Source file lrgrep_top.ml
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
open Utils
open Misc
open Fix.Indexing
open Kernel
open Info
let (!!) = Lazy.force
module Interpreter = Interpreter
module Message_file = Message_file
type 'a with_position = 'a * Lexing.position * Lexing.position
module type CUSTOM_GRAMMAR = sig
type g
val grammar : g grammar
val string_of_terminal : (g terminal index -> string) option
val string_of_sentence : (g terminal index list -> string) option
val sentence_lexer : (in_channel -> string with_position option *
g terminal index with_position list) option
end
module Lazy_grammar (Load : functor() -> CUSTOM_GRAMMAR) : sig
type g
module Force() : CUSTOM_GRAMMAR with type g = g
end = struct
type g
let loaded = lazy (
let m : (module CUSTOM_GRAMMAR) = (module Load()) in
let m' : (module CUSTOM_GRAMMAR with type g = g) = Obj.magic m in
m'
)
module Force() = (val !!loaded)
end
module Make(Language : sig
val name : string
val grammar : (module CUSTOM_GRAMMAR) option
val parser_name : string
end)() =
struct
let command_name =
match Language.name with
| "" -> "lrgrep"
| name -> "lrgrep for " ^ name
let binary_name = Sys.argv.(0)
let usage_prompt =
"Usage: " ^ binary_name ^ " <global-options> [ command <command-options> ]..."
let opt_terminal_printer = ref `Builtin
let opt_report_format = ref `Text
let badf fmt =
Printf.ksprintf (fun msg -> raise (Arg.Bad msg)) fmt
let usage_error_fun : (string -> unit) ref = ref failwith
let usage_error fmt =
Printf.ksprintf !usage_error_fun fmt
let fatal_error ?exn fmt =
Printf.ksprintf (fun msg ->
prerr_string "Fatal error: ";
prerr_string msg;
begin match exn with
| None -> ()
| Some exn ->
prerr_string " (";
prerr_string (Printexc.to_string exn);
prerr_string ")";
end;
prerr_newline ();
exit 1
) fmt
let with_output_file fmt =
Printf.ksprintf (fun str k ->
let oc = open_out_bin str in
k oc;
close_out oc;
) fmt
let opt_spec_file = ref None
let set_spec_file name =
match !opt_spec_file with
| Some name' ->
badf "unexpected argument %S: the specification is already set to %S"
name name'
| None ->
opt_spec_file := Some name
let get_spec_file () =
match !opt_spec_file with
| None ->
usage_error "no specification file has been set (pass -s <spec.lrgrep>)";
exit 1
| Some fname -> fname
let opt_grammar_file = ref None
let set_grammar_file name =
match !opt_grammar_file with
| Some name' ->
let message = Printf.sprintf
"unexpected argument %S: the grammar is already set to %S"
name name'
in
raise (Arg.Bad message)
| None ->
opt_grammar_file := Some name
let get_grammar_file () =
match !opt_grammar_file with
| None ->
usage_error "grammar file has not been set (pass -g <parser.cmly>)";
exit 1
| Some fname -> fname
let opt_dump_dot = ref false
let print_version_num () =
print_endline "0.2";
exit 0
let print_version_string () =
print_string "LRGrep toolkit for Menhir parsers";
if Language.name <> "" then (
print_string ", specialized for ";
print_string Language.name;
);
print_string ", version ";
print_version_num ()
let global_specs =
cons_if (Option.is_none Language.grammar)
("-g", Arg.String set_grammar_file,
" <file.cmly> Path to the compiled Menhir grammar to analyse")
@@ [
"-s", Arg.String set_spec_file,
" <file.lrgrep> Path to the error specification to process";
"-v", Arg.Unit (fun () -> incr Misc.verbosity_level),
" Increase output verbosity (for debugging/profiling)";
"--version", Arg.Unit print_version_string,
" Print version and exit";
"--vnum", Arg.Unit print_version_num,
