1 | //! Character specific parsers and combinators, streaming version |
2 | //! |
3 | //! Functions recognizing specific characters |
4 | |
5 | use crate::branch::alt; |
6 | use crate::combinator::opt; |
7 | use crate::error::ErrorKind; |
8 | use crate::error::ParseError; |
9 | use crate::internal::{Err, IResult, Needed}; |
10 | use crate::lib::std::ops::{Range, RangeFrom, RangeTo}; |
11 | use crate::traits::{ |
12 | AsChar, FindToken, InputIter, InputLength, InputTake, InputTakeAtPosition, Slice, |
13 | }; |
14 | use crate::traits::{Compare, CompareResult}; |
15 | |
16 | /// Recognizes one character. |
17 | /// |
18 | /// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data. |
19 | /// # Example |
20 | /// |
21 | /// ``` |
22 | /// # use nom::{Err, error::{ErrorKind, Error}, Needed, IResult}; |
23 | /// # use nom::character::streaming::char; |
24 | /// fn parser(i: &str) -> IResult<&str, char> { |
25 | /// char('a' )(i) |
26 | /// } |
27 | /// assert_eq!(parser("abc" ), Ok(("bc" , 'a' ))); |
28 | /// assert_eq!(parser("bc" ), Err(Err::Error(Error::new("bc" , ErrorKind::Char)))); |
29 | /// assert_eq!(parser("" ), Err(Err::Incomplete(Needed::new(1)))); |
30 | /// ``` |
31 | pub fn char<I, Error: ParseError<I>>(c: char) -> impl Fn(I) -> IResult<I, char, Error> |
32 | where |
33 | I: Slice<RangeFrom<usize>> + InputIter + InputLength, |
34 | <I as InputIter>::Item: AsChar, |
35 | { |
36 | move |i: I| match (i).iter_elements().next().map(|t: ::Item| { |
37 | let b: bool = t.as_char() == c; |
38 | (&c, b) |
39 | }) { |
40 | None => Err(Err::Incomplete(Needed::new(c.len() - i.input_len()))), |
41 | Some((_, false)) => Err(Err::Error(Error::from_char(input:i, c))), |
42 | Some((c: &char, true)) => Ok((i.slice(range:c.len()..), c.as_char())), |
43 | } |
44 | } |
45 | |
46 | /// Recognizes one character and checks that it satisfies a predicate |
47 | /// |
48 | /// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data. |
49 | /// # Example |
50 | /// |
51 | /// ``` |
52 | /// # use nom::{Err, error::{ErrorKind, Error}, Needed, IResult}; |
53 | /// # use nom::character::streaming::satisfy; |
54 | /// fn parser(i: &str) -> IResult<&str, char> { |
55 | /// satisfy(|c| c == 'a' || c == 'b' )(i) |
56 | /// } |
57 | /// assert_eq!(parser("abc" ), Ok(("bc" , 'a' ))); |
58 | /// assert_eq!(parser("cd" ), Err(Err::Error(Error::new("cd" , ErrorKind::Satisfy)))); |
59 | /// assert_eq!(parser("" ), Err(Err::Incomplete(Needed::Unknown))); |
60 | /// ``` |
61 | pub fn satisfy<F, I, Error: ParseError<I>>(cond: F) -> impl Fn(I) -> IResult<I, char, Error> |
62 | where |
63 | I: Slice<RangeFrom<usize>> + InputIter, |
64 | <I as InputIter>::Item: AsChar, |
65 | F: Fn(char) -> bool, |
66 | { |
67 | move |i: I| match (i).iter_elements().next().map(|t: ::Item| { |
68 | let c: char = t.as_char(); |
69 | let b: bool = cond(c); |
70 | (c, b) |
71 | }) { |
72 | None => Err(Err::Incomplete(Needed::Unknown)), |
73 | Some((_, false)) => Err(Err::Error(Error::from_error_kind(input:i, kind:ErrorKind::Satisfy))), |
74 | Some((c: char, true)) => Ok((i.slice(range:c.len()..), c)), |
75 | } |
76 | } |
77 | |
78 | /// Recognizes one of the provided characters. |
79 | /// |
80 | /// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data. |
81 | /// # Example |
82 | /// |
83 | /// ``` |
84 | /// # use nom::{Err, error::ErrorKind, Needed}; |
85 | /// # use nom::character::streaming::one_of; |
86 | /// assert_eq!(one_of::<_, _, (_, ErrorKind)>("abc" )("b" ), Ok(("" , 'b' ))); |
87 | /// assert_eq!(one_of::<_, _, (_, ErrorKind)>("a" )("bc" ), Err(Err::Error(("bc" , ErrorKind::OneOf)))); |
88 | /// assert_eq!(one_of::<_, _, (_, ErrorKind)>("a" )("" ), Err(Err::Incomplete(Needed::new(1)))); |
89 | /// ``` |
90 | pub fn one_of<I, T, Error: ParseError<I>>(list: T) -> impl Fn(I) -> IResult<I, char, Error> |
91 | where |
92 | I: Slice<RangeFrom<usize>> + InputIter, |
93 | <I as InputIter>::Item: AsChar + Copy, |
94 | T: FindToken<<I as InputIter>::Item>, |
95 | { |
96 | move |i: I| match (i).iter_elements().next().map(|c: ::Item| (c, list.find_token(c))) { |
97 | None => Err(Err::Incomplete(Needed::new(1))), |
98 | Some((_, false)) => Err(Err::Error(Error::from_error_kind(input:i, kind:ErrorKind::OneOf))), |
99 | Some((c: ::Item, true)) => Ok((i.slice(range:c.len()..), c.as_char())), |
100 | } |
101 | } |
102 | |
103 | /// Recognizes a character that is not in the provided characters. |
104 | /// |
105 | /// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data. |
106 | /// # Example |
107 | /// |
108 | /// ``` |
109 | /// # use nom::{Err, error::ErrorKind, Needed}; |
110 | /// # use nom::character::streaming::none_of; |
111 | /// assert_eq!(none_of::<_, _, (_, ErrorKind)>("abc" )("z" ), Ok(("" , 'z' ))); |
112 | /// assert_eq!(none_of::<_, _, (_, ErrorKind)>("ab" )("a" ), Err(Err::Error(("a" , ErrorKind::NoneOf)))); |
113 | /// assert_eq!(none_of::<_, _, (_, ErrorKind)>("a" )("" ), Err(Err::Incomplete(Needed::new(1)))); |
114 | /// ``` |
115 | pub fn none_of<I, T, Error: ParseError<I>>(list: T) -> impl Fn(I) -> IResult<I, char, Error> |
116 | where |
117 | I: Slice<RangeFrom<usize>> + InputIter, |
118 | <I as InputIter>::Item: AsChar + Copy, |
119 | T: FindToken<<I as InputIter>::Item>, |
120 | { |
121 | move |i: I| match (i).iter_elements().next().map(|c: ::Item| (c, !list.find_token(c))) { |
122 | None => Err(Err::Incomplete(Needed::new(1))), |
123 | Some((_, false)) => Err(Err::Error(Error::from_error_kind(input:i, kind:ErrorKind::NoneOf))), |
124 | Some((c: ::Item, true)) => Ok((i.slice(range:c.len()..), c.as_char())), |
125 | } |
126 | } |
127 | |
128 | /// Recognizes the string "\r\n". |
129 | /// |
130 | /// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data. |
131 | /// # Example |
132 | /// |
133 | /// ``` |
134 | /// # use nom::{Err, error::ErrorKind, IResult, Needed}; |
