separated_pair<I, O1, O2, O3, E: ParseError<I>, F, G, H>(first: F, sep: G, second: H) -> impl Fn(I) -> IResult<I, (O1, O3), E> +pub fn separated_pair<I, O1, O2, O3, E: ParseError<I>, F, G, H>( + mut first: F, + mut sep: G, + mut second: H, +) -> impl FnMut(I) -> IResult<I, (O1, O3), E> where - F: Fn(I) -> IResult<I, O1, E>, - G: Fn(I) -> IResult<I, O2, E>, - H: Fn(I) -> IResult<I, O3, E>, + F: Parser<I, O1, E>, + G: Parser<I, O2, E>, + H: Parser<I, O3, E>, { move |input: I| { - let (input, o1) = first(input)?; - let (input, _) = sep(input)?; - second(input).map(|(i, o2)| (i, (o1, o2))) + let (input, o1) = first.parse(input)?; + let (input, _) = sep.parse(input)?; + second.parse(input).map(|(i, o2)| (i, (o1, o2))) } } // this implementation is used for type inference issues in macros #[doc(hidden)] -pub fn separated_pairc<I, O1, O2, O3, E: ParseError<I>, F, G, H>(input: I, first: F, sep: G, second: H) -> IResult<I, (O1, O3), E> +pub fn separated_pairc<I, O1, O2, O3, E: ParseError<I>, F, G, H>( + input: I, + first: F, + sep: G, + second: H, +) -> IResult<I, (O1, O3), E> where F: Fn(I) -> IResult<I, O1, E>, G: Fn(I) -> IResult<I, O2, E>, @@ -171,63 +201,74 @@ where separated_pair(first, sep, second)(input) } -/// Matches an object from the first parser, -/// then gets an object from the sep_parser, -/// then matches another object from the second parser. +/// Matches an object from the first parser and discards it, +/// then gets an object from the second parser, +/// and finally matches an object from the third parser and discards it. /// /// # Arguments -/// * `first` The first parser to apply. -/// * `sep` The separator parser to apply. +/// * `first` The first parser to apply and discard. /// * `second` The second parser to apply. +/// * `third` The third parser to apply and discard. /// ```rust /// # use nom::{Err, error::ErrorKind, Needed}; /// # use nom::Needed::Size; /// use nom::sequence::delimited; /// use nom::bytes::complete::tag; /// -/// let parser = delimited(tag("abc"), tag("|"), tag("efg")); +/// let mut parser = delimited(tag("("), tag("abc"), tag(")")); /// -/// assert_eq!(parser("abc|efg"), Ok(("", "|"))); -/// assert_eq!(parser("abc|efghij"), Ok(("hij", "|"))); +/// assert_eq!(parser("(abc)"), Ok(("", "abc"))); +/// assert_eq!(parser("(abc)def"), Ok(("def", "abc"))); /// assert_eq!(parser(""), Err(Err::Error(("", ErrorKind::Tag)))); /// assert_eq!(parser("123"), Err(Err::Error(("123", ErrorKind::Tag)))); /// ``` -pub fn delimited<I, O1, O2, O3, E: ParseError<I>, F, G, H>(first: F, sep: G, second: H) -> impl Fn(I) -> IResult<I, O2, E> +pub fn delimited<I, O1, O2, O3, E: ParseError<I>, F, G, H>( + mut first: F, + mut second: G, + mut third: H, +) -> impl FnMut(I) -> IResult<I, O2, E> where - F: Fn(I) -> IResult<I, O1, E>, - G: Fn(I) -> IResult<I, O2, E>, - H: Fn(I) -> IResult<I, O3, E>, + F: Parser<I, O1, E>, + G: Parser<I, O2, E>, + H: Parser<I, O3, E>, { move |input: I| { - let (input, _) = first(input)?; - let (input, o2) = sep(input)?; - second(input).map(|(i, _)| (i, o2)) + let (input, _) = first.parse(input)?; + let (input, o2) = second.parse(input)?; + third.parse(input).map(|(i, _)| (i, o2)) } } // this implementation is used for type inference issues in macros #[doc(hidden)] -pub fn delimitedc<I, O1, O2, O3, E: ParseError<I>, F, G, H>(input: I, first: F, sep: G, second: H) -> IResult<I, O2, E> +pub fn delimitedc<I, O1, O2, O3, E: ParseError<I>, F, G, H>( + input: I, + first: F, + second: G, + third: H, +) -> IResult<I, O2, E> where F: Fn(I) -> IResult<I, O1, E>, G: Fn(I) -> IResult<I, O2, E>, H: Fn(I) -> IResult<I, O3, E>, { - delimited(first, sep, second)(input) + delimited(first, second, third)(input) } -/// helper trait for the tuple combinator +/// Helper trait for the tuple combinator. /// -/// this trait is implemented for tuples of parsers of up to 21 elements -pub trait Tuple<I,O,E> { - /// parses the input and returns a tuple of results of each parser - fn parse(&self, input: I) -> IResult<I,O,E>; +/// This trait is implemented for tuples of parsers of up to 21 elements. +pub trait Tuple<I, O, E> { + /// Parses the input and returns a tuple of results of each parser. + fn parse(&mut self, input: I) -> IResult<I, O, E>; } -impl<Input, Output, Error: ParseError<Input>, F: Fn(Input) -> IResult<Input, Output, Error> > Tuple<Input, (Output,), Error> for (F,) { - fn parse(&self, input: Input) -> IResult<Input,(Output,),Error> { - self.0(input).map(|(i,o)| (i, (o,))) - } +impl<Input, Output, Error: ParseError<Input>, F: Parser<Input, Output, Error>> + Tuple<Input, (Output,), Error> for (F,) +{ + fn parse(&mut self, input: Input) -> IResult<Input, (Output,), Error> { + self.0.parse(input).map(|(i, o)| (i, (o,))) + } } macro_rules! tuple_trait( @@ -248,10 +289,10 @@ macro_rules! tuple_trait_impl( ($($name:ident $ty: ident),+) => ( impl< Input: Clone, $($ty),+ , Error: ParseError<Input>, - $($name: Fn(Input) -> IResult<Input, $ty, Error>),+ + $($name: Parser<Input, $ty, Error>),+ > Tuple<Input, ( $($ty),+ ), Error> for ( $($name),+ ) { - fn parse(&self, input: Input) -> IResult<Input, ( $($ty),+ ), Error> { + fn parse(&mut self, input: Input) -> IResult<Input, ( $($ty),+ ), Error> { tuple_trait_inner!(0, self, input, (), $($name)+) } @@ -261,17 +302,17 @@ macro_rules! tuple_trait_impl( macro_rules! tuple_trait_inner( ($it:tt, $self:expr, $input:expr, (), $head:ident $($id:ident)+) => ({ - let (i, o) = $self.$it($input.clone())?; + let (i, o) = $self.$it.parse($input.clone())?; succ!($it, tuple_trait_inner!($self, i, ( o ), $($id)+)) }); ($it:tt, $self:expr, $input:expr, ($($parsed:tt)*), $head:ident $($id:ident)+) => ({ - let (i, o) = $self.$it($input.clone())?; + let (i, o) = $self.$it.parse($input.clone())?; succ!($it, tuple_trait_inner!($self, i, ($($parsed)* , o), $($id)+)) }); ($it:tt, $self:expr, $input:expr, ($($parsed:tt)*), $head:ident) => ({ - let (i, o) = $self.$it($input.clone())?; + let (i, o) = $self.$it.parse($input.clone())?; Ok((i, ($($parsed)* , o))) }); @@ -280,21 +321,21 @@ macro_rules! tuple_trait_inner( tuple_trait!(FnA A, FnB B, FnC C, FnD D, FnE E, FnF F, FnG G, FnH H, FnI I, FnJ J, FnK K, FnL L, FnM M, FnN N, FnO O, FnP P, FnQ Q, FnR R, FnS S, FnT T, FnU U); -/// applies a tuple of parsers one by one and returns their results as a tuple +///Applies a tuple of parsers one by one and returns their results as a tuple. /// /// ```rust /// # use nom::{Err, error::ErrorKind}; /// use nom::sequence::tuple; /// use nom::character::complete::{alpha1, digit1}; -/// let parser = tuple((alpha1, digit1, alpha1)); +/// let mut parser = tuple((alpha1, digit1, alpha1)); /// /// assert_eq!(parser("abc123def"), Ok(("", ("abc", "123", "def")))); /// assert_eq!(parser("123def"), Err(Err::Error(("123def", ErrorKind::Alpha)))); /// ``` -pub fn tuple<I: Clone, O, E: ParseError<I>, List: Tuple<I,O,E>>(l: List) -> impl Fn(I) -> IResult<I, O, E> { - move |i: I| { - l.parse(i) - } +pub fn tuple<I, O, E: ParseError<I>, List: Tuple<I, O, E>>( + mut l: List, +) -> impl FnMut(I) -> IResult<I, O, E> { + move |i: I| l.parse(i) } #[cfg(test)] @@ -303,11 +344,14 @@ mod tests { #[test] fn single_element_tuples() { - use crate::character::complete::{alpha1, digit1}; - use crate::{Err, error::ErrorKind}; + use crate::character::complete::alpha1; + use crate::{error::ErrorKind, Err}; - let parser = tuple((alpha1,)); + let mut parser = tuple((alpha1,)); assert_eq!(parser("abc123def"), Ok(("123def", ("abc",)))); - assert_eq!(parser("123def"), Err(Err::Error(("123def", ErrorKind::Alpha)))); + assert_eq!( + parser("123def"), + Err(Err::Error(("123def", ErrorKind::Alpha))) + ); } } @@ -1,6 +1,6 @@ #[cfg(test)] mod test { - use crate::{Err, error::ErrorKind, IResult}; + use crate::{error, error::ErrorKind, Err, IResult}; #[test] fn tagtr_succeed() { @@ -12,10 +12,7 @@ mod test { match test(INPUT) { Ok((extra, output)) => { - assert!( - extra == " World!", - "Parser `tag` consumed leftover input." - ); + assert!(extra == " World!", "Parser `tag` consumed leftover input."); assert!( output == TAG, "Parser `tag` doesn't return the tag it matched on success. \ @@ -37,7 +34,7 @@ mod test { const INPUT: &str = "Hello"; const TAG: &str = "Hello World!"; - let res: IResult<_,_,(_, ErrorKind)> = tag!(INPUT, TAG); + let res: IResult<_, _, error::Error<_>> = tag!(INPUT, TAG); match res { Err(Err::Incomplete(_)) => (), other => { @@ -55,7 +52,7 @@ mod test { const INPUT: &str = "Hello World!"; const TAG: &str = "Random"; // TAG must be closer than INPUT. - let res: IResult<_,_,(_, ErrorKind)> = tag!(INPUT, TAG); + let res: IResult<_, _, error::Error<_>> = tag!(INPUT, TAG); match res { Err(Err::Error(_)) => (), other => { @@ -73,7 +70,7 @@ mod test { const CONSUMED: &str = "Ī²ĆØĘĆ“ÅĆØĆĆĆ"; const LEFTOVER: &str = "Ć”ĘĘĆØÅ"; - let res: IResult<_,_,(_, ErrorKind)> = take!(INPUT, 9); + let res: IResult<_, _, error::Error<_>> = take!(INPUT, 9); match res { Ok((extra, output)) => { assert!( @@ -103,7 +100,7 @@ mod test { const CONSUMED: &str = "Ī²ĆØĘĆ“ÅĆØ"; const LEFTOVER: &str = "ĆĆĆāĆ”ĘĘĆØÅ"; - let res: IResult<_,_,(_, ErrorKind)> = take_until!(INPUT, FIND); + let res: IResult<_, _, (_, ErrorKind)> = take_until!(INPUT, FIND); match res { Ok((extra, output)) => { assert!( @@ -132,7 +129,7 @@ mod test { fn take_s_incomplete() { const INPUT: &str = "Ī²ĆØĘĆ“ÅĆØĆĆĆĆ”"; - let res: IResult<_,_,(_, ErrorKind)> = take!(INPUT, 13); + let res: IResult<_, _, (_, ErrorKind)> = take!(INPUT, 13); match res { Err(Err::Incomplete(_)) => (), other => panic!( @@ -157,8 +154,8 @@ mod test { let c = "abcd123"; let d = "123"; - assert_eq!(f(&a[..]), Err(Err::Incomplete(Needed::Size(1)))); - assert_eq!(f(&b[..]), Err(Err::Incomplete(Needed::Size(1)))); + assert_eq!(f(&a[..]), Err(Err::Incomplete(Needed::new(1)))); + assert_eq!(f(&b[..]), Err(Err::Incomplete(Needed::new(1)))); assert_eq!(f(&c[..]), Ok((&d[..], &b[..]))); assert_eq!(f(&d[..]), Ok((&d[..], &a[..]))); } @@ -171,8 +168,8 @@ mod test { let c = "abcd123"; let d = "123"; - assert_eq!(f(&a[..]), Err(Err::Incomplete(Needed::Size(1)))); - assert_eq!(f(&b[..]), Err(Err::Incomplete(Needed::Size(1)))); + assert_eq!(f(&a[..]), Err(Err::Incomplete(Needed::new(1)))); + assert_eq!(f(&b[..]), Err(Err::Incomplete(Needed::new(1)))); assert_eq!(f(&c[..]), Ok((&"123"[..], &b[..]))); assert_eq!