" Print version number and exit";
"--dump-dot", Arg.Set opt_dump_dot,
" debug: Dump internal automata to Graphviz .dot files";
"--terminal-display", Arg.String (function
| "name" -> opt_terminal_printer := `Name
| "alias" -> opt_terminal_printer := `Alias
| "builtin" -> opt_terminal_printer := `Builtin
| other ->
usage_error
"--terminal-display: unexpected argument %s\n\
Accepted values are:\n\
- name: print terminal names\n\
- alias: prefer string aliases when available\n\
- builtin: use builtin printer if available for the language\n"
other
),
"<name|alias|builtin> When reporting a terminal, display it by its name or its alias";
"--report-format", Arg.String (function
| "text" -> opt_report_format := `Text
| "json" -> opt_report_format := `Json
| other ->
usage_error
"--report-format: unexpected argument %s\n\
Accepted values are:\n\
- text: human-readable report\n\
- json: json-object easy to post-process\n"
other
),
"<text|json> Report serialization format";
]
include Lazy_grammar(functor () ->
(val
match Language.grammar with
| None ->
let path = get_grammar_file () in
begin try
let module G = MenhirSdk.Cmly_read.Read(struct let filename = path end) in
stopwatch 1 "Loaded grammar from disk (%d terminals, %d non-terminals, %d lr0 states, %d lr1 states)"
G.Terminal.count G.Nonterminal.count
G.Lr0.count G.Lr1.count;
let module I = Load_grammar(G) in
stopwatch 1 "Pre-processed grammar definition";
(module struct
include I
let string_of_terminal = None
let string_of_sentence = None
let sentence_lexer = None
end : CUSTOM_GRAMMAR)
with exn ->
fatal_error ~exn "cannot load grammar file %S" path
end
| Some g -> g
)
)
let grammar =
lazy (let module G = Force() in G.grammar)
let builtin_print_terminal =
lazy (let module G = Force() in
match G.string_of_terminal with
| Some f -> f
| None -> Terminal.to_string G.grammar)
let builtin_print_sentence =
lazy (let module G = Force() in G.string_of_sentence)
let builtin_sentence_lexer =
lazy (let module G = Force() in G.sentence_lexer)
let parser_name =
match Language.grammar with
| None ->
lazy (
String.capitalize_ascii
(Filename.remove_extension
(Filename.basename (get_grammar_file ())))
)
| Some _ -> lazy Language.parser_name
let spec = lazy (
let print_parse_error_and_exit lexbuf exn =
let bt = Printexc.get_raw_backtrace () in
match exn with
| Front.Parser.Error ->
let pos = Lexing.lexeme_start_p lexbuf in
Syntax.error pos "syntax error."
| Front.Lexer.Error {msg; pos} ->
Syntax.error pos "%s." msg
| _ -> Printexc.raise_with_backtrace exn bt
in
let parse_spec path =
match open_in_bin path with
| exception exn ->
fatal_error ~exn "cannot load specification file %s" path
| ic ->
let state = Front.Lexer.fresh_state () in
let lexbuf = Lexing.from_channel ~with_positions:true ic in
let lexbuf = Front.Lexer.prepare_lexbuf state lexbuf in
Lexing.set_filename lexbuf path;
let result =
try Front.Parser.parse_lexer_definition (Front.Lexer.main state) lexbuf
with exn -> print_parse_error_and_exit lexbuf exn
in
result
in
let file = get_spec_file () in
let result = parse_spec file in
stopwatch 1 "Loaded specification %s" file;
result
)
let reachability = lazy (
let result = Reachability.make !!grammar in
stopwatch 1 "Solved reachability on the LR(1) automaton";
result
)
let lrc = lazy (
let result = Lrc.make_minimal !!grammar !!reachability in
stopwatch 1 "Computed minimal LRC on the LR(1) automaton";
result
)
let red_closure = lazy (Redgraph.close_lr1_reductions !!grammar)
let red_table = lazy (Redpos.make !!grammar)
let red_trie, red_targets =
