-
Notifications
You must be signed in to change notification settings - Fork 1.3k
/
render.rs
1731 lines (1585 loc) · 63.7 KB
/
render.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
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
use std::{
cmp,
collections::{HashMap, HashSet},
fmt::Write,
mem::swap,
};
use super::{
build_tables::Tables,
grammars::{ExternalToken, LexicalGrammar, SyntaxGrammar, VariableType},
nfa::CharacterSet,
rules::{Alias, AliasMap, Symbol, SymbolType},
tables::{
AdvanceAction, FieldLocation, GotoAction, LexState, LexTable, ParseAction, ParseTable,
ParseTableEntry,
},
};
const SMALL_STATE_THRESHOLD: usize = 64;
const ABI_VERSION_MIN: usize = 13;
const ABI_VERSION_MAX: usize = tree_sitter::LANGUAGE_VERSION;
const ABI_VERSION_WITH_PRIMARY_STATES: usize = 14;
macro_rules! add {
($this: tt, $($arg: tt)*) => {{
$this.buffer.write_fmt(format_args!($($arg)*)).unwrap();
}}
}
macro_rules! add_whitespace {
($this:tt) => {{
for _ in 0..$this.indent_level {
write!(&mut $this.buffer, " ").unwrap();
}
}};
}
macro_rules! add_line {
($this: tt, $($arg: tt)*) => {
add_whitespace!($this);
$this.buffer.write_fmt(format_args!($($arg)*)).unwrap();
$this.buffer += "\n";
}
}
macro_rules! indent {
($this:tt) => {
$this.indent_level += 1;
};
}
macro_rules! dedent {
($this:tt) => {
assert_ne!($this.indent_level, 0);
$this.indent_level -= 1;
};
}
struct Generator {
buffer: String,
indent_level: usize,
language_name: String,
parse_table: ParseTable,
main_lex_table: LexTable,
keyword_lex_table: LexTable,
large_character_sets: Vec<(Option<Symbol>, CharacterSet)>,
large_character_set_info: Vec<LargeCharacterSetInfo>,
large_state_count: usize,
keyword_capture_token: Option<Symbol>,
syntax_grammar: SyntaxGrammar,
lexical_grammar: LexicalGrammar,
default_aliases: AliasMap,
symbol_order: HashMap<Symbol, usize>,
symbol_ids: HashMap<Symbol, String>,
alias_ids: HashMap<Alias, String>,
unique_aliases: Vec<Alias>,
symbol_map: HashMap<Symbol, Symbol>,
field_names: Vec<String>,
#[allow(unused)]
abi_version: usize,
}
struct LargeCharacterSetInfo {
constant_name: String,
is_used: bool,
}
impl Generator {
fn generate(mut self) -> String {
self.init();
self.add_includes();
self.add_pragmas();
self.add_stats();
self.add_symbol_enum();
self.add_symbol_names_list();
self.add_unique_symbol_map();
self.add_symbol_metadata_list();
if !self.field_names.is_empty() {
self.add_field_name_enum();
self.add_field_name_names_list();
self.add_field_sequences();
}
if !self.parse_table.production_infos.is_empty() {
self.add_alias_sequences();
}
self.add_non_terminal_alias_map();
if self.abi_version >= ABI_VERSION_WITH_PRIMARY_STATES {
self.add_primary_state_id_list();
}
let buffer_offset_before_lex_functions = self.buffer.len();
let mut main_lex_table = LexTable::default();
swap(&mut main_lex_table, &mut self.main_lex_table);
self.add_lex_function("ts_lex", main_lex_table);
if self.keyword_capture_token.is_some() {
let mut keyword_lex_table = LexTable::default();
swap(&mut keyword_lex_table, &mut self.keyword_lex_table);
self.add_lex_function("ts_lex_keywords", keyword_lex_table);
}
// Once the lex functions are generated, and we've determined which large
// character sets are actually used, we can generate the large character set
// constants. Insert them into the output buffer before the lex functions.