135 | /// # use nom::character::streaming::crlf; |
136 | /// assert_eq!(crlf::<_, (_, ErrorKind)>(" \r\nc" ), Ok(("c" , " \r\n" ))); |
137 | /// assert_eq!(crlf::<_, (_, ErrorKind)>("ab \r\nc" ), Err(Err::Error(("ab \r\nc" , ErrorKind::CrLf)))); |
138 | /// assert_eq!(crlf::<_, (_, ErrorKind)>("" ), Err(Err::Incomplete(Needed::new(2)))); |
139 | /// ``` |
140 | pub fn crlf<T, E: ParseError<T>>(input: T) -> IResult<T, T, E> |
141 | where |
142 | T: Slice<Range<usize>> + Slice<RangeFrom<usize>> + Slice<RangeTo<usize>>, |
143 | T: InputIter, |
144 | T: Compare<&'static str>, |
145 | { |
146 | match input.compare(" \r\n" ) { |
147 | //FIXME: is this the right index? |
148 | CompareResult::Ok => Ok((input.slice(range:2..), input.slice(range:0..2))), |
149 | CompareResult::Incomplete => Err(Err::Incomplete(Needed::new(2))), |
150 | CompareResult::Error => { |
151 | let e: ErrorKind = ErrorKind::CrLf; |
152 | Err(Err::Error(E::from_error_kind(input, kind:e))) |
153 | } |
154 | } |
155 | } |
156 | |
157 | /// Recognizes a string of any char except '\r\n' or '\n'. |
158 | /// |
159 | /// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data. |
160 | /// # Example |
161 | /// |
162 | /// ``` |
163 | /// # use nom::{Err, error::{Error, ErrorKind}, IResult, Needed}; |
164 | /// # use nom::character::streaming::not_line_ending; |
165 | /// assert_eq!(not_line_ending::<_, (_, ErrorKind)>("ab \r\nc" ), Ok((" \r\nc" , "ab" ))); |
166 | /// assert_eq!(not_line_ending::<_, (_, ErrorKind)>("abc" ), Err(Err::Incomplete(Needed::Unknown))); |
167 | /// assert_eq!(not_line_ending::<_, (_, ErrorKind)>("" ), Err(Err::Incomplete(Needed::Unknown))); |
168 | /// assert_eq!(not_line_ending::<_, (_, ErrorKind)>("a \rb \nc" ), Err(Err::Error(("a \rb \nc" , ErrorKind::Tag )))); |
169 | /// assert_eq!(not_line_ending::<_, (_, ErrorKind)>("a \rbc" ), Err(Err::Error(("a \rbc" , ErrorKind::Tag )))); |
170 | /// ``` |
171 | pub fn not_line_ending<T, E: ParseError<T>>(input: T) -> IResult<T, T, E> |
172 | where |
173 | T: Slice<Range<usize>> + Slice<RangeFrom<usize>> + Slice<RangeTo<usize>>, |
174 | T: InputIter + InputLength, |
175 | T: Compare<&'static str>, |
176 | <T as InputIter>::Item: AsChar, |
177 | <T as InputIter>::Item: AsChar, |
178 | { |
179 | match input.position(|item| { |
180 | let c = item.as_char(); |
181 | c == ' \r' || c == ' \n' |
182 | }) { |
183 | None => Err(Err::Incomplete(Needed::Unknown)), |
184 | Some(index) => { |
185 | let mut it = input.slice(index..).iter_elements(); |
186 | let nth = it.next().unwrap().as_char(); |
187 | if nth == ' \r' { |
188 | let sliced = input.slice(index..); |
189 | let comp = sliced.compare(" \r\n" ); |
190 | match comp { |
191 | //FIXME: calculate the right index |
192 | CompareResult::Incomplete => Err(Err::Incomplete(Needed::Unknown)), |
193 | CompareResult::Error => { |
194 | let e: ErrorKind = ErrorKind::Tag; |
195 | Err(Err::Error(E::from_error_kind(input, e))) |
196 | } |
197 | CompareResult::Ok => Ok((input.slice(index..), input.slice(..index))), |
198 | } |
199 | } else { |
200 | Ok((input.slice(index..), input.slice(..index))) |
201 | } |
202 | } |
203 | } |
204 | } |
205 | |
206 | /// Recognizes an end of line (both '\n' and '\r\n'). |
207 | /// |
208 | /// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data. |
209 | /// # Example |
210 | /// |
211 | /// ``` |
212 | /// # use nom::{Err, error::ErrorKind, IResult, Needed}; |
213 | /// # use nom::character::streaming::line_ending; |
214 | /// assert_eq!(line_ending::<_, (_, ErrorKind)>(" \r\nc" ), Ok(("c" , " \r\n" ))); |
215 | /// assert_eq!(line_ending::<_, (_, ErrorKind)>("ab \r\nc" ), Err(Err::Error(("ab \r\nc" , ErrorKind::CrLf)))); |
216 | /// assert_eq!(line_ending::<_, (_, ErrorKind)>("" ), Err(Err::Incomplete(Needed::new(1)))); |
217 | /// ``` |
218 | pub fn line_ending<T, E: ParseError<T>>(input: T) -> IResult<T, T, E> |
219 | where |
220 | T: Slice<Range<usize>> + Slice<RangeFrom<usize>> + Slice<RangeTo<usize>>, |
221 | T: InputIter + InputLength, |
222 | T: Compare<&'static str>, |
223 | { |
224 | match input.compare(" \n" ) { |
225 | CompareResult::Ok => Ok((input.slice(range:1..), input.slice(range:0..1))), |
226 | CompareResult::Incomplete => Err(Err::Incomplete(Needed::new(1))), |
227 | CompareResult::Error => { |
228 | match input.compare(" \r\n" ) { |
229 | //FIXME: is this the right index? |
230 | CompareResult::Ok => Ok((input.slice(range:2..), input.slice(range:0..2))), |
231 | CompareResult::Incomplete => Err(Err::Incomplete(Needed::new(2))), |
232 | CompareResult::Error => Err(Err::Error(E::from_error_kind(input, kind:ErrorKind::CrLf))), |
233 | } |
234 | } |
235 | } |
236 | } |
237 | |
238 | /// Matches a newline character '\\n'. |
239 | /// |
240 | /// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data. |
241 | /// # Example |
242 | /// |
243 | /// ``` |
244 | /// # use nom::{Err, error::ErrorKind, IResult, Needed}; |
245 | /// # use nom::character::streaming::newline; |
246 | /// assert_eq!(newline::<_, (_, ErrorKind)>(" \nc" ), Ok(("c" , ' \n' ))); |
247 | /// assert_eq!(newline::<_, (_, ErrorKind)>(" \r\nc" ), Err(Err::Error((" \r\nc" , ErrorKind::Char)))); |
248 | /// assert_eq!(newline::<_, (_, ErrorKind)>("" ), Err(Err::Incomplete(Needed::new(1)))); |
249 | /// ``` |
250 | pub fn newline<I, Error: ParseError<I>>(input: I) -> IResult<I, char, Error> |
251 | where |
252 | I: Slice<RangeFrom<usize>> + InputIter + InputLength, |
253 | <I as InputIter>::Item: AsChar, |
254 | { |
255 | char(' \n' )(input) |
256 | } |
257 | |
258 | /// Matches a tab character '\t'. |
259 | /// |
260 | /// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data. |
261 | /// # Example |
262 | /// |
263 | /// ``` |
264 | /// # use nom::{Err, error::ErrorKind, IResult, Needed}; |
265 | /// # use nom::character::streaming::tab; |
