1 | // pest. The Elegant Parser |
2 | // Copyright (c) 2018 DragoČ™ Tiselice |
3 | // |
4 | // Licensed under the Apache License, Version 2.0 |
5 | // <LICENSE-APACHE or http://www.apache.org/licenses/LICENSE-2.0> or the MIT |
6 | // license <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your |
7 | // option. All files in the project carrying such notice may not be copied, |
8 | // modified, or distributed except according to those terms. |
9 | |
10 | use alloc::format; |
11 | use alloc::rc::Rc; |
12 | use alloc::string::String; |
13 | use alloc::vec::Vec; |
14 | use core::fmt; |
15 | use core::hash::{Hash, Hasher}; |
16 | use core::iter::Filter; |
17 | use core::ptr; |
18 | use core::str; |
19 | |
20 | #[cfg (feature = "pretty-print" )] |
21 | use serde::ser::SerializeStruct; |
22 | |
23 | use super::flat_pairs::{self, FlatPairs}; |
24 | use super::line_index::LineIndex; |
25 | use super::pair::{self, Pair}; |
26 | use super::queueable_token::QueueableToken; |
27 | use super::tokens::{self, Tokens}; |
28 | use crate::RuleType; |
29 | |
30 | /// An iterator over [`Pair`]s. It is created by [`pest::state`] and [`Pair::into_inner`]. |
31 | /// |
32 | /// [`Pair`]: struct.Pair.html |
33 | /// [`pest::state`]: ../fn.state.html |
34 | /// [`Pair::into_inner`]: struct.Pair.html#method.into_inner |
35 | #[derive (Clone)] |
36 | pub struct Pairs<'i, R> { |
37 | queue: Rc<Vec<QueueableToken<'i, R>>>, |
38 | input: &'i str, |
39 | start: usize, |
40 | end: usize, |
41 | pairs_count: usize, |
42 | line_index: Rc<LineIndex>, |
43 | } |
44 | |
45 | pub fn new<'i, R: RuleType>( |
46 | queue: Rc<Vec<QueueableToken<'i, R>>>, |
47 | input: &'i str, |
48 | line_index: Option<Rc<LineIndex>>, |
49 | start: usize, |
50 | end: usize, |
51 | ) -> Pairs<'i, R> { |
52 | let line_index = match line_index { |
53 | Some(line_index) => line_index, |
54 | None => Rc::new(LineIndex::new(input)), |
55 | }; |
56 | |
57 | let mut pairs_count = 0; |
58 | let mut cursor = start; |
59 | while cursor < end { |
60 | cursor = match queue[cursor] { |
61 | QueueableToken::Start { |
62 | end_token_index, .. |
63 | } => end_token_index, |
64 | _ => unreachable!(), |
65 | } + 1; |
66 | pairs_count += 1; |
67 | } |
68 | |
69 | Pairs { |
70 | queue, |
71 | input, |
72 | start, |
73 | end, |
74 | pairs_count, |
75 | line_index, |
76 | } |
77 | } |
78 | |
79 | impl<'i, R: RuleType> Pairs<'i, R> { |
80 | /// Captures a slice from the `&str` defined by the starting position of the first token `Pair` |
81 | /// and the ending position of the last token `Pair` of the `Pairs`. This also captures |
82 | /// the input between those two token `Pair`s. |
83 | /// |
84 | /// # Examples |
85 | /// |
86 | /// ``` |
87 | /// # use std::rc::Rc; |
88 | /// # use pest; |
89 | /// # #[allow (non_camel_case_types)] |
90 | /// # #[derive(Clone, Copy, Debug, Eq, Hash, Ord, PartialEq, PartialOrd)] |
91 | /// enum Rule { |
92 | /// a, |
93 | /// b |
94 | /// } |
95 | /// |
96 | /// let input = "a b" ; |
97 | /// let pairs = pest::state(input, |state| { |
98 | /// // generating Token pairs with Rule::a and Rule::b ... |
