walFrames.ml
96.7 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
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
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
1554
1555
1556
1557
1558
1559
1560
1561
1562
1563
1564
1565
1566
1567
1568
1569
1570
1571
1572
1573
1574
1575
1576
1577
1578
1579
1580
1581
1582
1583
1584
1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
1598
1599
1600
1601
1602
1603
1604
1605
1606
1607
1608
1609
1610
1611
1612
1613
1614
1615
1616
1617
1618
1619
1620
1621
1622
1623
1624
1625
1626
1627
1628
1629
1630
1631
1632
1633
1634
1635
1636
1637
1638
1639
1640
1641
(*
* ENIAM: Categorial Syntactic-Semantic Parser for Polish
* Copyright (C) 2016 Wojciech Jaworski <wjaworski atSPAMfree mimuw dot edu dot pl>
* Copyright (C) 2016 Institute of Computer Science Polish Academy of Sciences
*
* This program is free software: you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation, either version 3 of the License, or
* (at your option) any later version.
*
* This program 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 General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program. If not, see <http://www.gnu.org/licenses/>.
*)
open WalTypes
open Xstd
let expands,compreps,comprep_reqs,subtypes,equivs = WalParser.load_realizations ()
(*let verb_frames = WalParser.load_frames (Paths.walenty_path ^ Paths.verb_filename)
let noun_frames = WalParser.load_frames (Paths.walenty_path ^ Paths.noun_filename)
let adj_frames = WalParser.load_frames (Paths.walenty_path ^ Paths.adj_filename)
let adv_frames = WalParser.load_frames (Paths.walenty_path ^ Paths.adv_filename) *)
let walenty = (*StringMap.empty*)WalTEI.load_walenty2 ()
(*let _ = StringMap.iter walenty (fun pos map ->
StringMap.iter map (fun lexeme frames ->
Printf.printf "%s %s %d\n%!" pos lexeme (Xlist.size frames)))*)
(*let all_frames =
["subst",noun_frames;
"adj",adj_frames;
"adv",adv_frames;
"ger",verb_frames;
"pact",verb_frames;
"ppas",verb_frames;
"fin",verb_frames;
"praet",verb_frames;
"impt",verb_frames;
"imps",verb_frames;
"inf",verb_frames;
"pcon",verb_frames]*)
let rec get_role_and_sense = function
Phrase(Lex "się") -> "Theme","", []
| PhraseAbbr(Xp "abl",_) -> "Location","Source", []
| PhraseAbbr(Xp "adl",_) -> "Location","Goal", []
| PhraseAbbr(Xp "caus",_) -> "Condition","", []
| PhraseAbbr(Xp "dest",_) -> "Purpose","", []
| PhraseAbbr(Xp "dur",_) -> "Duration","", []
| PhraseAbbr(Xp "instr",_) -> "Instrument","", []
| PhraseAbbr(Xp "locat",_) -> "Location","", []
| PhraseAbbr(Xp "mod",_) -> "Manner","", []
| PhraseAbbr(Xp "perl",_) -> "Path","", []
| PhraseAbbr(Xp "temp",_) -> "Time","", []
| PhraseAbbr(Advp "abl",_) -> "Location","Source", []
| PhraseAbbr(Advp "adl",_) -> "Location","Goal", []
| PhraseAbbr(Advp "dur",_) -> "Duration","", []
| PhraseAbbr(Advp "locat",_) -> "Location","", []
| PhraseAbbr(Advp "mod",_) -> "Manner","", []
| PhraseAbbr(Advp "perl",_) -> "Path","", []
| PhraseAbbr(Advp "temp",_) -> "Time","", []
(* | PhraseAbbr(Advp "pron",_) -> "Arg","", []
| PhraseAbbr(Advp "misc",_) -> "Arg","", []*)
| PhraseAbbr(Distrp,_) -> "Distributive","", [] (* FIXME: to jest kwantyfikator *)
| PhraseAbbr(Possp,_) -> "Possesive","", []
| LexPhraseMode("abl",_,_) -> "Location","Source", []
| LexPhraseMode("adl",_,_) -> "Location","Goal", []
| LexPhraseMode("caus",_,_) -> "Condition","", []
| LexPhraseMode("dest",_,_) -> "Purpose","", []
| LexPhraseMode("dur",_,_) -> "Duration","", []
| LexPhraseMode("instr",_,_) -> "Instrument","", []
| LexPhraseMode("locat",_,_) -> "Location","", []
| LexPhraseMode("mod",_,_) -> "Manner","", []
| LexPhraseMode("perl",_,_) -> "Path","", []
| LexPhraseMode("temp",_,_) -> "Time","", []
| _ -> "Arg","", []
(*let rec get_gf_role = function
[],Phrase(NP case) -> "C", "", ["T"]
| [],Phrase(AdjP case) -> "R", "", ["T"]
| [],Phrase(NumP(case,_)) -> "C", "", ["T"]
| [],Phrase(PrepNP _) -> "C", "", ["T"]
| [],Phrase(PrepAdjP _) -> "C", "", ["T"]
| [],Phrase(PrepNumP _) -> "C", "", ["T"]
| [],Phrase(ComprepNP _) -> "C", "", ["T"]
| [],Phrase(ComparP _) -> "C", "", ["T"]
| [],Phrase(CP _) -> "C", "", ["T"]
| [],Phrase(NCP(case,_,_)) -> "C", "", ["T"]
| [],Phrase(PrepNCP _) -> "C", "", ["T"]
| [],Phrase(InfP _) -> "C", "", ["T"]
| [],Phrase(FixedP _) -> "C", "", ["T"]
| [],Phrase Or -> "C", "", ["T"] (* FIXME: zbadać w walentym faktyczne użycia or, bo to nie tylko zdania, ale też np(nom) w cudzysłowach *)
| [],Phrase(Lex "się") -> "C", "Ptnt", ["T"]
| [],PhraseAbbr(Xp mode,_) -> "C", mode, ["T"]
| [],PhraseAbbr(Advp "pron",_) -> "R", "", ["T"]
| [],PhraseAbbr(Advp "misc",_) -> "R", "", ["T"]
| [],PhraseAbbr(Advp mode,_) -> "C", mode, ["T"]
| [],PhraseAbbr(Nonch,_) -> "C", "", ["T"]
| [],PhraseAbbr(Distrp,_) -> "C", "Distr", ["T"]
| [],PhraseAbbr(Possp,_) -> "C", "Poss", ["T"]
| [],LexPhraseMode(mode,_,_) -> "C", mode, ["T"]
| [],LexPhrase((SUBST(_,case),_) :: _,_) -> "C", "", ["T"]
| [],LexPhrase((PREP _,_) :: _,_) -> "C", "", ["T"]
| [],LexPhrase((NUM(case,_,_),_) :: _,_) -> "C", "", ["T"]
| [],LexPhrase((ADJ(_,case,_,_),_) :: _,_) -> "C", "", ["T"]
| [],LexPhrase((ADV _,_) :: _,_) -> "C", "", ["T"]
| [],LexPhrase((GER(_,case,_,_,_,_),_) :: _,_) -> "C", "", ["T"]
| [],LexPhrase((PACT(_,case,_,_,_,_),_) :: _,_) -> "C", "", ["T"]
| [],LexPhrase((PPAS(_,case,_,_,_),_) :: _,_) -> "C", "", ["T"]
| [],LexPhrase((INF _,_) :: _,_) -> "C", "", ["T"]
| [],LexPhrase((QUB,_) :: _,_) -> "C", "", ["T"]
| [],LexPhrase((COMPAR,_) :: _,_) -> "C", "", ["T"]
| [],LexPhrase((COMP _,_) :: _,_) -> "C", "", ["T"]
| [],morf -> print_endline(*failwith*) ("get_gf: []," ^ WalStringOf.morf morf);"","",[]
| _,Phrase(InfP _) -> "X", "", ["T"]
| _,Phrase(CP _) -> "X", "", ["T"] (* zwykle możliwa koordynacja z infp *)
| _,Phrase _ -> "X", "", ["T"]
| _,PhraseAbbr _ -> "X", "", ["T"]
| _,LexPhraseMode _ -> "X", "", ["T"]
| _,LexPhrase((INF _,_) :: _,_) -> "X", "", ["T"]
| _,LexPhrase _ -> "X", "", ["T"]
| _,morf -> failwith ("get_gf: _," ^ WalStringOf.morf morf)*)
(*let gf_rank = Xlist.fold [
"",1;
] StringMap.empty (fun gf_rank (gf,v) -> StringMap.add gf_rank gf v)*)
(*let agregate_gfs s gfs_roles =
(* fst (Xlist.fold gfs ("",0) (fun (best_gf,best_rank) gf ->
let rank = try StringMap.find gf_rank gf with Not_found -> failwith ("agregate_gfs: " ^ gf) in
if rank > best_rank then gf, rank else best_gf, best_rank))*)
(* let gfs,roles = List.split gfs_roles in
let gfs = StringSet.to_list (Xlist.fold gfs StringSet.empty StringSet.add) in
if Xlist.size gfs > 1 then print_endline ("agregate_gfs: " ^ String.concat " " gfs);
if Xlist.size roles > 1 then print_endline ("agregate_gfs: " ^ String.concat " " roles);*)
let gf,role,prefs = List.hd gfs_roles in
{s with gf=gf; role=role; prefs=prefs}
let rec make_gfs schema =
let schema = Xlist.map schema (function
{gf="subj"} as s -> {s with gf="SUBJ"; role="Agnt"; prefs=["T"]; morfs=make_gfs_morfs s.morfs}
| {gf="obj"} as s -> {s with gf="OBJ"; role="Ptnt"; prefs=["T"]; morfs=make_gfs_morfs s.morfs}
| {gf=""} as s -> agregate_gfs {s with morfs=make_gfs_morfs s.morfs} (Xlist.map s.morfs (fun morf -> get_gf_role (s.ce,morf)))
| {gf=t} -> failwith ("make_gfs: " ^ t)) in
(* let schema = List.rev (fst (Xlist.fold schema ([],StringMap.empty) (fun (schema,map) s ->
try
let n = StringMap.find map s.gf in
{s with gf=s.gf ^ string_of_int (n+1)} :: schema,
StringMap.add map s.gf (n+1)
with Not_found ->
s :: schema, StringMap.add map s.gf 1))) in*)
schema
and make_gfs_morfs morfs =
List.flatten (Xlist.map morfs (function
Phrase _ as morf -> [morf]
| PhraseAbbr(Advp _,[]) -> [Phrase AdvP]
| PhraseAbbr(_,[]) -> failwith "make_gfs_morfs"
| PhraseAbbr(_,morfs) -> make_gfs_morfs morfs
| LexPhrase(pos_lex,(restr,schema)) -> [LexPhrase(pos_lex,(restr,make_gfs schema))]
| LexPhraseMode(_,pos_lex,(restr,schema)) -> [LexPhrase(pos_lex,(restr,make_gfs schema))]
| _ -> failwith "make_gfs_morfs"))*)
let mark_nosem_morfs morfs =