266 | /// assert_eq!(tab::<_, (_, ErrorKind)>(" \tc" ), Ok(("c" , ' \t' ))); |
267 | /// assert_eq!(tab::<_, (_, ErrorKind)>(" \r\nc" ), Err(Err::Error((" \r\nc" , ErrorKind::Char)))); |
268 | /// assert_eq!(tab::<_, (_, ErrorKind)>("" ), Err(Err::Incomplete(Needed::new(1)))); |
269 | /// ``` |
270 | pub fn tab<I, Error: ParseError<I>>(input: I) -> IResult<I, char, Error> |
271 | where |
272 | I: Slice<RangeFrom<usize>> + InputIter + InputLength, |
273 | <I as InputIter>::Item: AsChar, |
274 | { |
275 | char(' \t' )(input) |
276 | } |
277 | |
278 | /// Matches one byte as a character. Note that the input type will |
279 | /// accept a `str`, but not a `&[u8]`, unlike many other nom parsers. |
280 | /// |
281 | /// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data. |
282 | /// # Example |
283 | /// |
284 | /// ``` |
285 | /// # use nom::{character::streaming::anychar, Err, error::ErrorKind, IResult, Needed}; |
286 | /// assert_eq!(anychar::<_, (_, ErrorKind)>("abc" ), Ok(("bc" ,'a' ))); |
287 | /// assert_eq!(anychar::<_, (_, ErrorKind)>("" ), Err(Err::Incomplete(Needed::new(1)))); |
288 | /// ``` |
289 | pub fn anychar<T, E: ParseError<T>>(input: T) -> IResult<T, char, E> |
290 | where |
291 | T: InputIter + InputLength + Slice<RangeFrom<usize>>, |
292 | <T as InputIter>::Item: AsChar, |
293 | { |
294 | let mut it: ::Iter = input.iter_indices(); |
295 | match it.next() { |
296 | None => Err(Err::Incomplete(Needed::new(1))), |
297 | Some((_, c: ::Item)) => match it.next() { |
298 | None => Ok((input.slice(range:input.input_len()..), c.as_char())), |
299 | Some((idx: usize, _)) => Ok((input.slice(range:idx..), c.as_char())), |
300 | }, |
301 | } |
302 | } |
303 | |
304 | /// Recognizes zero or more lowercase and uppercase ASCII alphabetic characters: a-z, A-Z |
305 | /// |
306 | /// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data, |
307 | /// or if no terminating token is found (a non alphabetic character). |
308 | /// # Example |
309 | /// |
310 | /// ``` |
311 | /// # use nom::{Err, error::ErrorKind, IResult, Needed}; |
312 | /// # use nom::character::streaming::alpha0; |
313 | /// assert_eq!(alpha0::<_, (_, ErrorKind)>("ab1c" ), Ok(("1c" , "ab" ))); |
314 | /// assert_eq!(alpha0::<_, (_, ErrorKind)>("1c" ), Ok(("1c" , "" ))); |
315 | /// assert_eq!(alpha0::<_, (_, ErrorKind)>("" ), Err(Err::Incomplete(Needed::new(1)))); |
316 | /// ``` |
317 | pub fn alpha0<T, E: ParseError<T>>(input: T) -> IResult<T, T, E> |
318 | where |
319 | T: InputTakeAtPosition, |
320 | <T as InputTakeAtPosition>::Item: AsChar, |
321 | { |
322 | input.split_at_position(|item: ::Item| !item.is_alpha()) |
323 | } |
324 | |
325 | /// Recognizes one or more lowercase and uppercase ASCII alphabetic characters: a-z, A-Z |
326 | /// |
327 | /// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data, |
328 | /// or if no terminating token is found (a non alphabetic character). |
329 | /// # Example |
330 | /// |
331 | /// ``` |
332 | /// # use nom::{Err, error::ErrorKind, IResult, Needed}; |
333 | /// # use nom::character::streaming::alpha1; |
334 | /// assert_eq!(alpha1::<_, (_, ErrorKind)>("aB1c" ), Ok(("1c" , "aB" ))); |
335 | /// assert_eq!(alpha1::<_, (_, ErrorKind)>("1c" ), Err(Err::Error(("1c" , ErrorKind::Alpha)))); |
336 | /// assert_eq!(alpha1::<_, (_, ErrorKind)>("" ), Err(Err::Incomplete(Needed::new(1)))); |
337 | /// ``` |
338 | pub fn alpha1<T, E: ParseError<T>>(input: T) -> IResult<T, T, E> |
339 | where |
340 | T: InputTakeAtPosition, |
341 | <T as InputTakeAtPosition>::Item: AsChar, |
342 | { |
343 | input.split_at_position1(|item| !item.is_alpha(), e:ErrorKind::Alpha) |
344 | } |
345 | |
346 | /// Recognizes zero or more ASCII numerical characters: 0-9 |
347 | /// |
348 | /// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data, |
349 | /// or if no terminating token is found (a non digit character). |
350 | /// # Example |
351 | /// |
352 | /// ``` |
353 | /// # use nom::{Err, error::ErrorKind, IResult, Needed}; |
354 | /// # use nom::character::streaming::digit0; |
355 | /// assert_eq!(digit0::<_, (_, ErrorKind)>("21c" ), Ok(("c" , "21" ))); |
356 | /// assert_eq!(digit0::<_, (_, ErrorKind)>("a21c" ), Ok(("a21c" , "" ))); |
357 | /// assert_eq!(digit0::<_, (_, ErrorKind)>("" ), Err(Err::Incomplete(Needed::new(1)))); |
358 | /// ``` |
359 | pub fn digit0<T, E: ParseError<T>>(input: T) -> IResult<T, T, E> |
360 | where |
361 | T: InputTakeAtPosition, |
362 | <T as InputTakeAtPosition>::Item: AsChar, |
363 | { |
364 | input.split_at_position(|item: ::Item| !item.is_dec_digit()) |
365 | } |
366 | |
367 | /// Recognizes one or more ASCII numerical characters: 0-9 |
368 | /// |
369 | /// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data, |
370 | /// or if no terminating token is found (a non digit character). |
371 | /// # Example |
372 | /// |
373 | /// ``` |
374 | /// # use nom::{Err, error::ErrorKind, IResult, Needed}; |
375 | /// # use nom::character::streaming::digit1; |
376 | /// assert_eq!(digit1::<_, (_, ErrorKind)>("21c" ), Ok(("c" , "21" ))); |
377 | /// assert_eq!(digit1::<_, (_, ErrorKind)>("c1" ), Err(Err::Error(("c1" , ErrorKind::Digit)))); |
378 | /// assert_eq!(digit1::<_, (_, ErrorKind)>("" ), Err(Err::Incomplete(Needed::new(1)))); |
379 | /// ``` |
380 | pub fn digit1<T, E: ParseError<T>>(input: T) -> IResult<T, T, E> |
381 | where |
382 | T: InputTakeAtPosition, |
383 | <T as InputTakeAtPosition>::Item: AsChar, |
384 | { |
385 | input.split_at_position1(|item| !item.is_dec_digit(), e:ErrorKind::Digit) |
386 | } |
387 | |
388 | /// Recognizes zero or more ASCII hexadecimal numerical characters: 0-9, A-F, a-f |
389 | /// |
390 | /// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data, |
391 | /// or if no terminating token is found (a non hexadecimal digit character). |
392 | /// # Example |
393 | /// |
394 | /// ``` |
