| 1 |
//! Character specific parsers and combinators |
| 2 |
//! |
| 3 |
//! Functions recognizing specific characters |
| 4 |
|
| 5 |
use core::marker::PhantomData; |
| 6 |
|
| 7 |
use crate::error::ErrorKind; |
| 8 |
use crate::FindToken; |
| 9 |
use crate::IsStreaming; |
| 10 |
use crate::Mode; |
| 11 |
use crate::{error::ParseError, AsChar, Err, IResult, Input, Needed, Parser}; |
| 12 |
|
| 13 |
#[cfg(test)] |
| 14 |
mod tests; |
| 15 |
|
| 16 |
pub mod complete; |
| 17 |
pub mod streaming; |
| 18 |
|
| 19 |
#[inline] |
| 20 |
#[doc(hidden)] |
| 21 |
#[deprecated(since = "8.0.0", note = "Replaced with `AsChar::is_alpha`")] |
| 22 |
pub fn is_alphabetic(chr: u8) -> bool { |
| 23 |
matches!(chr, 0x41..=0x5A | 0x61..=0x7A) |
| 24 |
} |
| 25 |
|
| 26 |
#[inline] |
| 27 |
#[doc(hidden)] |
| 28 |
#[deprecated(since = "8.0.0", note = "Replaced with `AsChar::is_dec_digit`")] |
| 29 |
pub fn is_digit(chr: u8) -> bool { |
| 30 |
matches!(chr, 0x30..=0x39) |
| 31 |
} |
| 32 |
|
| 33 |
#[inline] |
| 34 |
#[doc(hidden)] |
| 35 |
#[deprecated(since = "8.0.0", note = "Replaced with `AsChar::is_hex_digit`")] |
| 36 |
pub fn is_hex_digit(chr: u8) -> bool { |
| 37 |
matches!(chr, 0x30..=0x39 | 0x41..=0x46 | 0x61..=0x66) |
| 38 |
} |
| 39 |
|
| 40 |
#[inline] |
| 41 |
#[doc(hidden)] |
| 42 |
#[deprecated(since = "8.0.0", note = "Replaced with `AsChar::is_oct_digit`")] |
| 43 |
pub fn is_oct_digit(chr: u8) -> bool { |
| 44 |
matches!(chr, 0x30..=0x37) |
| 45 |
} |
| 46 |
|
| 47 |
/// Tests if byte is ASCII binary digit: 0-1 |
| 48 |
/// |
| 49 |
/// # Example |
| 50 |
/// |
| 51 |
/// ``` |
| 52 |
/// # use nom::character::is_bin_digit; |
| 53 |
/// assert_eq!(is_bin_digit(b'a'), false); |
| 54 |
/// assert_eq!(is_bin_digit(b'2'), false); |
| 55 |
/// assert_eq!(is_bin_digit(b'0'), true); |
| 56 |
/// assert_eq!(is_bin_digit(b'1'), true); |
| 57 |
/// ``` |
| 58 |
#[inline] |
| 59 |
pub fn is_bin_digit(chr: u8) -> bool { |
| 60 |
matches!(chr, 0x30..=0x31) |
| 61 |
} |
| 62 |
|
| 63 |
#[inline] |
| 64 |
#[doc(hidden)] |
| 65 |
#[deprecated(since = "8.0.0", note = "Replaced with `AsChar::is_alphanum`")] |
| 66 |
pub fn is_alphanumeric(chr: u8) -> bool { |
| 67 |
AsChar::is_alphanum(chr) |
| 68 |
} |
| 69 |
|
| 70 |
#[inline] |
| 71 |
#[doc(hidden)] |
| 72 |
#[deprecated(since = "8.0.0", note = "Replaced with `AsChar::is_space`")] |
| 73 |
pub fn is_space(chr: u8) -> bool { |
| 74 |
chr == b' ' || chr == b'\t' |
| 75 |
} |
| 76 |
|
| 77 |
#[inline] |
| 78 |
#[doc(hidden)] |
| 79 |
#[deprecated(since = "8.0.0", note = "Replaced with `AsChar::is_newline`")] |
| 80 |
pub fn is_newline(chr: u8) -> bool { |
| 81 |
chr == b'\n' |
| 82 |
} |
| 83 |
|
| 84 |
/// Recognizes one character. |
| 85 |
/// |
| 86 |
/// # Example |
| 87 |
/// |
| 88 |
/// ``` |
| 89 |
/// # use nom::{Err, error::{ErrorKind, Error}, Needed, IResult}; |
| 90 |