Xlist.map morfs (function
Phrase(PrepNP(_,prep,c)) -> Phrase(PrepNP(NoSem,prep,c))
| Phrase(PrepAdjP(_,prep,c)) -> Phrase(PrepAdjP(NoSem,prep,c))
| Phrase(PrepNumP(_,prep,c)) -> Phrase(PrepNumP(NoSem,prep,c))
(* | Phrase(ComprepNP(_,prep)) -> Phrase(ComprepNP(NoSem,prep)) *) (* FIXME: na razie ComprepNP są zawsze semantyczne *)
(* | Phrase(ComparNP(_,prep,c)) -> Phrase(ComparNP(NoSem,prep,c)) (* FIXME: pomijam niesemantyczny compar *)
| Phrase(ComparPP(_,prep)) -> Phrase(ComparPP(NoSem,prep))*)
| Phrase(PrepNCP(_,prep,c,ct,co)) -> Phrase(PrepNCP(NoSem,prep,c,ct,co))
| t -> t)
let agregate_role_and_sense s l =
let roles,senses = Xlist.fold l (StringSet.empty,StringSet.empty) (fun (roles,senses) (role,role_attr,sense) ->
StringSet.add roles (role ^ " " ^ role_attr),
Xlist.fold sense senses StringSet.add) in
let roles = if StringSet.size roles = 1 then roles else StringSet.remove roles "Arg " in
let role,role_attr =
match Str.split (Str.regexp " ") (StringSet.min_elt roles) with
[r;a] -> r,a
| [r] -> r,""
| _ -> failwith "agregate_role_and_sense" in
{s with role=role; role_attr=role_attr(*; sel_prefs=StringSet.to_list senses*)}
let rec assign_role_and_sense schema =
Xlist.map schema (function
{gf=SUBJ} as s ->
if s.role = "" then {s with role="Initiator"; sel_prefs=["ALL"]; morfs=assign_role_and_sense_morfs s.morfs}
else {s with morfs=assign_role_and_sense_morfs (mark_nosem_morfs s.morfs)}
| {gf=OBJ} as s ->
if s.role = "" then {s with role="Theme"; sel_prefs=["ALL"]; morfs=assign_role_and_sense_morfs s.morfs}
else {s with morfs=assign_role_and_sense_morfs (mark_nosem_morfs s.morfs)}
| {gf=ARG} as s ->
if s.role = "" then agregate_role_and_sense {s with sel_prefs=["ALL"]; morfs=assign_role_and_sense_morfs s.morfs}
(Xlist.map s.morfs (fun morf -> get_role_and_sense morf))
else {s with morfs=assign_role_and_sense_morfs (mark_nosem_morfs s.morfs)}
| _ -> failwith "assign_role_and_sense")
and assign_role_and_sense_morfs morfs =
List.flatten (Xlist.map morfs (function
Phrase _ as morf -> [morf]
| E _ as morf -> [morf]
| PhraseAbbr(Advp _,[]) -> [Phrase AdvP]
| PhraseAbbr(_,[]) -> failwith "assign_role_and_sense_morfs"
| PhraseAbbr(_,morfs) -> assign_role_and_sense_morfs morfs
| LexPhrase(pos_lex,(restr,schema)) -> [LexPhrase(pos_lex,(restr,assign_role_and_sense schema))]
| LexPhraseMode(_,pos_lex,(restr,schema)) -> [LexPhrase(pos_lex,(restr,assign_role_and_sense schema))]
| _ -> failwith "assign_role_and_sense_morfs"))
let rec assign_pro_args schema =
Xlist.map schema (fun s ->
let morfs = match s.morfs with
(E p) :: l -> E Pro :: (E p) :: l
| [LexPhrase _] as morfs -> morfs
| [Phrase(FixedP _)] as morfs -> morfs
| [Phrase(Lex _)] as morfs -> morfs
(* | [Phrase Refl] as morfs -> morfs
| [Phrase Recip] as morfs -> morfs*)
| Phrase Null :: _ as morfs -> morfs
| Phrase Pro :: _ as morfs -> morfs
| morfs -> if s.gf <> SUBJ && s.cr = [] && s.ce = [] then (Phrase Null) :: morfs else (Phrase Pro) :: morfs in (* FIXME: ustalić czy są inne przypadki uzgodnienia *)
(* let morfs = assign_pro_args_lex morfs in *) (* bez pro wewnątrz leksykalizacji *)
{s with morfs=morfs})
(*let assign_pro_args_lex morfs =
Xlist.map morfs (function
Lex(morf,specs,lex,restr) -> LexN(morf,specs,lex,assign_pro_args_restr restr)
| LexNum(morf,lex1,lex2,restr) -> LexNum(morf,lex1,lex2,assign_pro_args_restr restr)
| LexCompar(morf,l) -> LexCompar(morf,make_gfs_lex l)
| morf -> morf)
and assign_pro_args_restr = function
Natr -> Natr
| Ratr1 schema -> Ratr1(assign_pro_args schema)
| Atr1 schema -> Atr1(assign_pro_args schema)
| Ratr schema -> Ratr(assign_pro_args schema)
| Atr schema -> Atr(assign_pro_args schema)*)
(*let _ =
Xlist.iter walenty_filenames (fun filename ->
print_endline filename;
let frames = load_frames (walenty_path ^ filename) in
StringMap.iter frames (fun _ l ->
Xlist.iter l (fun (refl,opinion,negation,pred,aspect,schema) ->
ignore (process_opinion opinion);
ignore (process_negation [Text negation]);
ignore (process_pred [Text pred]);
ignore (process_aspect [Text aspect]);
ignore (assign_pro_args (make_gfs (process_schema expands subtypes equivs schema))))))*)
exception ImpossibleSchema
let rec reduce_comp lexemes = function
Comp s -> if StringMap.mem lexemes s then Comp s else raise Not_found
| Zeby -> if StringMap.mem lexemes "żeby" || StringMap.mem lexemes "że" then Zeby else raise Not_found
| Gdy -> if StringMap.mem lexemes "gdy" || StringMap.mem lexemes "gdyby" then Gdy else raise Not_found
| CompUndef -> failwith "reduce_comp"
let reduce_phrase lexemes = function
| PrepNP(_,prep,case) as phrase -> if StringMap.mem lexemes prep then phrase else raise Not_found
| PrepAdjP(_,prep,case) as phrase -> if StringMap.mem lexemes prep then phrase else raise Not_found
| PrepNumP(_,prep,case) as phrase -> if StringMap.mem lexemes prep then phrase else raise Not_found
| ComprepNP(_,prep) as phrase -> if Xlist.fold (try StringMap.find comprep_reqs prep with Not_found -> []) true (fun b s -> b && StringMap.mem lexemes s) then phrase else raise Not_found
| ComparNP(_,prep,case) as phrase -> if StringMap.mem lexemes prep then phrase else raise Not_found
| ComparPP(_,prep) as phrase -> if StringMap.mem lexemes prep then phrase else raise Not_found
| CP(ctype,comp) -> CP(ctype,reduce_comp lexemes comp)
| NCP(case,ctype,comp) -> if StringMap.mem lexemes "to" then NCP(case,ctype,reduce_comp lexemes comp) else raise Not_found
| PrepNCP(sem,prep,case,ctype,comp) -> if StringMap.mem lexemes prep && StringMap.mem lexemes "to" then PrepNCP(sem,prep,case,ctype,reduce_comp lexemes comp) else raise Not_found
| phrase -> phrase
let rec reduce_lex lexemes = function
Lexeme s -> if StringMap.mem lexemes s then Lexeme s else raise Not_found
| ORconcat l ->
let l = List.rev (Xlist.fold l [] (fun l lex -> try reduce_lex lexemes lex :: l with Not_found -> l)) in
(match l with
[] -> raise Not_found
| [x] -> x
| l -> ORconcat l)
| ORcoord l ->
let l = List.rev (Xlist.fold l [] (fun l lex -> try reduce_lex lexemes lex :: l with Not_found -> l)) in
(match l with
[] -> raise Not_found
| [x] -> x
| l -> ORcoord l)
| XOR l ->
let l = List.rev (Xlist.fold l [] (fun l lex -> try reduce_lex lexemes lex :: l with Not_found -> l)) in
(match l with
[] -> raise Not_found
| [x] -> x
| l -> XOR l)
| Elexeme gender -> Elexeme gender
let rec reduce_restr lexemes = function (* leksykalizacje wewnątrz leksykalizacji są w niektórych sytuacjach opcjonalne *)
Natr,[] -> Natr,[]
| Atr,[] -> Atr,[]
| Ratr,[] -> Ratr,[]
| Atr1,[] -> Atr1,[]
| Ratr1,[] -> Ratr1,[]
| Ratr1,schema -> let schema = reduce_schema2 lexemes schema in if schema = [] then raise Not_found else Ratr1,schema
| Atr1,schema -> let schema = reduce_schema2 lexemes schema in if schema = [] then Natr,[] else Atr1,schema
| Ratr,schema -> let schema = reduce_schema2 lexemes schema in if schema = [] then raise Not_found else Ratr,schema
| Atr,schema -> let schema = reduce_schema2 lexemes schema in if schema = [] then Natr,[] else Atr,schema
| Ratrs,schema -> Ratrs,reduce_schema lexemes schema
| _ -> failwith "reduce_restr"
and reduce_morf lexemes = function (* leksykalizacje, które się z czymś koordynują nie są obowiązakowe *)
Phrase phrase -> Phrase(reduce_phrase lexemes phrase)
| E phrases -> E phrases (* FIXME: uproszczenie *)
| LexPhrase(pos_lex,restr) -> LexPhrase(Xlist.map pos_lex (fun (pos,lex) -> pos, reduce_lex lexemes lex),reduce_restr lexemes restr)
| morf -> failwith ("reduce_morf: " ^ WalStringOf.morf morf)
and reduce_morfs lexemes = function
[] -> []
| morf :: l -> (try [reduce_morf lexemes morf] with Not_found -> []) @ reduce_morfs lexemes l
and reduce_schema2 lexemes = function
[] -> []
| s :: l ->
let morfs = reduce_morfs lexemes s.morfs in
if morfs = [] then reduce_schema2 lexemes l else
{s with morfs=morfs} :: reduce_schema2 lexemes l
and reduce_schema lexemes = function
[] -> []
| s :: l ->