395 | /// # use nom::{Err, error::ErrorKind, IResult, Needed}; |
396 | /// # use nom::character::streaming::hex_digit0; |
397 | /// assert_eq!(hex_digit0::<_, (_, ErrorKind)>("21cZ" ), Ok(("Z" , "21c" ))); |
398 | /// assert_eq!(hex_digit0::<_, (_, ErrorKind)>("Z21c" ), Ok(("Z21c" , "" ))); |
399 | /// assert_eq!(hex_digit0::<_, (_, ErrorKind)>("" ), Err(Err::Incomplete(Needed::new(1)))); |
400 | /// ``` |
401 | pub fn hex_digit0<T, E: ParseError<T>>(input: T) -> IResult<T, T, E> |
402 | where |
403 | T: InputTakeAtPosition, |
404 | <T as InputTakeAtPosition>::Item: AsChar, |
405 | { |
406 | input.split_at_position(|item: ::Item| !item.is_hex_digit()) |
407 | } |
408 | |
409 | /// Recognizes one or more ASCII hexadecimal numerical characters: 0-9, A-F, a-f |
410 | /// |
411 | /// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data, |
412 | /// or if no terminating token is found (a non hexadecimal digit character). |
413 | /// # Example |
414 | /// |
415 | /// ``` |
416 | /// # use nom::{Err, error::ErrorKind, IResult, Needed}; |
417 | /// # use nom::character::streaming::hex_digit1; |
418 | /// assert_eq!(hex_digit1::<_, (_, ErrorKind)>("21cZ" ), Ok(("Z" , "21c" ))); |
419 | /// assert_eq!(hex_digit1::<_, (_, ErrorKind)>("H2" ), Err(Err::Error(("H2" , ErrorKind::HexDigit)))); |
420 | /// assert_eq!(hex_digit1::<_, (_, ErrorKind)>("" ), Err(Err::Incomplete(Needed::new(1)))); |
421 | /// ``` |
422 | pub fn hex_digit1<T, E: ParseError<T>>(input: T) -> IResult<T, T, E> |
423 | where |
424 | T: InputTakeAtPosition, |
425 | <T as InputTakeAtPosition>::Item: AsChar, |
426 | { |
427 | input.split_at_position1(|item| !item.is_hex_digit(), e:ErrorKind::HexDigit) |
428 | } |
429 | |
430 | /// Recognizes zero or more octal characters: 0-7 |
431 | /// |
432 | /// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data, |
433 | /// or if no terminating token is found (a non octal digit character). |
434 | /// # Example |
435 | /// |
436 | /// ``` |
437 | /// # use nom::{Err, error::ErrorKind, IResult, Needed}; |
438 | /// # use nom::character::streaming::oct_digit0; |
439 | /// assert_eq!(oct_digit0::<_, (_, ErrorKind)>("21cZ" ), Ok(("cZ" , "21" ))); |
440 | /// assert_eq!(oct_digit0::<_, (_, ErrorKind)>("Z21c" ), Ok(("Z21c" , "" ))); |
441 | /// assert_eq!(oct_digit0::<_, (_, ErrorKind)>("" ), Err(Err::Incomplete(Needed::new(1)))); |
442 | /// ``` |
443 | pub fn oct_digit0<T, E: ParseError<T>>(input: T) -> IResult<T, T, E> |
444 | where |
445 | T: InputTakeAtPosition, |
446 | <T as InputTakeAtPosition>::Item: AsChar, |
447 | { |
448 | input.split_at_position(|item: ::Item| !item.is_oct_digit()) |
449 | } |
450 | |
451 | /// Recognizes one or more octal characters: 0-7 |
452 | /// |
453 | /// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data, |
454 | /// or if no terminating token is found (a non octal digit character). |
455 | /// # Example |
456 | /// |
457 | /// ``` |
458 | /// # use nom::{Err, error::ErrorKind, IResult, Needed}; |
459 | /// # use nom::character::streaming::oct_digit1; |
460 | /// assert_eq!(oct_digit1::<_, (_, ErrorKind)>("21cZ" ), Ok(("cZ" , "21" ))); |
461 | /// assert_eq!(oct_digit1::<_, (_, ErrorKind)>("H2" ), Err(Err::Error(("H2" , ErrorKind::OctDigit)))); |
462 | /// assert_eq!(oct_digit1::<_, (_, ErrorKind)>("" ), Err(Err::Incomplete(Needed::new(1)))); |
463 | /// ``` |
464 | pub fn oct_digit1<T, E: ParseError<T>>(input: T) -> IResult<T, T, E> |
465 | where |
466 | T: InputTakeAtPosition, |
467 | <T as InputTakeAtPosition>::Item: AsChar, |
468 | { |
469 | input.split_at_position1(|item| !item.is_oct_digit(), e:ErrorKind::OctDigit) |
470 | } |
471 | |
472 | /// Recognizes zero or more ASCII numerical and alphabetic characters: 0-9, a-z, A-Z |
473 | /// |
474 | /// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data, |
475 | /// or if no terminating token is found (a non alphanumerical character). |
476 | /// # Example |
477 | /// |
478 | /// ``` |
479 | /// # use nom::{Err, error::ErrorKind, IResult, Needed}; |
480 | /// # use nom::character::streaming::alphanumeric0; |
481 | /// assert_eq!(alphanumeric0::<_, (_, ErrorKind)>("21cZ%1" ), Ok(("%1" , "21cZ" ))); |
482 | /// assert_eq!(alphanumeric0::<_, (_, ErrorKind)>("&Z21c" ), Ok(("&Z21c" , "" ))); |
483 | /// assert_eq!(alphanumeric0::<_, (_, ErrorKind)>("" ), Err(Err::Incomplete(Needed::new(1)))); |
484 | /// ``` |
485 | pub fn alphanumeric0<T, E: ParseError<T>>(input: T) -> IResult<T, T, E> |
486 | where |
487 | T: InputTakeAtPosition, |
488 | <T as InputTakeAtPosition>::Item: AsChar, |
489 | { |
490 | input.split_at_position(|item: ::Item| !item.is_alphanum()) |
491 | } |
492 | |
493 | /// Recognizes one or more ASCII numerical and alphabetic characters: 0-9, a-z, A-Z |
494 | /// |
495 | /// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data, |
496 | /// or if no terminating token is found (a non alphanumerical character). |
497 | /// # Example |
498 | /// |
499 | /// ``` |
500 | /// # use nom::{Err, error::ErrorKind, IResult, Needed}; |
501 | /// # use nom::character::streaming::alphanumeric1; |
502 | /// assert_eq!(alphanumeric1::<_, (_, ErrorKind)>("21cZ%1" ), Ok(("%1" , "21cZ" ))); |
503 | /// assert_eq!(alphanumeric1::<_, (_, ErrorKind)>("&H2" ), Err(Err::Error(("&H2" , ErrorKind::AlphaNumeric)))); |
504 | /// assert_eq!(alphanumeric1::<_, (_, ErrorKind)>("" ), Err(Err::Incomplete(Needed::new(1)))); |
505 | /// ``` |
506 | pub fn alphanumeric1<T, E: ParseError<T>>(input: T) -> IResult<T, T, E> |
507 | where |
508 | T: InputTakeAtPosition, |
509 | <T as InputTakeAtPosition>::Item: AsChar, |
510 | { |
511 | input.split_at_position1(|item| !item.is_alphanum(), e:ErrorKind::AlphaNumeric) |
512 | } |
513 | |
514 | /// Recognizes zero or more spaces and tabs. |
515 | /// |
516 | /// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data, |