let lex_functions = self.buffer[buffer_offset_before_lex_functions..].to_string();
self.buffer.truncate(buffer_offset_before_lex_functions);
for ix in 0..self.large_character_sets.len() {
self.add_character_set(ix);
}
self.buffer.push_str(&lex_functions);
self.add_lex_modes_list();
self.add_parse_table();
if !self.syntax_grammar.external_tokens.is_empty() {
self.add_external_token_enum();
self.add_external_scanner_symbol_map();
self.add_external_scanner_states_list();
}
self.add_parser_export();
self.buffer
}
fn init(&mut self) {
let mut symbol_identifiers = HashSet::new();
for i in 0..self.parse_table.symbols.len() {
self.assign_symbol_id(self.parse_table.symbols[i], &mut symbol_identifiers);
}
self.symbol_ids.insert(
Symbol::end_of_nonterminal_extra(),
self.symbol_ids[&Symbol::end()].clone(),
);
self.symbol_map = HashMap::new();
for symbol in &self.parse_table.symbols {
let mut mapping = symbol;
// There can be multiple symbols in the grammar that have the same name and kind,
// due to simple aliases. When that happens, ensure that they map to the same
// public-facing symbol. If one of the symbols is not aliased, choose that one
// to be the public-facing symbol. Otherwise, pick the symbol with the lowest
// numeric value.
if let Some(alias) = self.default_aliases.get(symbol) {
let kind = alias.kind();
for other_symbol in &self.parse_table.symbols {
if let Some(other_alias) = self.default_aliases.get(other_symbol) {
if other_symbol < mapping && other_alias == alias {
mapping = other_symbol;
}
} else if self.metadata_for_symbol(*other_symbol) == (&alias.value, kind) {
mapping = other_symbol;
break;
}
}
}
// Two anonymous tokens with different flags but the same string value
// should be represented with the same symbol in the public API. Examples:
// * "<" and token(prec(1, "<"))
// * "(" and token.immediate("(")
else if symbol.is_terminal() {
let metadata = self.metadata_for_symbol(*symbol);
for other_symbol in &self.parse_table.symbols {
let other_metadata = self.metadata_for_symbol(*other_symbol);
if other_metadata == metadata {
if let Some(mapped) = self.symbol_map.get(other_symbol) {
if mapped == symbol {
break;
}
}
mapping = other_symbol;
break;
}
}
}
self.symbol_map.insert(*symbol, *mapping);
}
for production_info in &self.parse_table.production_infos {
// Build a list of all field names
for field_name in production_info.field_map.keys() {
if let Err(i) = self.field_names.binary_search(field_name) {
self.field_names.insert(i, field_name.clone());
}
}
for alias in &production_info.alias_sequence {
// Generate a mapping from aliases to C identifiers.
if let Some(alias) = &alias {
let existing_symbol = self.parse_table.symbols.iter().copied().find(|symbol| {
self.default_aliases.get(symbol).map_or_else(
|| {
let (name, kind) = self.metadata_for_symbol(*symbol);
name == alias.value && kind == alias.kind()
},
|default_alias| default_alias == alias,
)
});
// Some aliases match an existing symbol in the grammar.
let alias_id = if let Some(existing_symbol) = existing_symbol {
self.symbol_ids[&self.symbol_map[&existing_symbol]].clone()
}
// Other aliases don't match any existing symbol, and need their own
// identifiers.
else {
if let Err(i) = self.unique_aliases.binary_search(alias) {
self.unique_aliases.insert(i, alias.clone());
}
if alias.is_named {
format!("alias_sym_{}", self.sanitize_identifier(&alias.value))
} else {
format!("anon_alias_sym_{}", self.sanitize_identifier(&alias.value))
}
};
self.alias_ids.entry(alias.clone()).or_insert(alias_id);
}
}
}
for (ix, (symbol, _)) in self.large_character_sets.iter().enumerate() {
let count = self.large_character_sets[0..ix]
.iter()
.filter(|(sym, _)| sym == symbol)
.count()
+ 1;
let constant_name = if let Some(symbol) = symbol {
format!("{}_character_set_{}", self.symbol_ids[symbol], count)
} else {
format!("extras_character_set_{count}")
};
self.large_character_set_info.push(LargeCharacterSetInfo {
constant_name,
is_used: false,
});
}
// Determine which states should use the "small state" representation, and which should
// use the normal array representation.