/// # use nom::character::streaming::char; |
| 91 |
/// fn parser(i: &str) -> IResult<&str, char> { |
| 92 |
/// char('a')(i) |
| 93 |
/// } |
| 94 |
/// assert_eq!(parser("abc"), Ok(("bc", 'a'))); |
| 95 |
/// assert_eq!(parser("bc"), Err(Err::Error(Error::new("bc", ErrorKind::Char)))); |
| 96 |
/// assert_eq!(parser(""), Err(Err::Incomplete(Needed::new(1)))); |
| 97 |
/// ``` |
| 98 |
pub fn char<I, Error: ParseError<I>>(c: char) -> impl Parser<I, Output = char, Error = Error> |
| 99 |
where |
| 100 |
I: Input, |
| 101 |
<I as Input>::Item: AsChar, |
| 102 |
{ |
| 103 |
Char { c, e: PhantomData } |
| 104 |
} |
| 105 |
|
| 106 |
/// Parser implementation for [char()] |
| 107 |
pub struct Char<E> { |
| 108 |
c: char, |
| 109 |
e: PhantomData<E>, |
| 110 |
} |
| 111 |
|
| 112 |
impl<I, Error: ParseError<I>> Parser<I> for Char<Error> |
| 113 |
where |
| 114 |
I: Input, |
| 115 |
<I as Input>::Item: AsChar, |
| 116 |
{ |
| 117 |
type Output = char; |
| 118 |
type Error = Error; |
| 119 |
#[inline(always)] |
| 120 |
fn process<OM: crate::OutputMode>( |
| 121 |
&mut self, |
| 122 |
i: I, |
| 123 |
) -> crate::PResult<OM, I, Self::Output, Self::Error> { |
| 124 |
match (i).iter_elements().next().map(|t| { |
| 125 |
let b = t.as_char() == self.c; |
| 126 |
(&self.c, b) |
| 127 |
}) { |
| 128 |
None => { |
| 129 |
if OM::Incomplete::is_streaming() { |
| 130 |
Err(Err::Incomplete(Needed::new(self.c.len() - i.input_len()))) |
| 131 |
} else { |
| 132 |
Err(Err::Error(OM::Error::bind(|| Error::from_char(i, self.c)))) |
| 133 |
} |
| 134 |
} |
| 135 |
Some((_, false)) => Err(Err::Error(OM::Error::bind(|| Error::from_char(i, self.c)))), |
| 136 |
Some((c, true)) => Ok((i.take_from(c.len()), OM::Output::bind(|| c.as_char()))), |
| 137 |
} |
| 138 |
} |
| 139 |
} |
| 140 |
|
| 141 |
/// Recognizes one character and checks that it satisfies a predicate |
| 142 |
/// |
| 143 |
/// # Example |
| 144 |
/// |
| 145 |
/// ``` |
| 146 |
/// # use nom::{Err, error::{ErrorKind, Error}, Needed, IResult}; |
| 147 |
/// # use nom::character::complete::satisfy; |
| 148 |
/// fn parser(i: &str) -> IResult<&str, char> { |
| 149 |
/// satisfy(|c| c == 'a' || c == 'b')(i) |
| 150 |
/// } |
| 151 |
/// assert_eq!(parser("abc"), Ok(("bc", 'a'))); |
| 152 |
/// assert_eq!(parser("cd"), Err(Err::Error(Error::new("cd", ErrorKind::Satisfy)))); |
| 153 |
/// assert_eq!(parser(""), Err(Err::Error(Error::new("", ErrorKind::Satisfy)))); |
| 154 |
/// ``` |
| 155 |
pub fn satisfy<F, I, Error: ParseError<I>>( |
| 156 |
predicate: F, |
| 157 |
) -> impl Parser<I, Output = char, Error = Error> |
| 158 |
where |
| 159 |
I: Input, |
| 160 |
<I as Input>::Item: AsChar, |
| 161 |
F: Fn(char) -> bool, |
| 162 |
{ |
| 163 |
Satisfy { |
| 164 |
predicate, |
| 165 |
make_error: |i: I| Error::from_error_kind(i, ErrorKind::Satisfy), |
| 166 |
} |
| 167 |
} |
| 168 |
|
| 169 |