99 | /// # state.rule(Rule::a, |s| s.match_string("a" )).and_then(|s| s.skip(1)) |
100 | /// # .and_then(|s| s.rule(Rule::b, |s| s.match_string("b" ))) |
101 | /// }).unwrap(); |
102 | /// |
103 | /// assert_eq!(pairs.as_str(), "a b" ); |
104 | /// ``` |
105 | #[inline ] |
106 | pub fn as_str(&self) -> &'i str { |
107 | if self.start < self.end { |
108 | let start = self.pos(self.start); |
109 | let end = self.pos(self.end - 1); |
110 | // Generated positions always come from Positions and are UTF-8 borders. |
111 | &self.input[start..end] |
112 | } else { |
113 | "" |
114 | } |
115 | } |
116 | |
117 | /// Captures inner token `Pair`s and concatenates resulting `&str`s. This does not capture |
118 | /// the input between token `Pair`s. |
119 | /// |
120 | /// # Examples |
121 | /// |
122 | /// ``` |
123 | /// # use std::rc::Rc; |
124 | /// # use pest; |
125 | /// # #[allow (non_camel_case_types)] |
126 | /// # #[derive(Clone, Copy, Debug, Eq, Hash, Ord, PartialEq, PartialOrd)] |
127 | /// enum Rule { |
128 | /// a, |
129 | /// b |
130 | /// } |
131 | /// |
132 | /// let input = "a b" ; |
133 | /// let pairs = pest::state(input, |state| { |
134 | /// // generating Token pairs with Rule::a and Rule::b ... |
135 | /// # state.rule(Rule::a, |s| s.match_string("a" )).and_then(|s| s.skip(1)) |
136 | /// # .and_then(|s| s.rule(Rule::b, |s| s.match_string("b" ))) |
137 | /// }).unwrap(); |
138 | /// |
139 | /// assert_eq!(pairs.concat(), "ab" ); |
140 | /// ``` |
141 | #[inline ] |
142 | pub fn concat(&self) -> String { |
143 | self.clone() |
144 | .fold(String::new(), |string, pair| string + pair.as_str()) |
145 | } |
146 | |
147 | /// Flattens the `Pairs`. |
148 | /// |
149 | /// # Examples |
150 | /// |
151 | /// ``` |
152 | /// # use std::rc::Rc; |
153 | /// # use pest; |
154 | /// # #[allow (non_camel_case_types)] |
155 | /// # #[derive(Clone, Copy, Debug, Eq, Hash, Ord, PartialEq, PartialOrd)] |
156 | /// enum Rule { |
157 | /// a, |
158 | /// b |
159 | /// } |
160 | /// |
161 | /// let input = "" ; |
162 | /// let pairs = pest::state(input, |state| { |
163 | /// // generating nested Token pair with Rule::b inside Rule::a |
164 | /// # state.rule(Rule::a, |state| { |
165 | /// # state.rule(Rule::b, |s| Ok(s)) |
166 | /// # }) |
167 | /// }).unwrap(); |
168 | /// let tokens: Vec<_> = pairs.flatten().tokens().collect(); |
169 | /// |
170 | /// assert_eq!(tokens.len(), 4); |
171 | /// ``` |
172 | #[inline ] |
173 | pub fn flatten(self) -> FlatPairs<'i, R> { |
174 | unsafe { flat_pairs::new(self.queue, self.input, self.start, self.end) } |
175 | } |
176 | |
177 | /// Finds the first pair that has its node or branch tagged with the provided |
178 | /// label. |
179 | /// |
180 | /// # Examples |
181 | /// |
182 | /// Try to recognize the branch between add and mul |
183 | /// ``` |
184 | /// use pest::{state, ParseResult, ParserState}; |
185 | /// #[allow(non_camel_case_types)] |
186 | /// #[derive(Clone, Copy, Debug, Eq, Hash, Ord, PartialEq, PartialOrd)] |
187 | /// enum Rule { |
188 | /// number, // 0..9 |
189 | /// add, // num + num |
190 | /// mul, // num * num |
191 | /// } |
192 | /// fn mark_branch( |
193 | /// state: Box<ParserState<'_, Rule>>, |