let threshold = cmp::min(SMALL_STATE_THRESHOLD, self.parse_table.symbols.len() / 2);
self.large_state_count = self
.parse_table
.states
.iter()
.enumerate()
.take_while(|(i, s)| {
*i <= 1 || s.terminal_entries.len() + s.nonterminal_entries.len() > threshold
})
.count();
}
fn add_includes(&mut self) {
add_line!(self, "#include \"tree_sitter/parser.h\"");
add_line!(self, "");
}
fn add_pragmas(&mut self) {
add_line!(self, "#if defined(__GNUC__) || defined(__clang__)");
add_line!(
self,
"#pragma GCC diagnostic ignored \"-Wmissing-field-initializers\""
);
add_line!(self, "#endif");
add_line!(self, "");
// Compiling large lexer functions can be very slow. Disabling optimizations
// is not ideal, but only a very small fraction of overall parse time is
// spent lexing, so the performance impact of this is negligible.
if self.main_lex_table.states.len() > 300 {
add_line!(self, "#ifdef _MSC_VER");
add_line!(self, "#pragma optimize(\"\", off)");
add_line!(self, "#elif defined(__clang__)");
add_line!(self, "#pragma clang optimize off");
add_line!(self, "#elif defined(__GNUC__)");
add_line!(self, "#pragma GCC optimize (\"O0\")");
add_line!(self, "#endif");
add_line!(self, "");
}
}
fn add_stats(&mut self) {
let token_count = self
.parse_table
.symbols
.iter()
.filter(|symbol| {
if symbol.is_terminal() || symbol.is_eof() {
true
} else if symbol.is_external() {
self.syntax_grammar.external_tokens[symbol.index]
.corresponding_internal_token
.is_none()
} else {
false
}
})
.count();
add_line!(self, "#define LANGUAGE_VERSION {}", self.abi_version);
add_line!(
self,
"#define STATE_COUNT {}",
self.parse_table.states.len()
);
add_line!(self, "#define LARGE_STATE_COUNT {}", self.large_state_count);
add_line!(
self,
"#define SYMBOL_COUNT {}",
self.parse_table.symbols.len()
);
add_line!(self, "#define ALIAS_COUNT {}", self.unique_aliases.len());
add_line!(self, "#define TOKEN_COUNT {}", token_count);
add_line!(
self,
"#define EXTERNAL_TOKEN_COUNT {}",
self.syntax_grammar.external_tokens.len()
);
add_line!(self, "#define FIELD_COUNT {}", self.field_names.len());
add_line!(
self,
"#define MAX_ALIAS_SEQUENCE_LENGTH {}",
self.parse_table.max_aliased_production_length
);
add_line!(
self,
"#define PRODUCTION_ID_COUNT {}",
self.parse_table.production_infos.len()
);
add_line!(self, "");
}
fn add_symbol_enum(&mut self) {
add_line!(self, "enum ts_symbol_identifiers {{");
indent!(self);
self.symbol_order.insert(Symbol::end(), 0);
let mut i = 1;
for symbol in &self.parse_table.symbols {
if *symbol != Symbol::end() {
self.symbol_order.insert(*symbol, i);
add_line!(self, "{} = {i},", self.symbol_ids[symbol]);
i += 1;
}
}
for alias in &self.unique_aliases {
add_line!(self, "{} = {i},", self.alias_ids[alias]);
i += 1;
}
dedent!(self);
add_line!(self, "}};");
add_line!(self, "");
}
fn add_symbol_names_list(&mut self) {
add_line!(self, "static const char * const ts_symbol_names[] = {{");
indent!(self);
for symbol in &self.parse_table.symbols {
let name = self.sanitize_string(
self.default_aliases
.get(symbol)
.map_or(self.metadata_for_symbol(*symbol).0, |alias| {