( f(&d[..]), @@ -283,7 +280,15 @@ mod test { const CONSUMED: &str = "Ī²ĆØĘĆ“ÅĆØĆĆĆ"; const LEFTOVER: &str = "Ć”ĘĘĆØÅ"; fn while_s(c: char) -> bool { - c == 'Ī²' || c == 'ĆØ' || c == 'Ę' || c == 'Ć“' || c == 'Å' || c == 'ĆØ' || c == 'Ć' || c == 'Ć' || c == 'Ć' + c == 'Ī²' + || c == 'ĆØ' + || c == 'Ę' + || c == 'Ć“' + || c == 'Å' + || c == 'ĆØ' + || c == 'Ć' + || c == 'Ć' + || c == 'Ć' } fn test(input: &str) -> IResult<&str, &str> { take_while!(input, while_s) @@ -332,7 +337,15 @@ mod test { const CONSUMED: &str = "Ī²ĆØĘĆ“ÅĆØĆĆĆ"; const LEFTOVER: &str = "Ć”ĘĘĆØÅ"; fn while1_s(c: char) -> bool { - c == 'Ī²' || c == 'ĆØ' || c == 'Ę' || c == 'Ć“' || c == 'Å' || c == 'ĆØ' || c == 'Ć' || c == 'Ć' || c == 'Ć' + c == 'Ī²' + || c == 'ĆØ' + || c == 'Ę' + || c == 'Ć“' + || c == 'Å' + || c == 'ĆØ' + || c == 'Ć' + || c == 'Ć' + || c == 'Ć' } fn test(input: &str) -> IResult<&str, &str> { take_while1!(input, while1_s) @@ -364,7 +377,7 @@ mod test { const INPUT: &str = "Ī²ĆØĘĆ“ÅĆØ"; const FIND: &str = "Ī²ĆØĘĆ“ÅĆØĆĆĆ"; - let res: IResult<_,_,(_, ErrorKind)> = take_until!(INPUT, FIND); + let res: IResult<_, _, (_, ErrorKind)> = take_until!(INPUT, FIND); match res { Err(Err::Incomplete(_)) => (), other => panic!( @@ -446,7 +459,7 @@ mod test { const INPUT: &str = "Ī²ĆØĘĆ“ÅĆØĆĆĆĆ”ĘĘĆØÅ"; const FIND: &str = "RƔƱĪ“Ć“ā„"; - let res: IResult<_,_,(_, ErrorKind)> = take_until!(INPUT, FIND); + let res: IResult<_, _, (_, ErrorKind)> = take_until!(INPUT, FIND); match res { Err(Err::Incomplete(_)) => (), other => panic!( @@ -472,8 +485,8 @@ mod test { #[test] fn utf8_indexing() { named!(dot(&str) -> &str, - tag!(".") - ); + tag!(".") + ); let _ = dot("é»"); } diff --git a/src/traits.rs b/src/traits.rs index afd16ba..553b244 100644 --- a/src/traits.rs +++ b/src/traits.rs @@ -1,25 +1,25 @@ //! Traits input types have to implement to work with nom combinators -//! +use crate::error::{ErrorKind, ParseError}; use crate::internal::{Err, IResult, Needed}; -use crate::error::{ParseError, ErrorKind}; +use crate::lib::std::iter::{Copied, Enumerate}; use crate::lib::std::ops::{Range, RangeFrom, RangeFull, RangeTo}; -use crate::lib::std::iter::Enumerate; use crate::lib::std::slice::Iter; -use crate::lib::std::iter::Map; -use crate::lib::std::str::Chars; +use crate::lib::std::str::from_utf8; use crate::lib::std::str::CharIndices; +use crate::lib::std::str::Chars; use crate::lib::std::str::FromStr; -use crate::lib::std::str::from_utf8; -use memchr; #[cfg(feature = "alloc")] use crate::lib::std::string::String; #[cfg(feature = "alloc")] use crate::lib::std::vec::Vec; -/// abstract method to calculate the input length +#[cfg(feature = "bitvec")] +use bitvec::prelude::*; + +/// Abstract method to calculate the input length pub trait InputLength { - /// calculates the input length, as indicated by its name, + /// Calculates the input length, as indicated by its name, /// and the name of the trait itself fn input_len(&self) -> usize; } @@ -47,9 +47,21 @@ impl<'a> InputLength for (&'a [u8], usize) { } } -/// useful functions to calculate the offset between slices and show a hexdump of a slice +#[cfg(feature = "bitvec")] +impl<'a, O, T> InputLength for &'a BitSlice<O, T> +where + O: BitOrder, + T: 'a + BitStore, +{ + #[inline] + fn input_len(&self) -> usize { + self.len() + } +} + +/// Useful functions to calculate the offset between slices and show a hexdump of a slice pub trait Offset { - /// offset between the first byte of self and the first byte of the argument + /// Offset between the first byte of self and the first byte of the argument fn offset(&self, second: &Self) -> usize; } @@ -89,16 +101,40 @@ impl<'a> Offset for &'a str { } } +#[cfg(feature = "bitvec")] +impl<O, T> Offset for BitSlice<O, T> +where + O: BitOrder, + T: BitStore, +{ + #[inline(always)] + fn offset(&self, second: &Self) -> usize { + second.offset_from(self) as usize + } +} + +#[cfg(feature = "bitvec")] +impl<'a, O, T> Offset for &'a BitSlice<O, T> +where + O: BitOrder, + T: 'a + BitStore, +{ + #[inline(always)] + fn offset(&self, second: &Self) -> usize { + second.offset_from(self) as usize + } +} + /// Helper trait for types that can be viewed as a byte slice pub trait AsBytes { - /// casts the input type to a byte slice + /// Casts the input type to a byte slice fn as_bytes(&self) -> &[u8]; } impl<'a> AsBytes for &'a str { #[inline(always)] fn as_bytes(&self) -> &[u8] { - <str as AsBytes>::as_bytes(self) + (*self).as_bytes() } } @@ -123,6 +159,28 @@ impl AsBytes for [u8] { } } +#[cfg(feature = "bitvec")] +impl<'a, O> AsBytes for &'a BitSlice<O, u8> +where + O: BitOrder, +{ + #[inline(always)] + fn as_bytes(&self) -> &[u8] { + self.as_slice() + } +} + +#[cfg(feature = "bitvec")] +impl<O> AsBytes for BitSlice<O, u8> +where + O: BitOrder, +{ + #[inline(always)] + fn as_bytes(&self) -> &[u8] { + self.as_slice() + } +} + macro_rules! as_bytes_array_impls { ($($N:expr)+) => { $( @@ -139,6 +197,24 @@ macro_rules! as_bytes_array_impls { self } } + + #[cfg(feature = "bitvec")] + impl<'a, O> AsBytes for &'a BitArray<O, [u8; $N]> + where O: BitOrder { + #[inline(always)] + fn as_bytes(&self) -> &[u8] { + self.as_slice() + } + } + + #[cfg(feature = "bitvec")] + impl<O> AsBytes for BitArray<O, [u8; $N]> + where O: BitOrder { + #[inline(always)] + fn as_bytes(&self) -> &[u8] { + self.as_slice() + } + } )+ }; } @@ -150,27 +226,27 @@ as_bytes_array_impls! { 30 31 32 } -/// transforms common types to a char for basic token parsing +/// Transforms common types to a char for basic token parsing pub trait AsChar { /// makes a char from self fn as_char(self) -> char; - /// tests that self is an alphabetic character + /// Tests that self is an alphabetic character /// - /// warning: for `&str` it recognizes alphabetic + /// Warning: for `&str` it recognizes alphabetic /// characters outside of the 52 ASCII letters fn is_alpha(self) -> bool; - /// tests that self is an alphabetic character + /// Tests that self is an alphabetic character /// or a decimal digit fn is_alphanum(self) -> bool; - /// tests that self is a decimal digit + /// Tests that self is a decimal digit fn is_dec_digit(self) -> bool; - /// tests that self is an hex digit + /// Tests that self is an hex digit fn is_hex_digit(self) -> bool; - /// tests that self is an octal digit + /// Tests that self is an octal digit fn is_oct_digit(self) -> bool; - /// gets the len in bytes for self + /// Gets the len in bytes for self fn len(self) -> usize; } @@ -193,7 +269,9 @@ impl AsChar for u8 { } #[inline] fn is_hex_digit(self) -> bool { - (self >= 0x30 && self <= 0x39) || (self >= 0x41 && self <= 0x46) || (self >= 0x61 && self <= 0x66) + (self >= 0x30 && self <= 0x39) + || (self >= 0x41 && self <= 0x46) + || (self >= 0x61 && self <= 0x66) } #[inline] fn is_oct_digit(self) -> bool { @@ -223,7 +301,9 @@ impl<'a> AsChar for &'a u8 { } #[inline] fn is_hex_digit(self) -> bool { - (*self >= 0x30 && *self <= 0x39) || (*self >= 0x41 && *self <= 0x46) || (*self >= 0x61 && *self <= 0x66) + (*self >= 0x30 && *self <= 0x39) + || (*self >= 0x41 && *self <= 0x46) + || (*self >= 0x61 && *self <= 0x66) } #[inline] fn is_oct_digit(self) -> bool { @@ -297,48 +377,44 @@ impl<'a> AsChar for &'a char { } } -/// abstracts common iteration operations on the input type +/// Abstracts common iteration operations on the input type pub trait InputIter { - /// the current input type is a sequence of that `Item` type. + /// The current input type is a sequence of that `Item` type. /// - /// example: `u8` for `&[u8]` or `char` for &str` + /// Example: `u8` for `&[u8]` or `char` for `&str` type Item; - /// an iterator over the input type, producing the item and its position + /// An iterator over the input type, producing the item and its position /// for use with [Slice]. If we're iterating over `&str`, the position /// corresponds to the byte index of the character type Iter: Iterator<Item = (usize, Self::Item)>; - /// an iterator over the input type, producing the item + /// An iterator over the input type, producing the item type IterElem: Iterator<Item = Self::Item>; - /// returns an iterator over the elements and their byte offsets + /// Returns an iterator over the elements and their byte offsets fn iter_indices(&self) -> Self::Iter; - /// returns an iterator over the elements + /// Returns an iterator over the elements fn iter_elements(&self) -> Self::IterElem; - /// finds the byte position of the element + /// Finds the byte position of the element fn position<P>(&self, predicate: P) -> Option<usize> where P: Fn(Self::Item) -> bool; - /// get the byte offset from the element's position in the stream - fn slice_index(&self, count: usize) -> Option<usize>; + /// Get the byte offset from the element's position in the stream + fn slice_index(&self, count: usize) -> Result<usize, Needed>; } -/// abstracts slicing operations +/// Abstracts slicing operations pub trait InputTake: Sized { - /// returns a slice of `count` bytes. panics if count > length + /// Returns a slice of `count` bytes. panics if count > length fn take(&self, count: usize) -> Self; - /// split the stream at the `count` byte offset. panics if count > length + /// Split the stream at the `count` byte offset. panics if count > length fn take_split(&self, count: usize) -> (Self, Self); } -fn star(r_u8: &u8) -> u8 { - *r_u8 -} - impl<'a> InputIter for &'a [u8] { type Item = u8; type Iter = Enumerate<Self::IterElem>; - type IterElem = Map<Iter<'a, Self::Item>, fn(&u8) -> u8>; + type IterElem = Copied<Iter<'a, u8>>; #[inline] fn iter_indices(&self) -> Self::Iter { @@ -346,7 +422,7 @@ impl<'a> InputIter for &'a [u8] { } #[inline] fn iter_elements(&self) -> Self::IterElem { - self.iter().map(star) + self.iter().copied() } #[inline] fn position<P>(&self, predicate: P) -> Option<usize> @@ -356,11 +432,11 @@ impl<'a> InputIter for &'a [u8] { self.iter().position(|b| predicate(*b)) } #[inline] - fn slice_index(&self, count: usize) -> Option<usize> { + fn slice_index(&self, count: usize) -> Result<usize, Needed> { if self.len() >= count { - Some(count) + Ok(count) } else { - None + Err(Needed::new(count - self.len())) } } } @@ -401,18 +477,18 @@ impl<'a> InputIter for &'a str { None } #[inline] - fn slice_index(&self, count: usize) -> Option<usize> { + fn slice_index(&self, count: usize) -> Result<usize, Needed> { let mut cnt = 0; for (index, _) in self.char_indices() { if cnt == count { - return Some(index); + return Ok(index); } cnt += 1; } if cnt == count { - return Some(self.len()); + return Ok(self.len()); } - None + Err(Needed::Unknown) } } @@ -429,64 +505,133 @@ impl<'a> InputTake for &'a str { } } -/// Dummy trait used for default implementations (currently only used for `InputTakeAtPosition`). +#[cfg(feature = "bitvec")] +impl<'a, O, T> InputIter for &'a BitSlice<O, T> +where + O: BitOrder, + T: 'a + BitStore, +{ + type Item = bool; + type Iter = Enumerate<Self::IterElem>; + type IterElem = Copied<bitvec::slice::Iter<'a, O, T>>; + + #[inline] + fn iter_indices(&self) -> Self::Iter { + self.iter_elements().enumerate() + } + + #[inline] + fn iter_elements(&self) -> Self::IterElem { + self.iter().copied() + } + + #[inline] + fn position<P>(&self, predicate: P) -> Option<usize> + where + P: Fn(Self::Item) -> bool, + { + self.iter_elements().position(predicate) + } + + #[inline] + fn slice_index(&self, count: usize) -> Result<usize, Needed> { + if self.len() >= count { + Ok(count) + } else { + Err(Needed::new(count - self.len())) + } + } +} + +#[cfg(feature = "bitvec")] +impl<'a, O, T> InputTake for &'a BitSlice<O, T> +where + O: BitOrder, + T: 'a + BitStore, +{ + #[inline] + fn take(&self, count: usize) -> Self { + &self[..count] + } + + #[inline] + fn take_split(&self, count: usize) -> (Self, Self) { + let (a, b) = self.split_at(count); + (b, a) + } +} + +/// Dummy trait used for default implementations (currently only used for `InputTakeAtPosition` and `Compare`). /// /// When implementing a custom input type, it is possible to use directly the -/// default implementation: if the input type implements `InputLength`, `InputIter`, +/// default implementation: If the input type implements `InputLength`, `InputIter`, /// `InputTake` and `Clone`, you can implement `UnspecializedInput` and get -/// a default version of `InputTakeAtPosition`. +/// a default version of `InputTakeAtPosition` and `Compare`. /// /// For performance reasons, you might want to write a custom implementation of /// `InputTakeAtPosition` (like the one for `&[u8]`). pub trait UnspecializedInput {} -/// methods to take as much input as possible until the provided function returns true for the current element +/// Methods to take as much input as possible until the provided function returns true for the current element. /// -/// a large part of nom's basic parsers are built using this trait +/// A large part of nom's basic parsers are built using this trait. pub trait InputTakeAtPosition: Sized { - /// the current input type is a sequence of that `Item` type. + /// The current input type is a sequence of that `Item` type. /// - /// example: `u8` for `&[u8]` or `char` for &str` + /// Example: `u8` for `&[u8]` or `char` for `&str` type Item; - /// looks for the first element of the input type for which the condition returns true, - /// and returns the input up to this position + /// Looks for the first element of the input type for which the condition returns true, + /// and returns the input up to this position. /// - /// *streaming version*: if no element is found matching the condition, this will return `Incomplete` + /// *streaming version*: If no element is found matching the condition, this will return `Incomplete` fn split_at_position<P, E: ParseError<Self>>(&self, predicate: P) -> IResult<Self, Self, E> where P: Fn(Self::Item) -> bool; - /// looks for the first element of the input type for which the condition returns true - /// and returns the input up to this position + /// Looks for the first element of the input type for which the condition returns true + /// and returns the input up to this position. /// - /// fails if the produced slice is empty + /// Fails if the produced slice is empty. /// - /// *streaming version*: if no element is found matching the condition, this will return `Incomplete` - fn split_at_position1<P, E: ParseError<Self>>(&self, predicate: P, e: ErrorKind) -> IResult<Self, Self, E> + /// *streaming version*: If no element is found matching the condition, this will return `Incomplete` + fn split_at_position1<P, E: ParseError<Self>>( + &self, + predicate: P, + e: ErrorKind, + ) -> IResult<Self, Self, E> where P: Fn(Self::Item) -> bool; - /// looks for the first element of the input type for which the condition returns true, - /// and returns the input up to this position + /// Looks for the first element of the input type for which the condition returns true, + /// and returns the input up to this position. /// - /// *complete version*: if no element is found matching the condition, this will return the whole input - fn split_at_position_complete<P, E: ParseError<Self>>(&self, predicate: P) -> IResult<Self, Self, E> + /// *complete version*: If no element is found matching the condition, this will return the whole input + fn split_at_position_complete<P, E: ParseError<Self>>( + &self, + predicate: P, + ) -> IResult<Self, Self, E> where P: Fn(Self::Item) -> bool; - /// looks for the first element of the input type for which the condition returns true - /// and returns the input up to this position + /// Looks for the first element of the input type for which the condition returns true + /// and returns the input up to this position. /// - /// fails if the produced slice is empty + /// Fails if the produced slice is empty. /// - /// *complete version*: if no element is found matching the condition, this will return the whole input - fn split_at_position1_complete<P, E: ParseError<Self>>(&self, predicate: P, e: ErrorKind) -> IResult<Self, Self, E> + /// *complete version*: If no element is found matching the condition, this will return the whole input + fn split_at_position1_complete<P, E: ParseError<Self>>( + &self, + predicate: P, + e: ErrorKind, + ) -> IResult<Self, Self, E> where P: Fn(Self::Item) -> bool; } -impl<T: InputLength + InputIter + InputTake + Clone + UnspecializedInput> InputTakeAtPosition for T { +impl<T: InputLength + InputIter + InputTake + Clone + UnspecializedInput> InputTakeAtPosition + for T +{ type Item = <T as InputIter>::Item; fn split_at_position<P, E: ParseError<Self>>(&self, predicate: P) -> IResult<Self, Self, E> @@ -495,36 +640,53 @@ impl<T: InputLength + InputIter + InputTake + Clone + UnspecializedInput> InputT { match self.position(predicate) { Some(n) => Ok(self.take_split(n)), - None => Err(Err::Incomplete(Needed::Size(1))), + None => Err(Err::Incomplete(Needed::new(1))), } } - fn split_at_position1<P, E: ParseError<Self>>(&self, predicate: P, e: ErrorKind) -> IResult<Self, Self, E> + fn split_at_position1<P, E: ParseError<Self>>( + &self, + predicate: P, + e: ErrorKind, + ) -> IResult<Self, Self, E> where P: Fn(Self::Item) -> bool, { match self.position(predicate) { Some(0) => Err(Err::Error(E::from_error_kind(self.clone(), e))), Some(n) => Ok(self.take_split(n)), - None => Err(Err::Incomplete(Needed::Size(1))), + None => Err(Err::Incomplete(Needed::new(1))), } } - fn split_at_position_complete<P, E: ParseError<Self>>(&self, predicate: P) -> IResult<Self, Self, E> - where P: Fn(Self::Item) -> bool { + fn split_at_position_complete<P, E: ParseError<Self>>( + &self, + predicate: P, + ) -> IResult<Self, Self, E> + where + P: Fn(Self::Item) -> bool, + { match self.split_at_position(predicate) { Err(Err::Incomplete(_)) => Ok(self.take_split(self.input_len())), res => res, } } - fn split_at_position1_complete<P, E: ParseError<Self>>(&self, predicate: P, e: ErrorKind) -> IResult<Self, Self, E> - where P: Fn(Self::Item) -> bool { + fn split_at_position1_complete<P, E: ParseError<Self>>( + &self, + predicate: P, + e: ErrorKind, + ) -> IResult<Self, Self, E> + where + P: Fn(Self::Item) -> bool, + { match self.split_at_position1(predicate, e) { - Err(Err::Incomplete(_)) => if self.input_len() == 0 { - Err(Err::Error(E::from_error_kind(self.clone(), e))) - } else { - Ok(self.take_split(self.input_len())) + Err(Err::Incomplete(_)) => { + if self.input_len() == 0 { + Err(Err::Error(E::from_error_kind(self.clone(), e))) + } else { + Ok(self.take_split(self.input_len())) + } } res => res, } @@ -540,41 +702,56 @@ impl<'a> InputTakeAtPosition for &'a [u8] { { match (0..self.len()).find(|b| predicate(self[*b])) { Some(i) => Ok((&self[i..], &self[..i])), - None => Err(Err::Incomplete(Needed::Size(1))), + None => Err(Err::Incomplete(Needed::new(1))), } } - fn split_at_position1<P, E: ParseError<Self>>(&self, predicate: P, e: ErrorKind) -> IResult<Self, Self, E> + fn split_at_position1<P, E: ParseError<Self>>( + &self, + predicate: P, + e: ErrorKind, + ) -> IResult<Self, Self, E> where P: Fn(Self::Item) -> bool, { match (0..self.len()).find(|b| predicate(self[*b])) { Some(0) => Err(Err::Error(E::from_error_kind(self, e))), Some(i) => Ok((&self[i..], &self[..i])), - None => Err(Err::Incomplete(Needed::Size(1))), + None => Err(Err::Incomplete(Needed::new(1))), } } - fn split_at_position_complete<P, E: ParseError<Self>>(&self, predicate: P) -> IResult<Self, Self, E> - where P: Fn(Self::Item) -> bool { + fn split_at_position_complete<P, E: ParseError<Self>>( + &self, + predicate: P, + ) -> IResult<Self, Self, E> + where + P: Fn(Self::Item) -> bool, + { match (0..self.len()).find(|b| predicate(self[*b])) { Some(i) => Ok((&self[i..], &self[..i])), None => Ok(self.take_split(self.input_len())), } } - fn split_at_position1_complete<P, E: ParseError<Self>>(&self, predicate: P, e: ErrorKind) -> IResult<Self, Self, E> - where P: Fn(Self::Item) -> bool { + fn split_at_position1_complete<P, E: ParseError<Self>>( + &self, + predicate: P, + e: ErrorKind, + ) -> IResult<Self, Self, E> + where + P: Fn(Self::Item) -> bool, + { match (0..self.len()).find(|b| predicate(self[*b])) { Some(0) => Err(Err::Error(E::from_error_kind(self, e))), Some(i) => Ok((&self[i..], &self[..i])), None => { - if self.len() == 0 { + if self.is_empty() { Err(Err::Error(E::from_error_kind(self, e))) } else { Ok(self.take_split(self.input_len())) } - }, + } } } } @@ -588,71 +765,165 @@ impl<'a> InputTakeAtPosition for &'a str { { match self.find(predicate) { Some(i) => Ok((&self[i..], &self[..i])), - None => Err(Err::Incomplete(Needed::Size(1))), + None => Err(Err::Incomplete(Needed::new(1))), } } - fn