cfgrammar/yacc/
ast.rs

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
use std::{
    collections::{HashMap, HashSet},
    fmt,
};

use indexmap::{IndexMap, IndexSet};

use super::{
    parser::YaccParser, Precedence, YaccGrammarError, YaccGrammarErrorKind, YaccGrammarWarning,
    YaccGrammarWarningKind, YaccKind,
};

use crate::Span;
/// Contains a `GrammarAST` structure produced from a grammar source file.
/// As well as any errors which occurred during the construction of the AST.
pub struct ASTWithValidityInfo {
    ast: GrammarAST,
    errs: Vec<YaccGrammarError>,
}

impl ASTWithValidityInfo {
    /// Parses a source file into an AST, returning an ast and any errors that were
    /// encountered during the construction of it.  The `ASTWithValidityInfo` can be
    /// then unused to construct a `YaccGrammar`, which will either produce an
    /// `Ok(YaccGrammar)` or an `Err` which includes these errors.
    pub fn new(yacc_kind: YaccKind, s: &str) -> Self {
        let mut errs = Vec::new();
        let ast = match yacc_kind {
            YaccKind::Original(_) | YaccKind::Grmtools | YaccKind::Eco => {
                let mut yp = YaccParser::new(yacc_kind, s.to_string());
                yp.parse().map_err(|e| errs.extend(e)).ok();
                let mut ast = yp.ast();
                ast.complete_and_validate().map_err(|e| errs.push(e)).ok();
                ast
            }
        };
        ASTWithValidityInfo { ast, errs }
    }

    /// Returns a `GrammarAST` constructed as the result of parsing a source file.
    /// When errors have occurred and `is_valid` returns false, this AST is the
    /// subset of the source file which parsed correctly while not encountering
    /// any errors. As such even when an AST is not valid, it will return an AST.
    pub fn ast(&self) -> &GrammarAST {
        &self.ast
    }

    /// Returns whether any errors where encountered during the
    /// parsing and validation of the AST during it's construction.
    pub fn is_valid(&self) -> bool {
        self.errors().is_empty()
    }

    /// Returns all errors which were encountered during AST construction.
    pub fn errors(&self) -> &[YaccGrammarError] {
        self.errs.as_slice()
    }
}

/// An AST representing a grammar. This is built up gradually: when it is finished, the
/// `complete_and_validate` must be called exactly once in order to finish the set-up. At that
/// point, any further mutations made to the struct lead to undefined behaviour.
#[derive(Debug)]
pub struct GrammarAST {
    pub start: Option<(String, Span)>,
    // map from a rule name to indexes into prods
    pub rules: IndexMap<String, Rule>,
    pub prods: Vec<Production>,
    pub tokens: IndexSet<String>,
    pub spans: Vec<Span>,
    pub precs: HashMap<String, (Precedence, Span)>,
    pub avoid_insert: Option<HashMap<String, Span>>,
    pub implicit_tokens: Option<HashMap<String, Span>>,
    // Error pretty-printers,
    // The first span of the value is the span of the key,
    // The second span in the value, is the span of the values string.
    pub epp: HashMap<String, (Span, (String, Span))>,
    pub expect: Option<(usize, Span)>,
    pub expectrr: Option<(usize, Span)>,
    pub parse_param: Option<(String, String)>,
    pub programs: Option<String>,
    // The set of symbol names that, if unused in a
    // grammar, will not cause a warning or error.
    pub expect_unused: Vec<Symbol>,
}

#[derive(Debug)]
pub struct Rule {
    pub name: (String, Span),
    pub pidxs: Vec<usize>, // index into GrammarAST.prod
    pub actiont: Option<String>,
}

#[derive(Debug)]
#[cfg_attr(test, derive(Eq, PartialEq))]
pub struct Production {
    pub symbols: Vec<Symbol>,
    pub precedence: Option<String>,
    pub action: Option<String>,
}

#[derive(Clone, Debug)]
#[cfg_attr(test, derive(Eq, PartialEq))]
pub enum Symbol {
    Rule(String, Span),
    Token(String, Span),
}