let morfs = reduce_morfs lexemes s.morfs in
if morfs = [] then raise ImpossibleSchema else
{s with morfs=morfs} :: reduce_schema lexemes l
let reduce_schema_frame lexemes = function
Frame(atrs,schema) -> Frame(atrs,reduce_schema lexemes schema)
(* | ComprepFrame(s,morfs) ->
let morfs = reduce_morfs lexemes morfs in
if morfs = [] then raise ImpossibleSchema else ComprepFrame(s,morfs)*)
| _ -> failwith "reduce_schema_frame"
let remove_trivial_args schema =
Xlist.fold schema [] (fun l (_,_,_,morfs) ->
let morfs = Xlist.fold morfs [] (fun morfs -> function
Phrase(AdjP _) -> morfs
| Phrase(NP(Case "gen")) -> morfs
| Phrase(NCP(Case "gen",_,_)) -> morfs
| Phrase(PrepNP _) -> morfs
| Phrase(FixedP _) -> morfs
| LexPhrase([ADJ _,_],_) -> morfs
| LexPhrase([PPAS _,_],_) -> morfs
| LexPhrase([PACT _,_],_) -> morfs
| LexPhrase([SUBST(_,Case "gen"),_],_) -> morfs
| LexPhrase([PREP _,_;_],_) -> morfs
| morf -> morf :: morfs) in
if morfs = [] then l else morfs :: l)
(* leksykalizacje do zmiany struktury
lex([PREP(gen),'z';SUBST(sg,gen),'nazwa'],atr1[OBL{lex([QUB,'tylko'],natr[])}])
lex([PREP(loc),'na';SUBST(sg,loc),'papier'],atr1[OBL{lex([QUB,'tylko'],natr[])}])
lex([PREP(acc),'w';SUBST(pl,acc),'oko'],atr1[OBL-MOD{lex([ADV(pos),'prosto'],natr[])}])
lex([PREP(gen),'z';SUBST(sg,gen),'most'],ratr1[OBL-MOD{lex([ADV(pos),'prosto'],natr[])}])
lex([PREP(acc),'w';SUBST(pl,acc),'oko'],atr1[OBL-MOD{lex([ADV(pos),'prosto'],natr[])}])
lex([PREP(gen),'z';SUBST(sg,gen),'most'],ratr1[OBL-MOD{lex([ADV(pos),'prosto'],natr[])}])
lex([PREP(acc),'w';SUBST(pl,acc),'oko'],atr1[OBL-MOD{lex([ADV(pos),'prosto'],natr[])}])
lex([PREP(acc),'w';SUBST(pl,acc),'oko'],atr1[OBL-MOD{lex([ADV(pos),'prosto'],natr[])}])
lex([PREP(acc),'w';SUBST(pl,acc),'oko'],atr1[OBL-MOD{lex([ADV(pos),'prosto'],natr[])}])
lex([PREP(acc),'w';SUBST(pl,acc),'oko'],atr1[OBL-MOD{lex([ADV(pos),'prosto'],natr[])}])
lex([PREP(acc),'w';SUBST(pl,acc),'oko'],atr1[OBL-MOD{lex([ADV(pos),'prosto'],natr[])}])
*)
let rec split_elexeme = function
Lexeme s -> [],[Lexeme s]
| XOR l ->
let genders,l = Xlist.fold l ([],[]) (fun (genders,lexs) lex ->
let gender,lex = split_elexeme lex in
gender @ genders, lex @ lexs) in
genders,[XOR(List.rev l)]
| ORconcat l ->
let genders,l = Xlist.fold l ([],[]) (fun (genders,lexs) lex ->
let gender,lex = split_elexeme lex in
gender @ genders, lex @ lexs) in
genders,[ORconcat(List.rev l)]
| ORcoord l ->
let genders,l = Xlist.fold l ([],[]) (fun (genders,lexs) lex ->
let gender,lex = split_elexeme lex in
gender @ genders, lex @ lexs) in
genders,[ORcoord(List.rev l)]
| Elexeme gender -> [gender],[]
let prep_arg_schema_field morfs =
{gf=CORE; role="Ref"; role_attr=""; sel_prefs=["ALL"]; cr=[]; ce=[]; dir=Forward; morfs=morfs} (* FIXME: uporządkować sensy *)
let prep_arg_schema_field2 morfs =
{gf=CORE; role="Ref"; role_attr=""; sel_prefs=["ALL"]; cr=[]; ce=[]; dir=Forward; morfs=morfs} (* FIXME: uporządkować sensy *)
let num_arg_schema_field morfs =
{gf=CORE; role="QUANT-ARG"; role_attr=""; sel_prefs=["ALL"]; cr=[]; ce=[]; dir=Forward; morfs=morfs}
let std_arg_schema_field dir morfs =
{gf=ARG; role="Arg"; role_attr=""; sel_prefs=["ALL"]; cr=[]; ce=[]; dir=dir; morfs=morfs}
let simple_arg_schema_field morfs =
{gf=ARG; role=""; role_attr=""; sel_prefs=["ALL"]; cr=[]; ce=[]; dir=Both; morfs=morfs}
let nosem_refl_schema_field =
{gf=NOSEM; role=""; role_attr=""; sel_prefs=["ALL"]; cr=[]; ce=[]; dir=Both; morfs=[Phrase(Lex "się")]}
let rec expand_lexicalizations_schema schema =
Xlist.map schema (fun s ->
{s with morfs=expand_lexicalizations_morfs s.morfs})
and expand_lexicalizations_morfs morfs = (* uproszczenie polegające na zezwoleniu na koordynację przy zwiększaniu ilości LexPhrase *)
List.flatten (Xlist.map morfs (fun morf ->
let morf = match morf with
LexPhrase(pos_lex,(restr,schema)) -> LexPhrase(pos_lex,(restr,expand_lexicalizations_schema schema))
| morf -> morf in
match morf with
(* LexPhrase([ADV _,_],(_,_::_)) -> print_endline (WalStringOf.morf morf); [morf] *)
(* | LexPhrase([PREP _,_;SUBST _,_],(_,schema)) -> if remove_trivial_args schema <> [] then print_endline (WalStringOf.morf morf); [morf] *)
(* | LexPhrase([PREP _,_;GER _,_],(_,schema)) -> if remove_trivial_args schema <> [] then print_endline (WalStringOf.morf morf); [morf] *)
(* | LexPhrase([NUM _,_;_],(_,schema)) -> if remove_trivial_args schema <> [] then print_endline (WalStringOf.morf morf); [morf] *)
(* | LexPhrase([PREP _,_;NUM _,_;_],(_,schema)) -> if remove_trivial_args schema <> [] then print_endline (WalStringOf.morf morf); [morf] *)
(* | LexPhrase([PREP _,_;ADJ _,_],(_,_::_)) -> print_endline (WalStringOf.morf morf); [morf]
| LexPhrase([PREP _,_;PPAS _,_],(_,_::_)) -> print_endline (WalStringOf.morf morf); [morf]
| LexPhrase([PREP _,_;PACT _,_],(_,_::_)) -> print_endline (WalStringOf.morf morf); [morf] *)
| Phrase(PrepNumP(_,prep,case)) -> [LexPhrase([PREP case,Lexeme prep],(Ratrs,[prep_arg_schema_field2 [Phrase(NumP(case))]]))]
| LexPhrase([PREP pcase,plex;SUBST(n,c),slex],(Atr1,[{morfs=[LexPhrase([QUB,_],_)]} as s])) ->
(* print_endline (WalStringOf.morf morf); *)
[LexPhrase([PREP pcase,plex],(Ratrs,[prep_arg_schema_field [LexPhrase([SUBST(n,c),slex],(Natr,[]))]]));
LexPhrase([PREP pcase,plex],(Ratrs,[prep_arg_schema_field [LexPhrase([SUBST(n,c),slex],(Natr,[]))];{s with dir=Backward}]))]
| LexPhrase([PREP(pcase),plex;SUBST(n,c),slex],(Atr1,[{morfs=[LexPhrase([ADV _,_],_)]} as s])) ->
(* print_endline (WalStringOf.morf morf); *)
[LexPhrase([PREP pcase,plex],(Ratrs,[prep_arg_schema_field [LexPhrase([SUBST(n,c),slex],(Natr,[]))]]));
LexPhrase([PREP pcase,plex],(Ratrs,[prep_arg_schema_field [LexPhrase([SUBST(n,c),slex],(Natr,[]))];{s with dir=Backward}]))]
| LexPhrase([PREP pcase,plex;SUBST(n,c),slex],(Ratr1,[{morfs=[LexPhrase([ADV _,_],_)]} as s])) ->
(* print_endline (WalStringOf.morf morf); *)
[LexPhrase([PREP pcase,plex],(Ratrs,[prep_arg_schema_field [LexPhrase([SUBST(n,c),slex],(Natr,[]))];{s with dir=Backward}]))]
| LexPhrase([PREP pcase,plex;pos,lex],restr) ->
[LexPhrase([PREP pcase,plex],(Ratrs,[prep_arg_schema_field [LexPhrase([pos,lex],restr)]]))]
| LexPhrase([PREP pcase,plex;NUM(c,g,a),nlex;pos,lex],restr) ->
let genders,lexs = split_elexeme lex in
Xlist.map genders (fun gender ->
LexPhrase([PREP pcase,plex],(Ratrs,[prep_arg_schema_field [LexPhrase([NUM(c,gender,a),nlex],(Ratrs,[num_arg_schema_field [Phrase Pro]]))]]))) @
Xlist.map lexs (fun lex ->
LexPhrase([PREP pcase,plex],(Ratrs,[prep_arg_schema_field [LexPhrase([NUM(c,g,a),nlex],(Ratrs,[num_arg_schema_field [LexPhrase([pos,lex],restr)]]))]])))
| LexPhrase([NUM(c,g,a),nlex;pos,lex],restr) ->
let genders,lexs = split_elexeme lex in
Xlist.map genders (fun gender ->
LexPhrase([NUM(c,gender,a),nlex],(Ratrs,[num_arg_schema_field [Phrase Pro]]))) @
Xlist.map lexs (fun lex ->
LexPhrase([NUM(c,g,a),nlex],(Ratrs,[num_arg_schema_field [LexPhrase([pos,lex],restr)]])))
| LexPhrase([COMP ctype,clex;pos,lex],restr) ->
[LexPhrase([COMP ctype,clex],(Ratrs,[std_arg_schema_field Forward [LexPhrase([pos,lex],restr)]]))]
| LexPhrase([SUBST(n,c),slex;COMP ctype,clex;pos,lex],restr) ->
[LexPhrase([SUBST(n,c),slex],(Ratrs,[std_arg_schema_field Forward [LexPhrase([COMP ctype,clex],(Ratrs,[std_arg_schema_field Forward [LexPhrase([pos,lex],restr)]]))]]))] (* FIXME: poprawić po zrobieniu NCP *)
| LexPhrase(_::_::_,_) -> failwith ("expand_lexicalizations_morfs: " ^ WalStringOf.morf morf)
(* | LexPhrase([PREP pcase,plex;SUBST(n,c),slex],(Atr1,[gf,cr,ce,[LexPhrase([QUB,lex],arestr)]])) ->
(* print_endline (WalStringOf.morf morf); *)
[LexPhrase([PREP pcase,plex],(Ratrs,[("OBJ","Ref",["T"]),[],[],[LexPhrase([SUBST(n,c),slex],(Natr,[]))]]));
LexPhrase([PREP pcase,plex],(Ratrs,[("OBJ","Ref",["T"]),[],[],[LexPhrase([SUBST(n,c),slex],(Natr,[]))];gf,cr,ce,[LexPhrase([QUB,lex],arestr)]]))]
| LexPhrase([PREP(pcase),plex;SUBST(n,c),slex],(Atr1,[gf,cr,ce,[LexPhrase([ADV gr,lex],arestr)]])) ->