alias.value.as_str()
}),
);
add_line!(self, "[{}] = \"{name}\",", self.symbol_ids[symbol]);
}
for alias in &self.unique_aliases {
add_line!(
self,
"[{}] = \"{}\",",
self.alias_ids[alias],
self.sanitize_string(&alias.value)
);
}
dedent!(self);
add_line!(self, "}};");
add_line!(self, "");
}
fn add_unique_symbol_map(&mut self) {
add_line!(self, "static const TSSymbol ts_symbol_map[] = {{");
indent!(self);
for symbol in &self.parse_table.symbols {
add_line!(
self,
"[{}] = {},",
self.symbol_ids[symbol],
self.symbol_ids[&self.symbol_map[symbol]],
);
}
for alias in &self.unique_aliases {
add_line!(
self,
"[{}] = {},",
self.alias_ids[alias],
self.alias_ids[alias],
);
}
dedent!(self);
add_line!(self, "}};");
add_line!(self, "");
}
fn add_field_name_enum(&mut self) {
add_line!(self, "enum ts_field_identifiers {{");
indent!(self);
for (i, field_name) in self.field_names.iter().enumerate() {
add_line!(self, "{} = {},", self.field_id(field_name), i + 1);
}
dedent!(self);
add_line!(self, "}};");
add_line!(self, "");
}
fn add_field_name_names_list(&mut self) {
add_line!(self, "static const char * const ts_field_names[] = {{");
indent!(self);
add_line!(self, "[0] = NULL,");
for field_name in &self.field_names {
add_line!(self, "[{}] = \"{field_name}\",", self.field_id(field_name));
}
dedent!(self);
add_line!(self, "}};");
add_line!(self, "");
}
fn add_symbol_metadata_list(&mut self) {
add_line!(
self,
"static const TSSymbolMetadata ts_symbol_metadata[] = {{"
);
indent!(self);
for symbol in &self.parse_table.symbols {
add_line!(self, "[{}] = {{", self.symbol_ids[symbol]);
indent!(self);
if let Some(Alias { is_named, .. }) = self.default_aliases.get(symbol) {
add_line!(self, ".visible = true,");
add_line!(self, ".named = {is_named},");
} else {
match self.metadata_for_symbol(*symbol).1 {
VariableType::Named => {
add_line!(self, ".visible = true,");
add_line!(self, ".named = true,");
}
VariableType::Anonymous => {
add_line!(self, ".visible = true,");
add_line!(self, ".named = false,");
}
VariableType::Hidden => {
add_line!(self, ".visible = false,");
add_line!(self, ".named = true,");
if self.syntax_grammar.supertype_symbols.contains(symbol) {
add_line!(self, ".supertype = true,");
}
}
VariableType::Auxiliary => {
add_line!(self, ".visible = false,");
add_line!(self, ".named = false,");
}
}
}
dedent!(self);
add_line!(self, "}},");
}
for alias in &self.unique_aliases {
add_line!(self, "[{}] = {{", self.alias_ids[alias]);
indent!(self);
add_line!(self, ".visible = true,");
add_line!(self, ".named = {},", alias.is_named);
dedent!(self);
add_line!(self, "}},");
}
dedent!(self);
add_line!(self, "}};");
add_line!(self, "");
}
fn add_alias_sequences(&mut self) {
add_line!(
self,
"static const TSSymbol ts_alias_sequences[PRODUCTION_ID_COUNT][MAX_ALIAS_SEQUENCE_LENGTH] = {{",
);
indent!(self);
for (i, production_info) in self.parse_table.production_infos.iter().enumerate() {
if production_info.alias_sequence.is_empty() {
// Work around MSVC's intolerance of empty array initializers by
// explicitly zero-initializing the first element.