517 | /// or if no terminating token is found (a non space character). |
518 | /// # Example |
519 | /// |
520 | /// ``` |
521 | /// # use nom::{Err, error::ErrorKind, IResult, Needed}; |
522 | /// # use nom::character::streaming::space0; |
523 | /// assert_eq!(space0::<_, (_, ErrorKind)>(" \t21c" ), Ok(("21c" , " \t" ))); |
524 | /// assert_eq!(space0::<_, (_, ErrorKind)>("Z21c" ), Ok(("Z21c" , "" ))); |
525 | /// assert_eq!(space0::<_, (_, ErrorKind)>("" ), Err(Err::Incomplete(Needed::new(1)))); |
526 | /// ``` |
527 | pub fn space0<T, E: ParseError<T>>(input: T) -> IResult<T, T, E> |
528 | where |
529 | T: InputTakeAtPosition, |
530 | <T as InputTakeAtPosition>::Item: AsChar + Clone, |
531 | { |
532 | input.split_at_position(|item: ::Item| { |
533 | let c: char = item.as_char(); |
534 | !(c == ' ' || c == ' \t' ) |
535 | }) |
536 | } |
537 | /// Recognizes one or more spaces and tabs. |
538 | /// |
539 | /// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data, |
540 | /// or if no terminating token is found (a non space character). |
541 | /// # Example |
542 | /// |
543 | /// ``` |
544 | /// # use nom::{Err, error::ErrorKind, IResult, Needed}; |
545 | /// # use nom::character::streaming::space1; |
546 | /// assert_eq!(space1::<_, (_, ErrorKind)>(" \t21c" ), Ok(("21c" , " \t" ))); |
547 | /// assert_eq!(space1::<_, (_, ErrorKind)>("H2" ), Err(Err::Error(("H2" , ErrorKind::Space)))); |
548 | /// assert_eq!(space1::<_, (_, ErrorKind)>("" ), Err(Err::Incomplete(Needed::new(1)))); |
549 | /// ``` |
550 | pub fn space1<T, E: ParseError<T>>(input: T) -> IResult<T, T, E> |
551 | where |
552 | T: InputTakeAtPosition, |
553 | <T as InputTakeAtPosition>::Item: AsChar + Clone, |
554 | { |
555 | input.split_at_position1( |
556 | |item| { |
557 | let c = item.as_char(); |
558 | !(c == ' ' || c == ' \t' ) |
559 | }, |
560 | e:ErrorKind::Space, |
561 | ) |
562 | } |
563 | |
564 | /// Recognizes zero or more spaces, tabs, carriage returns and line feeds. |
565 | /// |
566 | /// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data, |
567 | /// or if no terminating token is found (a non space character). |
568 | /// # Example |
569 | /// |
570 | /// ``` |
571 | /// # use nom::{Err, error::ErrorKind, IResult, Needed}; |
572 | /// # use nom::character::streaming::multispace0; |
573 | /// assert_eq!(multispace0::<_, (_, ErrorKind)>(" \t\n\r21c" ), Ok(("21c" , " \t\n\r" ))); |
574 | /// assert_eq!(multispace0::<_, (_, ErrorKind)>("Z21c" ), Ok(("Z21c" , "" ))); |
575 | /// assert_eq!(multispace0::<_, (_, ErrorKind)>("" ), Err(Err::Incomplete(Needed::new(1)))); |
576 | /// ``` |
577 | pub fn multispace0<T, E: ParseError<T>>(input: T) -> IResult<T, T, E> |
578 | where |
579 | T: InputTakeAtPosition, |
580 | <T as InputTakeAtPosition>::Item: AsChar + Clone, |
581 | { |
582 | input.split_at_position(|item: ::Item| { |
583 | let c: char = item.as_char(); |
584 | !(c == ' ' || c == ' \t' || c == ' \r' || c == ' \n' ) |
585 | }) |
586 | } |
587 | |
588 | /// Recognizes one or more spaces, tabs, carriage returns and line feeds. |
589 | /// |
590 | /// *Streaming version*: Will return `Err(nom::Err::Incomplete(_))` if there's not enough input data, |
591 | /// or if no terminating token is found (a non space character). |
592 | /// # Example |
593 | /// |
594 | /// ``` |
595 | /// # use nom::{Err, error::ErrorKind, IResult, Needed}; |
596 | /// # use nom::character::streaming::multispace1; |
597 | /// assert_eq!(multispace1::<_, (_, ErrorKind)>(" \t\n\r21c" ), Ok(("21c" , " \t\n\r" ))); |
598 | /// assert_eq!(multispace1::<_, (_, ErrorKind)>("H2" ), Err(Err::Error(("H2" , ErrorKind::MultiSpace)))); |
599 | /// assert_eq!(multispace1::<_, (_, ErrorKind)>("" ), Err(Err::Incomplete(Needed::new(1)))); |
600 | /// ``` |
601 | pub fn multispace1<T, E: ParseError<T>>(input: T) -> IResult<T, T, E> |
602 | where |
603 | T: InputTakeAtPosition, |
604 | <T as InputTakeAtPosition>::Item: AsChar + Clone, |
605 | { |
606 | input.split_at_position1( |
607 | |item| { |
608 | let c = item.as_char(); |
609 | !(c == ' ' || c == ' \t' || c == ' \r' || c == ' \n' ) |
610 | }, |
611 | e:ErrorKind::MultiSpace, |
612 | ) |
613 | } |
614 | |
615 | pub(crate) fn sign<T, E: ParseError<T>>(input: T) -> IResult<T, bool, E> |
616 | where |
617 | T: Clone + InputTake + InputLength, |
618 | T: for<'a> Compare<&'a [u8]>, |
619 | { |
620 | use crate::bytes::streaming::tag; |
621 | use crate::combinator::value; |
622 | |
623 | let (i: T, opt_sign: Option) = opt(alt(( |
624 | value(val:false, parser:tag(&b"-" [..])), |
625 | value(val:true, parser:tag(&b"+" [..])), |
626 | )))(input)?; |
627 | let sign: bool = opt_sign.unwrap_or(default:true); |
628 | |
629 | Ok((i, sign)) |
630 | } |
631 | |
632 | #[doc (hidden)] |
633 | macro_rules! ints { |
634 | ($($t:tt)+) => { |
635 | $( |
636 | /// will parse a number in text form to a number |
637 | /// |
638 | /// *Complete version*: can parse until the end of input. |
639 | pub fn $t<T, E: ParseError<T>>(input: T) -> IResult<T, $t, E> |
640 | where |
641 | T: InputIter + Slice<RangeFrom<usize>> + InputLength + InputTake + Clone, |
642 | <T as InputIter>::Item: AsChar, |
643 | T: for <'a> Compare<&'a[u8]>, |
644 | { |
645 | let (i, sign) = sign(input.clone())?; |
646 | |
647 | if i.input_len() == 0 { |
648 | return Err(Err::Incomplete(Needed::new(1))); |
649 | } |
650 | |
651 | let mut value: $t = 0; |
652 | if sign { |
653 | for (pos, c) in i.iter_indices() { |
654 | match c.as_char().to_digit(10) { |
655 | None => { |
656 | if pos == 0 { |
657 | return Err(Err::Error(E::from_error_kind(input, ErrorKind::Digit))); |
658 | } else { |
659 | return Ok((i.slice(pos..), value)); |
660 | } |
661 | }, |
662 | Some(d) => match value.checked_mul(10).and_then(|v| v.checked_add(d as $t)) { |
663 | None => return Err(Err::Error(E::from_error_kind(input, ErrorKind::Digit))), |
664 | Some(v) => value = v, |