194 | /// ) -> ParseResult<Box<ParserState<'_, Rule>>> { |
195 | /// expr(state, Rule::mul, "*" ) |
196 | /// .and_then(|state| state.tag_node(std::borrow::Cow::Borrowed("mul" ))) |
197 | /// .or_else(|state| expr(state, Rule::add, "+" )) |
198 | /// .and_then(|state| state.tag_node(std::borrow::Cow::Borrowed("add" ))) |
199 | /// } |
200 | /// fn expr<'a>( |
201 | /// state: Box<ParserState<'a, Rule>>, |
202 | /// r: Rule, |
203 | /// o: &'static str, |
204 | /// ) -> ParseResult<Box<ParserState<'a, Rule>>> { |
205 | /// state.rule(r, |state| { |
206 | /// state.sequence(|state| { |
207 | /// number(state) |
208 | /// .and_then(|state| state.tag_node(std::borrow::Cow::Borrowed("lhs" ))) |
209 | /// .and_then(|state| state.match_string(o)) |
210 | /// .and_then(number) |
211 | /// .and_then(|state| state.tag_node(std::borrow::Cow::Borrowed("rhs" ))) |
212 | /// }) |
213 | /// }) |
214 | /// } |
215 | /// fn number(state: Box<ParserState<'_, Rule>>) -> ParseResult<Box<ParserState<'_, Rule>>> { |
216 | /// state.rule(Rule::number, |state| state.match_range('0' ..'9' )) |
217 | /// } |
218 | /// let input = "1+2" ; |
219 | /// let pairs = state(input, mark_branch).unwrap(); |
220 | /// assert_eq!(pairs.find_first_tagged("add" ).unwrap().as_rule(), Rule::add); |
221 | /// assert_eq!(pairs.find_first_tagged("mul" ), None); |
222 | /// ``` |
223 | #[inline ] |
224 | pub fn find_first_tagged(&self, tag: &'i str) -> Option<Pair<'i, R>> { |
225 | self.clone().find_tagged(tag).next() |
226 | } |
227 | |
228 | /// Returns the iterator over pairs that have their node or branch tagged |
229 | /// with the provided label. |
230 | /// |
231 | /// # Examples |
232 | /// |
233 | /// Try to recognize the node between left and right hand side |
234 | /// ``` |
235 | /// use pest::{state, ParseResult, ParserState}; |
236 | /// #[allow(non_camel_case_types)] |
237 | /// #[derive(Clone, Copy, Debug, Eq, Hash, Ord, PartialEq, PartialOrd)] |
238 | /// enum Rule { |
239 | /// number, // 0..9 |
240 | /// add, // num + num |
241 | /// mul, // num * num |
242 | /// } |
243 | /// fn mark_branch( |
244 | /// state: Box<ParserState<'_, Rule>>, |
245 | /// ) -> ParseResult<Box<ParserState<'_, Rule>>> { |
246 | /// expr(state, Rule::mul, "*" ) |
247 | /// .and_then(|state| state.tag_node(std::borrow::Cow::Borrowed("mul" ))) |
248 | /// .or_else(|state| expr(state, Rule::add, "+" )) |
249 | /// .and_then(|state| state.tag_node(std::borrow::Cow::Borrowed("add" ))) |
250 | /// } |
251 | /// fn expr<'a>( |
252 | /// state: Box<ParserState<'a, Rule>>, |
253 | /// r: Rule, |
254 | /// o: &'static str, |
255 | /// ) -> ParseResult<Box<ParserState<'a, Rule>>> { |
256 | /// state.rule(r, |state| { |
257 | /// state.sequence(|state| { |
258 | /// number(state) |
259 | /// .and_then(|state| state.tag_node(std::borrow::Cow::Borrowed("lhs" ))) |
260 | /// .and_then(|state| state.match_string(o)) |
261 | /// .and_then(number) |
262 | /// .and_then(|state| state.tag_node(std::borrow::Cow::Borrowed("rhs" ))) |
263 | /// }) |
264 | /// }) |
265 | /// } |
266 | /// fn number(state: Box<ParserState<'_, Rule>>) -> ParseResult<Box<ParserState<'_, Rule>>> { |
