ENIAMvalence.ml
35.7 KB
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
(*
* ENIAMlexSemantics is a library that assigns tokens with lexicosemantic information.
* Copyright (C) 2016-2017 Wojciech Jaworski <wjaworski atSPAMfree mimuw dot edu dot pl>
* Copyright (C) 2016-2017 Institute of Computer Science Polish Academy of Sciences
*
* This library is free software: you can redistribute it and/or modify
* it under the terms of the GNU Lesser General Public License as published by
* the Free Software Foundation, either version 3 of the License, or
* (at your option) any later version.
*
* This library is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public License
* along with this program. If not, see <http://www.gnu.org/licenses/>.
*)
open ENIAMwalTypes
open Xstd
let transform_zeby = function
Aff -> [Comp "że"]
| Negation -> [Comp "że";Comp "żeby"]
| NegationUndef -> [Comp "że";Comp "żeby"]
let transform_gdy = function
"indicative" -> [Comp "gdy"]
| "imperative" -> [Comp "gdy"]
| "conditional" -> [Comp "gdyby"]
| "gerundial" -> [Comp "gdy"]
| "no-subj" -> [Comp "gdy"]
| "" -> [Comp "gdy";Comp "gdyby"]
| s -> failwith ("transform_gdy: " ^ s)
let transform_comp negation mood = function
Comp comp -> [Comp comp]
| Zeby -> transform_zeby negation
| Gdy -> transform_gdy mood
| CompUndef -> [CompUndef]
let transform_str mood negation =
if mood = "gerundial" then [Case "gen"] else
match negation with
Aff -> [Case "acc"]
| Negation -> [Case "gen"]
| NegationUndef -> [Case "acc";Case "gen"]
let transform_phrase pos lemma = function
| CP(ctype,comp) as morf -> [morf]
| PrepNP _ as morf -> [morf]
| PrepFixed _ as morf -> [morf]
| PrepAdjP _ as morf -> [morf]
| ComprepNP _ as morf -> [morf]
| ComparP _ as morf -> [morf]
| PrepNCP _ as morf -> [morf]
| InfP _ as morf -> [morf]
| AdvP _ as morf -> [morf]
| XP as morf -> [morf]
| AdjA as morf -> [morf]
| PadvP as morf -> [morf]
| AdMod _ as morf -> [morf]
| Qub as morf -> [morf]
| FixedP _ as morf -> [morf]
| SymbolP as morf -> [morf]
| ColonP as morf -> [morf]
| Inclusion as morf -> [morf]
| Or as morf -> [morf]
| Pro as morf -> [morf]
| Null as morf -> [morf]
| morf -> failwith ("transform_phrase: " ^ lemma ^ " " ^ pos ^ " " ^ ENIAMwalStringOf.phrase morf)
let transform_noun_phrase lemma = function
NP(Case case) -> [NP(Case case)(*;NumP(Case case)*)]
| NP(CaseAgr) -> [NP(CaseAgr)(*;NumP(CaseAgr)*)]
| NPA(CaseAgr) -> [NPA(CaseAgr)(*;NumP(CaseAgr)*)]
| NCP(Case c,ctype,comp) -> [NCP(Case c,ctype,comp)]
| NP(CaseUndef) -> [NP(CaseUndef)(*;NumP(Case case)*)]
| AdjP(Case _) as morf -> [morf] (* tylko 'zagłada adjp(postp)' *)
| AdjP(CaseAgr) -> [AdjP(AllAgr)]
| AdjP(Str) -> [AdjP(AllAgr)] (* chyba błąd walentego, tylko 'barwa', 'bieda', 'głód', 'kolor', 'nędza', 'śmierć', 'usta' *)
| morf -> transform_phrase "noun" lemma morf
let transform_noun_pos lemma = function
| SUBST(_,Case _) as morf -> [morf]
| PPRON3(_,Case _) as morf -> [morf]
| SUBST(_,CaseAgr) as morf -> [morf]
| SUBST(n,Str) -> [ADJ(n,AllAgr,GenderUndef,Grad "pos")] (* błąd walentym: 'zła godzina' *)
| ADJ(_,Case _,_,_) as morf -> [morf]
| ADJ(n,CaseAgr,GenderAgr,gr) -> [ADJ(n,AllAgr,GenderAgr,gr)]
| PACT(n,CaseAgr,g,a,neg) -> [PACT(n,AllAgr,g,a,neg)]
| PPAS(_,Case _,_,_,_) as morf -> [morf]
| PPAS(n,CaseAgr,g,a,neg) -> [PPAS(n,AllAgr,g,a,neg)]
| ADJ(n,Str,g,gr) -> [ADJ(n,AllAgr,g,gr)]
| PPAS(n,Str,g,a,neg) -> [PPAS(n,AllAgr,g,a,neg)]
| PREP(Case _) as morf -> [morf]
| ADV _ as morf -> [morf] (* tu trafiają przysłówkowe realizacje *)
| COMP _ as morf -> [morf]
| QUB as morf -> [morf]
| PERS _ as morf -> [morf]
| pos -> failwith ("transform_noun_pos: " ^ lemma ^ " " ^ ENIAMwalStringOf.pos pos)
let transform_adj_phrase lemma = function
NP(Case case) -> [NP(Case case)(*;NumP(Case case)*)]
| NP(Part) -> [NP(Case "gen");NP(Case "acc")(*;NumP(Case "gen");NumP(Case "acc")*)] (* jedno wystąpienie 'krewny' *)
| NCP(Case c,ctype,comp) -> [NCP(Case c,ctype,comp)]
| AdjP(CaseAgr) -> [AdjP(AllAgr)] (* jedno wystąpienie 'cały szczęśliwy', może się przydać podniesienie typu *)
| morf -> transform_phrase "adj" lemma morf
let transform_adj_pos lemma = function
| ADJ(n,CaseAgr,g,gr) -> [ADJ(n,AllAgr,g,gr)]
| PREP(Case _) as morf -> [morf]
| ADV _ as morf -> [morf]