/// Specifies an index into a `GrammarAst.tokens` or a `GrammarAST.rules`.
/// Unlike `cfgrammar::Symbol` it is not parameterized by a `StorageT`.
#[derive(Eq, PartialEq, Debug, Copy, Clone)]
pub(crate) enum SymbolIdx {
    Rule(usize),
    Token(usize),
}

impl SymbolIdx {
    pub(crate) fn symbol(self, ast: &GrammarAST) -> Symbol {
        match self {
            SymbolIdx::Rule(idx) => {
                let (rule_name, rule_span) = &ast.rules[idx].name;
                Symbol::Rule(rule_name.clone(), *rule_span)
            }
            SymbolIdx::Token(idx) => {
                let tok = &ast.tokens[idx];
                Symbol::Token(tok.clone(), ast.spans[idx])
            }
        }
    }
}
impl fmt::Display for Symbol {
    fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
        match *self {
            Symbol::Rule(ref s, _) => write!(f, "{}", s),
            Symbol::Token(ref s, _) => write!(f, "{}", s),
        }
    }
}

impl GrammarAST {
    pub fn new() -> GrammarAST {
        GrammarAST {
            start: None,
            rules: IndexMap::new(), // Using an IndexMap means that we retain the order
            // of rules as they're found in the input file.
            prods: Vec::new(),
            spans: Vec::new(),
            tokens: IndexSet::new(),
            precs: HashMap::new(),
            avoid_insert: None,
            implicit_tokens: None,
            epp: HashMap::new(),
            expect: None,
            expectrr: None,
            parse_param: None,
            programs: None,
            expect_unused: Vec::new(),
        }
    }

    pub fn add_rule(&mut self, (name, name_span): (String, Span), actiont: Option<String>) {
        self.rules.insert(
            name.clone(),
            Rule {
                name: (name, name_span),
                pidxs: Vec::new(),
                actiont,
            },
        );
    }

    pub fn add_prod(
        &mut self,
        rule_name: String,
        symbols: Vec<Symbol>,
        precedence: Option<String>,
        action: Option<String>,
    ) {
        self.rules[&rule_name].pidxs.push(self.prods.len());
        self.prods.push(Production {
            symbols,
            precedence,
            action,
        });
    }

    #[deprecated(since = "0.10.2", note = "Please use set_programs instead")]
    pub fn add_programs(&mut self, s: String) {
        self.set_programs(s);
    }

    pub fn set_programs(&mut self, s: String) {
        self.programs = Some(s)
    }

    pub fn get_rule(&self, key: &str) -> Option<&Rule> {
        self.rules.get(key)
    }

    pub fn has_token(&self, s: &str) -> bool {
        self.tokens.contains(s)
    }

    /// After the AST has been populated, perform any final operations, and validate the grammar
    /// checking that:
    ///   1) The start rule references a rule in the grammar
    ///   2) Every rule reference references a rule in the grammar
    ///   3) Every token reference references a declared token
    ///   4) If a production has a precedence token, then it references a declared token
    ///   5) Every token declared with %epp matches a known token
    ///
    /// If the validation succeeds, None is returned.
    pub(crate) fn complete_and_validate(&mut self) -> Result<(), YaccGrammarError> {
        match self.start {
            None => {
                return Err(YaccGrammarError {
                    kind: YaccGrammarErrorKind::NoStartRule,
                    spans: vec![Span::new(0, 0)],
                });
            }
            Some((ref s, span)) => {
                if !self.rules.contains_key(s) {
                    return Err(YaccGrammarError {
                        kind: YaccGrammarErrorKind::InvalidStartRule(s.clone()),
                        spans: vec![span],
                    });
                }
            }
        }
        for rule in self.rules.values() {
            for &pidx in &rule.pidxs {
                let prod = &self.prods[pidx];
                if let Some(ref n) = prod.precedence {
                    if !self.tokens.contains(n) {
                        return Err(YaccGrammarError {
                            kind: YaccGrammarErrorKind::UnknownToken(n.clone()),
                            spans: vec![Span::new(0, 0)],
                        });
                    }
                    if !self.precs.contains_key(n) {
                        return Err(YaccGrammarError {
                            kind: YaccGrammarErrorKind::NoPrecForToken(n.clone()),
                            spans: vec![Span::new(0, 0)],
                        });
                    }
                }
                for sym in &prod.symbols {
                    match *sym {
                        Symbol::Rule(ref name, span) => {
                            if !self.rules.contains_key(name) {
                                return Err(YaccGrammarError {
                                    kind: YaccGrammarErrorKind::UnknownRuleRef(name.clone()),
                                    spans: vec![span],
                                });
                            }
                        }
                        Symbol::Token(ref name, span) => {
                            if !self.tokens.contains(name) {
                                return Err(YaccGrammarError {
                                    kind: YaccGrammarErrorKind::UnknownToken(name.clone()),
                                    spans: vec![span],
                                });
                            }
                        }
                    }
                }
            }
        }