if i == 0 {
add_line!(self, "[0] = {{0}},");
}
continue;
}
add_line!(self, "[{i}] = {{");
indent!(self);
for (j, alias) in production_info.alias_sequence.iter().enumerate() {
if let Some(alias) = alias {
add_line!(self, "[{j}] = {},", self.alias_ids[alias]);
}
}
dedent!(self);
add_line!(self, "}},");
}
dedent!(self);
add_line!(self, "}};");
add_line!(self, "");
}
fn add_non_terminal_alias_map(&mut self) {
let mut alias_ids_by_symbol = HashMap::new();
for variable in &self.syntax_grammar.variables {
for production in &variable.productions {
for step in &production.steps {
if let Some(alias) = &step.alias {
if step.symbol.is_non_terminal()
&& Some(alias) != self.default_aliases.get(&step.symbol)
&& self.symbol_ids.contains_key(&step.symbol)
{
if let Some(alias_id) = self.alias_ids.get(alias) {
let alias_ids =
alias_ids_by_symbol.entry(step.symbol).or_insert(Vec::new());
if let Err(i) = alias_ids.binary_search(&alias_id) {
alias_ids.insert(i, alias_id);
}
}
}
}
}
}
}
let mut alias_ids_by_symbol = alias_ids_by_symbol.iter().collect::<Vec<_>>();
alias_ids_by_symbol.sort_unstable_by_key(|e| e.0);
add_line!(
self,
"static const uint16_t ts_non_terminal_alias_map[] = {{"
);
indent!(self);
for (symbol, alias_ids) in alias_ids_by_symbol {
let symbol_id = &self.symbol_ids[symbol];
let public_symbol_id = &self.symbol_ids[&self.symbol_map[symbol]];
add_line!(self, "{symbol_id}, {},", 1 + alias_ids.len());
indent!(self);
add_line!(self, "{public_symbol_id},");
for alias_id in alias_ids {
add_line!(self, "{alias_id},");
}
dedent!(self);
}
add_line!(self, "0,");
dedent!(self);
add_line!(self, "}};");
add_line!(self, "");
}
/// Produces a list of the "primary state" for every state in the grammar.
///
/// The "primary state" for a given state is the first encountered state that behaves
/// identically with respect to query analysis. We derive this by keeping track of the `core_id`
/// for each state and treating the first state with a given `core_id` as primary.
fn add_primary_state_id_list(&mut self) {
add_line!(
self,
"static const TSStateId ts_primary_state_ids[STATE_COUNT] = {{"
);
indent!(self);
let mut first_state_for_each_core_id = HashMap::new();
for (idx, state) in self.parse_table.states.iter().enumerate() {
let primary_state = first_state_for_each_core_id
.entry(state.core_id)
.or_insert(idx);
add_line!(self, "[{idx}] = {primary_state},");
}
dedent!(self);
add_line!(self, "}};");
add_line!(self, "");
}
fn add_field_sequences(&mut self) {
let mut flat_field_maps = vec![];
let mut next_flat_field_map_index = 0;
self.get_field_map_id(
Vec::new(),
&mut flat_field_maps,
&mut next_flat_field_map_index,
);
let mut field_map_ids = Vec::new();
for production_info in &self.parse_table.production_infos {
if production_info.field_map.is_empty() {
field_map_ids.push((0, 0));
} else {
let mut flat_field_map = Vec::new();
for (field_name, locations) in &production_info.field_map {
for location in locations {
flat_field_map.push((field_name.clone(), *location));
}
}
field_map_ids.push((
self.get_field_map_id(
flat_field_map.clone(),
&mut flat_field_maps,
&mut next_flat_field_map_index,
),
flat_field_map.len(),
));
}
}
add_line!(
self,
"static const TSFieldMapSlice ts_field_map_slices[PRODUCTION_ID_COUNT] = {{",
);
indent!(self);
for (production_id, (row_id, length)) in field_map_ids.into_iter().enumerate() {
if length > 0 {
add_line!(
self,
"[{production_id}] = {{.index = {row_id}, .length = {length}}},",
);
}
}
dedent!(self);
add_line!(self, "}};");
add_line!(self, "");
add_line!(
self,
"static const TSFieldMapEntry ts_field_map_entries[] = {{",
);
indent!(self);
for (row_index, field_pairs) in flat_field_maps.into_iter().skip(1) {
add_line!(self, "[{row_index}] =");
indent!(self);
for (field_name, location) in field_pairs {
add_whitespace!(self);
add!(self, "{{{}, {}", self.field_id(&field_name), location.index);
if location.inherited {
add!(self, ", .inherited = true");
}
add!(self, "}},\n");
}
dedent!(self);
}
dedent!(self);
add_line!(self, "}};");
add_line!(self, "");
}
fn add_lex_function(&mut self, name: &str, lex_table: LexTable) {
add_line!(
self,
"static bool {name}(TSLexer *lexer, TSStateId state) {{",
);
indent!(self);
add_line!(self, "START_LEXER();");
add_line!(self, "eof = lexer->eof(lexer);");
add_line!(self, "switch (state) {{");
indent!(self);
for (i, state) in lex_table.states.into_iter().enumerate() {
add_line!(self, "case {i}:");
indent!(self);
self.add_lex_state(i, state);
dedent!(self);
}
add_line!(self, "default:");
indent!(self);
add_line!(self, "return false;");
dedent!(self);
dedent!(self);
add_line!(self, "}}");
dedent!(self);
add_line!(self, "}}");
add_line!(self, "");
}
fn add_lex_state(&mut self, _state_ix: usize, state: LexState) {
if let Some(accept_action) = state.accept_action {
add_line!(self, "ACCEPT_TOKEN({});", self.symbol_ids[&accept_action]);
}
if let Some(eof_action) = state.eof_action {
add_line!(self, "if (eof) ADVANCE({});", eof_action.state);
}
let mut chars_copy = CharacterSet::empty();
let mut large_set = CharacterSet::empty();
let mut ruled_out_chars = CharacterSet::empty();
// The transitions in a lex state are sorted with the single-character
// transitions first. If there are many single-character transitions,
// then implement them using an array of (lookahead character, state)
// pairs, instead of individual if statements, in order to reduce compile
// time.