| QUB as morf -> [morf]
| morf -> failwith ("transform_adj_pos: " ^ lemma ^ " " ^ ENIAMwalStringOf.pos morf)
let transform_adv_phrase lemma = function
NP(Case case) -> [NP(Case case)]
| NP(CaseUndef) -> [NP(CaseUndef)]
| NCP(Case c,ctype,comp) -> [NCP(Case c,ctype,comp)]
| morf -> transform_phrase "adv" lemma morf
let transform_adv_pos lemma = function
SUBST(_,Case _) as morf -> [morf]
(* | ADJ(_,CaseAgr,_,_) as morf -> [morf]*)
| COMP _ as morf -> [morf]
| PREP(Case _) as morf -> [morf]
| COMPAR _ as morf -> [morf]
| ADV _ as morf -> [morf] (* tu trafiają przysłówkowe realizacje *)
| morf -> failwith ("transform_adv_pos: " ^ lemma ^ " " ^ ENIAMwalStringOf.pos morf)
let transform_prep_phrase lemma = function
NP(Case case) -> [NP(Case case)]
| morf -> transform_phrase "prep" lemma morf
let transform_prep_pos lemma = function
| SUBST(_,Case _) as morf -> [morf]
| SIEBIE(Case _) as morf -> [morf]
| PPRON12(_,Case _) as morf -> [morf]
| PPRON3(_,Case _) as morf -> [morf]
| SUBST(n,Str) -> [SUBST(n,CaseAgr)]
| NUM(Case _,_) as morf -> [morf]
| ADJ(_,Case _,_,_) as morf -> [morf]
| GER(_,Case _,_,_,_) as morf -> [morf]
| PPAS(_,Case _,_,_,_) as morf -> [morf]
(* | ADV _ as morf -> [morf]*)
| QUB as morf -> [morf]
| pos -> failwith ("transform_prep_pos: " ^ lemma ^ " " ^ ENIAMwalStringOf.pos pos)
let transform_comprep_phrase lemma = function
NP(Case case) -> [NP(Case case)(*;NumP(Case case)*)]
| NCP(Case c,ctype,comp) -> [NCP(Case c,ctype,comp)]
| morf -> transform_phrase "comprep" lemma morf
let transform_comprep_pos lemma = function
| SUBST(_,Case _) as morf -> [morf]
(* | SUBST(n,Str) -> [SUBST(n,CaseAgr)]*)
| NUM(Case _,_) as morf -> [morf]
(* | ADJ(_,Case _,_,_) as morf -> [morf]
| GER(_,Case _,_,_,_,_) as morf -> [morf]
| PPAS(_,Case _,_,_,_) as morf -> [morf]
| ADV _ as morf -> [morf]
| QUB as morf -> [morf]*)
| pos -> failwith ("transform_comprep_pos: " ^ lemma ^ " " ^ ENIAMwalStringOf.pos pos)
let transform_compar_phrase lemma = function
| NP(Str) -> Xlist.map ["nom";"gen";"dat";"acc";"inst"] (fun case -> NP(Case case)) (* FIXME: sprawdzić kto kontroluje! *) (* FIXME: uzgodnić a komparatywem *)
| morf -> transform_phrase "compar" lemma morf
let transform_compar_pos lemma = function
| SUBST(_,Case _) as morf -> [morf]
| ADJ(_,Case _,_,_) as morf -> [morf]
| PREP(Case _) as morf -> [morf]
| SUBST(Number n,Str) -> [SUBST(Number n,CaseUndef)]
| SUBST(NumberAgr,Str) -> [SUBST(NumberUndef,CaseUndef)]
| SUBST(NumberUndef,Str) -> [SUBST(NumberUndef,CaseUndef)]
| PPAS(_,Case _,_,_,_) as morf -> [morf]
| PPAS(NumberAgr,Str,GenderAgr,a,neg) -> [PPAS(NumberUndef,CaseUndef,GenderUndef,a,neg)] (* FIXME: ta sama sytuacja co w "jako" *)
| PPAS(NumberAgr,CaseAgr,GenderAgr,a,neg) -> [PPAS(NumberUndef,CaseUndef,GenderUndef,a,neg)] (* FIXME: ta sama sytuacja co w "jako" *)
| ADJ(NumberAgr,Str,GenderAgr,gr) -> [ADJ(NumberUndef,CaseUndef,GenderUndef,gr)] (* FIXME: ta sama sytuacja co w "jako" *)
| ADJ(NumberAgr,CaseAgr,GenderAgr,gr) -> [ADJ(NumberUndef,CaseUndef,GenderUndef,gr)] (* FIXME: ta sama sytuacja co w "jako" *)
| NUM(Case _,_) as morf -> [morf]
| pos -> failwith ("transform_compar_pos: " ^ lemma ^ " " ^ ENIAMwalStringOf.pos pos)
let transform_comp_pos lemma = function
| PERS _ as morf -> [morf]
| pos -> failwith ("transform_comp_pos: " ^ lemma ^ " " ^ ENIAMwalStringOf.pos pos)
let transform_qub_pos lemma = function
| QUB as morf -> [morf]
| pos -> failwith ("transform_qub_pos: " ^ lemma ^ " " ^ ENIAMwalStringOf.pos pos)
let transform_interj_phrase lemma = function
NP(Case "nom") as morf -> [morf]
| morf -> transform_phrase "interj" lemma morf
let transform_interj_pos lemma = function
| pos -> failwith ("transform_interj_pos: " ^ lemma ^ " " ^ ENIAMwalStringOf.pos pos)
let transform_sinterj_pos lemma = function
| pos -> failwith ("transform_sinterj_pos: " ^ lemma ^ " " ^ ENIAMwalStringOf.pos pos)
let transform_aglt_pos lemma = function
| pos -> failwith ("transform_aglt_pos: " ^ lemma ^ " " ^ ENIAMwalStringOf.pos pos)
let transform_siebie_pos lemma = function
| ADJ(NumberAgr,CaseAgr,GenderAgr,gr) -> [ADJ(NumberAgr,AllAgr,GenderAgr,gr)]
| pos -> failwith ("transform_siebie_pos: " ^ lemma ^ " " ^ ENIAMwalStringOf.pos pos)