        for (k, (sp, _)) in self.epp.iter() {
            if self.tokens.contains(k) {
                continue;
            }
            if let Some(ref it) = self.implicit_tokens {
                if it.contains_key(k) {
                    continue;
                }
            }
            return Err(YaccGrammarError {
                kind: YaccGrammarErrorKind::UnknownEPP(k.clone()),
                spans: vec![*sp],
            });
        }

        for sym in &self.expect_unused {
            match sym {
                Symbol::Rule(sym_name, sym_span) => {
                    if self.get_rule(sym_name).is_none() {
                        return Err(YaccGrammarError {
                            kind: YaccGrammarErrorKind::UnknownRuleRef(sym_name.clone()),
                            spans: vec![*sym_span],
                        });
                    }
                }
                Symbol::Token(sym_name, sym_span) => {
                    if !self.has_token(sym_name) {
                        return Err(YaccGrammarError {
                            kind: YaccGrammarErrorKind::UnknownToken(sym_name.clone()),
                            spans: vec![*sym_span],
                        });
                    }
                }
            }
        }
        Ok(())
    }

    pub fn warnings(&self) -> Vec<YaccGrammarWarning> {
        self.unused_symbols()
            .map(|symidx| {
                let (kind, span) = match symidx.symbol(self) {
                    Symbol::Rule(_, span) => (YaccGrammarWarningKind::UnusedRule, span),
                    Symbol::Token(_, span) => (YaccGrammarWarningKind::UnusedToken, span),
                };
                YaccGrammarWarning {
                    kind,
                    spans: vec![span],
                }
            })
            .collect()
    }

    /// Return the indices of unexpectedly unused rules (relative to ast.rules)
    /// and tokens (relative to ast.tokens) as `SymbolIdx`s.
    pub(crate) fn unused_symbols(&self) -> impl Iterator<Item = SymbolIdx> + '_ {
        let start_rule_name = self.start.as_ref().map(|(name, _)| name.clone());
        let start_rule = self
            .rules
            .iter()
            .find(|(rule_name, _)| start_rule_name.as_ref() == Some(rule_name));
        let mut seen_rules = HashSet::new();
        let mut seen_tokens = HashSet::new();
        let mut expected_unused_tokens = HashSet::new();
        let mut expected_unused_rules = HashSet::new();
        for sym in &self.expect_unused {
            match sym {
                Symbol::Rule(sym_name, _) => {
                    expected_unused_rules.insert(sym_name);
                }
                Symbol::Token(sym_name, _) => {
                    expected_unused_tokens.insert(sym_name);
                }
            }
        }
        if let Some(implicit_tokens) = self.implicit_tokens.as_ref() {
            expected_unused_tokens.extend(implicit_tokens.keys())
        }
        if let Some((start_name, start_rule)) = start_rule {
            let mut todo = Vec::new();
            todo.extend(start_rule.pidxs.iter().copied());
            seen_rules.insert(start_name);

            while let Some(pidx) = todo.pop() {
                let prod = &self.prods[pidx];
                for sym in &prod.symbols {
                    match sym {
                        Symbol::Rule(name, _) => {
                            if seen_rules.insert(name) {
                                if let Some(rule) = self.rules.get(name) {
                                    todo.extend(&rule.pidxs);
                                }
                            }
                        }
                        Symbol::Token(name, _) => {
                            seen_tokens.insert(name);
                        }
                    };
                }
            }
        }
        self.rules
            .iter()
            .enumerate()
            .filter_map(move |(rule_id, (rule_name, _))| {
                if expected_unused_rules.contains(rule_name) || seen_rules.contains(rule_name) {
                    None
                } else {
                    Some(SymbolIdx::Rule(rule_id))
                }
            })
            .chain(
                self.tokens
                    .iter()
                    .enumerate()
                    .filter_map(move |(tok_idx, tok)| {
                        if expected_unused_tokens.contains(tok) || seen_tokens.contains(tok) {
                            None
                        } else {
                            Some(SymbolIdx::Token(tok_idx))
                        }
                    }),
            )
    }
}