/// Parser implementation for [satisfy] |
| 170 |
pub struct Satisfy<F, MakeError> { |
| 171 |
predicate: F, |
| 172 |
make_error: MakeError, |
| 173 |
} |
| 174 |
|
| 175 |
impl<I, Error: ParseError<I>, F, MakeError> Parser<I> for Satisfy<F, MakeError> |
| 176 |
where |
| 177 |
I: Input, |
| 178 |
<I as Input>::Item: AsChar, |
| 179 |
F: Fn(char) -> bool, |
| 180 |
MakeError: Fn(I) -> Error, |
| 181 |
{ |
| 182 |
type Output = char; |
| 183 |
type Error = Error; |
| 184 |
|
| 185 |
#[inline(always)] |
| 186 |
fn process<OM: crate::OutputMode>( |
| 187 |
&mut self, |
| 188 |
i: I, |
| 189 |
) -> crate::PResult<OM, I, Self::Output, Self::Error> { |
| 190 |
match (i).iter_elements().next().map(|t| { |
| 191 |
let c = t.as_char(); |
| 192 |
let b = (self.predicate)(c); |
| 193 |
(c, b) |
| 194 |
}) { |
| 195 |
None => { |
| 196 |
if OM::Incomplete::is_streaming() { |
| 197 |
Err(Err::Incomplete(Needed::Unknown)) |
| 198 |
} else { |
| 199 |
Err(Err::Error(OM::Error::bind(|| (self.make_error)(i)))) |
| 200 |
} |
| 201 |
} |
| 202 |
Some((_, false)) => Err(Err::Error(OM::Error::bind(|| (self.make_error)(i)))), |
| 203 |
Some((c, true)) => Ok((i.take_from(c.len()), OM::Output::bind(|| c.as_char()))), |
| 204 |
} |
| 205 |
} |
| 206 |
} |
| 207 |
|
| 208 |
/// Recognizes one of the provided characters. |
| 209 |
/// |
| 210 |
/// # Example |
| 211 |
/// |
| 212 |
/// ``` |
| 213 |
/// # use nom::{Err, error::ErrorKind}; |
| 214 |
/// # use nom::character::complete::one_of; |
| 215 |
/// assert_eq!(one_of::<_, _, (&str, ErrorKind)>("abc")("b"), Ok(("", 'b'))); |
| 216 |
/// assert_eq!(one_of::<_, _, (&str, ErrorKind)>("a")("bc"), Err(Err::Error(("bc", ErrorKind::OneOf)))); |
| 217 |
/// assert_eq!(one_of::<_, _, (&str, ErrorKind)>("a")(""), Err(Err::Error(("", ErrorKind::OneOf)))); |
| 218 |
/// ``` |
| 219 |
pub fn one_of<I, T, Error: ParseError<I>>(list: T) -> impl Parser<I, Output = char, Error = Error> |
| 220 |
where |
| 221 |
I: Input, |
| 222 |
<I as Input>::Item: AsChar, |
| 223 |
T: FindToken<char>, |
| 224 |
{ |
| 225 |
Satisfy { |
| 226 |
predicate: move |c: char| list.find_token(c), |
| 227 |
make_error: move |i| Error::from_error_kind(i, ErrorKind::OneOf), |
| 228 |
} |
| 229 |
} |
| 230 |
|
| 231 |
//. Recognizes a character that is not in the provided characters. |
| 232 |
/// |
| 233 |
/// # Example |
| 234 |
/// |
| 235 |
/// ``` |
| 236 |
/// # use nom::{Err, error::ErrorKind, Needed}; |
| 237 |
/// # use nom::character::streaming::none_of; |
| 238 |
/// assert_eq!(none_of::<_, _, (_, ErrorKind)>("abc")("z"), Ok(("", 'z'))); |
| 239 |
/// assert_eq!(none_of::<_, _, (_, ErrorKind)>("ab")("a"), Err(Err::Error(("a", ErrorKind::NoneOf)))); |
| 240 |
/// assert_eq!(none_of::<_, _, (_, ErrorKind)>("a")(""), Err(Err::Incomplete(Needed::Unknown))); |
| 241 |
/// ``` |
| 242 |