split_at_position1<P, E: ParseError<Self>>(&self, predicate: P, e: ErrorKind) -> IResult<Self, Self, E> + fn split_at_position1<P, E: ParseError<Self>>( + &self, + predicate: P, + e: ErrorKind, + ) -> IResult<Self, Self, E> where P: Fn(Self::Item) -> bool, { match self.find(predicate) { Some(0) => Err(Err::Error(E::from_error_kind(self, e))), Some(i) => Ok((&self[i..], &self[..i])), - None => Err(Err::Incomplete(Needed::Size(1))), + None => Err(Err::Incomplete(Needed::new(1))), } } - fn split_at_position_complete<P, E: ParseError<Self>>(&self, predicate: P) -> IResult<Self, Self, E> - where P: Fn(Self::Item) -> bool { + fn split_at_position_complete<P, E: ParseError<Self>>( + &self, + predicate: P, + ) -> IResult<Self, Self, E> + where + P: Fn(Self::Item) -> bool, + { match self.find(predicate) { Some(i) => Ok((&self[i..], &self[..i])), - None => Ok(self.take_split(self.input_len())) + None => Ok(self.take_split(self.input_len())), } } - fn split_at_position1_complete<P, E: ParseError<Self>>(&self, predicate: P, e: ErrorKind) -> IResult<Self, Self, E> - where P: Fn(Self::Item) -> bool { + fn split_at_position1_complete<P, E: ParseError<Self>>( + &self, + predicate: P, + e: ErrorKind, + ) -> IResult<Self, Self, E> + where + P: Fn(Self::Item) -> bool, + { match self.find(predicate) { Some(0) => Err(Err::Error(E::from_error_kind(self, e))), Some(i) => Ok((&self[i..], &self[..i])), None => { - if self.len() == 0 { + if self.is_empty() { Err(Err::Error(E::from_error_kind(self, e))) } else { Ok(self.take_split(self.input_len())) } - }, + } } } } -/// indicates wether a comparison was successful, an error, or +#[cfg(feature = "bitvec")] +impl<'a, O, T> InputTakeAtPosition for &'a BitSlice<O, T> +where + O: BitOrder, + T: 'a + BitStore, +{ + type Item = bool; + + fn split_at_position<P, E: ParseError<Self>>(&self, predicate: P) -> IResult<Self, Self, E> + where + P: Fn(Self::Item) -> bool, + { + self + .iter() + .copied() + .position(predicate) + .map(|i| self.split_at(i)) + .ok_or_else(|| Err::Incomplete(Needed::new(1))) + } + + fn split_at_position1<P, E: ParseError<Self>>( + &self, + predicate: P, + e: ErrorKind, + ) -> IResult<Self, Self, E> + where + P: Fn(Self::Item) -> bool, + { + match self.iter().copied().position(predicate) { + Some(0) => Err(Err::Error(E::from_error_kind(self, e))), + Some(i) => Ok(self.split_at(i)), + None => Err(Err::Incomplete(Needed::new(1))), + } + } + + fn split_at_position_complete<P, E: ParseError<Self>>( + &self, + predicate: P, + ) -> IResult<Self, Self, E> + where + P: Fn(Self::Item) -> bool, + { + self + .iter() + .position(|b| predicate(*b)) + .map(|i| self.split_at(i)) + .or_else(|| Some((self, Self::default()))) + .ok_or_else(|| unreachable!()) + } + + fn split_at_position1_complete<P, E: ParseError<Self>>( + &self, + predicate: P, + e: ErrorKind, + ) -> IResult<Self, Self, E> + where + P: Fn(Self::Item) -> bool, + { + match self.iter().copied().position(predicate) { + Some(0) => Err(Err::Error(E::from_error_kind(self, e))), + Some(i) => Ok(self.split_at(i)), + None => { + if self.is_empty() { + Err(Err::Error(E::from_error_kind(self, e))) + } else { + Ok((self, Self::default())) + } + } + } + } +} + +/// Indicates wether a comparison was successful, an error, or /// if more data was needed #[derive(Debug, PartialEq)] pub enum CompareResult { - /// comparison was successful + /// Comparison was successful Ok, - /// we need more data to be sure + /// We need more data to be sure Incomplete, - /// comparison failed + /// Comparison failed Error, } -/// abstracts comparison operations +/// Abstracts comparison operations pub trait Compare<T> { - /// compares self to another value for equality + /// Compares self to another value for equality fn compare(&self, t: T) -> CompareResult; - /// compares self to another value for equality + /// Compares self to another value for equality /// independently of the case. /// - /// warning: for `&str`, the comparison is done + /// Warning: for `&str`, the comparison is done /// by lowercasing both strings and comparing /// the result. This is a temporary solution until /// a better one appears fn compare_no_case(&self, t: T) -> CompareResult; } +fn lowercase_byte(c: u8) -> u8 { + match c { + b'A'..=b'Z' => c - b'A' + b'a', + _ => c, + } +} + impl<'a, 'b> Compare<&'b [u8]> for &'a [u8] { #[inline(always)] fn compare(&self, t: &'b [u8]) -> CompareResult { @@ -688,19 +959,53 @@ impl<'a, 'b> Compare<&'b [u8]> for &'a [u8] { #[inline(always)] fn compare_no_case(&self, t: &'b [u8]) -> CompareResult { - let len = self.len(); - let blen = t.len(); - let m = if len < blen { len } else { blen }; - let reduced = &self[..m]; - let other = &t[..m]; + if self + .iter() + .zip(t) + .any(|(a, b)| lowercase_byte(*a) != lowercase_byte(*b)) + { + CompareResult::Error + } else if self.len() < t.len() { + CompareResult::Incomplete + } else { + CompareResult::Ok + } + } +} - if !reduced.iter().zip(other).all(|(a, b)| match (*a, *b) { - (0..=64, 0..=64) | (91..=96, 91..=96) | (123..=255, 123..=255) => a == b, - (65..=90, 65..=90) | (97..=122, 97..=122) | (65..=90, 97..=122) | (97..=122, 65..=90) => *a | 0b00_10_00_00 == *b | 0b00_10_00_00, - _ => false, - }) { +impl< + T: InputLength + InputIter<Item = u8> + InputTake + UnspecializedInput, + O: InputLength + InputIter<Item = u8> + InputTake, + > Compare<O> for T +{ + #[inline(always)] + fn compare(&self, t: O) -> CompareResult { + let pos = self + .iter_elements() + .zip(t.iter_elements()) + .position(|(a, b)| a != b); + + match pos { + Some(_) => CompareResult::Error, + None => { + if self.input_len() >= t.input_len() { + CompareResult::Ok + } else { + CompareResult::Incomplete + } + } + } + } + + #[inline(always)] + fn compare_no_case(&self, t: O) -> CompareResult { + if self + .iter_elements() + .zip(t.iter_elements()) + .any(|(a, b)| lowercase_byte(a) != lowercase_byte(b)) + { CompareResult::Error - } else if m < blen { + } else if self.input_len() < t.input_len() { CompareResult::Incomplete } else { CompareResult::Ok @@ -722,7 +1027,16 @@ impl<'a, 'b> Compare<&'b str> for &'a [u8] { impl<'a, 'b> Compare<&'b str> for &'a str { #[inline(always)] fn compare(&self, t: &'b str) -> CompareResult { - let pos = self.chars().zip(t.chars()).position(|(a, b)| a != b); + self.as_bytes().compare(t.as_bytes()) + } + + //FIXME: this version is too simple and does not use the current locale + #[inline(always)] + fn compare_no_case(&self, t: &'b str) -> CompareResult { + let pos = self + .chars() + .zip(t.chars()) + .position(|(a, b)| a.to_lowercase().ne(b.to_lowercase())); match pos { Some(_) => CompareResult::Error, @@ -735,19 +1049,22 @@ impl<'a, 'b> Compare<&'b str> for &'a str { } } } +} - //FIXME: this version is too simple and does not use the current locale - #[inline(always)] - fn compare_no_case(&self, t: &'b str) -> CompareResult { - let pos = self - .chars() - .zip(t.chars()) - .position(|(a, b)| a.to_lowercase().zip(b.to_lowercase()).any(|(a, b)| a != b)); - - match pos { +#[cfg(feature = "bitvec")] +impl<'a, 'b, O1, O2, T1, T2> Compare<&'b BitSlice<O2, T2>> for &'a BitSlice<O1, T1> +where + O1: BitOrder, + O2: BitOrder, + T1: 'a + BitStore, + T2: 'a + BitStore, +{ + #[inline] + fn compare(&self, other: &'b BitSlice<O2, T2>) -> CompareResult { + match self.iter().zip(other.iter()).position(|(a, b)| a != b) { Some(_) => CompareResult::Error, None => { - if self.len() >= t.len() { + if self.len() >= other.len() { CompareResult::Ok } else { CompareResult::Incomplete @@ -755,11 +1072,16 @@ impl<'a, 'b> Compare<&'b str> for &'a str { } } } + + #[inline(always)] + fn compare_no_case(&self, other: &'b BitSlice<O2, T2>) -> CompareResult { + self.compare(other) + } } -/// look for a token in self +/// Look for a token in self pub trait FindToken<T> { - /// returns true if self contains the token + /// Returns true if self contains the token fn find_token(&self, token: T) -> bool; } @@ -777,7 +1099,7 @@ impl<'a> FindToken<u8> for &'a str { impl<'a, 'b> FindToken<&'a u8> for &'b [u8] { fn find_token(&self, token: &u8) -> bool { - memchr::memchr(*token, self).is_some() + self.find_token(*token) } } @@ -789,66 +1111,75 @@ impl<'a, 'b> FindToken<&'a u8> for &'b str { impl<'a> FindToken<char> for &'a [u8] { fn find_token(&self, token: char) -> bool { - for i in self.iter() { - if token as u8 == *i { - return true; - } - } - false + self.iter().any(|i| *i == token as u8) } } impl<'a> FindToken<char> for &'a str { fn find_token(&self, token: char) -> bool { - for i in self.chars() { - if token == i { - return true; - } - } - false + self.chars().any(|i| i == token) + } +} + +#[cfg(feature = "bitvec")] +impl<'a, O, T> FindToken<bool> for &'a BitSlice<O, T> +where + O: BitOrder, + T: 'a + BitStore, +{ + fn find_token(&self, token: bool) -> bool { + self.iter().copied().any(|i| i == token) + } +} + +#[cfg(feature = "bitvec")] +impl<'a, O, T> FindToken<(usize, bool)> for &'a BitSlice<O, T> +where + O: BitOrder, + T: 'a + BitStore, +{ + fn find_token(&self, token: (usize, bool)) -> bool { + self.iter().copied().enumerate().any(|i| i == token) } } -/// look for a substring in self +/// Look for a substring in self pub trait FindSubstring<T> { - /// returns the byte position of the substring if it is found + /// Returns the byte position of the substring if it is found fn find_substring(&self, substr: T) -> Option<usize>; } impl<'a, 'b> FindSubstring<&'b [u8]> for &'a [u8] { fn find_substring(&self, substr: &'b [u8]) -> Option<usize> { - let substr_len = substr.len(); + if substr.len() > self.len() { + return None; + } - if substr_len == 0 { + let (&substr_first, substr_rest) = match substr.split_first() { + Some(split) => split, // an empty substring is found at position 0 // This matches the behavior of str.find(""). - Some(0) - } else if substr_len == 1 { - memchr::memchr(substr[0], self) - } else if substr_len > self.len() { - None - } else { - let max = self.len() - substr_len; - let mut offset = 0; - let mut haystack = &self[..]; - - while let Some(position) = memchr::memchr(substr[0], haystack) { - offset += position; + None => return Some(0), + }; - if offset > max { - return None; - } + if substr_rest.is_empty() { + return memchr::memchr(substr_first, self); + } - if &haystack[position..position + substr_len] == substr { - return Some(offset); - } + let mut offset = 0; + let haystack = &self[..self.len() - substr_rest.len()]; - haystack = &haystack[position + 1..]; - offset += 1; + while let Some(position) = memchr::memchr(substr_first, &haystack[offset..]) { + offset += position; + let next_offset = offset + 1; + if &self[next_offset..][..substr_rest.len()] == substr_rest { + return Some(offset); } - None + offset = next_offset; } + + None } } @@ -865,10 +1196,33 @@ impl<'a, 'b> FindSubstring<&'b str> for &'a str { } } -/// used to integrate str's parse() method +#[cfg(feature = "bitvec")] +impl<'a, 'b, O1, O2, T1, T2> FindSubstring<&'b BitSlice<O2, T2>> for &'a BitSlice<O1, T1> +where + O1: BitOrder, + O2: BitOrder, + T1: 'a + BitStore, + T2: 'b + BitStore, +{ + fn find_substring(&self, substr: &'b BitSlice<O2, T2>) -> Option<usize> { + if substr.len() > self.len() { + return None; + } + + if substr.is_empty() { + return Some(0); + } + + self + .windows(substr.len()) + .position(|window| window == substr) + } +} + +/// Used to integrate `str`'s `parse()` method pub trait ParseTo<R> { - /// succeeds if `parse()` succeeded. The byte slice implementation - /// will first convert it to a &str, then apply the `parse()` function + /// Succeeds if `parse()` succeeded. The byte slice implementation + /// will first convert it to a `&str`, then apply the `parse()` function fn parse_to(&self) -> Option<R>; } @@ -884,55 +1238,73 @@ impl<'a, R: FromStr> ParseTo<R> for &'a str { } } -/// slicing operations using ranges +/// Slicing operations using ranges. /// -/// this trait is loosely based on +/// This trait is loosely based on /// `Index`, but can actually return /// something else than a `&[T]` or `&str` pub trait Slice<R> { - /// slices self according to the range argument + /// Slices self according to the range argument fn slice(&self, range: R) -> Self; } macro_rules! impl_fn_slice { - ( $ty:ty ) => { - fn slice(&self, range:$ty) -> Self { - &self[range] - } + ( $ty:ty ) => { + fn slice(&self, range: $ty) -> Self { + &self[range] } + }; } macro_rules! slice_range_impl { - ( [ $for_type:ident ], $ty:ty ) => { - impl<'a, $for_type> Slice<$ty> for &'a [$for_type] { - impl_fn_slice!( $ty ); - } - }; - ( $for_type:ty, $ty:ty ) => { - impl<'a> Slice<$ty> for &'a $for_type { - impl_fn_slice!( $ty ); - } + ( BitSlice, $ty:ty ) => { + impl<'a, O, T> Slice<$ty> for &'a BitSlice<O, T> + where + O: BitOrder, + T: BitStore, + { + impl_fn_slice!($ty); + } + }; + ( [ $for_type:ident ], $ty:ty ) => { + impl<'a, $for_type> Slice<$ty> for &'a [$for_type] { + impl_fn_slice!($ty); + } + }; + ( $for_type:ty, $ty:ty ) => { + impl<'a> Slice<$ty> for &'a $for_type { + impl_fn_slice!($ty); } + }; } macro_rules! slice_ranges_impl { - ( [ $for_type:ident ] ) => { - slice_range_impl! {[$for_type], Range<usize>} - slice_range_impl! {[$for_type], RangeTo<usize>} - slice_range_impl! {[$for_type], RangeFrom<usize>} - slice_range_impl! {[$for_type], RangeFull} - }; - ( $for_type:ty ) => { - slice_range_impl! {$for_type, Range<usize>} - slice_range_impl! {$for_type, RangeTo<usize>} - slice_range_impl! {$for_type, RangeFrom<usize>} - slice_range_impl! {$for_type, RangeFull} - } + ( BitSlice ) => { + slice_range_impl! {BitSlice, Range<usize>} + slice_range_impl! {BitSlice, RangeTo<usize>} + slice_range_impl! {BitSlice, RangeFrom<usize>} + slice_range_impl! {BitSlice, RangeFull} + }; + ( [ $for_type:ident ] ) => { + slice_range_impl! {[$for_type], Range<usize>} + slice_range_impl! {[$for_type], RangeTo<usize>} + slice_range_impl! {[$for_type], RangeFrom<usize>} + slice_range_impl! {[$for_type], RangeFull} + }; + ( $for_type:ty ) => { + slice_range_impl! {$for_type, Range<usize>} + slice_range_impl! {$for_type, RangeTo<usize>} + slice_range_impl! {$for_type, RangeFrom<usize>} + slice_range_impl! {$for_type, RangeFull} + }; } slice_ranges_impl! {str} slice_ranges_impl! {[T]} +#[cfg(feature = "bitvec")] +slice_ranges_impl! {BitSlice} + macro_rules! array_impls { ($($N:expr)+) => { $( @@ -950,6 +1322,29 @@ macro_rules! array_impls { } } + impl<'a> InputIter for &'a [u8; $N] { + type Item = u8; + type Iter = Enumerate<Self::IterElem>; + type IterElem = Copied<Iter<'a, u8>>; + + fn iter_indices(&self) -> Self::Iter { + (&self[..]).iter_indices() + } + + fn iter_elements(&self) -> Self::IterElem { + (&self[..]).iter_elements() + } + + fn position<P>(&self, predicate: P) -> Option<usize> + where P: Fn(Self::Item) -> bool { + (&self[..]).position(predicate) + } + + fn slice_index(&self, count: usize) -> Result<usize, Needed> { + (&self[..]).slice_index(count) + } + } + impl<'a> Compare<[u8; $N]> for &'a [u8] { #[inline(always)] fn compare(&self, t: [u8; $N]) -> CompareResult { @@ -982,7 +1377,7 @@ macro_rules! array_impls { impl<'a> FindToken<&'a u8> for [u8; $N] { fn find_token(&self, token: &u8) -> bool { - memchr::memchr(*token, &self[..]).is_some() + self.find_token(*token) } } )+ @@ -996,21 +1391,20 @@ array_impls! { 30 31 32 } -/// abstracts something which can extend an `Extend` -/// used to build modified input slices in `escaped_transform` +/// Abstracts something which can extend an `Extend`. +/// Used to build modified input slices in `escaped_transform` pub trait ExtendInto { - - /// the current input type is a sequence of that `Item` type. + /// The current input type is a sequence of that `Item` type. /// - /// example: `u8` for `&[u8]` or `char` for &str` + /// Example: `u8` for `&[u8]` or `char` for `&str` type Item; - /// the type that will be produced - type Extender: Extend<Self::Item>; + /// The type that will be produced + type Extender; - /// create a new `Extend` of the correct type + /// Create a new `Extend` of the correct type fn new_builder(&self) -> Self::Extender; - /// accumulate the input into an accumulator + /// Accumulate the input into an accumulator fn extend_into(&self, acc: &mut Self::Extender); } @@ -1040,11 +1434,10 @@ impl ExtendInto for &[u8] { } #[inline] fn extend_into(&self, acc: &mut Vec<u8>) { - acc.extend(self.iter().cloned()); + acc.extend_from_slice(self); } } - #[cfg(feature = "alloc")] impl ExtendInto for str { type Item = char; @@ -1090,9 +1483,49 @@ impl ExtendInto for char { } } -/// Helper trait to convert numbers to usize +#[cfg(all(feature = "alloc", feature = "bitvec"))] +impl<O, T> ExtendInto for BitSlice<O, T> +where + O: BitOrder, + T: BitStore, +{ + type Item = bool; + type Extender = BitVec<O, T>; + + #[inline] + fn new_builder(&self) -> BitVec<O, T> { + BitVec::new() + } + + #[inline] + fn extend_into(&self, acc: &mut Self::Extender) { + acc.extend(self.iter()); + } +} + +#[cfg(all(feature = "alloc", feature = "bitvec"))] +impl<'a, O, T> ExtendInto for &'a BitSlice<O, T> +where + O: BitOrder, + T: 'a + BitStore, +{ + type Item = bool; + type Extender = BitVec<O, T>; + + #[inline] + fn new_builder(&self) -> BitVec<O, T> { + BitVec::new() + } + + #[inline] + fn extend_into(&self, acc: &mut Self::Extender) { + acc.extend(self.iter()); + } +} + +/// Helper trait to convert numbers to usize. /// -/// by default, usize implements `From<u8>` and `From<u16>` but not +/// By default, usize implements `From<u8>` and `From<u16>` but not /// `From<u32>` and `From<u64>` because that would be invalid on some /// platforms. This trait implements the conversion for platforms /// with 32 and 64 bits pointer platforms @@ -1138,9 +1571,9 @@ impl ToUsize for u64 { } } -/// equivalent From implementation to avoid orphan rules in bits parsers +/// Equivalent From implementation to avoid orphan rules in bits parsers pub trait ErrorConvert<E> { - /// transform to another error type + /// Transform to another error type fn convert(self) -> E; } @@ -1156,6 +1589,45 @@ impl<I> ErrorConvert<((I, usize), ErrorKind)> for (I, ErrorKind) { } } +use crate::error; +impl<I> ErrorConvert<error::Error<I>> for error::Error<(I, usize)> { + fn convert(self) -> error::Error<I> { + error::Error { + input: self.input.0, + code: self.code, + } + } +} + +impl<I> ErrorConvert<error::Error<(I, usize)>> for error::Error<I> { + fn convert(self) -> error::Error<(I, usize)> { + error::Error { + input: (self.input, 0), + code: self.code, + } + } +} + +#[cfg(feature = "alloc")] +#[cfg_attr(feature = "docsrs", doc(cfg(feature = "alloc")))] +impl<I> ErrorConvert<error::VerboseError<I>> for error::VerboseError<(I, usize)> { + fn convert(self) -> error::VerboseError<I> { + error::VerboseError { + errors: self.errors.into_iter().map(|(i, e)| (i.0, e)).collect(), + } + } +} + +#[cfg(feature = "alloc")] +#[cfg_attr(feature = "docsrs", doc(cfg(feature = "alloc")))] +impl<I> ErrorConvert<error::VerboseError<(I, usize)>> for error::VerboseError<I> { + fn convert(self) -> error::VerboseError<(I, usize)> { + error::VerboseError { + errors: self.errors.into_iter().map(|(i, e)| ((i, 0), e)).collect(), + } + } +} + #[cfg(test)] mod tests { use super::*; diff --git a/src/util.rs b/src/util.rs index 488bbfb..c48f808 100644 --- a/src/util.rs +++ b/src/util.rs @@ -4,19 +4,20 @@ use crate::internal::IResult; use std::fmt::Debug; #[cfg(feature = "std")] +#[cfg_attr(feature = "docsrs", doc(cfg(feature = "std")))] /// Helper trait to show a byte slice as a hex dump pub trait HexDisplay { /// Converts the value of `self` to a hex dump, returning the owned - /// string. + /// `String`. fn to_hex(&self, chunk_size: usize) -> String; /// Converts the value of `self` to a hex dump beginning at `from` address, returning the owned - /// string. + /// `String`. fn to_hex_from(&self, chunk_size: usize, from: usize) -> String; } #[cfg(feature = "std")] -static CHARS: &'static [u8] = b"0123456789abcdef"; +static CHARS: &[u8] = b"0123456789abcdef"; #[cfg(feature = "std")] impl HexDisplay for [u8] { @@ -97,8 +98,7 @@ macro_rules! nom_stringify ( ($($args:tt)*) => (stringify!($($args)*)); ); - -/// Prints a message if the parser fails +/// Prints a message if the parser fails. /// /// The message prints the `Error` or `Incomplete` /// and the parser's calling code @@ -106,7 +106,7 @@ macro_rules! nom_stringify ( /// ``` /// # #[macro_use] extern crate nom; /// # fn main() { -/// named!(f, dbg!( tag!( "abcd" ) ) ); +/// named!(f, dbg_basic!( tag!