267 | /// state.rule(Rule::number, |state| state.match_range('0' ..'9' )) |
268 | /// } |
269 | /// |
270 | /// let input = "1+2" ; |
271 | /// let pairs = state(input, mark_branch).unwrap(); |
272 | /// let mut left_numbers = pairs.find_tagged("lhs" ); |
273 | /// assert_eq!(left_numbers.next().unwrap().as_str(), "1" ); |
274 | /// assert_eq!(left_numbers.next(), None); |
275 | /// ``` |
276 | #[inline ] |
277 | pub fn find_tagged( |
278 | self, |
279 | tag: &'i str, |
280 | ) -> Filter<FlatPairs<'i, R>, impl FnMut(&Pair<'i, R>) -> bool + '_> { |
281 | self.flatten() |
282 | .filter(move |pair: &Pair<'i, R>| matches!(pair.as_node_tag(), Some(nt) if nt == tag)) |
283 | } |
284 | |
285 | /// Returns the `Tokens` for the `Pairs`. |
286 | /// |
287 | /// # Examples |
288 | /// |
289 | /// ``` |
290 | /// # use std::rc::Rc; |
291 | /// # use pest; |
292 | /// # #[allow (non_camel_case_types)] |
293 | /// # #[derive(Clone, Copy, Debug, Eq, Hash, Ord, PartialEq, PartialOrd)] |
294 | /// enum Rule { |
295 | /// a |
296 | /// } |
297 | /// |
298 | /// let input = "" ; |
299 | /// let pairs = pest::state(input, |state| { |
300 | /// // generating Token pair with Rule::a ... |
301 | /// # state.rule(Rule::a, |s| Ok(s)) |
302 | /// }).unwrap(); |
303 | /// let tokens: Vec<_> = pairs.tokens().collect(); |
304 | /// |
305 | /// assert_eq!(tokens.len(), 2); |
306 | /// ``` |
307 | #[inline ] |
308 | pub fn tokens(self) -> Tokens<'i, R> { |
309 | tokens::new(self.queue, self.input, self.start, self.end) |
310 | } |
311 | |
312 | /// Peek at the first inner `Pair` without changing the position of this iterator. |
313 | #[inline ] |
314 | pub fn peek(&self) -> Option<Pair<'i, R>> { |
315 | if self.start < self.end { |
316 | Some(unsafe { |
317 | pair::new( |
318 | Rc::clone(&self.queue), |
319 | self.input, |
320 | Rc::clone(&self.line_index), |
321 | self.start, |
322 | ) |
323 | }) |
324 | } else { |
325 | None |
326 | } |
327 | } |
328 | |
329 | /// Generates a string that stores the lexical information of `self` in |
330 | /// a pretty-printed JSON format. |
331 | #[cfg (feature = "pretty-print" )] |
332 | pub fn to_json(&self) -> String { |
333 | ::serde_json::to_string_pretty(self).expect("Failed to pretty-print Pairs to json." ) |
334 | } |
335 | |
336 | fn pair(&self) -> usize { |
337 | match self.queue[self.start] { |
338 | QueueableToken::Start { |
339 | end_token_index, .. |
340 | } => end_token_index, |
341 | _ => unreachable!(), |
342 | } |
343 | } |
344 | |
345 | fn pair_from_end(&self) -> usize { |
346 | match self.queue[self.end - 1] { |
347 | QueueableToken::End { |
348 | start_token_index, .. |
349 | } => start_token_index, |
350 | _ => unreachable!(), |
351 | } |
352 | } |
353 | |
354 | fn pos(&self, index: usize) -> usize { |
355 | match self.queue[index] { |
356 | QueueableToken::Start { input_pos, .. } | QueueableToken::End { input_pos, .. } => { |
357 | input_pos |
358 | } |
359 | } |
360 | } |
361 | } |
362 | |
363 | impl<'i, R: RuleType> ExactSizeIterator for Pairs<'i, R> { |
364 | #[inline ] |
365 | fn len(&self) -> usize { |
366 | self.pairs_count |
367 | } |
368 | } |