#[cfg(test)]
mod test {
    use super::{
        super::{AssocKind, Precedence},
        GrammarAST, Span, Symbol, YaccGrammarError, YaccGrammarErrorKind,
    };

    fn rule(n: &str) -> Symbol {
        Symbol::Rule(n.to_string(), Span::new(0, 0))
    }

    fn token(n: &str) -> Symbol {
        Symbol::Token(n.to_string(), Span::new(0, 0))
    }

    #[test]
    fn test_empty_grammar() {
        let mut grm = GrammarAST::new();
        match grm.complete_and_validate() {
            Err(YaccGrammarError {
                kind: YaccGrammarErrorKind::NoStartRule,
                ..
            }) => (),
            _ => panic!("Validation error"),
        }
    }

    #[test]
    fn test_invalid_start_rule() {
        let mut grm = GrammarAST::new();
        let empty_span = Span::new(0, 0);
        grm.start = Some(("A".to_string(), empty_span));
        grm.add_rule(("B".to_string(), empty_span), None);
        grm.add_prod("B".to_string(), vec![], None, None);
        match grm.complete_and_validate() {
            Err(YaccGrammarError {
                kind: YaccGrammarErrorKind::InvalidStartRule(_),
                ..
            }) => (),
            _ => panic!("Validation error"),
        }
    }

    #[test]
    fn test_valid_start_rule() {
        let mut grm = GrammarAST::new();
        let empty_span = Span::new(0, 0);
        grm.start = Some(("A".to_string(), empty_span));
        grm.add_rule(("A".to_string(), empty_span), None);
        grm.add_prod("A".to_string(), vec![], None, None);
        assert!(grm.complete_and_validate().is_ok());
    }

    #[test]
    fn test_valid_rule_ref() {
        let mut grm = GrammarAST::new();
        let empty_span = Span::new(0, 0);
        grm.start = Some(("A".to_string(), empty_span));
        grm.add_rule(("A".to_string(), empty_span), None);
        grm.add_rule(("B".to_string(), empty_span), None);
        grm.add_prod("A".to_string(), vec![rule("B")], None, None);
        grm.add_prod("B".to_string(), vec![], None, None);
        assert!(grm.complete_and_validate().is_ok());
    }

    #[test]
    fn test_invalid_rule_ref() {
        let mut grm = GrammarAST::new();
        let empty_span = Span::new(0, 0);
        grm.start = Some(("A".to_string(), empty_span));
        grm.add_rule(("A".to_string(), empty_span), None);
        grm.add_prod("A".to_string(), vec![rule("B")], None, None);
        match grm.complete_and_validate() {
            Err(YaccGrammarError {
                kind: YaccGrammarErrorKind::UnknownRuleRef(_),
                ..
            }) => (),
            _ => panic!("Validation error"),
        }
    }

    #[test]
    fn test_valid_token_ref() {
        let mut grm = GrammarAST::new();
        let empty_span = Span::new(0, 0);
        grm.tokens.insert("b".to_string());
        grm.start = Some(("A".to_string(), empty_span));
        grm.add_rule(("A".to_string(), empty_span), None);
        grm.add_prod("A".to_string(), vec![token("b")], None, None);
        assert!(grm.complete_and_validate().is_ok());
    }

    #[test]
    fn test_redefine_rules_as_tokens() {
        // for now we won't support the YACC feature that allows
        // to redefine rules as tokens by adding them to '%token'
        let mut grm = GrammarAST::new();
        let empty_span = Span::new(0, 0);
        grm.tokens.insert("b".to_string());
        grm.start = Some(("A".to_string(), empty_span));
        grm.add_rule(("A".to_string(), empty_span), None);
        grm.add_prod("A".to_string(), vec![rule("b")], None, None);
        assert!(grm.complete_and_validate().is_err());
    }