( "abcd" ) ) ); /// /// let a = &b"efgh"[..]; /// @@ -116,7 +116,7 @@ macro_rules! nom_stringify ( /// # } /// ``` #[macro_export(local_inner_macros)] -macro_rules! dbg ( +macro_rules! dbg_basic ( ($i: expr, $submac:ident!( $($args:tt)* )) => ( { use $crate::lib::std::result::Result::*; @@ -132,11 +132,11 @@ macro_rules! dbg ( ); ($i:expr, $f:ident) => ( - dbg!($i, call!($f)); + dbg_basic!($i, call!($f)); ); ); -/// Prints a message and the input if the parser fails +/// Prints a message and the input if the parser fails. /// /// The message prints the `Error` or `Incomplete` /// and the parser's calling code. @@ -158,20 +158,24 @@ macro_rules! dbg ( /// f(a); /// ``` #[cfg(feature = "std")] -pub fn dbg_dmp<'a, F, O, E: Debug>(f: F, context: &'static str) -> impl Fn(&'a [u8]) -> IResult<&'a [u8], O, E> - where F: Fn(&'a [u8]) -> IResult<&'a [u8], O, E> { - move |i: &'a [u8]| { - match f(i) { - Err(e) => { - println!("{}: Error({:?}) at:\n{}", context, e, i.to_hex(8)); - Err(e) - }, - a => a, - } +#[cfg_attr(feature = "docsrs", doc(cfg(feature = "std")))] +pub fn dbg_dmp<'a, F, O, E: Debug>( + f: F, + context: &'static str, +) -> impl Fn(&'a [u8]) -> IResult<&'a [u8], O, E> +where + F: Fn(&'a [u8]) -> IResult<&'a [u8], O, E>, +{ + move |i: &'a [u8]| match f(i) { + Err(e) => { + println!("{}: Error({:?}) at:\n{}", context, e, i.to_hex(8)); + Err(e) + } + a => a, } } -/// Prints a message and the input if the parser fails +/// Prints a message and the input if the parser fails. /// /// The message prints the `Error` or `Incomplete` /// and the parser's calling code. @@ -192,6 +196,7 @@ pub fn dbg_dmp<'a, F, O, E: Debug>(f: F, context: &'static str) -> impl Fn(&'a [ /// # } #[macro_export(local_inner_macros)] #[cfg(feature = "std")] +#[cfg_attr(feature = "docsrs", doc(cfg(feature = "std")))] macro_rules! dbg_dmp ( ($i: expr, $submac:ident!( $($args:tt)* )) => ( { @@ -211,4 +216,3 @@ macro_rules! dbg_dmp ( dbg_dmp!($i, call!($f)); ); ); - diff --git a/src/whitespace.rs b/src/whitespace.rs deleted file mode 100644 index 9ed3f55..0000000 --- a/src/whitespace.rs +++ /dev/null @@ -1,1077 +0,0 @@ -//! Support for whitespace delimited formats -//! -//! a lot of textual formats allows spaces and other -//! types of separators between tokens. Handling it -//! manually with nom means wrapping all parsers -//! like this: -//! -//! ```ignore -//! named!(token, delimited!(space, tk, space)); -//! ``` -//! -//! To ease the development of such parsers, you -//! can use the whitespace parsing facility, which works -//! as follows: -//! -//! ``` -//! # #[macro_use] extern crate nom; -//! # fn main() { -//! named!(tuple<&[u8], (&[u8], &[u8]) >, -//! ws!(tuple!( take!(3), tag!("de") )) -//! ); -//! -//! assert_eq!( -//! tuple(&b" \t abc de fg"[..]), -//! Ok((&b"fg"[..], (&b"abc"[..], &b"de"[..]))) -//! ); -//! # } -//! ``` -//! -//! The `ws!` combinator will modify the parser to -//! intersperse space parsers everywhere. By default, -//! it will consume the following characters: `" \t\r\n"`. -//! -//! If you want to modify that behaviour, you can make -//! your own whitespace wrapper. As an example, if -//! you don't want to consume ends of lines, only -//! spaces and tabs, you can do it like this: -//! -//! ``` -//! # #[macro_use] extern crate nom; -//! named!(pub space, eat_separator!(&b" \t"[..])); -//! -//! #[macro_export] -//! macro_rules! sp ( -//! ($i:expr, $($args:tt)*) => ( -//! { -//! use nom::Err; -//! -//! match sep!($i, space, $($args)*) { -//! Err(e) => Err(e), -//! Ok((i1,o)) => { -//! match space(i1) { -//! Err(e) => Err(Err::convert(e)), -//! Ok((i2,_)) => Ok((i2, o)) -//! } -//! } -//! } -//! } -//! ) -//! ); -//! -//! # fn main() { -//! named!(tuple<&[u8], (&[u8], &[u8]) >, -//! sp!(tuple!( take!(3), tag!("de") )) -//! ); -//! -//! assert_eq!( -//! tuple(&b" \t abc de fg"[..]), -//! Ok((&b"fg"[..], (&b"abc"[..], &b"de"[..]))) -//! ); -//! # } -//! ``` -//! -//! This combinator works by replacing each combinator with -//! a version that supports wrapping with separator parsers. -//! It will not support the combinators you wrote in your -//! own code. You can still manually wrap them with the separator -//! you want, or you can copy the macros defined in src/whitespace.rs -//! and modify them to support a new combinator: -//! -//! * copy the combinator's code here, add the _sep suffix -//! * add the `$separator:expr` as second argument -//! * wrap any sub parsers with sep!($separator, $submac!($($args)*)) -//! * reference it in the definition of `sep!` as follows: -//! -//! ```ignore -//! ($i:expr, $separator:path, my_combinator ! ($($rest:tt)*) ) => { -//! wrap_sep!($i, -//! $separator, -//! my_combinator_sep!($separator, $($rest)*) -//! ) -//! }; -//! ``` -//! - -/// applies the separator parser before the other parser -#[macro_export(local_inner_macros)] -macro_rules! wrap_sep ( - ($i:expr, $separator:expr, $submac:ident!( $($args:tt)* )) => ({ - use $crate::lib::std::result::Result::*; - use $crate::{Err,IResult}; - - fn unify_types<I,O,P,E>(_: &IResult<I,O,E>, _: &IResult<I,P,E>) {} - - let sep_res = ($separator)($i); - match sep_res { - Ok((i1,_)) => { - let res = $submac!(i1, $($args)*); - unify_types(&sep_res, &res); - res - }, - Err(e) => Err(Err::convert(e)), - } - }); - ($i:expr, $separator:expr, $f:expr) => ( - wrap_sep!($i, $separator, call!($f)) - ); -); - -#[doc(hidden)] -#[macro_export(local_inner_macros)] -macro_rules! pair_sep ( - ($i:expr, $separator:path, $submac:ident!( $($args:tt)* ), $submac2:ident!( $($args2:tt)* )) => ( - tuple!( - $i, - sep!($separator, $submac!($($args)*)), - sep!($separator, $submac2!($($args2)*)) - ) - ); - ($i:expr, $separator:path, $submac:ident!( $($args:tt)* ), $g:expr) => ( - pair_sep!($i, $separator, $submac!($($args)*), call!($g)); - ); - ($i:expr, $separator:path, $f:expr, $submac:ident!( $($args:tt)* )) => ( - pair_sep!($i, $separator, call!($f), $submac!($($args)*)); - ); - ($i:expr, $separator:path, $f:expr, $g:expr) => ( - pair_sep!($i, $separator, call!($f), call!($g)); - ); -); - -#[doc(hidden)] -#[macro_export(local_inner_macros)] -macro_rules! delimited_sep ( - ($i:expr, $separator:path, $submac1:ident!( $($args1:tt)* ), $($rest:tt)+) => ({ - use $crate::lib::std::result::Result::*; - - match tuple_sep!($i, $separator, (), $submac1!($($args1)*), $($rest)+) { - Err(e) => Err(e), - Ok((remaining, (_,o,_))) => { - Ok((remaining, o)) - } - } - }); - ($i:expr, $separator:path, $f:expr, $($rest:tt)+) => ( - delimited_sep!($i, $separator, call!($f), $($rest)+); - ); -); - -#[doc(hidden)] -#[macro_export(local_inner_macros)] -macro_rules! separated_pair_sep ( - ($i:expr, $separator:path, $submac1:ident!( $($args1:tt)* ), $($rest:tt)+) => ({ - use $crate::lib::std::result::Result::*; - - match tuple_sep!($i, $separator, (), $submac1!($($args1)*), $($rest)+) { - Err(e) => Err(e), - Ok((remaining, (o1,_,o2))) => { - Ok((remaining, (o1,o2))) - } - } - }); - ($i:expr, $separator:path, $f:expr, $($rest:tt)+) => ( - separated_pair_sep!($i, $separator, call!($f), $($rest)+); - ); -); - -#[doc(hidden)] -#[macro_export(local_inner_macros)] -macro_rules! preceded_sep ( - ($i:expr, $separator:path, $submac:ident!( $($args:tt)* ), $submac2:ident!( $($args2:tt)* )) => ({ - use $crate::lib::std::result::Result::*; - - match pair_sep!($i, $separator, $submac!($($args)*), $submac2!($($args2)*)) { - Err(e) => Err(e), - Ok((remaining, (_,o))) => { - Ok((remaining, o)) - } - } - }); - ($i:expr, $separator:path, $submac:ident!( $($args:tt)* ), $g:expr) => ( - preceded_sep!($i, $separator, $submac!($($args)*), call!($g)); - ); - ($i:expr, $separator:path, $f:expr, $submac:ident!( $($args:tt)* )) => ( - preceded_sep!($i, $separator, call!($f), $submac!($($args)*)); - ); - ($i:expr, $separator:path, $f:expr, $g:expr) => ( - preceded_sep!($i, $separator, call!($f), call!($g)); - ); -); - -#[doc(hidden)] -#[macro_export(local_inner_macros)] -macro_rules! terminated_sep ( - ($i:expr, $separator:path, $submac:ident!( $($args:tt)* ), $submac2:ident!( $($args2:tt)* )) => ({ - use $crate::lib::std::result::Result::*; - - match pair_sep!($i, $separator, $submac!($($args)*), $submac2!($($args2)*)) { - Err(e) => Err(e), - Ok((remaining, (o,_))) => { - Ok((remaining, o)) - } - } - }); - ($i:expr, $separator:path, $submac:ident!( $($args:tt)* ), $g:expr) => ( - terminated_sep!($i, $separator, $submac!($($args)*), call!($g)); - ); - ($i:expr, $separator:path, $f:expr, $submac:ident!( $($args:tt)* )) => ( - terminated_sep!($i, $separator, call!($f), $submac!($($args)*)); - ); - ($i:expr, $separator:path, $f:expr, $g:expr) => ( - terminated_sep!($i, $separator, call!($f), call!($g)); - ); -); - -/// Internal parser, do not use directly -#[doc(hidden)] -#[macro_export(local_inner_macros)] -macro_rules! tuple_sep ( - ($i:expr, $separator:path, ($($parsed:tt),*), $e:path, $($rest:tt)*) => ( - tuple_sep!($i, $separator, ($($parsed),*), call!($e), $($rest)*); - ); - ($i:expr, $separator:path, (), $submac:ident!( $($args:tt)* ), $($rest:tt)*) => ( - { - use $crate::lib::std::result::Result::*; - - match sep!($i, $separator, $submac!($($args)*)) { - Err(e) => Err(e), - Ok((i,o)) => { - tuple_sep!(i, $separator, (o), $($rest)*) - } - } - } - ); - ($i:expr, $separator:path, ($($parsed:tt)*), $submac:ident!( $($args:tt)* ), $($rest:tt)*) => ( - { - use $crate::lib::std::result::Result::*; - - match sep!($i, $separator, $submac!($($args)*)) { - Err(e) => Err(e), - Ok((i,o)) => { - tuple_sep!(i, $separator, ($($parsed)* , o), $($rest)*) - } - } - } - ); - ($i:expr, $separator:path, ($($parsed:tt),*), $e:path) => ( - tuple_sep!($i, $separator, ($($parsed),*), call!($e)); - ); - ($i:expr, $separator:path, (), $submac:ident!( $($args:tt)* )) => ( - { - use $crate::lib::std::result::Result::*; - - match sep!($i, $separator, $submac!($($args)*)) { - Err(e) => Err(e), - Ok((i,o)) => { - Ok((i, (o))) - } - } - } - ); - ($i:expr, $separator:path, ($($parsed:expr),*), $submac:ident!( $($args:tt)* )) => ( - { - use $crate::lib::std::result::Result::*; - - match sep!($i, $separator, $submac!($($args)*)) { - Err(e) => Err(e), - Ok((i,o)) => { - Ok((i, ($($parsed),* , o))) - } - } - } - ); - ($i:expr, $separator:path, ($($parsed:expr),*)) => ( - { - ::sts::result::Result::Ok(($i, ($($parsed),*))) - } - ); -); - -#[doc(hidden)] -#[macro_export(local_inner_macros)] -macro_rules! do_parse_sep ( - (__impl $i:expr, $separator:path, ( $($rest:expr),* )) => ( - $crate::lib::std::result::Result::Ok(($i, ( $($rest),* ))) - ); - - (__impl $i:expr, $separator:path, $e:ident >> $($rest:tt)*) => ( - do_parse_sep!