369 | |
370 | impl<'i, R: RuleType> Iterator for Pairs<'i, R> { |
371 | type Item = Pair<'i, R>; |
372 | |
373 | fn next(&mut self) -> Option<Self::Item> { |
374 | let pair: Pair<'_, R> = self.peek()?; |
375 | |
376 | self.start = self.pair() + 1; |
377 | self.pairs_count -= 1; |
378 | Some(pair) |
379 | } |
380 | |
381 | fn size_hint(&self) -> (usize, Option<usize>) { |
382 | let len: usize = <Self as ExactSizeIterator>::len(self); |
383 | (len, Some(len)) |
384 | } |
385 | } |
386 | |
387 | impl<'i, R: RuleType> DoubleEndedIterator for Pairs<'i, R> { |
388 | fn next_back(&mut self) -> Option<Self::Item> { |
389 | if self.end <= self.start { |
390 | return None; |
391 | } |
392 | |
393 | self.end = self.pair_from_end(); |
394 | self.pairs_count -= 1; |
395 | |
396 | let pair: Pair<'_, R> = unsafe { |
397 | pair::new( |
398 | queue:Rc::clone(&self.queue), |
399 | self.input, |
400 | line_index:Rc::clone(&self.line_index), |
401 | self.end, |
402 | ) |
403 | }; |
404 | |
405 | Some(pair) |
406 | } |
407 | } |
408 | |
409 | impl<'i, R: RuleType> fmt::Debug for Pairs<'i, R> { |
410 | fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result { |
411 | f.debug_list().entries(self.clone()).finish() |
412 | } |
413 | } |
414 | |
415 | impl<'i, R: RuleType> fmt::Display for Pairs<'i, R> { |
416 | fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result { |
417 | write!( |
418 | f, |
419 | "[ {}]" , |
420 | self.clone() |
421 | .map(|pair| format!(" {}" , pair)) |
422 | .collect::<Vec<_>>() |
423 | .join(", " ) |
424 | ) |
425 | } |
426 | } |
427 | |
428 | impl<'i, R: PartialEq> PartialEq for Pairs<'i, R> { |
429 | fn eq(&self, other: &Pairs<'i, R>) -> bool { |
430 | Rc::ptr_eq(&self.queue, &other.queue) |
431 | && ptr::eq(self.input, b:other.input) |
432 | && self.start == other.start |
433 | && self.end == other.end |
434 | } |
435 | } |
436 | |
437 | impl<'i, R: Eq> Eq for Pairs<'i, R> {} |
438 | |
439 | impl<'i, R: Hash> Hash for Pairs<'i, R> { |
440 | fn hash<H: Hasher>(&self, state: &mut H) { |
441 | (&*self.queue as *const Vec<QueueableToken<'i, R>>).hash(state); |
442 | (self.input as *const str).hash(state); |
443 | self.start.hash(state); |
444 | self.end.hash(state); |
445 | } |
446 | } |
447 | |
448 | #[cfg (feature = "pretty-print" )] |
449 | impl<'i, R: RuleType> ::serde::Serialize for Pairs<'i, R> { |
450 | fn serialize<S>(&self, serializer: S) -> Result<S::Ok, S::Error> |
451 | where |
452 | S: ::serde::Serializer, |
453 | { |
454 | let start = self.pos(self.start); |
455 | let end = self.pos(self.end - 1); |
456 | let pairs = self.clone().collect::<Vec<_>>(); |
457 | |
458 | let mut ser = serializer.serialize_struct("Pairs" , 2)?; |
459 | ser.serialize_field("pos" , &(start, end))?; |
460 | ser.serialize_field("pairs" , &pairs)?; |
461 | ser.end() |
462 | } |
463 | } |
464 | |
465 | #[cfg (test)] |
466 | mod tests { |
467 | use super::super::super::macros::tests::*; |
468 | use super::super::super::Parser; |
469 | use alloc::borrow::ToOwned; |
470 | use alloc::boxed::Box; |
471 | use alloc::format; |
472 | use alloc::vec; |
473 | use alloc::vec::Vec; |
474 | |
475 | #[test ] |