let red_index = lazy (Redgraph.index_targets !!grammar !!red_closure) in
lazy (fst (Lazy.force red_index)),
lazy (snd (Lazy.force red_index))
let red_graph = lazy (
let result = Redgraph.make !!grammar !!red_closure !!red_targets in
stopwatch 1 "Computed viable reductions";
result
)
let indices = lazy (
let result = Transl.Indices.make !!grammar in
stopwatch 1 "Indexed items and symbols for translation";
result
)
let lrc_from_entrypoints =
let cache = Hashtbl.create 7 in
fun from_entrypoints ->
match Hashtbl.find_opt cache from_entrypoints with
| Some ep -> ep
| None ->
let lr1s =
if IndexSet.is_empty from_entrypoints
then Lr1.entrypoints !!grammar
else from_entrypoints
in
let lrcs = IndexSet.bind lr1s (Fix.Indexing.Vector.get !!lrc.lrcs_of) in
let ep = Lrc.from_entrypoints !!grammar !!lrc lrcs in
Hashtbl.add cache from_entrypoints ep;
stopwatch 2 "Computed LRC subset reachable from entrypoints";
ep
let translate_entrypoints symbols =
let unknown = ref [] in
let initial_states =
List.filter_map (fun (sym, pos) ->
let result = Hashtbl.find_opt (Lr1.entrypoint_table !!grammar) sym in
if Option.is_none result then
push unknown (sym, pos);
result
) symbols
|> IndexSet.of_list
in
begin match List.rev !unknown with
| [] -> ()
| (_, pos) :: _ as unknowns ->
let names = String.concat ", " (List.map fst unknowns) in
let candidates = Hashtbl.to_seq_keys (Lr1.entrypoint_table !!grammar) in
Syntax.error pos
"unknown start symbols %s.\n\
Start symbols of this grammar are:\n\
%s\n"
names (String.concat ", " (List.of_seq candidates))
end;
initial_states
let make_stacks (subset : _ Lrc.entrypoints) ~error_only =
let domain = Vector.length !!lrc.lr1_of in
let tops = if error_only then subset.wait else subset.reachable in
let prev = Vector.get subset.predecessors in
let label = Vector.get !!lrc.lr1_of in
{Automata. domain; tops; prev; label}
let opt_compile_output = ref None
let opt_compile_cover_report = ref ""
let opt_compile_cover_error = ref false
let compiler_cover_output = lazy (
if !opt_compile_cover_report = "" then
None
else
try Some (open_out_bin !opt_compile_cover_report)
with exn ->
Syntax.error Lexing.dummy_pos
"cannot open %S for reporting coverage: %S"
!opt_compile_cover_report
(Printexc.to_string exn)
)
let print_terminal g t =
match !opt_terminal_printer with
| `Name -> Terminal.to_string g t
| `Alias -> begin match Terminal.alias g t with
| None -> Terminal.to_string g t
| Some alias -> alias
end
| `Builtin -> !!builtin_print_terminal t
let print_sentence = lazy (
match !!builtin_print_sentence with
| Some f -> f
| None -> string_concat_map " " (print_terminal !!grammar)
)
type 'g report_sample = {
stack: 'g lr1 index list;
sentence: g terminal index list;
failing: 'g terminal indexset;
patterns: 'g lr0 indexset;
}
let prepare_sample grammar (stacks : _ Automata.stacks)
~some_prefix ~get_lrc ~reached ~all (tactic, failing)
=
let suffix = List.rev_map get_lrc tactic in
let _, prefix = some_prefix (List.hd suffix) in
let stack =
list_rev_mappend stacks.label prefix @@
List.map stacks.label suffix
in
let sentence = Sentence_generation.sentence_of_stack grammar !!reachability stack in
let patterns =
if all then
List.fold_left (fun acc n -> IndexSet.union (reached n) acc) IndexSet.empty tactic
else
IndexSet.empty
in
{stack; sentence; failing; patterns}
let report_text_sample oc grammar main i {stack; sentence; failing; patterns} =
Printf.fprintf oc "### Sample %d\n\n" i;
Printf.fprintf oc "Sentence:\n```\n%s\n```\n" (!!print_sentence sentence);