let transform_pers_subj_phrase lemma negation mood = function (* FIXME: prepnp(na,loc) *)
| NP(Str) -> [NP(NomAgr);NP(VocAgr)(*;NumP(NomAgr)*)]
| NP(Part) -> [NP(Case "gen");NP(Case "acc")(*;NumP(Case "gen");NumP(Case "acc")*)] (* tylko w 'nalewać', 'nalać', 'ponalewać', 'najechać','uzbierać' *)
| NCP(Str,ctype,comp) -> [NCP(NomAgr,ctype,comp);NCP(VocAgr,ctype,comp)]
| Pro -> [ProNG]
| morf -> transform_phrase "pers_subj" lemma morf
let transform_pers_subj_pos lemma negation mood = function
(* COMP _ as morf -> [morf]*)
| SUBST(n,Str) -> [SUBST(n,NomAgr)]
| SUBST(n,Case "nom") -> [SUBST(n,NomAgr)] (* wygląda na błąd Walentego, ale nie ma znaczenia *)
| NUM(Str,g) -> [NUM(NomAgr,g)]
| NUM(Case "nom",g) -> [NUM(NomAgr,g)]
(* | ADJ(n,Str,g,gr) -> [ADJ(n,NomAgr,g,gr)]*)
| morf -> failwith ("transform_pers_subj_pos: " ^ lemma ^ " " ^ ENIAMwalStringOf.pos morf)
let transform_ger_subj_phrase lemma negation mood control = function
| NP(Str) -> [NP(Case "gen");PrepNP("przez",Case "acc")(*;NumP(Case "gen")*)(*;PrepNumP("przez",Case "acc")*)] (* FIXME: czy przez:acc jest możliwe? *)
| NP(Part) -> [NP(Case "gen")(*;NP(Case "acc")*)(*;NumP(Case "gen");NumP(Case "acc")*)]
| NCP(Str,ctype,comp) -> [NCP(Case "gen",ctype,comp);PrepNCP("przez",Case "acc",ctype,comp)] (* FIXME: czy przez:acc jest możliwe? *)
| Pro -> if control then [Pro] else [Null]
| morf -> transform_phrase "ger_subj" lemma morf
let transform_ger_subj_pos lemma negation mood = function (* FIXME: ADV(_) *)
(* COMP _ as morf -> [morf] (* FIXME: czy to jest możliwe? *)*)
| SUBST(n,Str) -> [SUBST(n,Case "gen")]
| SUBST(n,Case "nom") -> [SUBST(n,Case "gen")] (* wygląda na błąd Walentego, ale nie ma znaczenia *)
| NUM(Str,g) -> [NUM(Case "gen",g)]
| NUM(Case "nom",g) -> [NUM(Case "gen",g)]
(* | ADJ(n,Str,g,gr) -> [ADJ(n,Case "gen",g,gr)]*)
| morf -> failwith ("transform_pers_subj_pos: " ^ lemma ^ " " ^ ENIAMwalStringOf.pos morf)
let transform_ppas_subj_phrase lemma negation mood control = function
| NP(Str) -> [PrepNP("przez",Case "acc")(*;PrepNumP("przez",Case "acc")*)]
| NCP(Str,ctype,comp) -> [PrepNCP("przez",Case "acc",ctype,comp)]
| Pro -> if control then [Pro] else [Null]
| morf -> transform_phrase "ppas_subj" lemma morf
let transform_pers_phrase lemma negation mood = function
| NP(Str) -> List.flatten (Xlist.map (transform_str mood negation) (fun case -> [NP case(*;NumP(case)*)]))
| NP(Part) -> [NP(Case "gen")] @ (if mood = "gerundial" then [] else [NP(Case "acc")(*;NumP(Case "gen");NumP(Case "acc")*)])
| NP(Case case) -> [NP(Case case)(*;NumP(Case case)*)]
| NP(CaseUndef) -> [NP(CaseUndef)(*;NumP(Case case)*)]
| NCP(Str,ctype,comp) -> List.flatten (Xlist.map (transform_str mood negation) (fun case -> [NCP(case,ctype,comp)]))
| NCP(Part,ctype,comp) -> List.flatten (Xlist.map (transform_str mood negation) (fun case -> [NCP(case,ctype,comp)]))
| NCP(Case case,ctype,comp) -> [NCP(Case case,ctype,comp)]
| AdjP(Str) -> Xlist.map (transform_str mood negation) (fun case -> AdjP case) (* FIXME: pomijam uzgadnianie liczby i rodzaju - wykonalne za pomocą kontroli *)
| AdjP CaseAgr as morf -> if mood = "gerundial" then [AdjP AllAgr] else (failwith ("transform_pers_phrase2: " ^ lemma ^ " " ^ ENIAMwalStringOf.phrase morf))
| AdjP(Case _) as morf -> [morf] (* FIXME: pomijam uzgadnianie liczby i rodzaju - wykonalne za pomocą kontroli *)
| AdjP(NomAgr) as morf -> if mood = "no-subj" then [AdjP(Case "nom")] else [morf]
| AdvP "misc" -> if mood = "gerundial" then [AdjP AllAgr] else [AdvP "misc"]
| AdvP "mod" -> if mood = "gerundial" then [AdjP AllAgr] else [AdvP "mod"]
| morf -> transform_phrase "pers" lemma morf
let transform_pers_pos lemma negation mood = function
| SUBST(n,Str) -> Xlist.map (transform_str mood negation) (fun case -> SUBST(n,case))
| PPRON12(n,Str) -> Xlist.map (transform_str mood negation) (fun case -> PPRON12(n,case))
| PPRON3(n,Str) -> Xlist.map (transform_str mood negation) (fun case -> PPRON3(n,case))
| SIEBIE(Str) -> Xlist.map (transform_str mood negation) (fun case -> SIEBIE(case))
| NUM(Str,g) -> Xlist.map (transform_str mood negation) (fun case -> NUM(case,g))
| ADJ(n,Str,g,gr) -> Xlist.map (transform_str mood negation) (fun case -> ADJ(n,case,g,gr))
(* | PPAS(n,Str,g,a,neg) -> Xlist.map (transform_str negation) (fun case -> PPAS(n,Str,g,a,neg))*)