let mut leading_simple_transition_count = 0;
let mut leading_simple_transition_range_count = 0;
for (chars, action) in &state.advance_actions {
if action.in_main_token
&& chars.ranges().all(|r| {
let start = *r.start() as u32;
let end = *r.end() as u32;
end <= start + 1 && end <= u16::MAX as u32
})
{
leading_simple_transition_count += 1;
leading_simple_transition_range_count += chars.range_count();
} else {
break;
}
}
if leading_simple_transition_range_count >= 8 {
add_line!(self, "ADVANCE_MAP(");
indent!(self);
for (chars, action) in &state.advance_actions[0..leading_simple_transition_count] {
for range in chars.ranges() {
add_whitespace!(self);
self.add_character(*range.start());
add!(self, ", {},\n", action.state);
if range.end() > range.start() {
add_whitespace!(self);
self.add_character(*range.end());
add!(self, ", {},\n", action.state);
}
}
ruled_out_chars = ruled_out_chars.add(chars);
}
dedent!(self);
add_line!(self, ");");
} else {
leading_simple_transition_count = 0;
}
for (chars, action) in &state.advance_actions[leading_simple_transition_count..] {
add_whitespace!(self);
// The lex state's advance actions are represented with disjoint
// sets of characters. When translating these disjoint sets into a
// sequence of checks, we don't need to re-check conditions that
// have already been checked due to previous transitions.
//
// Note that this simplification may result in an empty character set.
// That means that the transition is guaranteed (nothing further needs to
// be checked), not that this transition is impossible.
let simplified_chars = chars.simplify_ignoring(&ruled_out_chars);
// For large character sets, find the best matching character set from
// a pre-selected list of large character sets, which are based on the
// state transitions for invidual tokens. This transition may not exactly
// match one of the pre-selected character sets. In that case, determine
// the additional checks that need to be performed to match this transition.
let mut best_large_char_set: Option<(usize, CharacterSet, CharacterSet)> = None;
if simplified_chars.range_count() >= super::build_tables::LARGE_CHARACTER_RANGE_COUNT {
for (ix, (_, set)) in self.large_character_sets.iter().enumerate() {
chars_copy.assign(&simplified_chars);
large_set.assign(set);
let intersection = chars_copy.remove_intersection(&mut large_set);
if !intersection.is_empty() {
let additions = chars_copy.simplify_ignoring(&ruled_out_chars);
let removals = large_set.simplify_ignoring(&ruled_out_chars);
let total_range_count = additions.range_count() + removals.range_count();
if total_range_count >= simplified_chars.range_count() {
continue;
}
if let Some((_, best_additions, best_removals)) = &best_large_char_set {
let best_range_count =
best_additions.range_count() + best_removals.range_count();
if best_range_count < total_range_count {
continue;
}
}
best_large_char_set = Some((ix, additions, removals));
}
}
}
// Add this transition's character set to the set of ruled out characters,
// which don't need to be checked for subsequent transitions in this state.