Printf.fprintf oc "Stack:\n```\n%s\n```\n"
(string_concat_map " " (Lr1.symbol_to_string grammar) stack);
Printf.fprintf oc "Rejected when looking ahead at any of the terminals in:\n\
```\n%s\n```\n"
(String.concat " " (List.rev_map (print_terminal grammar) (IndexSet.elements failing)));
let patterns = IndexSet.remove main patterns in
if IndexSet.is_not_empty patterns then begin
Printf.fprintf oc "Also covered by the following patterns:\n```\n";
IndexSet.iter begin fun lr0 ->
List.iter (Printf.fprintf oc "%s\n") (Coverage.print_pattern grammar lr0)
end patterns;
Printf.fprintf oc "```\n";
end;
Printf.fprintf oc "\n"
let report_text oc grammar cases =
begin match header with
| `Rule name -> Printf.fprintf oc "# Rule %s\n" name
| `Enum_maximal -> Printf.fprintf oc "# Maximal patterns\n\n"
| `Enum_other -> Printf.fprintf oc "# Other reduce-filter patterns\n\n";
end;
Seq.iteri begin fun i (lr0, samples) ->
Printf.fprintf oc "## Pattern %d\n\n```\n" i;
List.iter (Printf.fprintf oc "%s\n") (Coverage.print_pattern grammar lr0);
Printf.fprintf oc "```\n\n";
Seq.iteri (report_text_sample oc grammar lr0) samples
end cases
let escape_json_string oc string =
escape_json_string (output_substring oc) string
let print_json_string oc string =
output_char oc '"';
escape_json_string oc string;
output_char oc '"'
let print_comma_sep () =
let first = ref true in
fun oc ->
if !first then
first := false
else
output_char oc ','
let report_json_sample oc grammar acc i {stack; sentence; failing; patterns} =
if i > 0 then
output_char oc ',';
output_string oc "{\"sentence\":";
print_json_string oc (!!print_sentence sentence);
output_string oc ",\"stack\":\"";
List.iteri begin fun i sym ->
if i > 0 then output_char oc ' ';
escape_json_string oc (Lr1.symbol_to_string grammar sym);
end stack;
output_string oc "\",\"failing\":[";
let print_comma = print_comma_sep () in
IndexSet.rev_iter begin fun term ->
print_comma oc;
print_json_string oc (print_terminal grammar term)
end failing;
if IndexSet.is_not_empty patterns then begin
acc := IndexSet.union patterns !acc;
output_string oc "],\"patterns\":[";
let print_comma = print_comma_sep () in
IndexSet.iter begin fun lr0 ->
print_comma oc;
Printf.fprintf oc "\"%d\"" (Index.to_int lr0);
end patterns;
end;
output_string oc "]}"
let report_json_patterns oc grammar patterns =
output_string oc "{";
let print_comma = print_comma_sep () in
IndexSet.iter begin fun lr0 ->
print_comma oc;
Printf.fprintf oc "\"%d\":\"" (Index.to_int lr0);
List.iteri begin fun i text ->
if i > 0 then output_string oc "\\n";
escape_json_string oc text;
end (Coverage.print_pattern grammar lr0);
output_string oc "\""
end patterns;
output_string oc "}"
let report_json oc grammar cases acc =
output_string oc "{\"kind\":";
begin match header with
| `Rule name ->
output_string oc "\"coverage\",\"name\":";
print_json_string oc name;
| `Enum_maximal ->
output_string oc "\"enumeration\",\"name\":\"maximal\""
| `Enum_other ->
output_string oc "\"enumeration\",\"name\":\"other\""
end;
output_string oc ",\"cases\":[";
Seq.iteri begin fun i (lr0, samples) ->
if i > 0 then output_char oc ',';
acc := IndexSet.add lr0 !acc;
Printf.fprintf oc "{\"pattern\":\"%d\",\"sentences\":[" (Index.to_int lr0);
Seq.iteri (report_json_sample oc grammar acc) samples;
output_string oc "]}";
end cases;
output_string oc "]}"
let begin_report oc grammar =
match !opt_report_format with
| `Json ->
output_string oc "{\"reports\":[";
let print_comma = print_comma_sep () in
let patterns = ref IndexSet.empty in