| SUBST(n,Part) -> [SUBST(n,Case "gen");SUBST(n,Case "acc")]
| ADJ(n,Part,g,gr) -> [ADJ(n,Case "gen",g,gr);ADJ(n,Case "acc",g,gr)]
| ADJ(n,CaseAgr,g,gr) as morf -> if lemma = "siedzieć" then [ADJ(n,AllAgr,g,gr)] else (failwith ("transform_pers_pos2: " ^ lemma ^ " " ^ ENIAMwalStringOf.pos morf)) (* FIXME *)
| SUBST(_,Case _) as morf -> [morf]
| PPRON12(_,Case _) as morf -> [morf]
| PPRON3(_,Case _) as morf -> [morf]
| SIEBIE(Case _) as morf -> [morf]
| NUM(Case _,_) as morf -> [morf]
| PREP _ as morf -> [morf]
| ADJ(_,Case _,_,_) as morf -> [morf]
| PPAS(_,Case _,_,_,_) as morf -> [morf]
(* | SUBST(n,CaseAgr) -> Xlist.map ["nom";"gen";"dat";"acc";"inst"] (fun case -> SUBST(n,Case case)) (* FIXME: sprawdzić kto kontroluje! *)
| ADJ(n,CaseAgr,g,gr) -> Xlist.map ["nom";"gen";"dat";"acc";"inst"] (fun case -> ADJ(n,Case case,g,gr)) (* FIXME: sprawdzić kto kontroluje! *)*)
| COMPAR _ as morf -> [morf]
| COMP _ as morf -> [morf]
| INF _ as morf -> [morf]
| QUB as morf -> [morf]
| ADV grad -> (*if mood = "gerundial" then [ADJ(NumberAgr,AllAgr,GenderAgr,grad)] else*) [ADV grad] (* FIXME: to nie poprawi lematu *)
| PERS _ as morf -> [morf]
| morf -> failwith ("transform_pers_pos: " ^ lemma ^ " " ^ ENIAMwalStringOf.pos morf)
let rec transform_comps negation mood = function
| CP(ctype,comp) -> Xlist.map (transform_comp negation mood comp) (fun comp -> CP(ctype,comp))
| NCP(case,ctype,comp) -> Xlist.map (transform_comp negation mood comp) (fun comp -> NCP(case,ctype,comp))
| PrepNCP(prep,case,ctype,comp) -> Xlist.map (transform_comp negation mood comp) (fun comp -> PrepNCP(prep,case,ctype,comp))
| E phrase -> Xlist.map (transform_comps negation mood phrase) (fun phrase -> E phrase)
| morf -> [morf]
let compars = StringSet.of_list ["jak"; "jako"; "niż"; "niczym" ;"niby"; "co"; "zamiast"]
let is_compar lex = StringSet.mem compars lex
(* FIXME: pomijam uzgadnianie przypadku, liczby i rodzaju - wykonalne za pomocą kontroli *)
let transform_preps morf =
let morf = match morf with
| LexArg(id,lex,PREP c) -> if is_compar lex then LexArg(id,lex,COMPAR c) else LexArg(id,lex,PREP c)
| SimpleLexArg(lex,PREP c) -> if is_compar lex then SimpleLexArg(lex,COMPAR c) else SimpleLexArg(lex,PREP c)
| PrepNP(prep,c) -> if is_compar prep then ComparP(prep,c) else PrepNP(prep,c)
| PrepAdjP(prep,c) -> if is_compar prep then ComparP(prep,c) else PrepAdjP(prep,c)
| PrepNCP(prep,case,ctype,comp) as morf -> if is_compar prep then failwith "transform_preps 1" else morf
| morf -> morf in
match morf with
| ComparP(prep,Str) -> Xlist.map ["nom";"gen";"dat";"acc";"inst";"postp"] (fun case -> ComparP(prep,Case case))
| ComparP(_,CaseUndef) as morf -> [morf]
| ComparP(_,Case _) as morf -> [morf]
| LexArg(id,lex,COMPAR Str) -> Xlist.map ["nom";"gen";"dat";"acc";"inst";"postp"] (fun case -> LexArg(id,lex,COMPAR (Case case)))
| SimpleLexArg(lex,COMPAR Str) -> Xlist.map ["nom";"gen";"dat";"acc";"inst";"postp"] (fun case -> SimpleLexArg(lex,COMPAR (Case case)))
| LexArg(id,lex,COMPAR (Case _)) as morf -> [morf]
| SimpleLexArg(lex,COMPAR (Case _)) as morf -> [morf]
| LexArg(id,lex,COMPAR _) -> failwith "transform_preps 3"
| SimpleLexArg(lex,COMPAR _) -> failwith "transform_preps 4"
| PrepNP("per",Str) -> [PrepNP("per",Case "nom");PrepNP("per",Case "voc")] (* FIXME: voc do poprawienie w leksykonie *)
| PrepNP(_,Case _) as morf -> [morf]
| PrepAdjP(_,Case _) as morf -> [morf]
| PrepNCP(_,Case _,_,_) as morf -> [morf]
| PrepNP(_,CaseUndef) as morf -> [morf]
| PrepNP _ as morf -> failwith ("transform_preps 5: " ^ ENIAMwalStringOf.phrase morf)
| PrepAdjP _ -> failwith "transform_preps 6"
| PrepNCP _ -> failwith "transform_preps 7"
| LexArg(id,"w",PREP Str) -> [LexArg(id,"w",PREP (Case "acc"));LexArg(id,"w",PREP (Case "loc"));]
| SimpleLexArg("w",PREP Str) -> [SimpleLexArg("w",PREP (Case "acc"));SimpleLexArg("w",PREP (Case "loc"))]
| LexArg(id,lex,PREP (Case _)) as morf -> [morf]
| SimpleLexArg(lex,PREP (Case _)) as morf -> [morf]
| LexArg(id,lex,PREP _) -> failwith "transform_preps 8"
| SimpleLexArg(lex,PREP _) -> failwith "transform_preps 9"
| morf -> [morf]
let transform_pers_schema lemma negation mood schema =
Xlist.map schema (fun s ->
{s with morfs = (