pub fn none_of<I, T, Error: ParseError<I>>(list: T) -> impl Parser<I, Output = char, Error = Error> |
| 243 |
where |
| 244 |
I: Input, |
| 245 |
<I as Input>::Item: AsChar, |
| 246 |
T: FindToken<char>, |
| 247 |
{ |
| 248 |
Satisfy { |
| 249 |
predicate: move |c: char| !list.find_token(c), |
| 250 |
make_error: move |i| Error::from_error_kind(i, ErrorKind::NoneOf), |
| 251 |
} |
| 252 |
} |
| 253 |
|
| 254 |
// Matches one byte as a character. Note that the input type will |
| 255 |
/// accept a `str`, but not a `&[u8]`, unlike many other nom parsers. |
| 256 |
/// |
| 257 |
/// # Example |
| 258 |
/// |
| 259 |
/// ``` |
| 260 |
/// # use nom::{character::complete::anychar, Err, error::{Error, ErrorKind}, IResult}; |
| 261 |
/// fn parser(input: &str) -> IResult<&str, char> { |
| 262 |
/// anychar(input) |
| 263 |
/// } |
| 264 |
/// |
| 265 |
/// assert_eq!(parser("abc"), Ok(("bc",'a'))); |
| 266 |
/// assert_eq!(parser(""), Err(Err::Error(Error::new("", ErrorKind::Eof)))); |
| 267 |
/// ``` |
| 268 |
pub fn anychar<T, E: ParseError<T>>(input: T) -> IResult<T, char, E> |
| 269 |
where |
| 270 |
T: Input, |
| 271 |
<T as Input>::Item: AsChar, |
| 272 |
{ |
| 273 |
let mut it = input.iter_elements(); |
| 274 |
match it.next() { |
| 275 |
None => Err(Err::Error(E::from_error_kind(input, ErrorKind::Eof))), |
| 276 |
Some(c) => Ok((input.take_from(c.len()), c.as_char())), |
| 277 |
} |
| 278 |
} |
| 279 |
|
| 280 |
/// Parser implementation for char |
| 281 |
pub struct AnyChar<E> { |
| 282 |
e: PhantomData<E>, |
| 283 |
} |
| 284 |
|
| 285 |
impl<I, Error: ParseError<I>> Parser<I> for AnyChar<Error> |
| 286 |
where |
| 287 |
I: Input, |
| 288 |
<I as Input>::Item: AsChar, |
| 289 |
{ |
| 290 |
type Output = char; |
| 291 |
type Error = Error; |
| 292 |
|
| 293 |
fn process<OM: crate::OutputMode>( |
| 294 |
&mut self, |
| 295 |
i: I, |
| 296 |
) -> crate::PResult<OM, I, Self::Output, Self::Error> { |
| 297 |
match (i).iter_elements().next() { |
| 298 |
None => { |
| 299 |
if OM::Incomplete::is_streaming() { |
| 300 |
Err(Err::Incomplete(Needed::new(1))) |
| 301 |
} else { |
| 302 |
Err(Err::Error(OM::Error::bind(|| { |
| 303 |
Error::from_error_kind(i, ErrorKind::Eof) |
| 304 |
}))) |
| 305 |
} |
| 306 |
} |
| 307 |
Some(c) => Ok((i.take_from(c.len()), OM::Output::bind(|| c.as_char()))), |
| 308 |
} |
| 309 |
} |
| 310 |
} |
| 311 |
|
| 312 |
/// Recognizes one or more ASCII numerical characters: 0-9 |
| 313 |
/// |
| 314 |
/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data, |
| 315 |
/// or if no terminating token is found (a non digit character). |
| 316 |
/// # Example |
| 317 |
/// |
| 318 |
/// ``` |
| 319 |
/// # use nom::{Err, error::ErrorKind, IResult, Needed}; |
| 320 |
/// # use nom::character::streaming::digit1; |
| 321 |
/// assert_eq!(digit1::<_, (_, ErrorKind)>("21c"), Ok(("c", "21"))); |
| 322 |
/// assert_eq!(digit1::<_, (_, ErrorKind)>("c1"), Err(Err::Error(("c1", ErrorKind::Digit)))); |