((fun cases ->
print_comma oc;
report_json oc grammar header cases patterns),
(fun () ->
output_string oc "]";
output_string oc ",\"patterns\":";
report_json_patterns oc grammar !patterns;
output_string oc "}"))
| `Text -> (report_text oc grammar, ignore)
let do_compile spec (cp : Code_printer.t option) =
let grammar = !!grammar in
Codegen.output_header grammar spec cp;
let cover_report = match compiler_cover_output with
| lazy None -> None
| lazy (Some oc) -> Some (begin_report oc grammar)
in
List.iter begin fun (rule : Syntax.rule) ->
let entrypoints = lrc_from_entrypoints (translate_entrypoints rule.startsymbols) in
let stacks = make_stacks entrypoints ~error_only:(fst rule.error) in
let Spec.Rule (clauses, branches) =
Spec.import_rule grammar !!red_graph !!indices !!red_trie rule
in
let nfa = Automata.NFA.from_branches grammar !!red_graph branches in
Vector.iteri (fun br nfa ->
if !opt_dump_dot then
with_output_file "%s_%s_br_%d_line_%d.dot" !!parser_name rule.name
(br : _ index :> int) branches.expr.:(br).position.pos_lnum
(Automata.NFA.dump grammar nfa);
) nfa;
stopwatch 1 "constructed NFA";
let Automata.DFA.T dfa =
Automata.DFA.determinize grammar branches stacks nfa in
if !opt_dump_dot then
with_output_file "%s_%s_dfa.dot" !!parser_name rule.name
(Automata.DFA.dump grammar dfa !!red_graph);
begin
let states = Vector.length_as_int dfa.states in
let branches = ref 0 in
Vector.iter (fun (Automata.DFA.Packed st) ->
branches := !branches + Vector.length_as_int st.branches
) dfa.states;
stopwatch 1 "determinization (%d states, %d branches, average %.02f branch/state)"
states !branches (float !branches /. float states);
end;
let dataflow = Automata.Dataflow.make branches dfa in
stopwatch 1 "dataflow analysis";
if !opt_dump_dot then
with_output_file "%s_%s_dataflow.dot" !!parser_name rule.name
(Automata.Dataflow.dump grammar dfa dataflow);
let Automata.Machine.T machine =
Automata.Machine.minimize branches dfa dataflow in
stopwatch 1 "machine minimization";
if !opt_dump_dot then
with_output_file "%s_%s_machine.dot" !!parser_name rule.name
(Automata.Machine.dump grammar machine);
Codegen.output_rule grammar spec rule clauses branches machine cp;
stopwatch 1 "table & code generation";
if fst rule.error then (
let rcs = !!red_closure in
let rtable = Redpos.make grammar in
let open Coverage in
let Andor.Graph agr = Andor.make grammar rcs stacks rtable in
let Deter.Graph dgr = Deter.make grammar rcs stacks rtable agr in
let Cover.Graph cgr = Cover.coverage grammar branches machine stacks agr dgr in
if not (list_is_empty cgr.sinks) then (
begin match cover_report with
| None -> ()
| (Some (report, _)) ->
let get_lrc st =
match Sum.prj stacks.domain (cgr.position st) with
| L lrc -> lrc
| R enu -> Deter.get_lrc agr dgr.nodes.:(enu)
in
let reached st =
match Sum.prj stacks.domain (cgr.position st) with
| L _ -> IndexSet.empty
| R enu -> Deter.get_lr0 grammar rcs stacks rtable agr dgr enu
in
let iter_sinks f =
List.iter (fun sink -> f sink (fst (entrypoints.some_prefix (get_lrc sink))))
cgr.sinks
in
let maximals = Extract.compute_maximal_prefixes
~graph:cgr.enum
~iter_sinks
~reached
in
let locals, _globals =
Report.emit_all
~goals:(Lr0.cardinal grammar)
~graph:cgr.enum
~maximals
~globals:[||]
~reached
in
let prepare = prepare_sample grammar stacks
~some_prefix:entrypoints.some_prefix
~get_lrc ~reached ~all:false
in
report (`Rule rule.name)
(Seq.map (fun (lr0, sentences) -> (lr0, Seq.map prepare sentences)) locals);
end;
stopwatch 1 "coverage report";