let morfs = List.flatten (Xlist.map s.morfs (transform_comps negation mood)) in
(* Printf.printf "A %s\n" (String.concat " " (Xlist.map morfs ENIAMwalStringOf.phrase)); *)
let morfs = List.flatten (Xlist.map morfs transform_preps) in
(* Printf.printf "B %s\n" (String.concat " " (Xlist.map morfs ENIAMwalStringOf.phrase)); *)
let morfs = if s.gf = SUBJ then List.flatten (Xlist.map morfs (function
| E phrase -> Xlist.map (transform_pers_subj_phrase lemma negation mood phrase) (fun phrase -> E phrase)
| LexArg(id,lex,pos) -> Xlist.map (transform_pers_subj_pos lemma negation mood pos) (fun pos -> LexArg(id,lex,pos))
| SimpleLexArg(lex,pos) -> Xlist.map (transform_pers_subj_pos lemma negation mood pos) (fun pos -> SimpleLexArg(lex,pos))
| phrase -> transform_pers_subj_phrase lemma negation mood phrase))
else List.flatten (Xlist.map morfs (function
| LexArg(id,lex,pos) -> Xlist.map (transform_pers_pos lemma negation mood pos) (fun pos -> LexArg(id,lex,pos))
| SimpleLexArg(lex,pos) -> Xlist.map (transform_pers_pos lemma negation mood pos) (fun pos -> SimpleLexArg(lex,pos))
| phrase -> transform_pers_phrase lemma negation mood phrase)) in
(* Printf.printf "C %s\n" (String.concat " " (Xlist.map morfs ENIAMwalStringOf.phrase)); *)
morfs)})
let transform_nosubj_schema lemma negation mood schema =
Xlist.map schema (fun s ->
{s with morfs =
let morfs = List.flatten (Xlist.map s.morfs (transform_comps negation mood)) in
let morfs = List.flatten (Xlist.map morfs transform_preps) in
if s.gf = SUBJ then [Null]
else List.flatten (Xlist.map morfs (function
| LexArg(id,lex,pos) -> Xlist.map (transform_pers_pos lemma negation mood pos) (fun pos -> LexArg(id,lex,pos))
| SimpleLexArg(lex,pos) -> Xlist.map (transform_pers_pos lemma negation mood pos) (fun pos -> SimpleLexArg(lex,pos))
| phrase -> transform_pers_phrase lemma negation mood phrase))})
(* let transform_ger_adv_lex = function
| s -> print_endline ("transform_ger_adv_lex: " ^ s); s
let transform_ger_adv_pos = function
| LexArg(id,lex,ADV grad) -> LexArg(id,transform_ger_adv_lex lex,ADJ(NumberAgr,AllAgr,GenderAgr,grad))
| SimpleLexArg(lex,ADV grad) -> SimpleLexArg(transform_ger_adv_lex lex,ADJ(NumberAgr,AllAgr,GenderAgr,grad))
| morf -> morf *)
let transform_ger_schema lemma negation schema = (* FIXME: zakładam, że ger zeruje mood, czy to prawda? *)
Xlist.map schema (fun s ->
{s with morfs =
let morfs = List.flatten (Xlist.map s.morfs (transform_comps negation "gerundial")) in
let morfs = List.flatten (Xlist.map morfs transform_preps) in
(* let morfs = Xlist.map morfs transform_ger_adv_pos in *)
if s.gf = SUBJ then List.flatten (Xlist.map morfs (function
| E phrase -> Xlist.map (transform_ger_subj_phrase lemma negation "gerundial" (s.cr <> [] || s.ce <> []) phrase) (fun phrase -> E phrase)
| LexArg(id,lex,pos) -> Xlist.map (transform_ger_subj_pos lemma negation "gerundial" pos) (fun pos -> LexArg(id,lex,pos))
| SimpleLexArg(lex,pos) -> Xlist.map (transform_ger_subj_pos lemma negation "gerundial" pos) (fun pos -> SimpleLexArg(lex,pos))
| phrase -> transform_ger_subj_phrase lemma negation "gerundial" (s.cr <> [] || s.ce <> []) phrase))
else List.flatten (Xlist.map morfs (function
| LexArg(id,lex,pos) -> Xlist.map (transform_pers_pos lemma negation "gerundial" pos) (fun pos -> LexArg(id,lex,pos))
| SimpleLexArg(lex,pos) -> Xlist.map (transform_pers_pos lemma negation "gerundial" pos) (fun pos -> SimpleLexArg(lex,pos))
| phrase -> transform_pers_phrase lemma negation "gerundial" phrase))})
let transform_ppas_schema lemma negation mood schema =
if not (Xlist.fold schema false (fun b p -> if p.gf = OBJ then true else b)) then
(*failwith ("transform_ppas_schema: attempt to make ppas schema for lemma " ^ lemma ^ "without OBJ arg")*)raise Not_found else
Xlist.map schema (fun s ->
let morfs = List.flatten (Xlist.map s.morfs (transform_comps negation mood)) in
let morfs = List.flatten (Xlist.map morfs transform_preps) in
{s with morfs =
if s.gf = OBJ then [Null] else
if s.gf = SUBJ then List.flatten (Xlist.map morfs (function
| E phrase -> raise Not_found (* tylko 'obladzać' i 'oblodzić', chyba błąd *)
| LexArg(id,lex,SUBST(n,Str)) -> raise Not_found (* FIXME!!! *)
| SimpleLexArg(lex,SUBST(n,Str)) -> raise Not_found (* FIXME!!! *)