ruled_out_chars = ruled_out_chars.add(chars);
let mut large_char_set_ix = None;
let mut asserted_chars = simplified_chars;
let mut negated_chars = CharacterSet::empty();
if let Some((char_set_ix, additions, removals)) = best_large_char_set {
asserted_chars = additions;
negated_chars = removals;
large_char_set_ix = Some(char_set_ix);
}
let mut line_break = "\n".to_string();
for _ in 0..self.indent_level + 2 {
line_break.push_str(" ");
}
let has_positive_condition = large_char_set_ix.is_some() || !asserted_chars.is_empty();
let has_negative_condition = !negated_chars.is_empty();
let has_condition = has_positive_condition || has_negative_condition;
if has_condition {
add!(self, "if (");
if has_positive_condition && has_negative_condition {
add!(self, "(");
}
}
if let Some(large_char_set_ix) = large_char_set_ix {
let large_set = &self.large_character_sets[large_char_set_ix].1;
// If the character set contains the null character, check that we
// are not at the end of the file.
let check_eof = large_set.contains('\0');
if check_eof {
add!(self, "(!eof && ")
}
let char_set_info = &mut self.large_character_set_info[large_char_set_ix];
char_set_info.is_used = true;
add!(
self,
"set_contains({}, {}, lookahead)",
&char_set_info.constant_name,
large_set.range_count(),
);
if check_eof {
add!(self, ")");
}
}
if !asserted_chars.is_empty() {
if large_char_set_ix.is_some() {
add!(self, " ||{line_break}");
}
// If the character set contains the max character, than it probably
// corresponds to a negated character class in a regex, so it will be more
// concise and readable to express it in terms of negated ranges.
let is_included = !asserted_chars.contains(char::MAX);
if !is_included {
asserted_chars = asserted_chars.negate().add_char('\0');
}
self.add_character_range_conditions(&asserted_chars, is_included, &line_break);
}
if has_negative_condition {
if has_positive_condition {
add!(self, ") &&{line_break}");
}
self.add_character_range_conditions(&negated_chars, false, &line_break);
}
if has_condition {
add!(self, ") ");
}
self.add_advance_action(action);
add!(self, "\n");
}
add_line!(self, "END_STATE();");
}
fn add_character_range_conditions(
&mut self,
characters: &CharacterSet,
is_included: bool,
line_break: &str,
) {
for (i, range) in characters.ranges().enumerate() {
let start = *range.start();
let end = *range.end();
if is_included {
if i > 0 {
add!(self, " ||{line_break}");
}
if start == '\0' {
add!(self, "(!eof && ");
if end == '\0' {
add!(self, "lookahead == 0");
} else {
add!(self, "lookahead <= ");
}
self.add_character(end);
add!(self, ")");
continue;
} else if end == start {
add!(self, "lookahead == ");
self.add_character(start);
} else if end as u32 == start as u32 + 1 {
add!(self, "lookahead == ");
self.add_character(start);
add!(self, " ||{line_break}lookahead == ");
self.add_character(end);
} else {
add!(self, "(");
self.add_character(start);
add!(self, " <= lookahead && lookahead <= ");
self.add_character(end);
add!(self, ")");
}
} else {
if i > 0 {
add!(self, " &&{line_break}");
}
if end == start {
add!(self, "lookahead != ");
self.add_character(start);
} else if end as u32 == start as u32 + 1 {
add!(self, "lookahead != ");
self.add_character(start);
add!(self, " &&{line_break}lookahead != ");
self.add_character(end);
} else if start != '\0' {
add!(self, "(lookahead < ");
self.add_character(start);
add!(self, " || ");
self.add_character(end);
add!(self, " < lookahead)");
} else {
add!(self, "lookahead > ");
self.add_character(end);
}
}
}
}
fn add_character_set(&mut self, ix: usize) {
let characters = self.large_character_sets[ix].1.clone();
let info = &self.large_character_set_info[ix];
if !info.is_used {
return;
}
add_line!(
self,
"static TSCharacterRange {}[] = {{",
info.constant_name
);
indent!(self);
for (ix, range) in characters.ranges().enumerate() {
let column = ix % 8;
if column == 0 {
if ix > 0 {
add!(self, "\n");
}
add_whitespace!(self);
} else {
add!(self, " ");
}
add!(self, "{{");
self.add_character(*range.start());
add!(self, ", ");
self.add_character(*range.end());
add!(self, "}},");
}
add!(self, "\n");
dedent!(self);
add_line!(self, "}};");
add_line!(self, "");