(__impl $i, $separator, call!($e) >> $($rest)*); - ); - (__impl $i:expr, $separator:path, $submac:ident!( $($args:tt)* ) >> $($rest:tt)*) => ( - { - use $crate::lib::std::result::Result::*; - - match sep!($i, $separator, $submac!($($args)*)) { - Err(e) => Err(e), - Ok((i,_)) => { - do_parse_sep!(__impl i, $separator, $($rest)*) - }, - } - } - ); - - (__impl $i:expr, $separator:path, $field:ident : $e:ident >> $($rest:tt)*) => ( - do_parse_sep!(__impl $i, $separator, $field: call!($e) >> $($rest)*); - ); - - (__impl $i:expr, $separator:path, $field:ident : $submac:ident!( $($args:tt)* ) >> $($rest:tt)*) => ( - { - use $crate::lib::std::result::Result::*; - - match sep!($i, $separator, $submac!($($args)*)) { - Err(e) => Err(e), - Ok((i,o)) => { - let $field = o; - do_parse_sep!(__impl i, $separator, $($rest)*) - }, - } - } - ); - - // ending the chain - (__impl $i:expr, $separator:path, $e:ident >> ( $($rest:tt)* )) => ( - do_parse_sep!(__impl $i, $separator, call!($e) >> ( $($rest)* )); - ); - - (__impl $i:expr, $separator:path, $submac:ident!( $($args:tt)* ) >> ( $($rest:tt)* )) => ({ - use $crate::lib::std::result::Result::*; - - match sep!($i, $separator, $submac!($($args)*)) { - Err(e) => Err(e), - Ok((i,_)) => { - Ok((i, ( $($rest)* ))) - }, - } - }); - - (__impl $i:expr, $separator:path, $field:ident : $e:ident >> ( $($rest:tt)* )) => ( - do_parse_sep!(__impl $i, $separator, $field: call!($e) >> ( $($rest)* ) ); - ); - - (__impl $i:expr, $separator:path, $field:ident : $submac:ident!( $($args:tt)* ) >> ( $($rest:tt)* )) => ({ - use $crate::lib::std::result::Result::*; - - match sep!($i, $separator, $submac!($($args)*)) { - Err(e) => Err(e), - Ok((i,o)) => { - let $field = o; - Ok((i, ( $($rest)* ))) - }, - } - }); - - ($i:expr, $separator:path, $($rest:tt)*) => ( - { - do_parse_sep!(__impl $i, $separator, $($rest)*) - } - ); -); - -#[doc(hidden)] -#[macro_export(local_inner_macros)] -macro_rules! permutation_sep ( - ($i:expr, $separator:path, $($rest:tt)*) => ( - { - use $crate::lib::std::result::Result::*; - use $crate::lib::std::option::Option::*; - use $crate::{Err,error::ErrorKind}; - - let mut res = permutation_init!((), $($rest)*); - let mut input = $i; - let mut error = None; - let mut needed = None; - - loop { - let mut all_done = true; - permutation_iterator_sep!(0, input, $separator, all_done, needed, res, $($rest)*); - - //if we reach that part, it means none of the parsers were able to read anything - if !all_done { - //FIXME: should wrap the error returned by the child parser - error = Option::Some(error_position!(input, ErrorKind::Permutation)); - } - break; - } - - if let Some(need) = needed { - Err(Err::convert(need)) - } else { - if let Some(unwrapped_res) = { permutation_unwrap!(0, (), res, $($rest)*) } { - Ok((input, unwrapped_res)) - } else { - if let Some(e) = error { - Err(Err::Error(error_node_position!($i, ErrorKind::Permutation, e))) - } else { - Err(Err::Error(error_position!($i, ErrorKind::Permutation))) - } - } - } - } - ); -); - -#[doc(hidden)] -#[macro_export(local_inner_macros)] -macro_rules! permutation_iterator_sep ( - ($it:tt,$i:expr, $separator:path, $all_done:expr, $needed:expr, $res:expr, $e:ident?, $($rest:tt)*) => ( - permutation_iterator_sep!($it, $i, $separator, $all_done, $needed, $res, call!($e), $($rest)*); - ); - ($it:tt,$i:expr, $separator:path, $all_done:expr, $needed:expr, $res:expr, $e:ident, $($rest:tt)*) => ( - permutation_iterator_sep!($it, $i, $separator, $all_done, $needed, $res, call!($e), $($rest)*); - ); - - ($it:tt, $i:expr, $separator:path, $all_done:expr, $needed:expr, $res:expr, $submac:ident!( $($args:tt)* )?, $($rest:tt)*) => ({ - permutation_iterator_sep!($it, $i, $separator, $all_done, $needed, $res, $submac!($($args)*), $($rest)*); - }); - ($it:tt, $i:expr, $separator:path, $all_done:expr, $needed:expr, $res:expr, $submac:ident!( $($args:tt)* ), $($rest:tt)*) => ({ - use $crate::lib::std::result::Result::*; - use $crate::Err; - - if $res.$it == $crate::lib::std::option::Option::None { - match {sep!($i, $separator, $submac!($($args)*))} { - Ok((i,o)) => { - $i = i; - $res.$it = $crate::lib::std::option::Option::Some(o); - continue; - }, - Err(Err::Error(_)) => { - $all_done = false; - }, - Err(e) => { - $needed = $crate::lib::std::option::Option::Some(e); - break; - } - }; - } - succ!($it, permutation_iterator_sep!($i, $separator, $all_done, $needed, $res, $($rest)*)); - }); - - ($it:tt,$i:expr, $separator:path, $all_done:expr, $needed:expr, $res:expr, $e:ident?) => ( - permutation_iterator_sep!($it, $i, $separator, $all_done, $res, call!($e)); - ); - ($it:tt,$i:expr, $separator:path, $all_done:expr, $needed:expr, $res:expr, $e:ident) => ( - permutation_iterator_sep!($it, $i, $separator, $all_done, $res, call!($e)); - ); - - ($it:tt, $i:expr, $separator:path, $all_done:expr, $needed:expr, $res:expr, $submac:ident!( $($args:tt)* )?) => ({ - permutation_iterator_sep!($it, $i, $separator, $all_done, $needed, $res, $submac!($($args)*)); - }); - ($it:tt, $i:expr, $separator:path, $all_done:expr, $needed:expr, $res:expr, $submac:ident!( $($args:tt)* )) => ({ - use $crate::lib::std::result::Result::*; - use $crate::Err; - - if $res.$it == $crate::lib::std::option::Option::None { - match sep!($i, $separator, $submac!($($args)*)) { - Ok((i,o)) => { - $i = i; - $res.$it = $crate::lib::std::option::Option::Some(o); - continue; - }, - Err(Err::Error(_)) => { - $all_done = false; - }, - Err(e) => { - $needed = $crate::lib::std::option::Option::Some(e); - break; - } - }; - } - }); -); - -#[doc(hidden)] -#[macro_export(local_inner_macros)] -macro_rules! alt_sep ( - (__impl $i:expr, $separator:path, $e:path | $($rest:tt)*) => ( - alt_sep!(__impl $i, $separator, call!($e) | $($rest)*); - ); - - (__impl $i:expr, $separator:path, $subrule:ident!( $($args:tt)*) | $($rest:tt)*) => ( - { - use $crate::lib::std::result::Result::*; - use $crate::Err; - - let res = sep!($i, $separator, $subrule!($($args)*)); - match res { - Ok((_,_)) => res, - Err(Err::Error(_)) => alt_sep!(__impl $i, $separator, $($rest)*), - Err(e) => Err(e), - } - } - ); - - (__impl $i:expr, $separator:path, $subrule:ident!( $($args:tt)* ) => { $gen:expr } | $($rest:tt)+) => ( - { - use $crate::lib::std::result::Result::*; - use $crate::Err; - - match sep!($i, $separator, $subrule!( $($args)* )) { - Ok((i,o)) => Ok((i,$gen(o))), - Err(Err::Error(_)) => { - alt_sep!(__impl $i, $separator, $($rest)+) - }, - Err(e) => Err(e), - } - } - ); - - (__impl $i:expr, $separator:path, $e:path => { $gen:expr } | $($rest:tt)*) => ( - alt_sep!(__impl $i, $separator, call!($e) => { $gen } | $($rest)*); - ); - - (__impl $i:expr, $separator:path, $e:path => { $gen:expr }) => ( - alt_sep!(__impl $i, $separator, call!($e) => { $gen }); - ); - - (__impl $i:expr, $separator:path, $subrule:ident!( $($args:tt)* ) => { $gen:expr }) => ( - { - use $crate::lib::std::result::Result::*; - use $crate::Err; - - match sep!($i, $separator, $subrule!( $($args)* )) { - Ok((i,o)) => Ok((i,$gen(o))), - Err(Err::Error(e)) => { - fn unify_types<T>(_: &T, _: &T) {} - let e2 = error_position!($i, $crate::error::ErrorKind::Alt); - unify_types(&e, &e2); - Err(Err::Error(e2)) - }, - Err(e) => Err(e), - } - } - ); - - (__impl $i:expr, $separator:path, $e:path) => ( - alt_sep!(__impl $i, $separator, call!($e)); - ); - - (__impl $i:expr, $separator:path, $subrule:ident!( $($args:tt)*)) => ( - { - use $crate::lib::std::result::Result::*; - use $crate::Err; - - match sep!($i, $separator, $subrule!( $($args)* )) { - Ok((i,o)) => Ok((i,o)), - Err(Err::Error(e)) => { - fn unify_types<T>(_: &T, _: &T) {} - let e2 = error_position!($i, $crate::error::ErrorKind::Alt); - unify_types(&e, &e2); - Err(Err::Error(e2)) - }, - Err(e) => Err(e), - } - } - ); - - (__impl $i:expr) => ({ - use $crate::lib::std::result::Result::*; - use $crate::{Err,Needed,IResult}; - - Err(Err::Error(error_position!($i, $crate::error::ErrorKind::Alt))) - }); - - (__impl $i:expr, $separator:path) => ({ - use $crate::lib::std::result::Result::*; - use $crate::{Err,Needed,IResult}; - - Err(Err::Error(error_position!($i, $crate::error::ErrorKind::Alt))) - }); - - ($i:expr, $separator:path, $($rest:tt)*) => ( - { - alt_sep!(__impl $i, $separator, $($rest)*) - } - ); -); - -#[doc(hidden)] -#[macro_export(local_inner_macros)] -macro_rules! switch_sep ( - (__impl $i:expr, $separator:path, $submac:ident!( $($args:tt)* ), $($p:pat => $subrule:ident!( $($args2:tt)* ))|* ) => ( - { - use $crate::lib::std::result::Result::*; - use $crate::Err; - - match sep!($i, $separator, $submac!($($args)*)) { - Err(Err::Error(e)) => Err(Err::Error(error_node_position!( - $i, $crate::error::ErrorKind::Switch, e - ))), - Err(Err::Failure(e)) => Err(Err::Failure( - error_node_position!($i, $crate::error::ErrorKind::Switch, e))), - Err(e) => Err(e), - Ok((i, o)) => { - match o { - $($p => match sep!(i, $separator, $subrule!($($args2)*)) { - Err(Err::Error(e)) => Err(Err::Error(error_node_position!( - $i, $crate::error::ErrorKind::Switch, e - ))), - Err(Err::Failure(e)) => Err(Err::Failure( - error_node_position!($i, $crate::error::ErrorKind::Switch, e))), - a => a, - }),*, - _ => Err(Err::Error(error_position!($i, $crate::error::ErrorKind::Switch))) - } - } - } - } - ); - ($i:expr, $separator:path, $submac:ident!( $($args:tt)*), $($rest:tt)*) => ( - { - switch_sep!(__impl $i, $separator, $submac!($($args)*), $($rest)*) - } - ); - ($i:expr, $separator:path, $e:path, $($rest:tt)*) => ( - { - switch_sep!(__impl $i, $separator, call!($e), $($rest)*) - } - ); -); - -#[doc(hidden)] -#[cfg(feature = "alloc")] -#[macro_export(local_inner_macros)] -macro_rules! separated_list_sep ( - ($i:expr, $separator:path, $submac:ident!( $($args:tt)* ), $submac2:ident!( $($args2:tt)* )) => ( - separated_list!( - $i, - sep!($separator, $submac!($($args)*)), - sep!($separator, $submac2!($($args2)*)) - ) - ); - ($i:expr, $separator:path, $submac:ident!( $($args:tt)* ), $g:expr) => ( - separated_list_sep!($i, $separator, $submac!($($args)*), call!($g)); - ); - ($i:expr, $separator:path, $f:expr, $submac:ident!