| phrase -> transform_ppas_subj_phrase lemma negation mood (s.cr <> [] || s.ce <> []) phrase))
else List.flatten (Xlist.map morfs (function
| LexArg(id,lex,pos) -> Xlist.map (transform_pers_pos lemma negation mood pos) (fun pos -> LexArg(id,lex,pos))
| SimpleLexArg(lex,pos) -> Xlist.map (transform_pers_pos lemma negation mood pos) (fun pos -> SimpleLexArg(lex,pos))
| phrase -> transform_pers_phrase lemma negation mood phrase))})
let transform_num_schema acm schema =
Xlist.map schema (fun s ->
{s with morfs=List.flatten (Xlist.map s.morfs (function
| Null -> [Null]
| FixedP _ as morf -> [morf]
| LexArg(id,lex,SUBST(NumberUndef,CaseUndef)) ->
(match acm with
"rec" -> [LexArg(id,lex,SUBST(NumberUndef,GenAgr))]
| "congr" -> [LexArg(id,lex,SUBST(NumberUndef,AllAgr))]
| _ -> failwith "transform_num_schema")
| SimpleLexArg(lex,SUBST(NumberUndef,CaseUndef)) ->
(match acm with
"rec" -> [SimpleLexArg(lex,SUBST(NumberUndef,GenAgr))]
| "congr" -> [SimpleLexArg(lex,SUBST(NumberUndef,AllAgr))]
| _ -> failwith "transform_num_schema")
| morf -> failwith ("transform_num_schema: " ^ ENIAMwalStringOf.phrase morf)))})
let transform_schema pos lemma schema =
let phrase_fun,pos_fun = match pos with
"subst" -> transform_noun_phrase,transform_noun_pos
| "adj" -> transform_adj_phrase,transform_adj_pos
| "adv" -> transform_adv_phrase,transform_adv_pos
| "prep" | "x" -> transform_prep_phrase,transform_prep_pos
| "comprep" -> transform_comprep_phrase,transform_comprep_pos
| "compar" -> transform_compar_phrase,transform_compar_pos
| "comp" -> transform_phrase "comp",transform_comp_pos
| "qub" -> transform_phrase "qub",transform_qub_pos
| "siebie" -> transform_phrase "siebie",transform_siebie_pos
| "interj" -> transform_interj_phrase,transform_interj_pos
| "sinterj" -> transform_phrase "sinterj",transform_interj_pos
| "aglt" -> transform_phrase "aglt",transform_interj_pos
| _ -> failwith "transform_schema"
in
Xlist.map schema (fun s ->
let morfs = List.flatten (Xlist.map s.morfs (transform_comps NegationUndef "")) in (* FIXME: zależność od trybu warunkowego i negacji *)
let morfs = List.flatten (Xlist.map morfs transform_preps) in
{s with morfs=List.flatten (Xlist.map morfs (function
LexArg(id,lex,pos) -> Xlist.map (pos_fun lemma pos) (fun pos -> LexArg(id,lex,pos))
| SimpleLexArg(lex,pos) -> Xlist.map (pos_fun lemma pos) (fun pos -> SimpleLexArg(lex,pos))
| phrase -> phrase_fun lemma phrase))})
let rec remove_adj_agr = function
[] -> []
| {morfs=[Null;AdjP(CaseAgr)]} :: l -> remove_adj_agr l
| {morfs=[Null;AdjP(Part)]} :: l -> remove_adj_agr l
| s :: l -> (*print_endline (ENIAMwalStringOf.schema [s]);*) s :: (remove_adj_agr l)
let rec get_role gf = function
[] -> raise Not_found
| s :: l -> if s.gf = gf then s.role,s.role_attr else get_role gf l
let expand_negation = function
Negation -> [Negation]
| Aff -> [Aff]
| NegationUndef -> [Negation;Aff]
let expand_aspect = function
Aspect s -> [Aspect s]
| AspectUndef -> [Aspect "imperf";Aspect "perf"]
let aspect_sel = function
Aspect s -> [ENIAM_LCGlexiconTypes.Aspect,ENIAM_LCGlexiconTypes.Eq,[s]]
| AspectUndef -> []
open ENIAM_LCGlexiconTypes
let transform_entry pos lemma negation pred aspect schema =
(* print_endline "transform_entry: start"; *)
match pos with
"subst" | "depr" | "year" | "day" | "day-month" | "date" | "date-interval" |
"hour-minute" | "hour" | "hour-minute-interval" | "hour-interval" |
"day-interval" | "day-month-interval" | "month-interval" | "year-interval" | "initial" | "fixed" | "unk" ->
if negation <> NegationUndef || pred <> PredFalse || aspect <> AspectUndef then failwith ("transform_entry 1");
[[],transform_schema "subst" lemma schema]
| "adj" | "adja" | "adjc" |"adjp" ->
if negation <> NegationUndef || aspect <> AspectUndef then failwith ("transform_entry 2");
let sel = match pred with PredTrue -> [Case,Eq,["pred"]] | _ -> [] in
[sel,transform_schema "adj" lemma schema]
| "adv" | "prep" | "comprep" | "comp" | "compar" | "qub" | "siebie" | "x" ->
if negation <> NegationUndef || (*pred <> PredFalse ||*) aspect <> AspectUndef then failwith ("transform_entry 3"); (* FIXME: typy przysłówków *)
[[],transform_schema pos lemma schema]
| _ ->
if pred <> PredFalse then failwith ("transform_entry 4") else