(* print_endline (WalStringOf.morf morf); *)
[LexPhrase([PREP pcase,plex],(Ratrs,[("OBJ","Ref",["T"]),[],[],[LexPhrase([SUBST(n,c),slex],(Natr,[]))]]));
LexPhrase([PREP pcase,plex],(Ratrs,[("OBJ","Ref",["T"]),[],[],[LexPhrase([SUBST(n,c),slex],(Natr,[]))];gf,cr,ce,[LexPhrase([ADV gr,lex],arestr)]]))]
| LexPhrase([PREP pcase,plex;SUBST(n,c),slex],(Ratr1,[gf,cr,ce,[LexPhrase([ADV gr,lex],arestr)]])) ->
(* print_endline (WalStringOf.morf morf); *)
[LexPhrase([PREP pcase,plex],(Ratrs,[("OBJ","Ref",["T"]),[],[],[LexPhrase([SUBST(n,c),slex],(Natr,[]))];gf,cr,ce,[LexPhrase([ADV gr,lex],arestr)]]))]
| LexPhrase([PREP pcase,plex;pos,lex],restr) ->
[LexPhrase([PREP pcase,plex],(Ratrs,[("OBJ","Ref",["T"]),[],[],[LexPhrase([pos,lex],restr)]]))]
| LexPhrase([PREP pcase,plex;NUM(c,g,a),nlex;pos,lex],restr) ->
let genders,lexs = split_elexeme lex in
Xlist.map genders (fun gender ->
LexPhrase([PREP pcase,plex],(Ratrs,[("OBJ","Ref",["T"]),[],[],[LexPhrase([NUM(c,gender,a),nlex],(Ratrs,[("OBJ","QUANT-ARG",["T"]),[],[],[Phrase Pro]]))]]))) @
Xlist.map lexs (fun lex ->
LexPhrase([PREP pcase,plex],(Ratrs,[("OBJ","Ref",["T"]),[],[],[LexPhrase([NUM(c,g,a),nlex],(Ratrs,[("OBJ","QUANT-ARG",["T"]),[],[],[LexPhrase([pos,lex],restr)]]))]])))
| LexPhrase([NUM(c,g,a),nlex;pos,lex],restr) ->
let genders,lexs = split_elexeme lex in
Xlist.map genders (fun gender ->
LexPhrase([NUM(c,gender,a),nlex],(Ratrs,[("OBJ","QUANT-ARG",["T"]),[],[],[Phrase Pro]]))) @
Xlist.map lexs (fun lex ->
LexPhrase([NUM(c,g,a),nlex],(Ratrs,[("OBJ","QUANT-ARG",["T"]),[],[],[LexPhrase([pos,lex],restr)]])))
| LexPhrase([COMP ctype,clex;pos,lex],restr) ->
[LexPhrase([COMP ctype,clex],(Ratrs,[("C","",["T"]),[],[],[LexPhrase([pos,lex],restr)]]))]
| LexPhrase([SUBST(n,c),slex;COMP ctype,clex;pos,lex],restr) ->
[LexPhrase([SUBST(n,c),slex],(Ratrs,[("OBJ","",["T"]),[],[],[LexPhrase([COMP ctype,clex],(Ratrs,[("C","",["T"]),[],[],[LexPhrase([pos,lex],restr)]]))]]))]
| LexPhrase(_::_::_,_) -> failwith ("expand_lexicalizations_morfs: " ^ WalStringOf.morf morf)*)
| morf -> [morf]))
let expand_lexicalizations = function
Frame(atrs,schema) -> Frame(atrs,expand_lexicalizations_schema schema)
(* ComprepFrame(s,morfs) -> ComprepFrame(atrs,expand_lexicalizations_morfs morfs) *)
| _ -> failwith "expand_lexicalizations"
let lex_id_counter = ref 0
let get_lex_id () =
incr lex_id_counter;
string_of_int (!lex_id_counter)
let get_pos lex = function
SUBST _ ->
(match lex with
"ja" -> ["ppron12"]
| "my" -> ["ppron12"]
| "ty" -> ["ppron12"]
| "wy" -> ["ppron12"]
| "on" -> ["ppron3"]
| "siebie" -> ["siebie"]
| "się" -> ["qub"]
| _ -> ["subst"])
| PREP _ -> ["prep"]
| NUM _ -> ["num"]
| ADV _ -> ["adv"]
| ADJ _ -> ["adj"]
| GER _ -> ["ger"]
| PPAS _ -> ["ppas"]
| PACT _ -> ["pact"]
| PERS _ -> ["fin";"praet";"winien"(*;"impt";"imps"*);"pred"]
| INF _ -> ["inf"]
| QUB -> ["qub"]
| COMPAR -> ["compar"]
| COMP _ -> ["comp"]
let rec extract_lex_frames lexeme p frames = function
Frame(atrs,schema) ->
let schema,frames = Xlist.fold schema ([],frames) (fun (schema,frames) s ->
let morfs,frames = Xlist.fold s.morfs ([],frames) extract_lex_morf in
{s with morfs=List.rev morfs} :: schema, frames) in
(lexeme,p,Frame(atrs,List.rev schema)) :: frames
| LexFrame(id,pos,restr,schema) ->
let schema,frames = Xlist.fold schema ([],frames) (fun (schema,frames) s ->
let morfs,frames = Xlist.fold s.morfs ([],frames) extract_lex_morf in
{s with morfs=List.rev morfs} :: schema, frames) in
(lexeme,p,LexFrame(id,pos,restr,List.rev schema)) :: frames
| ComprepFrame(s,pos,restr,schema) ->
let schema,frames = Xlist.fold schema ([],frames) (fun (schema,frames) s ->
let morfs,frames = Xlist.fold s.morfs ([],frames) extract_lex_morf in
{s with morfs=List.rev morfs} :: schema, frames) in
(lexeme,p,ComprepFrame(s,pos,restr,List.rev schema)) :: frames
(* | _ -> failwith "extract_lex_frames" *)
and extract_lex_morf (morfs,frames) = function
LexPhrase([pos,lex],(restr,schema)) ->
let id = get_lex_id () in
let lexemes = WalParser.get_lexemes lex in
let frames = Xlist.fold lexemes frames (fun frames lexeme ->
let poss = get_pos lexeme pos in
Xlist.fold poss frames (fun frames p ->
extract_lex_frames lexeme p frames (LexFrame(id,pos,restr,schema)))) in
LexPhraseId(id,pos,lex) :: morfs, frames
| LexPhrase _ -> failwith "extract_lex_morf"
| morf -> morf :: morfs, frames
let split_xor schema =
Xlist.multiply_list (Xlist.map schema (fun s ->
Xlist.map (Xlist.multiply_list (Xlist.map s.morfs (function
LexPhraseId(id,pos,XOR l) -> Xlist.map l (fun lex -> LexPhraseId(id,pos,lex))
| LexPhraseId(id,pos,lex) -> [LexPhraseId(id,pos,lex)]
| morf -> [morf]))) (fun morfs -> {s with morfs=morfs})))
let split_or_coord schema =
Xlist.map schema (fun s ->
{s with morfs=List.flatten (Xlist.map s.morfs (function
LexPhraseId(id,pos,ORcoord l) -> Xlist.map l (fun lex -> LexPhraseId(id,pos,lex))
| LexPhraseId(id,pos,ORconcat l) -> Xlist.map l (fun lex -> LexPhraseId(id,pos,lex)) (* FIXME: koordynacja zamiast konkatenacji *)
| LexPhraseId(id,pos,lex) -> [LexPhraseId(id,pos,lex)]
| morf -> [morf]))})
let simplify_lex schemas =
Xlist.map schemas (fun schema ->
Xlist.map schema (fun s ->
{s with morfs=Xlist.map s.morfs (function
LexPhraseId(id,pos,Lexeme lex) -> LexArg(id,pos,lex)
| LexPhraseId _ as morf -> failwith ("simplify_lex: " ^ WalStringOf.morf morf)
| morf -> morf)}))
let prepare_schema_comprep expands subtypes equivs schema =
assign_pro_args (assign_role_and_sense (WalParser.expand_equivs_schema equivs (WalParser.expand_subtypes subtypes (WalParser.expand_schema expands schema))))
let prepare_schema expands subtypes equivs schema =
prepare_schema_comprep expands subtypes equivs (WalParser.parse_schema schema)
let prepare_schema_sem expands subtypes equivs schema =
prepare_schema_comprep expands subtypes equivs schema
let default_frames = Xlist.fold [ (* FIXME: poprawić domyślne ramki po ustaleniu adjunctów *)
"verb",(ReflEmpty,Domyslny,NegationUndef,PredNA,AspectUndef,"subj{np(str)}+obj{np(str)}"); (* FIXME: dodać ramkę z refl *)
"noun",(ReflEmpty,Domyslny,NegationNA,PredNA,AspectNA,"{possp}+{adjp(agr)}");
"adj",(ReflEmpty,Domyslny,NegationNA,PredNA,AspectNA,"");
"adv",(ReflEmpty,Domyslny,NegationNA,PredNA,AspectNA,"");
"empty",(ReflEmpty,Domyslny,NegationNA,PredNA,AspectNA,"");
"date",(ReflEmpty,Domyslny,NegationNA,PredNA,AspectNA,"{null;lex(np(gen),sg,'rok',natr)}");
"date2",(ReflEmpty,Domyslny,NegationNA,PredNA,AspectNA,"{null;lex(np(gen),sg,'rok',atr1({adjp(agr)}))}"); (* FIXME: wskazać możliwe podrzędniki *)
"day",(ReflEmpty,Domyslny,NegationNA,PredNA,AspectNA,""
(*"{lex(np(gen),sg,XOR('styczeń','luty','marzec','kwiecień','maj','czerwiec','lipiec','sierpień','wrzesień','październik','litopad','grudzień'),atr1({np(gen)}))}"*)); (* FIXME: wskazać możliwe podrzędniki *)
"hour",(ReflEmpty,Domyslny,NegationNA,PredNA,AspectNA,"{null;lex(advp(temp),pos,'rano',natr)}");
] StringMap.empty (fun map (k,(refl,opinion,negation,pred,aspect,schema)) ->
StringMap.add map k (Frame(DefaultAtrs([],refl,opinion,negation,pred,aspect),prepare_schema expands subtypes equivs schema)))
let adjunct_schema_field role dir morfs =
{gf=ADJUNCT; role=role; role_attr=""; sel_prefs=["ALL"]; cr=[]; ce=[]; dir=dir; morfs=morfs}
let verb_prep_adjunct_schema_field lemma case =
{gf=ADJUNCT; role="Manner"; role_attr=""; sel_prefs=["ALL"]; cr=[]; ce=[]; dir=Both; morfs=[
Phrase Null;
Phrase(PrepNP(Sem,lemma,Case case));
Phrase(PrepAdjP(Sem,lemma,Case case));
Phrase(PrepNumP(Sem,lemma,Case case))]}
let verb_comprep_adjunct_schema_field lemma =
{gf=ADJUNCT; role="Manner"; role_attr=""; sel_prefs=["ALL"]; cr=[]; ce=[]; dir=Both; morfs=[
Phrase Null;
Phrase(ComprepNP(Sem,lemma))]}
let verb_compar_adjunct_schema_field lemma =
{gf=ADJUNCT; role="Manner"; role_attr=""; sel_prefs=["ALL"]; cr=[]; ce=[]; dir=Both; morfs=[
Phrase Null;
Phrase(ComparPP(Sem,lemma))] @