let report =
if !opt_compile_cover_error
then Syntax.error
else Syntax.warn
in
report Lexing.dummy_pos
"rule %s has only partial coverage%s" rule.name
(if !opt_compile_cover_report <> "" then ""
else " (use --cover-report <file> to get more information)");
)
)
end spec.lexer_definition.rules;
begin match cover_report with
| None -> ()
| Some (_, end_report) -> end_report ()
end;
Codegen.output_trailer grammar spec cp
let compile_command () =
let output_file = match !opt_compile_output with
| Some o -> o
| None -> Filename.remove_extension (get_spec_file ()) ^ ".ml"
in
match open_out_bin output_file with
| exception exn ->
fatal_error ~exn "cannot open output file %S" output_file
| oc ->
let cp = Code_printer.create ~filename:output_file (output_string oc) in
do_compile {parser_name = !!parser_name; lexer_definition= !!spec} (Some cp);
close_out oc
let opt_enum_all = ref false
let opt_enum_entrypoints = ref []
let enumerate_command () =
let grammar = !!grammar in
let initial_states =
!opt_enum_entrypoints
|> List.rev_map (fun name -> name, Lexing.dummy_pos)
|> translate_entrypoints
in
let subset = lrc_from_entrypoints initial_states in
let rcs = !!red_closure in
let rtbl = !!red_table in
let stacks = make_stacks subset ~error_only:true in
let open Coverage in
let Andor.Graph agr = Andor.make grammar rcs stacks rtbl in
let Deter.Graph dgr = Deter.make grammar rcs stacks rtbl agr in
let graph, maximals = Enum.prepare grammar agr dgr subset.some_prefix in
let reached = Deter.get_lr0 grammar rcs stacks rtbl agr dgr in
let globals =
if !opt_enum_all then
Extract.compute_global_prefixes
~graph ~maximals ~reached
else [||]
in
let locals, globals =
Report.emit_all
~goals:(Lr0.cardinal grammar)
~graph ~maximals ~globals ~reached
in
let prepare_sample =
prepare_sample grammar stacks
~some_prefix:subset.some_prefix
~get_lrc:(fun node -> Deter.get_lrc agr dgr.Deter.nodes.:(node))
~reached:(Deter.get_lr0 grammar rcs stacks rtbl agr dgr)
~all:!opt_enum_all
in
let prepare_samples seq =
Seq.map (fun (lr0, sentences) -> (lr0, Seq.map prepare_sample sentences))
seq
in
let report, end_report = begin_report stdout grammar in
report `Enum_maximal (prepare_samples locals);
if !opt_enum_all then
report `Enum_other (prepare_samples globals);
end_report ()
let opt_import_file = ref ""
let opt_import_output = ref ""
let opt_import_shortest = ref false
let set_import_message_file path =
if !opt_import_file <> "" then
Printf.eprintf "unexpected argument %S: message file already set to %S\n"
path !opt_import_file;
opt_import_file := path
let import_command () =
if !opt_import_file = "" then (
prerr_endline "No .message file specified";
exit 1
);
if !opt_import_output = "" then (
prerr_endline "Use -o <file.lrgrep> to specify the output";
exit 1
);
let ic = open_in_bin !opt_import_file in
let rec lines () =
match input_line ic with
| exception End_of_file -> Seq.Nil
| line -> Seq.Cons (line, lines)
in
let parser = Message_file.parse_sentence !!grammar in
let blocks =
lines
|> Message_file.extract_pre_block
|> Message_file.extract_block
|> Seq.map (Message_file.map_block (fun block ->
let sentences = List.map (Message_file.map_line (fun sentence ->
parser (Interpreter.lift_sentence !!grammar sentence)
)) block.Message_file.sentences
in
{block with sentences}
))
|> List.of_seq |> List.to_seq
in
close_in ic;
let oc = open_out_bin !opt_import_output in
Seq.iter
(fun line -> output_string oc line; output_char oc '\n')
(Message_file.blocks_to_file
!!grammar ~shortest:!opt_import_shortest blocks);