if pos = "num" || pos = "intnum" then (
if negation <> NegationUndef || aspect <> AspectUndef then failwith ("transform_entry 5");
Xlist.map ["congr";"rec"] (fun acm ->
[Acm,Eq,[acm]],transform_num_schema acm schema)) else
if pos = "interj" then (
if negation <> NegationUndef || pred <> PredFalse || aspect <> AspectUndef then failwith ("transform_entry 6");
[[],transform_schema "interj" lemma schema]) else
if (pos = "conj" || pos = "interp" || pos = "ppron12" || pos = "ppron3" || pos = "numcomp" || pos = "realnum" ||
pos = "intnum-interval" || pos = "realnum-interval" || pos = "symbol" || pos = "ordnum" || pos = "roman" ||
pos = "roman-interval" || pos = "roman-ordnum" || pos = "match-result" || pos = "url" || pos = "email" ||
pos = "phone-number" || pos = "postal-code" || pos = "obj-id" || pos = "building-number" || pos = "list-item" ||
pos = "aglt" || pos = "pacta" || pos = "part" || pos = "conj" || pos = "sinterj" || pos = "burk" ||
pos = "interp" || pos = "xxx" || pos = "unk" || pos = "html-tag" || pos = "apron" || pos = "x" ||
pos = "other") && schema = [] then [[],[]] else
List.flatten (Xlist.map (expand_negation negation) (fun negation ->
let sel = [Negation,Eq,[ENIAMwalStringOf.negation negation]] @ aspect_sel aspect in
if pos = "fin" || pos = "bedzie" then
[sel @ [Mood,Eq,["indicative"]],transform_pers_schema lemma negation "indicative" schema;
sel @ [Mood,Eq,["imperative"]],transform_pers_schema lemma negation "imperative" schema] else
if pos = "praet" || pos = "winien" then
[sel @ [Mood,Eq,["indicative"]],transform_pers_schema lemma negation "indicative" schema;
sel @ [Mood,Eq,["conditional"]],transform_pers_schema lemma negation "conditional" schema] else
if pos = "impt" then
[sel @ [Mood,Eq,["imperative"]],transform_nosubj_schema lemma negation "imperative" schema] else
if pos = "imps" then
[sel @ [Mood,Eq,["indicative"]],transform_nosubj_schema lemma negation "indicative" schema] else
if pos = "pred" then
[sel @ [Mood,Eq,["indicative"]],transform_pers_schema lemma negation "indicative" schema] else
if pos = "pcon" || pos = "pant" || pos = "inf" || pos = "pact" then
(* let role,role_attr = try get_role SUBJ schema with Not_found -> "Initiator","" in *)
[sel, transform_nosubj_schema lemma negation "no-subj" schema] else
if pos = "ppas" then
try
(* let role,role_attr = try get_role OBJ schema with Not_found -> "Theme","" in *)
[sel, transform_ppas_schema lemma negation "indicative" schema]
with Not_found -> [] else
if pos = "ger" then
[sel,transform_ger_schema lemma negation schema] else
(* if schema = [] then (print_endline "transform_entry: empty"; [[],[]]) else *)
failwith ("transform_entry: " ^ pos)))
let transform_lex_entry pos lemma = function
SimpleLexEntry(lemma,pos) -> [[],SimpleLexEntry(lemma,pos)]
| LexEntry(id,lemma,pos,NoRestr,schema) ->
Xlist.map (transform_entry pos lemma NegationUndef PredFalse AspectUndef schema) (fun (sel,schema) ->
sel,LexEntry(id,lemma,pos,NoRestr,schema))
| ComprepNPEntry(s,NoRestr,schema) ->
Xlist.map (transform_entry "comprep" s NegationUndef PredFalse AspectUndef schema) (fun (sel,schema) ->
sel,ComprepNPEntry(s,NoRestr,schema))
| LexEntry(id,lemma,pos,_,[]) as entry -> [[],entry]
| entry -> failwith ("transform_lex_entry:" ^ ENIAMwalStringOf.lex_entry entry)
let prepare_all_valence phrases schemata entries =
let schemata = ENIAMwalReduce.merge_entries phrases schemata in
Entries.fold schemata Entries.empty (fun schemata pos lemma (opinion,neg,pred,aspect,schema) ->
match pos with
"noun" -> Entries.add_inc_list schemata "subst" lemma (transform_entry "subst" lemma neg pred aspect schema)
| "adj" -> Entries.add_inc_list schemata "adj" lemma (transform_entry "adj" lemma neg pred aspect schema)
| "adv" -> Entries.add_inc_list schemata "adv" lemma (transform_entry "adv" lemma neg pred aspect schema)
| "verb" ->
let schemata = Entries.add_inc_list schemata "fin" lemma (transform_entry "fin" lemma neg pred aspect schema) in
(*Entries.iter schemata (fun _ _ (_,s) -> print_endline ("E " ^ ENIAMwalStringOf.schema s)); *)
let schemata = Entries.add_inc_list schemata "praet" lemma (transform_entry "praet" lemma neg pred aspect schema) in