Xlist.map ["nom";"gen";"dat";"acc";"inst"] (fun case -> Phrase(ComparNP(Sem,lemma,Case case)))}
let noun_prep_adjunct_schema_field preps compreps =
{gf=ADJUNCT; role="Attribute"; role_attr=""; sel_prefs=["ALL"]; cr=[]; ce=[]; dir=Both; morfs=
let l = Xlist.fold preps [Phrase Null] (fun l (lemma,case) ->
[Phrase(PrepNP(Sem,lemma,Case case));
Phrase(PrepAdjP(Sem,lemma,Case case));
Phrase(PrepNumP(Sem,lemma,Case case))] @ l) in
Xlist.fold compreps l (fun l lemma ->
Phrase(ComprepNP(Sem,lemma)) :: l)}
let noun_compar_adjunct_schema_field compars =
{gf=ADJUNCT; role="Attribute"; role_attr=""; sel_prefs=["ALL"]; cr=[]; ce=[]; dir=Both; morfs=
Xlist.fold compars [Phrase Null] (fun l lemma ->
[Phrase(ComparPP(Sem,lemma))] @ Xlist.map ["nom";"gen";"dat";"acc";"inst"] (fun case -> Phrase(ComparNP(Sem,lemma,Case case))) @ l)}
let adj_compar_adjunct_schema_field compars =
{gf=ADJUNCT; role="Manner"; role_attr=""; sel_prefs=["ALL"]; cr=[]; ce=[]; dir=Both; morfs=
Xlist.fold compars [Phrase Null] (fun l lemma ->
[Phrase(ComparPP(Sem,lemma))] @ Xlist.map ["nom"] (fun case -> Phrase(ComparNP(Sem,lemma,Case case))) @ l)}
(*let nogf_schema_field dir morfs =
{gf=NOGF; role=""; role_attr=""; sel_prefs=[]; cr=[]; ce=[]; dir=dir; morfs=morfs} *)
let schema_field gf role dir morfs =
{gf=gf; role=role; role_attr=""; sel_prefs=["ALL"]; cr=[]; ce=[]; dir=dir; morfs=morfs}
(*let verb_adjuncts = [
adjunct_schema_field "R" "" Both [Phrase AdvP];
adjunct_schema_field "R" "" Both [Phrase PrepP]; (* FIXME: Trzeba będzie uzgodnić PrepNP, PrepAdjP, PrepNumP z PrepP i XP *)
]
let noun_adjuncts = [
adjunct_schema_field "C" "poss" Both [Phrase(NP(Case "gen"))];
adjunct_schema_field "C" "=" Both [Phrase(NP(Case "nom"))];
adjunct_schema_field "C" "=" Both [Phrase(NP(CaseAgr))];
adjunct_schema_field "R" "" Backward [Multi[AdjP AllAgr]];
adjunct_schema_field "R" "" Forward [Multi[AdjP AllAgr]];
adjunct_schema_field "R" "" Both [Phrase PrepP];
]
let adj_adjuncts = [
adjunct_schema_field "R" "" Both [Phrase PrepP];
]*)
let verb_adjuncts = [
(* adjunct_schema_field "" Both [Phrase Null;Phrase AdvP];
adjunct_schema_field "" Both [Phrase Null;Phrase PrepP]; (* FIXME: Trzeba będzie uzgodnić PrepNP, PrepAdjP, PrepNumP z PrepP i XP *)
adjunct_schema_field "Topic" Forward [Phrase Null;Phrase (CP(CompTypeUndef,CompUndef))]; (* poprawić semantykę *) (* FIXME: to powinno być jako ostatnia lista argumentów *)*)
]
(* FIXME: pozycje dublują się z domyślną ramką "noun" *)
let noun_adjuncts = [ (* FIXME: usuniecie noun_adjuncts pozostawia poss dla 'Witoldzie' *)
(* adjunct_schema_field "poss" Both [Phrase Null;Phrase(NP(Case "gen"))];
adjunct_schema_field "=" Both [Phrase Null;Phrase(NP(Case "nom"))];
adjunct_schema_field "=" Both [Phrase Null;Phrase(NP(CaseAgr))];
adjunct_schema_field "" Backward [(*Phrase Null;Phrase(AdjP AllAgr)*)Multi[AdjP AllAgr]]; (* FIXME: za pomocą Multi można zrobić konkatenowane leksykalizacje *)
adjunct_schema_field "" Forward [Phrase Null;Phrase(AdjP AllAgr)];
adjunct_schema_field "" Both [Phrase Null;Phrase PrepP];*)
]
let adj_adjuncts = [
(* adjunct_schema_field "" Both [Phrase Null;Phrase AdvP]; *)
]
let verb_adjuncts_simp = [
adjunct_schema_field "Manner" Both [Phrase Null;Phrase AdvP];
adjunct_schema_field "Recipent" Both [Phrase Null;Phrase (NP(Case "dat"));Phrase (NumP(Case "dat"));Phrase (NCP(Case "dat",CompTypeUndef,CompUndef))];
adjunct_schema_field "Instrument" Both [Phrase Null;Phrase (NP(Case "inst"));Phrase (NumP(Case "inst"));Phrase (NCP(Case "inst",CompTypeUndef,CompUndef))];
adjunct_schema_field "Time" Both [Phrase Null;Phrase (Lex "date");Phrase (Lex "day-lex");Phrase (Lex "day-month");Phrase (Lex "day")];
(* adjunct_schema_field "" Both [Phrase Null;Phrase PrepP]; (* FIXME: Trzeba będzie uzgodnić PrepNP, PrepAdjP, PrepNumP z PrepP i XP *) *)
adjunct_schema_field "Condition" Forward [Phrase Null;Phrase (CP(CompTypeUndef,CompUndef))]; (* poprawić semantykę *) (* FIXME: to powinno być jako ostatnia lista argumentów *)
adjunct_schema_field "Theme" Both [Phrase Null;Phrase Or];
]
let noun_adjuncts_simp = [ (* FIXME: usuniecie noun_adjuncts pozostawia poss dla 'Witoldzie' *)
adjunct_schema_field "Possesive" Both [Phrase Null;Phrase(NP(Case "gen"));Phrase(NumP(Case "gen"))];
adjunct_schema_field "Aposition" Forward [Phrase Null;Phrase(NP(Case "nom"));Phrase(NumP(Case "nom"));Phrase Null;Phrase(NP(CaseAgr));Phrase(NumP(CaseAgr))];
adjunct_schema_field "Attribute" Backward [(*Phrase Null;Phrase(AdjP AllAgr)*)Multi[AdjP AllAgr]]; (* FIXME: za pomocą Multi można zrobić konkatenowane leksykalizacje *)
adjunct_schema_field "Base" Forward [Phrase Null;Phrase(AdjP AllAgr)];
(* adjunct_schema_field "" Both [Phrase Null;Phrase PrepP]; *)
]
let noun_measure_adjuncts_simp = [ (* FIXME: usuniecie noun_adjuncts pozostawia poss dla 'Witoldzie' *)
adjunct_schema_field "Attribute" Backward [(*Phrase Null;Phrase(AdjP AllAgr)*)Multi[AdjP AllAgr]]; (* FIXME: za pomocą Multi można zrobić konkatenowane leksykalizacje *)
adjunct_schema_field "Base" Forward [Phrase Null;Phrase(AdjP AllAgr)];
(* adjunct_schema_field "" Both [Phrase Null;Phrase PrepP]; *)
]
let adj_adjuncts_simp = [
adjunct_schema_field "Manner" Both [Phrase Null;Phrase AdvP];
]
let adv_adjuncts_simp = [
adjunct_schema_field "Manner" Both [Phrase Null;Phrase AdvP];
]
let convert_frame expands subtypes equivs lexemes valence lexeme pos (refl,opinion,negation,pred,aspect,schema) =
(* Printf.printf "convert_frame %s %s\n" lexeme pos; *)
try
if refl = "się" && not (StringMap.mem lexemes "się") then raise ImpossibleSchema else
let frame =
try StringMap.find default_frames refl (* w refl jest przekazywana informacja o typie domyślnej ramki *)
with Not_found ->
Frame(DefaultAtrs([],WalParser.parse_refl [Text refl],
WalParser.parse_opinion opinion,
WalParser.parse_negation [Text negation],
WalParser.parse_pred [Text pred],
WalParser.parse_aspect [Text aspect]),
prepare_schema expands subtypes equivs schema) in
let frame = if StringMap.is_empty lexemes then frame else reduce_schema_frame lexemes frame in
let frame = expand_lexicalizations frame in
Xlist.fold (extract_lex_frames lexeme pos [] frame) valence (fun valence -> function
lexeme,pos,Frame(atrs,schema) ->
let schemas = simplify_lex (split_xor (split_or_coord schema)) in
Xlist.fold schemas valence (fun valence schema ->
let poss = try StringMap.find valence lexeme with Not_found -> StringMap.empty in
let poss = StringMap.add_inc poss pos [Frame(atrs,schema)] (fun l -> Frame(atrs,schema) :: l) in
StringMap.add valence lexeme poss)
| lexeme,pos,LexFrame(id,pos2,restr,schema) ->
let schemas = simplify_lex (split_xor (split_or_coord schema)) in
Xlist.fold schemas valence (fun valence schema ->
let poss = try StringMap.find valence lexeme with Not_found -> StringMap.empty in
let poss = StringMap.add_inc poss pos [LexFrame(id,pos2,restr,schema)] (fun l -> LexFrame(id,pos2,restr,schema) :: l) in
StringMap.add valence lexeme poss)
| _ -> failwith "convert_frame")
with ImpossibleSchema -> valence
let convert_frame_sem expands subtypes equivs lexemes valence lexeme pos = function
Frame(DefaultAtrs(meanings,refl,opinion,negation,pred,aspect),positions) ->
(* Printf.printf "convert_frame_sem %s\n" (WalStringOf.frame lexeme (Frame(DefaultAtrs(meanings,refl,opinion,negation,pred,aspect),positions))); *)
(try
if refl = ReflSie && not (StringMap.mem lexemes "się") then raise ImpossibleSchema else
let frame =
Frame(DefaultAtrs(meanings,refl,opinion,negation,pred,aspect),
prepare_schema_sem expands subtypes equivs positions) in
let frame = if StringMap.is_empty lexemes then frame else reduce_schema_frame lexemes frame in
let frame = expand_lexicalizations frame in
Xlist.fold (extract_lex_frames lexeme pos [] frame) valence (fun valence -> function
lexeme,pos,Frame(atrs,schema) ->
let schemas = simplify_lex (split_xor (split_or_coord schema)) in
Xlist.fold schemas valence (fun valence schema ->