665 | } |
666 | } |
667 | } |
668 | } else { |
669 | for (pos, c) in i.iter_indices() { |
670 | match c.as_char().to_digit(10) { |
671 | None => { |
672 | if pos == 0 { |
673 | return Err(Err::Error(E::from_error_kind(input, ErrorKind::Digit))); |
674 | } else { |
675 | return Ok((i.slice(pos..), value)); |
676 | } |
677 | }, |
678 | Some(d) => match value.checked_mul(10).and_then(|v| v.checked_sub(d as $t)) { |
679 | None => return Err(Err::Error(E::from_error_kind(input, ErrorKind::Digit))), |
680 | Some(v) => value = v, |
681 | } |
682 | } |
683 | } |
684 | } |
685 | |
686 | Err(Err::Incomplete(Needed::new(1))) |
687 | } |
688 | )+ |
689 | } |
690 | } |
691 | |
692 | ints! { i8 i16 i32 i64 i128 } |
693 | |
694 | #[doc (hidden)] |
695 | macro_rules! uints { |
696 | ($($t:tt)+) => { |
697 | $( |
698 | /// will parse a number in text form to a number |
699 | /// |
700 | /// *Complete version*: can parse until the end of input. |
701 | pub fn $t<T, E: ParseError<T>>(input: T) -> IResult<T, $t, E> |
702 | where |
703 | T: InputIter + Slice<RangeFrom<usize>> + InputLength, |
704 | <T as InputIter>::Item: AsChar, |
705 | { |
706 | let i = input; |
707 | |
708 | if i.input_len() == 0 { |
709 | return Err(Err::Incomplete(Needed::new(1))); |
710 | } |
711 | |
712 | let mut value: $t = 0; |
713 | for (pos, c) in i.iter_indices() { |
714 | match c.as_char().to_digit(10) { |
715 | None => { |
716 | if pos == 0 { |
717 | return Err(Err::Error(E::from_error_kind(i, ErrorKind::Digit))); |
718 | } else { |
719 | return Ok((i.slice(pos..), value)); |
720 | } |
721 | }, |
722 | Some(d) => match value.checked_mul(10).and_then(|v| v.checked_add(d as $t)) { |
723 | None => return Err(Err::Error(E::from_error_kind(i, ErrorKind::Digit))), |
724 | Some(v) => value = v, |
725 | } |
726 | } |
727 | } |
728 | |
729 | Err(Err::Incomplete(Needed::new(1))) |
730 | } |
731 | )+ |
732 | } |
733 | } |
734 | |
735 | uints! { u8 u16 u32 u64 u128 } |
736 | |
737 | #[cfg (test)] |
738 | mod tests { |
739 | use super::*; |
740 | use crate::error::ErrorKind; |
741 | use crate::internal::{Err, Needed}; |
742 | use crate::sequence::pair; |
743 | use crate::traits::ParseTo; |
744 | use proptest::prelude::*; |
745 | |
746 | macro_rules! assert_parse( |
747 | ($left: expr, $right: expr) => { |
748 | let res: $crate::IResult<_, _, (_, ErrorKind)> = $left; |
749 | assert_eq!(res, $right); |
750 | }; |
751 | ); |
752 | |
753 | #[test ] |
754 | fn anychar_str() { |
755 | use super::anychar; |
756 | assert_eq!(anychar::<_, (&str, ErrorKind)>("Ә" ), Ok(("" , 'Ә' ))); |
757 | } |
758 | |
759 | #[test ] |
760 | fn character() { |
761 | let a: &[u8] = b"abcd" ; |
762 | let b: &[u8] = b"1234" ; |
763 | let c: &[u8] = b"a123" ; |
764 | let d: &[u8] = "azé12" .as_bytes(); |
765 | let e: &[u8] = b" " ; |
766 | let f: &[u8] = b" ;" ; |
767 | //assert_eq!(alpha1::<_, (_, ErrorKind)>(a), Err(Err::Incomplete(Needed::new(1)))); |
768 | assert_parse!(alpha1(a), Err(Err::Incomplete(Needed::new(1)))); |
769 | assert_eq!(alpha1(b), Err(Err::Error((b, ErrorKind::Alpha)))); |
770 | assert_eq!(alpha1::<_, (_, ErrorKind)>(c), Ok((&c[1..], &b"a" [..]))); |
771 | assert_eq!( |
772 | alpha1::<_, (_, ErrorKind)>(d), |
773 | Ok(("é12" .as_bytes(), &b"az" [..])) |
774 | ); |
775 | assert_eq!(digit1(a), Err(Err::Error((a, ErrorKind::Digit)))); |
776 | assert_eq!( |
777 | digit1::<_, (_, ErrorKind)>(b), |
778 | Err(Err::Incomplete(Needed::new(1))) |
779 | ); |
780 | assert_eq!(digit1(c), Err(Err::Error((c, ErrorKind::Digit)))); |
781 | assert_eq!(digit1(d), Err(Err::Error((d, ErrorKind::Digit)))); |
782 | assert_eq!( |
783 | hex_digit1::<_, (_, ErrorKind)>(a), |
784 | Err(Err::Incomplete(Needed::new(1))) |
785 | ); |
786 | assert_eq!( |
787 | hex_digit1::<_, (_, ErrorKind)>(b), |
788 | Err(Err::Incomplete(Needed::new(1))) |
789 | ); |
790 | assert_eq!( |
791 | hex_digit1::<_, (_, ErrorKind)>(c), |
792 | Err(Err::Incomplete(Needed::new(1))) |
793 | ); |
794 | assert_eq!( |
795 | hex_digit1::<_, (_, ErrorKind)>(d), |
796 | Ok(("zé12" .as_bytes(), &b"a" [..])) |
797 | ); |
798 | assert_eq!(hex_digit1(e), Err(Err::Error((e, ErrorKind::HexDigit)))); |
799 | assert_eq!(oct_digit1(a), Err(Err::Error((a, ErrorKind::OctDigit)))); |
800 | assert_eq!( |
801 | oct_digit1::<_, (_, ErrorKind)>(b), |
802 | Err(Err::Incomplete(Needed::new(1))) |
803 | ); |
804 | assert_eq!(oct_digit1(c), Err(Err::Error((c, ErrorKind::OctDigit)))); |
805 | assert_eq!(oct_digit1(d), Err(Err::Error((d, ErrorKind::OctDigit)))); |
806 | assert_eq!( |
807 | alphanumeric1::<_, (_, ErrorKind)>(a), |
808 | Err(Err::Incomplete(Needed::new(1))) |
809 | ); |
810 | //assert_eq!(fix_error!(b,(), alphanumeric1), Ok((empty, b))); |
811 | assert_eq!( |
812 | alphanumeric1::<_, (_, ErrorKind)>(c), |
813 | Err(Err::Incomplete(Needed::new(1))) |
814 | ); |
815 | assert_eq!( |
816 | alphanumeric1::<_, (_, ErrorKind)>(d), |
817 | Ok(("é12" .as_bytes(), &b"az" [..])) |
818 | ); |
819 | assert_eq!( |
820 | space1::<_, (_, ErrorKind)>(e), |
821 | Err(Err::Incomplete(Needed::new(1))) |
822 | ); |
823 | assert_eq!(space1::<_, (_, ErrorKind)>(f), Ok((&b";" [..], &b" " [..]))); |
824 | } |
825 | |
826 | #[cfg (feature = "alloc" )] |
827 | #[test ] |
828 | fn character_s() { |
829 | let a = "abcd" ; |
830 | let b = "1234" ; |
831 | let c = "a123" ; |
832 | let d = "azé12" ; |
833 | let e = " " ; |
834 | assert_eq!( |
835 | alpha1::<_, (_, ErrorKind)>(a), |
836 | Err(Err::Incomplete(Needed::new(1))) |
837 | ); |
838 | assert_eq!(alpha1(b), Err(Err::Error((b, ErrorKind::Alpha)))); |
839 | assert_eq!(alpha1::<_, (_, ErrorKind)>(c), Ok((&c[1..], &"a" [..]))); |
840 | assert_eq!(alpha1::<_, (_, ErrorKind)>(d), Ok(("é12" , &"az" [..]))); |
841 | assert_eq!(digit1(a), Err(Err::Error((a, ErrorKind::Digit)))); |
842 | assert_eq!( |
843 | digit1::<_, (_, ErrorKind)>(b), |