let schemata = Entries.add_inc_list schemata "impt" lemma (transform_entry "impt" lemma neg pred aspect schema) in
let schemata = Entries.add_inc_list schemata "imps" lemma (transform_entry "imps" lemma neg pred aspect schema) in
let schemata = Entries.add_inc_list schemata "ger" lemma (transform_entry "ger" lemma neg pred aspect schema) in
let schemata = Entries.add_inc_list schemata "pact" lemma (transform_entry "pact" lemma neg pred aspect schema) in
let schemata = Entries.add_inc_list schemata "ppas" lemma (transform_entry "ppas" lemma neg pred aspect schema) in
let schemata = Entries.add_inc_list schemata "inf" lemma (transform_entry "inf" lemma neg pred aspect schema) in
let schemata = Entries.add_inc_list schemata "pcon" lemma (transform_entry "pcon" lemma neg pred aspect schema) in
let schemata = Entries.add_inc_list schemata "pant" lemma (transform_entry "pant" lemma neg pred aspect schema) in
let schemata = Entries.add_inc_list schemata "bedzie" lemma (transform_entry "bedzie" lemma neg pred aspect schema) in
let schemata = Entries.add_inc_list schemata "winien" lemma (transform_entry "winien" lemma neg pred aspect schema) in
let schemata = Entries.add_inc_list schemata "pred" lemma (transform_entry "pred" lemma neg pred aspect schema) in
schemata
| _ -> failwith "unknown pos"),
(Entries.flatten_map entries transform_lex_entry)
(* let _ = prepare_all_valence ENIAMwalParser.phrases ENIAMwalParser.schemata ENIAMwalParser.entries *)
(* let prepare_selected_valence schemata entries lexemes =
Entries.fold schemata Entries.empty (fun schemata pos2 lemma (opinion,neg,pred,aspect,schema) ->
Xlist.fold (Entries.find pos2 lemma) schemata (function pos ->
Entries.add_inc_list schemata pos lemma (transform_entry pos lemma neg pred aspect schema))),
Entries.flatten_map entries transform_lex_entry *)
let get_default_valence = function
"verb" -> [Nieokreslony,NegationUndef,PredFalse,AspectUndef,[{empty_position with gf=SUBJ; morfs=[NP(Str);NCP(Str,CompTypeUndef,CompUndef)]};
{empty_position with gf=OBJ; morfs=[NP(Str);NCP(Str,CompTypeUndef,CompUndef)]}];
Nieokreslony,NegationUndef,PredFalse,AspectUndef,[{empty_position with gf=SUBJ; morfs=[NP(Str);NCP(Str,CompTypeUndef,CompUndef)]};
{empty_position with mode=["lemma"]; morfs=[SimpleLexArg("się",QUB)]}];]
| "noun" -> [Nieokreslony,NegationUndef,PredFalse,AspectUndef,[]]
| "adj" -> [Nieokreslony,NegationUndef,PredFalse,AspectUndef,[]]
| "adv" -> [Nieokreslony,NegationUndef,PredFalse,AspectUndef,[]]
| _ -> []
open ENIAMcategoriesPL
let get_aroles schema lemma = function
"pact" -> [Xlist.fold schema ([],"Arg","",true) (fun (sel,arole,arole_attr,arev) p ->
if p.gf = SUBJ then sel,p.role,p.role_attr,arev else sel,arole,arole_attr,arev)]
| "ppas" -> [Xlist.fold schema ([],"Arg","",true) (fun (sel,arole,arole_attr,arev) p ->
if p.gf = OBJ then sel,p.role,p.role_attr,arev else sel,arole,arole_attr,arev)]
| "subst" -> [
[Case,Eq,["dat"]],"Recipent","",false;
[Case,Eq,["inst"]],(if StringSet.mem ENIAMlexSemanticsData.subst_inst_time lemma then "Time" else "Instrument"),"",false;
[Case,Neq,["dat";"inst"]],"","",false]
| "adj" | "adjc" | "adjp" -> (* FIXME czy adjc i adjp mogą być adjunctami? *)
let l = try StringMap.find ENIAMlexSemanticsData.adj_roles lemma with Not_found -> ["Attribute",""] in
Xlist.map l (fun (role,role_attr) -> [],role,role_attr,false)
| "adv" ->
let modes = ENIAMcategoriesPL.adv_mode lemma in
let roles = try StringMap.find ENIAMlexSemanticsData.adv_roles lemma with Not_found -> ["Manner",""] in
Xlist.fold modes [] (fun l -> function
"mod" -> Xlist.fold roles l (fun l (role,role_attr) -> ([Mode,Eq,["mod"]],role,role_attr,false) :: l)
| "abl" -> ([Mode,Eq,["abl"]],"Location","Souce",false) :: l
| "adl" -> ([Mode,Eq,["adl"]],"Location","Goal",false) :: l
| "locat" -> ([Mode,Eq,["locat"]],"Location","",false) :: l
| "perl" -> ([Mode,Eq,["perl"]],"Path","",false) :: l
| "dur" -> ([Mode,Eq,["dur"]],"Duration","",false) :: l
| "temp" -> ([Mode,Eq,["temp"]],"Time","",false) :: l
| "pron" -> ([Mode,Eq,["mod"]],"Manner","",false) :: l
| _ -> failwith "get_aroles")
| "qub" ->
let l = try StringMap.find ENIAMlexSemanticsData.qub_roles lemma with Not_found -> ["Arg",""] in
Xlist.map l (fun (role,role_attr) -> [],role,role_attr,false)
| _ -> [[],"","",false]