( $($args:tt)* )) => ( - separated_list_sep!($i, $separator, call!($f), $submac!($($args)*)); - ); - ($i:expr, $separator:path, $f:expr, $g:expr) => ( - separated_list_sep!($i, $separator, call!($f), call!($g)); - ); -); - -/// helper macros to build a separator parser -/// -/// ``` -/// # #[macro_use] extern crate nom; -/// named!(pub space, eat_separator!(&b" \t"[..])); -/// # fn main() {} -/// ``` -#[macro_export(local_inner_macros)] -macro_rules! eat_separator ( - ($i:expr, $arr:expr) => ( - { - use $crate::{FindToken, InputTakeAtPosition}; - let input = $i; - input.split_at_position(|c| !$arr.find_token(c)) - } - ); -); - -/// sep is the parser rewriting macro for whitespace separated formats -/// -/// it takes as argument a space eating function and a parser tree, -/// and will intersperse the space parser everywhere -/// -/// ```ignore -/// #[macro_export(local_inner_macros)] -/// macro_rules! ws ( -/// ($i:expr, $($args:tt)*) => ( -/// { -/// use sp; -/// sep!($i, sp, $($args)*) -/// } -/// ) -/// ); -/// ``` -#[macro_export(local_inner_macros)] -macro_rules! sep ( - ($i:expr, $separator:path, tuple ! ($($rest:tt)*) ) => { - tuple_sep!($i, $separator, (), $($rest)*) - }; - ($i:expr, $separator:path, pair ! ($($rest:tt)*) ) => { - wrap_sep!($i, - $separator, - pair_sep!($separator, $($rest)*) - ) - }; - ($i:expr, $separator:path, delimited ! ($($rest:tt)*) ) => { - wrap_sep!($i, - $separator, - delimited_sep!($separator, $($rest)*) - ) - }; - ($i:expr, $separator:path, separated_pair ! ($($rest:tt)*) ) => { - wrap_sep!($i, - $separator, - separated_pair_sep!($separator, $($rest)*) - ) - }; - ($i:expr, $separator:path, preceded ! ($($rest:tt)*) ) => { - wrap_sep!($i, - $separator, - preceded_sep!($separator, $($rest)*) - ) - }; - ($i:expr, $separator:path, terminated ! ($($rest:tt)*) ) => { - wrap_sep!($i, - $separator, - terminated_sep!($separator, $($rest)*) - ) - }; - ($i:expr, $separator:path, do_parse ! ($($rest:tt)*) ) => { - wrap_sep!($i, - $separator, - do_parse_sep!($separator, $($rest)*) - ) - }; - ($i:expr, $separator:path, permutation ! ($($rest:tt)*) ) => { - wrap_sep!($i, - $separator, - permutation_sep!($separator, $($rest)*) - ) - }; - ($i:expr, $separator:path, alt ! ($($rest:tt)*) ) => { - wrap_sep!($i, - $separator, - alt_sep!($separator, $($rest)*) - ) - }; - ($i:expr, $separator:path, switch ! ($($rest:tt)*) ) => { - wrap_sep!($i, - $separator, - switch_sep!($separator, $($rest)*) - ) - }; - ($i:expr, $separator:path, separated_list ! ($($rest:tt)*) ) => { - wrap_sep!($i, - $separator, - separated_list_sep!($separator, $($rest)*) - ) - }; - ($i:expr, $separator:path, many0 ! ($($rest:tt)*) ) => { - many0!($i, wrap_sep!($separator, $($rest)*)) - }; - ($i:expr, $separator:path, many1 ! ($($rest:tt)*) ) => { - many1!($i, wrap_sep!($separator, $($rest)*)) - }; - ($i:expr, $separator:path, return_error!( $($args:tt)* )) => { - return_error!($i, wrap_sep!($separator, $($args)*)) - }; -//FIXME: missing separated_nonempty_list, -// many_till, many_m_n, count, count_fixed, fold_many0, fold_many1, -// fold_many_m_n - ($i:expr, $separator:path, $submac:ident!( $($args:tt)* )) => { - wrap_sep!($i, $separator, $submac!($($args)*)) - }; - ($i:expr, $separator:path, $f:expr) => { - wrap_sep!($i, $separator, call!($f)) - }; -); - -/// `ws!(I -> IResult<I,O>) => I -> IResult<I, O>` -/// -/// transforms a parser to automatically consume -/// whitespace between each token. By default, -/// it takes the following characters: `" \t\r\n"`. -/// -/// If you need a whitespace parser consuming a -/// different set of characters, you can make -/// your own by reusing the `sep!` combinator. -/// -/// To use `ws!`, pass your parser as argument: -/// -/// ``` -/// # #[macro_use] extern crate nom; -/// # fn main() { -/// named!(tuple<&[u8], (&[u8], &[u8]) >, -/// ws!(tuple!( take!(3), tag!("de") )) -/// ); -/// -/// assert_eq!( -/// tuple(&b" \t abc de fg"[..]), -/// Ok((&b"fg"[..], (&b"abc"[..], &b"de"[..]))) -/// ); -/// # } -/// ``` -/// -#[macro_export(local_inner_macros)] -#[deprecated(since = "5.0.0", note = "whitespace parsing only works with macros and will not be updated anymore")] -macro_rules! ws ( - ($i:expr, $($args:tt)*) => ( - { - use $crate::Err; - use $crate::lib::std::result::Result::*; - use $crate::character::complete::multispace0; - - match sep!($i, multispace0, $($args)*) { - Err(e) => Err(e), - Ok((i1,o)) => { - match (multispace0)(i1) { - Err(e) => Err(Err::convert(e)), - Ok((i2,_)) => Ok((i2, o)) - } - } - } - } - ) -); - -#[cfg(test)] -#[allow(dead_code)] -mod tests { - #[cfg(feature = "alloc")] - use crate::{ - error::ParseError, - lib::std::{ - string::{String, ToString}, - fmt::Debug - } - }; - use crate::internal::{Err, IResult, Needed}; - use crate::character::complete::multispace0 as sp; - use crate::error::ErrorKind; - - #[test] - fn spaaaaace() { - assert_eq!(sp::<_,(_,ErrorKind)>(&b" \t abc "[..]), Ok((&b"abc "[..], &b" \t "[..]))); - } - - #[test] - fn tag() { - named!(abc, ws!(tag!("abc"))); - - assert_eq!(abc(&b" \t abc def"[..]), Ok((&b"def"[..], &b"abc"[..]))); - } - - #[test] - fn pair() { - named!(pair_2<&[u8], (&[u8], &[u8]) >, - ws!(pair!( take!(3), tag!("de") )) - ); - - assert_eq!( - pair_2(&b" \t abc de fg"[..]), - Ok((&b"fg"[..], (&b"abc"[..], &b"de"[..]))) - ); - } - - #[test] - fn preceded() { - named!(prec<&[u8], &[u8] >, - ws!(preceded!( take!(3), tag!("de") )) - ); - - assert_eq!(prec(&b" \t abc de fg"[..]), Ok((&b"fg"[..], &b"de"[..]))); - } - - #[test] - fn terminated() { - named!(term<&[u8], &[u8] >, - ws!(terminated!( take!(3), tag!("de") )) - ); - - assert_eq!(term(&b" \t abc de fg"[..]), Ok((&b"fg"[..], &b"abc"[..]))); - } - - #[test] - fn tuple() { - //trace_macros!(true); - named!(tuple_2<&[u8], (&[u8], &[u8]) >, - ws!(tuple!( take!(3), tag!("de") )) - ); - //trace_macros!(false); - - assert_eq!( - tuple_2(&b" \t abc de fg"[..]), - Ok((&b"fg"[..], (&b"abc"[..], &b"de"[..]))) - ); - } - - #[test] - fn levels() { - //trace_macros!(true); - named!(level_2<&[u8], (&[u8], (&[u8], &[u8])) >, - ws!(pair!(take!(3), tuple!( tag!("de"), tag!("fg ") ))) - ); - //trace_macros!(false); - - assert_eq!( - level_2(&b" \t abc de fg \t hi "[..]), - Ok((&b"hi "[..], (&b"abc"[..], (&b"de"[..], &b"fg "[..])))) - ); - } - - #[test] - fn do_parse() { - fn ret_int1(i: &[u8]) -> IResult<&[u8], u8> { - Ok((i, 1)) - }; - fn ret_int2(i: &[u8]) -> IResult<&[u8], u8> { - Ok((i, 2)) - }; - - //trace_macros!(true); - named!(do_parser<&[u8], (u8, u8)>, - ws!(do_parse!( - tag!("abcd") >> - opt!(tag!("abcd")) >> - aa: ret_int1 >> - tag!("efgh") >> - bb: ret_int2 >> - tag!("efgh") >> - (aa, bb) - )) - ); - - //trace_macros!(false); - - assert_eq!( - do_parser(&b"abcd abcd\tefghefghX"[..]), - Ok((&b"X"[..], (1, 2))) - ); - assert_eq!( - do_parser(&b"abcd\tefgh efgh X"[..]), - Ok((&b"X"[..], (1, 2))) - ); - assert_eq!( - do_parser(&b"abcd ab"[..]), - Err(Err::Incomplete(Needed::Size(4))) - ); - assert_eq!( - do_parser(&b" abcd\tefgh\tef"[..]), - Err(Err::Incomplete(Needed::Size(4))) - ); - } - - #[test] - fn permutation() { - //trace_macros!(true); - named!( - perm<(&[u8], &[u8], &[u8])>, - ws!(permutation!(tag!("abcd"), tag!("efg"), tag!("hi"))) - ); - //trace_macros!(false); - - let expected = (&b"abcd"[..], &b"efg"[..], &b"hi"[..]); - - let a = &b"abcd\tefg \thijk"[..]; - assert_eq!(perm(a), Ok((&b"jk"[..], expected))); - let b = &b" efg \tabcdhi jk"[..]; - assert_eq!(perm(b), Ok((&b"jk"[..], expected))); - let c = &b" hi efg\tabcdjk"[..]; - assert_eq!(perm(c), Ok((&b"jk"[..], expected))); - - let d = &b"efg xyzabcdefghi"[..]; - assert_eq!( - perm(d), - Err(Err::Error(error_node_position!( - &b"efg xyzabcdefghi"[..], - ErrorKind::Permutation, - error_position!(&b" xyzabcdefghi"[..], ErrorKind::Permutation) - ))) - ); - - let e = &b" efg \tabc"[..]; - assert_eq!(perm(e), Err(Err::Incomplete(Needed::Size(4)))); - } - - #[cfg(feature = "alloc")] - #[derive(Debug, Clone, PartialEq)] - pub struct ErrorStr(String); - - #[cfg(feature = "alloc")] - impl<'a> From<(&'a[u8], ErrorKind)> for ErrorStr { - fn from(i: (&'a[u8], ErrorKind)) -> Self { - ErrorStr(format!("custom error code: {:?}", i)) - } - } - - #[cfg(feature = "alloc")] - impl<'a> From<(&'a str, ErrorKind)> for ErrorStr { - fn from(i: (&'a str, ErrorKind)) -> Self { - ErrorStr(format!("custom error message: {:?}", i)) - } - } - - #[cfg(feature = "alloc")] - impl<I: Debug> ParseError<I> for ErrorStr { - fn from_error_kind(input: I, kind: ErrorKind) -> Self { - ErrorStr(format!("custom error message: ({:?}, {:?})", input, kind)) - } - - fn append(input: I, kind: ErrorKind, other: Self) -> Self { - ErrorStr(format!("custom error message: ({:?}, {:?}) - {:?}", input, kind, other)) - } - } - - #[cfg(feature = "alloc")] - #[test] - fn alt() { - fn work(input: &[u8]) -> IResult<&[u8], &[u8], ErrorStr> { - Ok((&b""[..], input)) - } - - #[allow(unused_variables)] - fn dont_work(input: &[u8]) -> IResult<&[u8], &[u8], ErrorStr> { - Err(Err::Error(ErrorStr("abcd".to_string()))) - } - - fn work2(input: &[u8]) -> IResult<&[u8], &[u8], ErrorStr> { - Ok((input, &b""[..])) - } - - fn alt1(i: &[u8]) -> IResult<&[u8], &[u8], ErrorStr> { - alt!(i, dont_work | dont_work) - } - fn alt2(i: &[u8]) -> IResult<&[u8], &[u8], ErrorStr> { - alt!(i, dont_work | work) - } - fn alt3(i: &[u8]) -> IResult<&[u8], &[u8], ErrorStr> { - alt!(i, dont_work | dont_work | work2 | dont_work) - } - - let a = &b"\tabcd"[..]; - assert_eq!( - alt1(a), - Err(Err::Error(error_position!(a, ErrorKind::Alt))) - ); - assert_eq!(alt2(a), Ok((&b""[..], a))); - assert_eq!(alt3(a), Ok((a, &b""[..]))); - - } - - named!(str_parse(&str) -> &str, ws!(tag!("test"))); - #[allow(unused_variables)] - #[test] - fn str_test() { - assert_eq!(str_parse(" \n test\t a\nb"), Ok(("a\nb", "test"))); - } - - // test whitespace parser generation for alt - named!(space, tag!(" ")); - #[cfg(feature = "alloc")] - named!(pipeline_statement<&[u8], ()>, - ws!( - do_parse!( - tag!("pipeline") >> - attributes: delimited!(char!('{'), - separated_list!(char!(','), alt!( - space | - space - )), - char!('}')) >> - - ({ - let _ = attributes; - () - }) - ) - ) - ); - - #[cfg(feature = "alloc")] - named!( - fail<&[u8]>, - map!(many_till!(take!(1), ws!(tag!("."))), |(r, _)| r[0]) - ); -} |