let poss = try StringMap.find valence lexeme with Not_found -> StringMap.empty in
let poss = StringMap.add_inc poss pos [Frame(atrs,schema)] (fun l -> Frame(atrs,schema) :: l) in
StringMap.add valence lexeme poss)
| lexeme,pos,LexFrame(id,pos2,restr,schema) ->
let schemas = simplify_lex (split_xor (split_or_coord schema)) in
Xlist.fold schemas valence (fun valence schema ->
let poss = try StringMap.find valence lexeme with Not_found -> StringMap.empty in
let poss = StringMap.add_inc poss pos [LexFrame(id,pos2,restr,schema)] (fun l -> LexFrame(id,pos2,restr,schema) :: l) in
StringMap.add valence lexeme poss)
| _ -> failwith "convert_frame_sem")
with ImpossibleSchema -> valence)
| _ -> failwith "convert_frame_sem"
let make_comprep_frames_of_schema s = function
[{cr=[];ce=[]; morfs=[LexPhrase([pos,Lexeme lex],(restr,schema))]}] ->
lex,
(match get_pos lex pos with [pos] -> pos | _ -> failwith "make_comprep_frame_of_schema 2"),
ComprepFrame(s,pos,restr,schema)
| schema -> failwith ("make_comprep_frame_of_schema: " ^ WalStringOf.schema schema)
let convert_comprep_frame expands subtypes equivs lexemes valence lexeme pos (s,morf) =
try
let schema = prepare_schema_comprep expands subtypes equivs [simple_arg_schema_field [morf]] in
let schema = if StringMap.is_empty lexemes then schema else reduce_schema lexemes schema in
let schema = expand_lexicalizations_schema schema in
let lexeme,pos,frame = make_comprep_frames_of_schema s schema in
Xlist.fold (extract_lex_frames lexeme pos [] frame) valence (fun valence -> function
lexeme,pos,ComprepFrame(s,pos2,restr,schema) ->
let schemas = simplify_lex (split_xor (split_or_coord schema)) in
Xlist.fold schemas valence (fun valence schema ->
let poss = try StringMap.find valence lexeme with Not_found -> StringMap.empty in
let poss = StringMap.add_inc poss pos [ComprepFrame(s,pos2,restr,schema)] (fun l -> ComprepFrame(s,pos2,restr,schema) :: l) in
StringMap.add valence lexeme poss)
| lexeme,pos,LexFrame(id,pos2,restr,schema) ->
let schemas = simplify_lex (split_xor (split_or_coord schema)) in
Xlist.fold schemas valence (fun valence schema ->
let poss = try StringMap.find valence lexeme with Not_found -> StringMap.empty in
let poss = StringMap.add_inc poss pos [LexFrame(id,pos2,restr,schema)] (fun l -> LexFrame(id,pos2,restr,schema) :: l) in
StringMap.add valence lexeme poss)
| _ -> failwith "convert_comprep_frame")
with ImpossibleSchema -> valence
let remove_pro_args schema = (* FIXME: sprawdzić czy Pro i Null są zawsze na początku *)
List.rev (Xlist.fold schema [] (fun schema -> function
{morfs=[Phrase Pro]} -> schema
| {morfs=(Phrase Pro) :: morfs} as s -> {s with morfs=morfs} :: schema
| {morfs=[Phrase Null]} -> schema
| {morfs=(Phrase Null) :: morfs} as s -> {s with morfs=morfs} :: schema
| s -> s :: schema))
let rec expand_restr valence lexeme pos = function
LexFrame(id,pos2,Natr,[]) -> [LexFrame(id,pos2,NoRestr,[])]
| LexFrame(id,pos2,Natr,_) -> failwith "expand_restr"
| LexFrame(id,pos2,restr,[]) ->
(* print_endline "expand_restr"; *)
let frames = try StringMap.find (StringMap.find valence lexeme) pos with Not_found -> failwith ("expand_restr:" ^ lexeme ^ " " ^ pos) in
(* Printf.printf "%s %s %d\n" lexeme pos (Xlist.size frames);
Xlist.iter frames (fun frame -> print_endline (WalStringOf.frame lexeme frame));
print_endline "";*)
(if restr = Atr || restr = Atr1 then [LexFrame(id,pos2,NoRestr,[])] else []) @
(Xlist.fold frames [] (fun frames -> function
Frame(_,schema) ->
let schema = remove_pro_args schema in
if schema = [] then frames else
(expand_restr valence lexeme pos (LexFrame(id,pos2,restr,schema))) @ frames
| _ -> frames))
| LexFrame(id,pos2,Atr,schema) ->
let schemas = Xlist.map (Xlist.multiply_list (Xlist.map schema (fun x -> [[x];[]]))) List.flatten in
Xlist.map schemas (fun schema -> LexFrame(id,pos2,NoRestr,schema))
| LexFrame(id,pos2,Atr1,schema) ->
LexFrame(id,pos2,NoRestr,[]) :: (Xlist.map schema (fun x -> LexFrame(id,pos2,NoRestr,[x])))
| LexFrame(id,pos2,Ratr,schema) ->
let schemas = Xlist.map (Xlist.multiply_list (Xlist.map schema (fun x -> [[x];[]]))) List.flatten in
Xlist.fold schemas [] (fun schemas schema -> if schema = [] then schemas else LexFrame(id,pos2,NoRestr,schema) :: schemas)
| LexFrame(id,pos2,Ratr1,schema) ->
Xlist.map schema (fun x -> LexFrame(id,pos2,NoRestr,[x]))
| LexFrame(id,pos2,Ratrs,schema) -> [LexFrame(id,pos2,NoRestr,schema)]
| LexFrame(id,pos2,NoRestr,_) -> failwith "expand_restr"
| ComprepFrame(s,pos2,Natr,[]) -> [ComprepFrame(s,pos2,NoRestr,[])]
| ComprepFrame(s,pos2,Natr,_) -> failwith "expand_restr"
| ComprepFrame(s,pos2,restr,[]) as frame -> failwith ("expand_restr: " ^ WalStringOf.frame lexeme frame)
| ComprepFrame(s,pos2,Atr,schema) ->
let schemas = Xlist.map (Xlist.multiply_list (Xlist.map schema (fun x -> [[x];[]]))) List.flatten in
Xlist.map schemas (fun schema -> ComprepFrame(s,pos2,NoRestr,schema))
| ComprepFrame(s,pos2,Atr1,schema) ->
ComprepFrame(s,pos2,NoRestr,[]) :: (Xlist.map schema (fun x -> ComprepFrame(s,pos2,NoRestr,[x])))
| ComprepFrame(s,pos2,Ratr,schema) ->
let schemas = Xlist.map (Xlist.multiply_list (Xlist.map schema (fun x -> [[x];[]]))) List.flatten in
Xlist.fold schemas [] (fun schemas schema -> if schema = [] then schemas else ComprepFrame(s,pos2,NoRestr,schema) :: schemas)
| ComprepFrame(s,pos2,Ratr1,schema) ->
Xlist.map schema (fun x -> ComprepFrame(s,pos2,NoRestr,[x]))
| ComprepFrame(s,pos2,Ratrs,schema) -> [ComprepFrame(s,pos2,NoRestr,schema)]
| ComprepFrame(s,pos2,NoRestr,_) -> failwith "expand_restr"
| Frame _ as frame -> [frame]
(* | _ -> failwith "expand_restr" *)
let simplify_pos = function
"subst" -> "noun"
| "depr" -> "noun"
| "psubst" -> "noun"
| "pdepr" -> "noun"
| "adj" -> "adj"
| "adjc" -> "adj"
| "ger" -> "verb"
| "pact" -> "verb"
| "ppas" -> "verb"
| "fin" -> "verb"
| "bedzie" -> "verb"
| "praet" -> "verb"
| "winien" -> "verb"
| "impt" -> "verb"
| "imps" -> "verb"
| "inf" -> "verb"
| "pcon" -> "verb"
| "pant" -> "verb"
| "pred" -> "verb"
| "ppron12" -> "pron"
| "ppron3" -> "pron"
| "siebie" -> "pron"
| s -> s
let transform_zeby = function
Aff -> [Comp "że"]
| Negation -> [Comp "że";Comp "żeby"]
| NegationUndef -> [Comp "że";Comp "żeby"]
| _ -> failwith "transform_zeby"
let transform_gdy = function
"indicative" -> [Comp "gdy"]
| "imperative" -> [Comp "gdy"]
| "conditional" -> [Comp "gdyby"]
| "gerundial" -> [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](*failwith "transform_comp"*)
let transform_str = function
Aff -> [Case "acc"]
| Negation -> [Case "gen"]
| NegationUndef -> [Case "acc";Case "gen"]
| _ -> failwith "transform_str"
(* FIXME: wstawić wszędzie adj jako wariant PrepNP, ComprepNP i NP *)
let transform_np_phrase = function
NP(Case case) -> [NP(Case case)(*;NumP(Case case)*)]
| NP(CaseAgr) -> [NP(CaseAgr)(*;NumP(CaseAgr)*)]
| AdjP(Case _) as morf -> [morf]
| AdjP(CaseAgr) -> [AdjP(AllAgr)]
| AdjP(AllAgr) -> [AdjP(AllAgr)]
| AdjP(Str) -> [AdjP(AllAgr)]
| PrepNP(sem,prep,Case case) -> [PrepNP(sem,prep,Case case)(*;PrepNumP(prep,Case case)*)]
(* | PrepNumP(_,Case _) as morf -> [morf] *)
| ComprepNP _ as morf -> [morf]
| ComparNP(sem,prep,Str) -> Xlist.map ["nom";"gen";"dat";"acc";"inst"] (fun case -> ComparNP(sem,prep,Case case))
| ComparPP _ as morf -> [morf]
| CP(ctype,comp) -> Xlist.map (transform_comp NegationUndef "" comp) (fun comp -> CP(ctype,comp)) (* FIXME zależność od trybu warunkowego*) (* FIXME zależność od negacji*)
| NCP(Case c,ctype,comp) -> Xlist.map (transform_comp NegationUndef "" comp) (fun comp -> NCP(Case c,ctype,comp)) (* FIXME zależność od trybu warunkowego*) (* FIXME zależność od negacji*)
| PrepNCP(sem,prep,Case case,ctype,comp) -> Xlist.map (transform_comp NegationUndef "" comp) (fun comp -> PrepNCP(sem,prep,Case case,ctype,comp)) (* FIXME zależność od trybu warunkowego*) (* FIXME zależność od negacji *)
| PrepAdjP(sem,_,Case _) as morf -> [morf] (* to wygląda seryjny błąd w Walentym xp(abl[prepadjp(z,gen)]) *)