476 | #[cfg (feature = "pretty-print" )] |
477 | fn test_pretty_print() { |
478 | let pairs = AbcParser::parse(Rule::a, "abcde" ).unwrap(); |
479 | |
480 | let expected = r#"{ |
481 | "pos": [ |
482 | 0, |
483 | 5 |
484 | ], |
485 | "pairs": [ |
486 | { |
487 | "pos": [ |
488 | 0, |
489 | 3 |
490 | ], |
491 | "rule": "a", |
492 | "inner": { |
493 | "pos": [ |
494 | 1, |
495 | 2 |
496 | ], |
497 | "pairs": [ |
498 | { |
499 | "pos": [ |
500 | 1, |
501 | 2 |
502 | ], |
503 | "rule": "b", |
504 | "inner": "b" |
505 | } |
506 | ] |
507 | } |
508 | }, |
509 | { |
510 | "pos": [ |
511 | 4, |
512 | 5 |
513 | ], |
514 | "rule": "c", |
515 | "inner": "e" |
516 | } |
517 | ] |
518 | }"# ; |
519 | |
520 | assert_eq!(expected, pairs.to_json()); |
521 | } |
522 | |
523 | #[test ] |
524 | fn as_str() { |
525 | let pairs = AbcParser::parse(Rule::a, "abcde" ).unwrap(); |
526 | |
527 | assert_eq!(pairs.as_str(), "abcde" ); |
528 | } |
529 | |
530 | #[test ] |
531 | fn as_str_empty() { |
532 | let mut pairs = AbcParser::parse(Rule::a, "abcde" ).unwrap(); |
533 | |
534 | assert_eq!(pairs.nth(1).unwrap().into_inner().as_str(), "" ); |
535 | } |
536 | |
537 | #[test ] |
538 | fn concat() { |
539 | let pairs = AbcParser::parse(Rule::a, "abcde" ).unwrap(); |
540 | |
541 | assert_eq!(pairs.concat(), "abce" ); |
542 | } |
543 | |
544 | #[test ] |
545 | fn pairs_debug() { |
546 | let pairs = AbcParser::parse(Rule::a, "abcde" ).unwrap(); |
547 | |
548 | #[rustfmt::skip] |
549 | assert_eq!( |
550 | format!(" {:?}" , pairs), |
551 | "[\ |
552 | Pair { rule: a, span: Span { str: \"abc \", start: 0, end: 3 }, inner: [\ |
553 | Pair { rule: b, span: Span { str: \"b \", start: 1, end: 2 }, inner: [] }\ |
554 | ] }, \ |
555 | Pair { rule: c, span: Span { str: \"e \", start: 4, end: 5 }, inner: [] }\ |
556 | ]" |
557 | .to_owned() |
558 | ); |
559 | } |
560 | |
561 | #[test ] |
562 | fn pairs_display() { |
563 | let pairs = AbcParser::parse(Rule::a, "abcde" ).unwrap(); |
564 | |
565 | assert_eq!( |
566 | format!(" {}" , pairs), |
567 | "[a(0, 3, [b(1, 2)]), c(4, 5)]" .to_owned() |
568 | ); |
569 | } |
570 | |
571 | #[test ] |
572 | fn iter_for_pairs() { |
573 | let pairs = AbcParser::parse(Rule::a, "abcde" ).unwrap(); |
574 | assert_eq!( |
575 | pairs.map(|p| p.as_rule()).collect::<Vec<Rule>>(), |
576 | vec![Rule::a, Rule::c] |
577 | ); |
578 | } |
579 | |
580 | #[test ] |
581 | fn double_ended_iter_for_pairs() { |
582 | let pairs = AbcParser::parse(Rule::a, "abcde" ).unwrap(); |
583 | assert_eq!( |
584 | pairs.rev().map(|p| p.as_rule()).collect::<Vec<Rule>>(), |
585 | vec![Rule::c, Rule::a] |
586 | ); |
587 | } |
588 | |
589 | #[test ] |
590 | fn test_line_col() { |
591 | let mut pairs = AbcParser::parse(Rule::a, "abc \nefgh" ).unwrap(); |
592 | let pair = pairs.next().unwrap(); |
593 | assert_eq!(pair.as_str(), "abc" ); |
594 | assert_eq!(pair.line_col(), (1, 1)); |
595 | |
596 | let pair = pairs.next().unwrap(); |
597 | assert_eq!(pair.as_str(), "e" ); |
598 | assert_eq!(pair.line_col(), (2, 1)); |
599 | |
600 | let pair = pairs.next().unwrap(); |
601 | assert_eq!(pair.as_str(), "fgh" ); |
602 | assert_eq!(pair.line_col(), (2, 2)); |