844 | Err(Err::Incomplete(Needed::new(1))) |
845 | ); |
846 | assert_eq!(digit1(c), Err(Err::Error((c, ErrorKind::Digit)))); |
847 | assert_eq!(digit1(d), Err(Err::Error((d, ErrorKind::Digit)))); |
848 | assert_eq!( |
849 | hex_digit1::<_, (_, ErrorKind)>(a), |
850 | Err(Err::Incomplete(Needed::new(1))) |
851 | ); |
852 | assert_eq!( |
853 | hex_digit1::<_, (_, ErrorKind)>(b), |
854 | Err(Err::Incomplete(Needed::new(1))) |
855 | ); |
856 | assert_eq!( |
857 | hex_digit1::<_, (_, ErrorKind)>(c), |
858 | Err(Err::Incomplete(Needed::new(1))) |
859 | ); |
860 | assert_eq!(hex_digit1::<_, (_, ErrorKind)>(d), Ok(("zé12" , &"a" [..]))); |
861 | assert_eq!(hex_digit1(e), Err(Err::Error((e, ErrorKind::HexDigit)))); |
862 | assert_eq!(oct_digit1(a), Err(Err::Error((a, ErrorKind::OctDigit)))); |
863 | assert_eq!( |
864 | oct_digit1::<_, (_, ErrorKind)>(b), |
865 | Err(Err::Incomplete(Needed::new(1))) |
866 | ); |
867 | assert_eq!(oct_digit1(c), Err(Err::Error((c, ErrorKind::OctDigit)))); |
868 | assert_eq!(oct_digit1(d), Err(Err::Error((d, ErrorKind::OctDigit)))); |
869 | assert_eq!( |
870 | alphanumeric1::<_, (_, ErrorKind)>(a), |
871 | Err(Err::Incomplete(Needed::new(1))) |
872 | ); |
873 | //assert_eq!(fix_error!(b,(), alphanumeric1), Ok((empty, b))); |
874 | assert_eq!( |
875 | alphanumeric1::<_, (_, ErrorKind)>(c), |
876 | Err(Err::Incomplete(Needed::new(1))) |
877 | ); |
878 | assert_eq!(alphanumeric1::<_, (_, ErrorKind)>(d), Ok(("é12" , "az" ))); |
879 | assert_eq!( |
880 | space1::<_, (_, ErrorKind)>(e), |
881 | Err(Err::Incomplete(Needed::new(1))) |
882 | ); |
883 | } |
884 | |
885 | use crate::traits::Offset; |
886 | #[test ] |
887 | fn offset() { |
888 | let a = &b"abcd;" [..]; |
889 | let b = &b"1234;" [..]; |
890 | let c = &b"a123;" [..]; |
891 | let d = &b" \t;" [..]; |
892 | let e = &b" \t\r\n;" [..]; |
893 | let f = &b"123abcDEF;" [..]; |
894 | |
895 | match alpha1::<_, (_, ErrorKind)>(a) { |
896 | Ok((i, _)) => { |
897 | assert_eq!(a.offset(i) + i.len(), a.len()); |
898 | } |
899 | _ => panic!("wrong return type in offset test for alpha" ), |
900 | } |
901 | match digit1::<_, (_, ErrorKind)>(b) { |
902 | Ok((i, _)) => { |
903 | assert_eq!(b.offset(i) + i.len(), b.len()); |
904 | } |
905 | _ => panic!("wrong return type in offset test for digit" ), |
906 | } |
907 | match alphanumeric1::<_, (_, ErrorKind)>(c) { |
908 | Ok((i, _)) => { |
909 | assert_eq!(c.offset(i) + i.len(), c.len()); |
910 | } |
911 | _ => panic!("wrong return type in offset test for alphanumeric" ), |
912 | } |
913 | match space1::<_, (_, ErrorKind)>(d) { |
914 | Ok((i, _)) => { |
915 | assert_eq!(d.offset(i) + i.len(), d.len()); |
916 | } |
917 | _ => panic!("wrong return type in offset test for space" ), |
918 | } |
919 | match multispace1::<_, (_, ErrorKind)>(e) { |
920 | Ok((i, _)) => { |
921 | assert_eq!(e.offset(i) + i.len(), e.len()); |
922 | } |
923 | _ => panic!("wrong return type in offset test for multispace" ), |
924 | } |
925 | match hex_digit1::<_, (_, ErrorKind)>(f) { |
926 | Ok((i, _)) => { |
927 | assert_eq!(f.offset(i) + i.len(), f.len()); |
928 | } |
929 | _ => panic!("wrong return type in offset test for hex_digit" ), |
930 | } |
931 | match oct_digit1::<_, (_, ErrorKind)>(f) { |
932 | Ok((i, _)) => { |
933 | assert_eq!(f.offset(i) + i.len(), f.len()); |
934 | } |
935 | _ => panic!("wrong return type in offset test for oct_digit" ), |
936 | } |
937 | } |
938 | |
939 | #[test ] |
940 | fn is_not_line_ending_bytes() { |
941 | let a: &[u8] = b"ab12cd \nefgh" ; |
942 | assert_eq!( |
943 | not_line_ending::<_, (_, ErrorKind)>(a), |
944 | Ok((&b" \nefgh" [..], &b"ab12cd" [..])) |
945 | ); |
946 | |
947 | let b: &[u8] = b"ab12cd \nefgh \nijkl" ; |
948 | assert_eq!( |
949 | not_line_ending::<_, (_, ErrorKind)>(b), |
950 | Ok((&b" \nefgh \nijkl" [..], &b"ab12cd" [..])) |
951 | ); |
952 | |
953 | let c: &[u8] = b"ab12cd \r\nefgh \nijkl" ; |
954 | assert_eq!( |
955 | not_line_ending::<_, (_, ErrorKind)>(c), |
956 | Ok((&b" \r\nefgh \nijkl" [..], &b"ab12cd" [..])) |
957 | ); |
958 | |
959 | let d: &[u8] = b"ab12cd" ; |
960 | assert_eq!( |
961 | not_line_ending::<_, (_, ErrorKind)>(d), |
962 | Err(Err::Incomplete(Needed::Unknown)) |
963 | ); |
964 | } |
965 | |
966 | #[test ] |
967 | fn is_not_line_ending_str() { |
968 | /* |
969 | let a: &str = "ab12cd\nefgh"; |
970 | assert_eq!(not_line_ending(a), Ok((&"\nefgh"[..], &"ab12cd"[..]))); |
971 | |
972 | let b: &str = "ab12cd\nefgh\nijkl"; |
973 | assert_eq!(not_line_ending(b), Ok((&"\nefgh\nijkl"[..], &"ab12cd"[..]))); |
974 | |
975 | let c: &str = "ab12cd\r\nefgh\nijkl"; |
976 | assert_eq!(not_line_ending(c), Ok((&"\r\nefgh\nijkl"[..], &"ab12cd"[..]))); |
977 | |
978 | let d = "βèƒôřè\nÂßÇáƒƭèř"; |
979 | assert_eq!(not_line_ending(d), Ok((&"\nÂßÇáƒƭèř"[..], &"βèƒôřè"[..]))); |
980 | |
981 | let e = "βèƒôřè\r\nÂßÇáƒƭèř"; |
982 | assert_eq!(not_line_ending(e), Ok((&"\r\nÂßÇáƒƭèř"[..], &"βèƒôřè"[..]))); |
983 | */ |
984 | |
985 | let f = "βèƒôřè \rÂßÇáƒƭèř" ; |
986 | assert_eq!(not_line_ending(f), Err(Err::Error((f, ErrorKind::Tag)))); |
987 | |
988 | let g2: &str = "ab12cd" ; |
989 | assert_eq!( |
990 | not_line_ending::<_, (_, ErrorKind)>(g2), |
991 | Err(Err::Incomplete(Needed::Unknown)) |
992 | ); |
993 | } |
994 | |
995 | #[test ] |
996 | fn hex_digit_test() { |
997 | let i = &b"0123456789abcdefABCDEF;" [..]; |
998 | assert_parse!(hex_digit1(i), Ok((&b";" [..], &i[..i.len() - 1]))); |
999 | |
1000 | let i = &b"g" [..]; |
1001 | assert_parse!( |
1002 | hex_digit1(i), |
1003 | Err(Err::Error(error_position!(i, ErrorKind::HexDigit))) |
1004 | ); |
1005 | |
1006 | let i = &b"G" [..]; |
1007 | assert_parse!( |
1008 | hex_digit1(i), |
1009 | Err(Err::Error(error_position!(i, ErrorKind::HexDigit))) |
1010 | ); |
1011 | |
1012 | assert!(crate::character::is_hex_digit(b'0' )); |
1013 | assert!(crate::character::is_hex_digit(b'9' )); |