| PrepNP(sem,prep,Str) -> List.flatten (Xlist.map ["nom";"gen";"dat";"acc";"inst"] (fun case -> [PrepNP(sem,prep,Case case)(*;PrepNumP(prep,Case case)*)])) (* FIXME: sprawdzić kto kontroluje! *) (* FIXME: pomijam uzgodnienie liczby i rodzaju *) (* zakładam, że nie jest kontrolowany przez SUBJ w czasowikach z OBJ *)
| PrepAdjP(sem,prep,Str) -> Xlist.map ["nom";"gen";"dat";"acc";"inst"] (fun case -> PrepAdjP(sem,prep,Case case)) (* FIXME: sprawdzić kto kontroluje! *) (* FIXME: pomijam uzgodnienie liczby i rodzaju *) (* zakładam, że nie jest kontrolowany przez SUBJ w czasowikach z OBJ *)
| AdvP as morf -> [morf] (* FIXME: tu trafiają przysłówkowe realizacje, trzeba by je przetłumaczyć na przymiotniki *)
| FixedP _ as morf -> [morf]
| PrepP as morf -> [morf]
| Or as morf -> [morf]
| Pro as morf -> [morf]
| Null as morf -> [morf]
| phrase -> print_endline ("transform_np_phrase: " ^ WalStringOf.phrase phrase); [phrase]
let transform_np_pos = function
| SUBST(_,Case _) as morf -> [morf]
| SUBST(_,CaseAgr) as morf -> [morf]
| ADJ(_,Case _,_,_) as morf -> [morf]
| ADJ(n,CaseAgr,g,gr) -> [ADJ(n,AllAgr,g,gr)]
| PACT(n,CaseAgr,g,a,neg,r) -> [PACT(n,AllAgr,g,a,neg,r)]
| 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] (* FIXME: tu trafiają przysłówkowe realizacje, trzeba by je przetłumaczyć na przymiotniki *)
| COMP _ as morf -> [morf]
| QUB as morf -> [morf]
| pos -> print_endline ("transform_np_pos: " ^ WalStringOf.pos pos); [pos]
let transform_adj_phrase = function
NP(Case case) -> [NP(Case case)(*;NumP(Case case)*)]
| NP(Part) -> [NP(Case "gen");NP(Case "acc")(*;NumP(Case "gen");NumP(Case "acc")*)]
| AdjP(CaseAgr) -> [AdjP(AllAgr)] (* jedno wystąpienie 'cały szczęśliwy', może się przydać podniesienie typu *)
| PrepNP(sem,prep,Case case) -> [PrepNP(sem,prep,Case case)(*;PrepNumP(prep,Case case)*)]
| ComprepNP _ as morf -> [morf]
| ComparNP(sem,prep,Str) -> Xlist.map ["nom";"gen";"dat";"acc";"inst"] (fun case -> ComparNP(sem,prep,Case case))
| ComparPP _ as morf -> [morf]
| CP(ctype,comp) -> Xlist.map (transform_comp NegationUndef "" comp) (fun comp -> CP(ctype,comp)) (* FIXME zależność od trybu warunkowego*) (* FIXME zależność od negacji*)
| NCP(Case c,ctype,comp) -> Xlist.map (transform_comp NegationUndef "" comp) (fun comp -> NCP(Case c,ctype,comp)) (* FIXME zależność od trybu warunkowego*) (* FIXME zależność od negacji*)
| PrepNCP(sem,prep,Case case,ctype,comp) -> Xlist.map (transform_comp NegationUndef "" comp) (fun comp -> PrepNCP(sem,prep,Case case,ctype,comp)) (* FIXME zależność od trybu warunkowego*) (* FIXME zależność od negacji *)
| PrepAdjP(sem,_,Case _) as morf -> [morf]
| PrepNP(sem,prep,Str) -> List.flatten (Xlist.map ["nom";"gen";"dat";"acc";"inst"] (fun case -> [PrepNP(sem,prep,Case case)(*;PrepNumP(prep,Case case)*)])) (* FIXME: sprawdzić kto kontroluje! *) (* FIXME: pomijam uzgodnienie liczby i rodzaju *) (* zakładam, że nie jest kontrolowany przez SUBJ w czasowikach z OBJ *)
| PrepAdjP(sem,prep,Str) -> Xlist.map ["nom";"gen";"dat";"acc";"inst"] (fun case -> PrepAdjP(sem,prep,Case case)) (* FIXME: sprawdzić kto kontroluje! *) (* FIXME: pomijam uzgodnienie liczby i rodzaju *) (* zakładam, że nie jest kontrolowany przez SUBJ w czasowikach z OBJ *)
| InfP _ as morf -> [morf]
| AdvP as morf -> [morf]
| FixedP _ as morf -> [morf]
| PrepP as morf -> [morf]
| Or as morf -> [morf]
| Pro as morf -> [morf]
| Null as morf -> [morf]
| morf -> print_endline ("transform_adj_phrase: " ^ WalStringOf.phrase morf); [morf]
let transform_adj_pos = function
| SUBST(_,Case _) as morf -> [morf]
| ADJ(n,CaseAgr,g,gr) -> [ADJ(n,AllAgr,g,gr)]
| PREP(Case _) as morf -> [morf]
| ADV _ as morf -> [morf]
| morf -> print_endline ("transform_adj_pos: " ^ WalStringOf.pos morf); [morf]
let transform_prep_pos = 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 -> print_endline ("transform_prep_pos: " ^ WalStringOf.pos pos); [pos]
let transform_compar_phrase = function
NP(Str) -> [NP CaseUndef(*;NumP(CaseUndef)*)] (* FIXME: ta sama sytuacja co w "jako" *)
| FixedP _ as morf -> [morf]
| phrase -> print_endline ("transform_compar_phrase: " ^ WalStringOf.phrase phrase); [phrase]
let transform_compar_pos = function
| SUBST(_,Case _) as morf -> [morf]
| ADJ(_,Case _,_,_) as morf -> [morf]
| PREP(Case _) as morf -> [morf]
| PPAS(_,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(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 -> print_endline ("transform_compar_pos: " ^ WalStringOf.pos pos); [pos]
let transform_adv_phrase = function
NP(Case case) -> [NP(Case case)(*;NumP(Case case)*)]
| PrepNP(sem,prep,Case case) -> [PrepNP(sem,prep,Case case)(*;PrepNumP(prep,Case case)*)]
| PrepNCP(sem,prep,Case case,ctype,comp) -> Xlist.map (transform_comp NegationUndef "" comp) (fun comp -> PrepNCP(sem,prep,Case case,ctype,comp)) (* FIXME zależność od trybu warunkowego*) (* FIXME zależność od negacji *)
| ComprepNP _ as morf -> [morf]
| CP(ctype,comp) -> Xlist.map (transform_comp NegationUndef "" comp) (fun comp -> CP(ctype,comp)) (* FIXME zależność od trybu warunkowego*) (* FIXME zależność od negacji*)
| InfP _ as morf -> [morf]
| AdvP as morf -> [morf]
| Or as morf -> [morf]
| Pro as morf -> [morf]
| Null as morf -> [morf]
| PrepAdjP(sem,_,Case _) as morf -> [morf]
| PrepNP(sem,prep,Str) -> List.flatten (Xlist.map ["nom";"gen";"dat";"acc";"inst"] (fun case -> [PrepNP(sem,prep,Case case)(*;PrepNumP(prep,Case case)*)])) (* FIXME: sprawdzić kto kontroluje! *) (* FIXME: pomijam uzgodnienie liczby i rodzaju *) (* zakładam, że nie jest kontrolowany przez SUBJ w czasowikach z OBJ *)
| PrepAdjP(sem,prep,Str) -> Xlist.map ["nom";"gen";"dat";"acc";"inst"] (fun case -> PrepAdjP(sem,prep,Case case)) (* FIXME: sprawdzić kto kontroluje! *) (* FIXME: pomijam uzgodnienie liczby i rodzaju *) (* zakładam, że nie jest kontrolowany przez SUBJ w czasowikach z OBJ *)
| ComparNP(sem,prep,Str) -> Xlist.map ["nom";"gen";"dat";"acc";"inst"] (fun case -> ComparNP(sem,prep,Case case))
| ComparPP _ as morf -> [morf]
(* | AdjP(CaseAgr) as morf -> [morf] *)
(* | NCP(Case c,ctype,comp) -> Xlist.map (transform_comp NegationUndef "" comp) (fun comp -> NCP(Case c,ctype,comp)) (* FIXME zależność od trybu warunkowego*) (* FIXME zależność od negacji*)
| PrepNCP(prep,Case case,ctype,comp) -> Xlist.map (transform_comp NegationUndef "" comp) (fun comp -> PrepNCP(prep,Case case,ctype,comp)) (* FIXME zależność od trybu warunkowego*) (* FIXME zależność od negacji *)
| FixedP _ as morf -> [morf]*)
| morf -> print_endline ("transform_adv_phrase: " ^ WalStringOf.phrase morf); [morf]
let transform_adv_pos = function
(* | SUBST(_,Case _) as morf -> [morf]
| ADJ(_,CaseAgr,_,_) as morf -> [morf]*)
COMP _ as morf -> [morf]
| PREP(Case _) as morf -> [morf]
| ADV _ as morf -> [morf]
| morf -> print_endline ("transform_adv_pos: " ^ WalStringOf.pos morf); [morf]
(*| Prepnp("jako",Str) as morf -> morf
| Prepnp("jak",Str) as morf -> morf
| Prepnp("niczym",Str) as morf -> morf
| Prepadjp("jako",Str) as morf -> morf
| Prepadjp("jak",Str) as morf -> morf
| Prepadjp("niczym",Str) as morf -> morf
| Compar "jako" as morf -> morf
| Compar "jak" as morf -> morf
| Compar "niczym" as morf -> morf
| Compar "niż" as morf -> morf*)
let transform_pers_subj_phrase negation mood = function (* FIXME: prepnp(na,loc) *)
| NP(Str) -> [NP(NomAgr)(*;NumP(NomAgr)*)]
| NCP(Str,ctype,comp) -> Xlist.map (transform_comp negation mood comp) (fun comp -> NCP(NomAgr,ctype,comp))
| CP(ctype,comp) -> Xlist.map (transform_comp negation mood comp) (fun comp -> CP(ctype,comp))
| InfP _ as morf -> [morf]
| Or as morf -> [morf]
| NP(Part) -> [NP(Case "gen")(*;NP(Case "acc")*)(*;NumP(Case "gen");NumP(Case "acc")*)]
| Pro -> [ProNG]
| morf -> print_endline ("transform_pers_subj_phrase: " ^ WalStringOf.phrase morf); [morf]
let transform_pers_subj_pos 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,AcmUndef) -> [NUM(NomAgr,g,AcmUndef)]