close_out oc
let opt_interpret_patterns = ref true
let opt_interpret_items = ref false
let interpret_command () =
let grammar = !!grammar in
let parser = Interpreter.make_parser grammar in
let config =
Interpreter.config
~print_reduce_filter:!opt_interpret_patterns
~print_stack_items:!opt_interpret_items
()
in
let rec loop () =
match
match !!builtin_sentence_lexer with
| None ->
output_string stdout "$ ";
flush stdout;
Interpreter.lift_sentence grammar (input_line stdin)
| Some f ->
let entrypoint, symbols = f stdin in
let entrypoint = Option.map (Interpreter.lift_entrypoint grammar) entrypoint in
{entrypoint; symbols}
with
| exception Exit ->
flush_all ();
loop ()
| exception End_of_file -> ()
| sentence ->
let entrypoint =
match sentence.entrypoint with
| None -> Option.get (IndexSet.minimum (Lr1.entrypoints grammar))
| Some (ep, _, _) -> ep
in
let stack, _final_stack, remainder =
Interpreter.parse_sentence parser
(entrypoint, Lexing.dummy_pos, Lexing.dummy_pos)
(List.to_seq sentence.symbols)
in
let remainder = List.map (fun (x, _, _) -> x) (List.of_seq remainder) in
Interpreter.analyze_stack ~stack ~remainder
grammar !!red_closure config;
loop ()
in
loop ()
let commands =
Subarg.[
command "compile" "Translate a specification to an OCaml module" [
"-o", Arg.String (fun x -> opt_compile_output := Some x),
"<file.ml> Set output file name to <file> (defaults to <spec>.ml)";
"--cover-all", Arg.Set opt_compile_cover_error,
" Exit with a failure if coverage of errors is not exhaustive";
"--cover-report", Arg.Set_string opt_compile_cover_report,
"<report.md> Write a detailed report of uncovered cases (if any)";
] ~commit:compile_command;
command "interpret" "Parse a sentence and suggest patterns that can match it" [
"--no-patterns", Arg.Clear opt_interpret_patterns, " Do not suggest patterns";
"--items", Arg.Set opt_interpret_items," Annotate each state with its items";
]
~commit:interpret_command;
command "enumerate" "Generate a negative testsuite (sentences that cover possible failures)" [
"-a", Arg.Set opt_enum_all, " Cover all filter-reduce patterns";
"-e", Arg.String (push opt_enum_entrypoints),
"<entrypoint> Enumerate sentences from this entrypoint (multiple allowed)";
] ~commit:enumerate_command;
command "import-messages" "<file.messages> Generate a .lrgrep file from a .messages file" [
"-o", Arg.Set_string opt_import_output, "<file.lrgrep> Output destination";
"--shortest", Arg.Set opt_import_shortest, " Shortest matching strategy";
] ~anon:set_import_message_file ~commit:import_command;
]
let () =
usage_error_fun := (fun msg ->
Subarg.usage global_specs commands
(command_name ^ ": " ^ msg ^ "\n\n" ^ usage_prompt)
);
Subarg.parse global_specs commands
(fun arg ->
if Sys.file_exists arg then (
set_spec_file arg;
Printf.eprintf "warning: directly passing %S is deprecated, use -s %S\n"
arg arg
)
)
usage_prompt
~no_subcommand:(fun () -> usage_error "expecting at least one command")
end
let run_lrgrep () =
let open Make(struct
let name = ""
let grammar = None
let parser_name = ""
end)()
in
()
let run_custom_lrgrep
(type g)
~language_name ~parser_module_name ~(grammar:g grammar)
?string_of_terminal
?string_of_sentence
?sentence_lexer
()
=
let open Make(struct
let name = language_name
let grammar = Some (module struct
type nonrec g = g
let grammar = grammar
let string_of_terminal = string_of_terminal
let string_of_sentence = string_of_sentence
let sentence_lexer = sentence_lexer
end : CUSTOM_GRAMMAR)
let parser_name = parser_module_name
end)()
in
()