603 | } |
604 | |
605 | #[test ] |
606 | fn test_rev_iter_line_col() { |
607 | let mut pairs = AbcParser::parse(Rule::a, "abc \nefgh" ).unwrap().rev(); |
608 | let pair = pairs.next().unwrap(); |
609 | assert_eq!(pair.as_str(), "fgh" ); |
610 | assert_eq!(pair.line_col(), (2, 2)); |
611 | |
612 | let pair = pairs.next().unwrap(); |
613 | assert_eq!(pair.as_str(), "e" ); |
614 | assert_eq!(pair.line_col(), (2, 1)); |
615 | |
616 | let pair = pairs.next().unwrap(); |
617 | assert_eq!(pair.as_str(), "abc" ); |
618 | assert_eq!(pair.line_col(), (1, 1)); |
619 | } |
620 | |
621 | #[test ] |
622 | fn test_tag_node_branch() { |
623 | use crate::{state, ParseResult, ParserState}; |
624 | #[allow (non_camel_case_types)] |
625 | #[derive (Clone, Copy, Debug, Eq, Hash, Ord, PartialEq, PartialOrd)] |
626 | enum Rule { |
627 | number, // 0..9 |
628 | add, // num + num |
629 | mul, // num * num |
630 | } |
631 | fn mark_branch( |
632 | state: Box<ParserState<'_, Rule>>, |
633 | ) -> ParseResult<Box<ParserState<'_, Rule>>> { |
634 | expr(state, Rule::mul, "*" ) |
635 | .and_then(|state| state.tag_node(alloc::borrow::Cow::Borrowed("mul" ))) |
636 | .or_else(|state| expr(state, Rule::add, "+" )) |
637 | .and_then(|state| state.tag_node(alloc::borrow::Cow::Borrowed("add" ))) |
638 | } |
639 | fn expr<'a>( |
640 | state: Box<ParserState<'a, Rule>>, |
641 | r: Rule, |
642 | o: &'static str, |
643 | ) -> ParseResult<Box<ParserState<'a, Rule>>> { |
644 | state.rule(r, |state| { |
645 | state.sequence(|state| { |
646 | number(state) |
647 | .and_then(|state| state.tag_node(alloc::borrow::Cow::Borrowed("lhs" ))) |
648 | .and_then(|state| state.match_string(o)) |
649 | .and_then(number) |
650 | .and_then(|state| state.tag_node(alloc::borrow::Cow::Borrowed("rhs" ))) |
651 | }) |
652 | }) |
653 | } |
654 | fn number(state: Box<ParserState<'_, Rule>>) -> ParseResult<Box<ParserState<'_, Rule>>> { |
655 | state.rule(Rule::number, |state| state.match_range('0' ..'9' )) |
656 | } |
657 | let input = "1+2" ; |
658 | let pairs = state(input, mark_branch).unwrap(); |
659 | assert_eq!(pairs.find_first_tagged("add" ).unwrap().as_rule(), Rule::add); |
660 | assert_eq!(pairs.find_first_tagged("mul" ), None); |
661 | |
662 | let mut left_numbers = pairs.clone().find_tagged("lhs" ); |
663 | |
664 | assert_eq!(left_numbers.next().unwrap().as_str(), "1" ); |
665 | assert_eq!(left_numbers.next(), None); |
666 | let mut right_numbers = pairs.find_tagged("rhs" ); |
667 | |
668 | assert_eq!(right_numbers.next().unwrap().as_str(), "2" ); |
669 | assert_eq!(right_numbers.next(), None); |
670 | } |
671 | |
672 | #[test ] |
673 | fn exact_size_iter_for_pairs() { |
674 | let pairs = AbcParser::parse(Rule::a, "abc \nefgh" ).unwrap(); |
675 | assert_eq!(pairs.len(), pairs.count()); |
676 | |
677 | let pairs = AbcParser::parse(Rule::a, "abc \nefgh" ).unwrap().rev(); |
678 | assert_eq!(pairs.len(), pairs.count()); |
679 | |
680 | let mut pairs = AbcParser::parse(Rule::a, "abc \nefgh" ).unwrap(); |
681 | let pairs_len = pairs.len(); |
682 | let _ = pairs.next().unwrap(); |
683 | assert_eq!(pairs.count() + 1, pairs_len); |
684 | } |
685 | } |
686 | |