    #[test]
    fn test_invalid_token_ref() {
        let mut grm = GrammarAST::new();
        let empty_span = Span::new(0, 0);
        grm.start = Some(("A".to_string(), empty_span));
        grm.add_rule(("A".to_string(), empty_span), None);
        grm.add_prod("A".to_string(), vec![token("b")], None, None);
        match grm.complete_and_validate() {
            Err(YaccGrammarError {
                kind: YaccGrammarErrorKind::UnknownToken(_),
                ..
            }) => (),
            _ => panic!("Validation error"),
        }
    }

    #[test]
    fn test_invalid_rule_forgotten_token() {
        let mut grm = GrammarAST::new();
        let empty_span = Span::new(0, 0);
        grm.start = Some(("A".to_string(), empty_span));
        grm.add_rule(("A".to_string(), empty_span), None);
        grm.add_prod("A".to_string(), vec![rule("b"), token("b")], None, None);
        match grm.complete_and_validate() {
            Err(YaccGrammarError {
                kind: YaccGrammarErrorKind::UnknownRuleRef(_),
                ..
            }) => (),
            _ => panic!("Validation error"),
        }
    }

    #[test]
    fn test_invalid_epp() {
        let mut grm = GrammarAST::new();
        let empty_span = Span::new(2, 3);
        grm.start = Some(("A".to_string(), empty_span));
        grm.add_rule(("A".to_string(), empty_span), None);
        grm.add_prod("A".to_string(), vec![], None, None);
        grm.epp
            .insert("k".to_owned(), (empty_span, ("v".to_owned(), empty_span)));
        match grm.complete_and_validate() {
            Err(YaccGrammarError {
                kind: YaccGrammarErrorKind::UnknownEPP(_),
                spans,
            }) if spans.len() == 1 && spans[0] == Span::new(2, 3) => (),
            _ => panic!("Validation error"),
        }
    }

    #[test]
    fn test_precedence_override() {
        let mut grm = GrammarAST::new();
        let empty_span = Span::new(0, 0);
        grm.precs.insert(
            "b".to_string(),
            (
                Precedence {
                    level: 1,
                    kind: AssocKind::Left,
                },
                Span::new(0, 0),
            ),
        );
        grm.start = Some(("A".to_string(), empty_span));
        grm.tokens.insert("b".to_string());
        grm.add_rule(("A".to_string(), empty_span), None);
        grm.add_prod(
            "A".to_string(),
            vec![token("b")],
            Some("b".to_string()),
            None,
        );
        assert!(grm.complete_and_validate().is_ok());
    }

    #[test]
    fn test_invalid_precedence_override() {
        let mut grm = GrammarAST::new();
        let empty_span = Span::new(0, 0);
        grm.start = Some(("A".to_string(), empty_span));
        grm.add_rule(("A".to_string(), empty_span), None);
        grm.add_prod(
            "A".to_string(),
            vec![token("b")],
            Some("b".to_string()),
            None,
        );
        match grm.complete_and_validate() {
            Err(YaccGrammarError {
                kind: YaccGrammarErrorKind::UnknownToken(_),
                ..
            }) => (),
            _ => panic!("Validation error"),
        }
        grm.tokens.insert("b".to_string());
        match grm.complete_and_validate() {
            Err(YaccGrammarError {
                kind: YaccGrammarErrorKind::NoPrecForToken(_),
                ..
            }) => (),
            _ => panic!("Validation error"),
        }
    }

    #[test]
    fn test_ast_unused_symbols() {
        let mut grm = GrammarAST::new();
        let empty_span = Span::new(0, 0);
        grm.start = Some(("A".to_string(), empty_span));
        grm.add_rule(("A".to_string(), empty_span), None);
        grm.add_prod("A".to_string(), vec![], None, None);
        grm.tokens.insert("b".to_string());
        grm.spans.push(Span::new(4, 5));
        grm.add_rule(("B".to_string(), Span::new(1, 2)), None);
        grm.add_prod("B".to_string(), vec![token("b")], None, None);

        assert_eq!(
            grm.unused_symbols()
                .map(|sym_idx| sym_idx.symbol(&grm))
                .collect::<Vec<Symbol>>()
                .as_slice(),
            &[
                Symbol::Rule("B".to_string(), Span::new(1, 2)),
                Symbol::Token("b".to_string(), Span::new(4, 5))
            ]
        )
    }
}