1014 | assert!(crate::character::is_hex_digit(b'a' )); |
1015 | assert!(crate::character::is_hex_digit(b'f' )); |
1016 | assert!(crate::character::is_hex_digit(b'A' )); |
1017 | assert!(crate::character::is_hex_digit(b'F' )); |
1018 | assert!(!crate::character::is_hex_digit(b'g' )); |
1019 | assert!(!crate::character::is_hex_digit(b'G' )); |
1020 | assert!(!crate::character::is_hex_digit(b'/' )); |
1021 | assert!(!crate::character::is_hex_digit(b':' )); |
1022 | assert!(!crate::character::is_hex_digit(b'@' )); |
1023 | assert!(!crate::character::is_hex_digit(b' \x60' )); |
1024 | } |
1025 | |
1026 | #[test ] |
1027 | fn oct_digit_test() { |
1028 | let i = &b"01234567;" [..]; |
1029 | assert_parse!(oct_digit1(i), Ok((&b";" [..], &i[..i.len() - 1]))); |
1030 | |
1031 | let i = &b"8" [..]; |
1032 | assert_parse!( |
1033 | oct_digit1(i), |
1034 | Err(Err::Error(error_position!(i, ErrorKind::OctDigit))) |
1035 | ); |
1036 | |
1037 | assert!(crate::character::is_oct_digit(b'0' )); |
1038 | assert!(crate::character::is_oct_digit(b'7' )); |
1039 | assert!(!crate::character::is_oct_digit(b'8' )); |
1040 | assert!(!crate::character::is_oct_digit(b'9' )); |
1041 | assert!(!crate::character::is_oct_digit(b'a' )); |
1042 | assert!(!crate::character::is_oct_digit(b'A' )); |
1043 | assert!(!crate::character::is_oct_digit(b'/' )); |
1044 | assert!(!crate::character::is_oct_digit(b':' )); |
1045 | assert!(!crate::character::is_oct_digit(b'@' )); |
1046 | assert!(!crate::character::is_oct_digit(b' \x60' )); |
1047 | } |
1048 | |
1049 | #[test ] |
1050 | fn full_line_windows() { |
1051 | fn take_full_line(i: &[u8]) -> IResult<&[u8], (&[u8], &[u8])> { |
1052 | pair(not_line_ending, line_ending)(i) |
1053 | } |
1054 | let input = b"abc \r\n" ; |
1055 | let output = take_full_line(input); |
1056 | assert_eq!(output, Ok((&b"" [..], (&b"abc" [..], &b" \r\n" [..])))); |
1057 | } |
1058 | |
1059 | #[test ] |
1060 | fn full_line_unix() { |
1061 | fn take_full_line(i: &[u8]) -> IResult<&[u8], (&[u8], &[u8])> { |
1062 | pair(not_line_ending, line_ending)(i) |
1063 | } |
1064 | let input = b"abc \n" ; |
1065 | let output = take_full_line(input); |
1066 | assert_eq!(output, Ok((&b"" [..], (&b"abc" [..], &b" \n" [..])))); |
1067 | } |
1068 | |
1069 | #[test ] |
1070 | fn check_windows_lineending() { |
1071 | let input = b" \r\n" ; |
1072 | let output = line_ending(&input[..]); |
1073 | assert_parse!(output, Ok((&b"" [..], &b" \r\n" [..]))); |
1074 | } |
1075 | |
1076 | #[test ] |
1077 | fn check_unix_lineending() { |
1078 | let input = b" \n" ; |
1079 | let output = line_ending(&input[..]); |
1080 | assert_parse!(output, Ok((&b"" [..], &b" \n" [..]))); |
1081 | } |
1082 | |
1083 | #[test ] |
1084 | fn cr_lf() { |
1085 | assert_parse!(crlf(&b" \r\na" [..]), Ok((&b"a" [..], &b" \r\n" [..]))); |
1086 | assert_parse!(crlf(&b" \r" [..]), Err(Err::Incomplete(Needed::new(2)))); |
1087 | assert_parse!( |
1088 | crlf(&b" \ra" [..]), |
1089 | Err(Err::Error(error_position!(&b" \ra" [..], ErrorKind::CrLf))) |
1090 | ); |
1091 | |
1092 | assert_parse!(crlf(" \r\na" ), Ok(("a" , " \r\n" ))); |
1093 | assert_parse!(crlf(" \r" ), Err(Err::Incomplete(Needed::new(2)))); |
1094 | assert_parse!( |
1095 | crlf(" \ra" ), |
1096 | Err(Err::Error(error_position!(" \ra" , ErrorKind::CrLf))) |
1097 | ); |
1098 | } |
1099 | |
1100 | #[test ] |
1101 | fn end_of_line() { |
1102 | assert_parse!(line_ending(&b" \na" [..]), Ok((&b"a" [..], &b" \n" [..]))); |
1103 | assert_parse!(line_ending(&b" \r\na" [..]), Ok((&b"a" [..], &b" \r\n" [..]))); |
1104 | assert_parse!( |
1105 | line_ending(&b" \r" [..]), |
1106 | Err(Err::Incomplete(Needed::new(2))) |
1107 | ); |
1108 | assert_parse!( |
1109 | line_ending(&b" \ra" [..]), |
1110 | Err(Err::Error(error_position!(&b" \ra" [..], ErrorKind::CrLf))) |
1111 | ); |
1112 | |
1113 | assert_parse!(line_ending(" \na" ), Ok(("a" , " \n" ))); |
1114 | assert_parse!(line_ending(" \r\na" ), Ok(("a" , " \r\n" ))); |
1115 | assert_parse!(line_ending(" \r" ), Err(Err::Incomplete(Needed::new(2)))); |
1116 | assert_parse!( |
1117 | line_ending(" \ra" ), |
1118 | Err(Err::Error(error_position!(" \ra" , ErrorKind::CrLf))) |
1119 | ); |
1120 | } |
1121 | |
1122 | fn digit_to_i16(input: &str) -> IResult<&str, i16> { |
1123 | let i = input; |
1124 | let (i, opt_sign) = opt(alt((char('+' ), char('-' ))))(i)?; |
1125 | let sign = match opt_sign { |
1126 | Some('+' ) => true, |
1127 | Some('-' ) => false, |
1128 | _ => true, |
1129 | }; |
1130 | |
1131 | let (i, s) = match digit1::<_, crate::error::Error<_>>(i) { |
1132 | Ok((i, s)) => (i, s), |
1133 | Err(Err::Incomplete(i)) => return Err(Err::Incomplete(i)), |
1134 | Err(_) => { |
1135 | return Err(Err::Error(crate::error::Error::from_error_kind( |
1136 | input, |
1137 | ErrorKind::Digit, |
1138 | ))) |
1139 | } |
1140 | }; |
1141 | match s.parse_to() { |
1142 | Some(n) => { |
1143 | if sign { |
1144 | Ok((i, n)) |
1145 | } else { |
1146 | Ok((i, -n)) |
1147 | } |
1148 | } |
1149 | None => Err(Err::Error(crate::error::Error::from_error_kind( |
1150 | i, |
1151 | ErrorKind::Digit, |
1152 | ))), |
1153 | } |
1154 | } |
1155 | |
1156 | fn digit_to_u32(i: &str) -> IResult<&str, u32> { |
1157 | let (i, s) = digit1(i)?; |
1158 | match s.parse_to() { |
1159 | Some(n) => Ok((i, n)), |
1160 | None => Err(Err::Error(crate::error::Error::from_error_kind( |
1161 | i, |
1162 | ErrorKind::Digit, |
1163 | ))), |
1164 | } |
1165 | } |
1166 | |
1167 | proptest! { |
1168 | #[test] |
1169 | fn ints(s in " \\PC*" ) { |
1170 | let res1 = digit_to_i16(&s); |
1171 | let res2 = i16(s.as_str()); |
1172 | assert_eq!(res1, res2); |
1173 | } |
1174 | |
1175 | #[test] |
1176 | fn uints(s in " \\PC*" ) { |
1177 | let res1 = digit_to_u32(&s); |
1178 | let res2 = u32(s.as_str()); |
1179 | assert_eq!(res1, res2); |
1180 | } |
1181 | } |
1182 | } |
1183 | |