| 323 |
/// assert_eq!(digit1::<_, (_, ErrorKind)>(""), Err(Err::Incomplete(Needed::new(1)))); |
| 324 |
/// ``` |
| 325 |
pub fn digit1<T, E: ParseError<T>>() -> impl Parser<T, Output = T, Error = E> |
| 326 |
where |
| 327 |
T: Input, |
| 328 |
<T as Input>::Item: AsChar, |
| 329 |
{ |
| 330 |
Digit1 { e: PhantomData } |
| 331 |
} |
| 332 |
|
| 333 |
/// Parser implementation for [digit1] |
| 334 |
pub struct Digit1<E> { |
| 335 |
e: PhantomData<E>, |
| 336 |
} |
| 337 |
|
| 338 |
impl<I: Input, E: ParseError<I>> Parser<I> for Digit1<E> |
| 339 |
where |
| 340 |
<I as Input>::Item: AsChar, |
| 341 |
{ |
| 342 |
type Output = I; |
| 343 |
|
| 344 |
type Error = E; |
| 345 |
|
| 346 |
#[inline] |
| 347 |
fn process<OM: crate::OutputMode>( |
| 348 |
&mut self, |
| 349 |
input: I, |
| 350 |
) -> crate::PResult<OM, I, Self::Output, Self::Error> { |
| 351 |
input.split_at_position_mode1::<OM, _, _>(|item| !item.is_dec_digit(), ErrorKind::Digit) |
| 352 |
} |
| 353 |
} |
| 354 |
|
| 355 |
/// Recognizes zero or more spaces, tabs, carriage returns and line feeds. |
| 356 |
/// |
| 357 |
/// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data, |
| 358 |
/// or if no terminating token is found (a non space character). |
| 359 |
/// # Example |
| 360 |
/// |
| 361 |
/// ``` |
| 362 |
/// # use nom::{Err, error::ErrorKind, IResult, Needed}; |
| 363 |
/// # use nom::character::streaming::multispace0; |
| 364 |
/// assert_eq!(multispace0::<_, (_, ErrorKind)>(" \t\n\r21c"), Ok(("21c", " \t\n\r"))); |
| 365 |
/// assert_eq!(multispace0::<_, (_, ErrorKind)>("Z21c"), Ok(("Z21c", ""))); |
| 366 |
/// assert_eq!(multispace0::<_, (_, ErrorKind)>(""), Err(Err::Incomplete(Needed::new(1)))); |
| 367 |
/// ``` |
| 368 |
pub fn multispace0<T, E: ParseError<T>>() -> impl Parser<T, Output = T, Error = E> |
| 369 |
where |
| 370 |
T: Input, |
| 371 |
<T as Input>::Item: AsChar, |
| 372 |
{ |
| 373 |
MultiSpace0 { e: PhantomData } |
| 374 |
/*input.split_at_position(|item| { |
| 375 |
let c = item.as_char(); |
| 376 |
!(c == ' ' || c == '\t' || c == '\r' || c == '\n') |
| 377 |
})*/ |
| 378 |
} |
| 379 |
|
| 380 |
/// Parser implementation for [multispace0()] |
| 381 |
pub struct MultiSpace0<E> { |
| 382 |
e: PhantomData<E>, |
| 383 |
} |
| 384 |
|
| 385 |
impl<I, Error: ParseError<I>> Parser<I> for MultiSpace0<Error> |
| 386 |
where |
| 387 |
I: Input, |
| 388 |
<I as Input>::Item: AsChar, |
| 389 |
{ |
| 390 |
type Output = I; |
| 391 |
type Error = Error; |
| 392 |
|
| 393 |
fn process<OM: crate::OutputMode>( |
| 394 |
&mut self, |
| 395 |
i: I, |
| 396 |
) -> crate::PResult<OM, I, Self::Output, Self::Error> { |
| 397 |
i.split_at_position_mode::<OM, _, _>(|item| { |
| 398 |
let c = item.as_char(); |
| 399 |
!(c == ' ' || c == '\t' || c == '\r' || c == '\n') |
| 400 |
}) |
| 401 |
} |
| 402 |
} |