| ADJ(n,Str,g,gr) -> [ADJ(n,NomAgr,g,gr)]
| morf -> print_endline ("transform_ger_subj_pos: " ^ WalStringOf.pos morf); [morf]
let transform_ger_subj_phrase negation mood control = function
| NP(Str) -> [NP(Case "gen");PrepNP(NoSem,"przez",Case "acc")(*;NumP(Case "gen")*)(*;PrepNumP("przez",Case "acc")*)] (* FIXME: czy przez:acc jest możliwe? *)
| NCP(Str,ctype,comp) -> List.flatten (Xlist.map (transform_comp negation mood comp) (fun comp -> [NCP(Case "gen",ctype,comp);PrepNCP(NoSem,"przez",Case "acc",ctype,comp)])) (* FIXME: czy przez:acc jest możliwe? *)
| CP(ctype,comp) -> Xlist.map (transform_comp negation mood comp) (fun comp -> CP(ctype,comp)) (* FIXME: czy to jest możliwe? *)
| InfP _ as morf -> [morf] (* FIXME: czy to jest możliwe? *)
| Or as morf -> [morf]
| NP(Part) -> [NP(Case "gen")(*;NP(Case "acc")*)(*;NumP(Case "gen");NumP(Case "acc")*)]
| Pro -> if control then [Pro] else [Null]
| morf -> print_endline ("transform_ger_subj_phrase: " ^ WalStringOf.phrase morf); [morf]
let transform_ger_subj_pos 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,AcmUndef) -> [NUM(Case "gen",g,AcmUndef)]
| ADJ(n,Str,g,gr) -> [ADJ(n,Case "gen",g,gr)]
| morf -> print_endline ("transform_pers_subj_pos: " ^ WalStringOf.pos morf); [morf]
let transform_ppas_subj_phrase negation mood control = function
| NP(Str) -> [PrepNP(NoSem,"przez",Case "acc")(*;PrepNumP("przez",Case "acc")*)]
| NCP(Str,ctype,comp) -> Xlist.map (transform_comp negation mood comp) (fun comp -> PrepNCP(NoSem,"przez",Case "acc",ctype,comp))
| CP(ctype,comp) -> [Null] (* zakładam, że w ramie jest też NCP *)
| Pro -> if control then [Pro] else [Null]
| morf -> print_endline ("transform_ppas_subj_phrase: " ^ WalStringOf.phrase morf); [morf]
let transform_pers_phrase negation mood = function
| NP(Str) -> List.flatten (Xlist.map (transform_str negation) (fun case -> [NP case(*;NumP(case)*)]))
| AdjP(Str) -> Xlist.map (transform_str negation) (fun case -> AdjP case) (* FIXME: pomijam uzgadnianie liczby i rodzaju - wykonalne za pomocą kontroli *)
| NCP(Str,ctype,comp) -> List.flatten (Xlist.map (transform_str negation) (fun case -> Xlist.map (transform_comp negation mood comp) (fun comp -> NCP(case,ctype,comp))))
| NP(Part) -> [NP(Case "gen");NP(Case "acc")(*;NumP(Case "gen");NumP(Case "acc")*)]
| NCP(Part,ctype,comp) -> List.flatten (Xlist.map (transform_comp negation mood comp) (fun comp -> [NCP(Case "gen",ctype,comp);NCP(Case "acc",ctype,comp)]))
| NP(Case case) -> [NP(Case case)(*;NumP(Case case)*)]
| PrepNP(sem,prep,Case case) -> [PrepNP(sem,prep,Case case)(*;PrepNumP(prep,Case case)*)]
(* | PrepNumP(_,Case _) as morf -> [morf] *)
| ComprepNP _ as morf -> [morf]
| NCP(Case case,ctype,comp) -> Xlist.map (transform_comp negation mood comp) (fun comp -> NCP(Case case,ctype,comp))
| PrepNCP(sem,prep,Case case,ctype,comp) -> Xlist.map (transform_comp negation mood comp) (fun comp -> PrepNCP(sem,prep,Case case,ctype,comp))
| AdjP(Case _) as morf -> [morf] (* FIXME: pomijam uzgadnianie liczby i rodzaju - wykonalne za pomocą kontroli *)
| PrepAdjP(sem,_,Case _) as morf -> [morf] (* FIXME: pomijam uzgadnianie liczby i rodzaju - wykonalne za pomocą kontroli *)
| PrepNP(sem,prep,Str) -> List.flatten (Xlist.map ["nom";"gen";"dat";"acc";"inst"] (fun case -> [PrepNP(sem,prep,Case case)(*;PrepNumP(prep,Case case)*)])) (* FIXME: sprawdzić kto kontroluje! *) (* FIXME: pomijam uzgodnienie liczby i rodzaju *) (* zakładam, że nie jest kontrolowany przez SUBJ w czasowikach z OBJ *)
| PrepAdjP(sem,prep,Str) -> Xlist.map ["nom";"gen";"dat";"acc";"inst"] (fun case -> PrepAdjP(sem,prep,Case case)) (* FIXME: sprawdzić kto kontroluje! *) (* FIXME: pomijam uzgodnienie liczby i rodzaju *) (* zakładam, że nie jest kontrolowany przez SUBJ w czasowikach z OBJ *)
| ComparNP(sem,prep,Str) -> Xlist.map ["nom";"gen";"dat";"acc";"inst"] (fun case -> ComparNP(sem,prep,Case case))
| ComparPP _ as morf -> [morf]
| CP(ctype,comp) -> Xlist.map (transform_comp negation mood comp) (fun comp -> CP(ctype,comp))
| InfP _ as morf -> [morf]
| PadvP as morf -> [morf]
| AdvP -> if mood = "gerundial" then [AdjP AllAgr] else [AdvP]
| FixedP _ as morf -> [morf]
| PrepP as morf -> [morf]
| Or as morf -> [morf]
| Lex "się" as morf -> [morf]
(* | Refl as morf -> [morf] *)
(* | Recip as morf -> [morf] *)
| Pro as morf -> [morf]
| Null as morf -> [morf]
| morf -> print_endline ("transform_pers_phrase: " ^ WalStringOf.phrase morf); [morf]
let transform_pers_pos negation mood = function
| SUBST(n,Str) -> Xlist.map (transform_str negation) (fun case -> SUBST(n,case))
| NUM(Str,g,a) -> Xlist.map (transform_str negation) (fun case -> NUM(case,g,a))
| ADJ(n,Str,g,gr) -> Xlist.map (transform_str 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")]
| SUBST(_,Case _) as morf -> [morf]
| NUM(Case _,_,_) as morf -> [morf]
| PREP(Case _) as morf -> [morf]
| ADJ(_,Case _,_,_) as morf -> [morf]
| PREP(Str) -> Xlist.map ["nom";"gen";"dat";"acc";"inst"] (fun case -> PREP(Case case)) (* FIXME: sprawdzić kto kontroluje! *) (* FIXME: pomijam uzgodnienie liczby i rodzaju *) (* zakładam, że nie jest kontrolowany przez SUBJ w czasowikach z OBJ *)
| 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]
| ADV grad -> if mood = "gerundial" then [ADJ(NumberAgr,AllAgr,GenderAgr,grad)] else [ADV grad]
| morf -> print_endline ("transform_pers_pos: " ^ WalStringOf.pos morf); [morf]
let transform_pers_schema negation mood schema =
Xlist.map schema (fun s ->
{s with morfs =
if s.gf = SUBJ then List.flatten (Xlist.map s.morfs (function
Phrase phrase -> Xlist.map (transform_pers_subj_phrase negation mood phrase) (fun phrase -> Phrase phrase)
| E phrase -> Xlist.map (transform_pers_subj_phrase negation mood phrase) (fun phrase -> E phrase)
| LexArg(id,pos,lex) -> Xlist.map (transform_pers_subj_pos negation mood pos) (fun pos -> LexArg(id,pos,lex))
| _ -> failwith "transform_fin_schema"))
else List.flatten (Xlist.map s.morfs (function
Phrase phrase -> Xlist.map (transform_pers_phrase negation mood phrase) (fun phrase -> Phrase phrase)
| E phrase -> [Phrase Null] (*E(List.flatten (Xlist.map phrases (transform_pers_phrase negation mood)))*) (* FIXME *)
| LexArg(id,pos,lex) -> Xlist.map (transform_pers_pos negation mood pos) (fun pos -> LexArg(id,pos,lex))
| _ -> failwith "transform_fin_schema"))})
let transform_impt_schema negation mood schema =
Xlist.map schema (fun s ->
{s with morfs =
if s.gf = SUBJ then [Phrase ProNG]
else List.flatten (Xlist.map s.morfs (function
Phrase phrase -> Xlist.map (transform_pers_phrase negation mood phrase) (fun phrase -> Phrase phrase)
| E phrase -> [Phrase Null] (*E(List.flatten (Xlist.map phrases (transform_pers_phrase negation mood)))*) (* FIXME *)
| LexArg(id,pos,lex) -> Xlist.map (transform_pers_pos negation mood pos) (fun pos -> LexArg(id,pos,lex))
| _ -> failwith "transform_impt_schema"))})
let transform_imps_schema negation mood schema =
Xlist.map schema (fun s ->
{s with morfs =
if s.gf = SUBJ then [Phrase Pro]
else List.flatten (Xlist.map s.morfs (function
Phrase phrase -> Xlist.map (transform_pers_phrase negation mood phrase) (fun phrase -> Phrase phrase)
| E phrase -> [Phrase Null] (*E(List.flatten (Xlist.map phrases (transform_pers_phrase negation mood)))*) (* FIXME *)
| LexArg(id,pos,lex) -> Xlist.map (transform_pers_pos negation mood pos) (fun pos -> LexArg(id,pos,lex))
| _ -> failwith "transform_imps_chema"))})
let transform_ger_schema negation schema = (* FIXME: zakładam, że ger zeruje mood, czy to prawda? *)
Xlist.map schema (fun s ->
{s with morfs =
if s.gf = SUBJ then List.flatten (Xlist.map s.morfs (function
Phrase phrase -> Xlist.map (transform_ger_subj_phrase negation "gerundial" (s.cr <> [] || s.ce <> []) phrase) (fun phrase -> Phrase phrase)
| E phrase -> Xlist.map (transform_ger_subj_phrase negation "gerundial" (s.cr <> [] || s.ce <> []) phrase) (fun phrase -> E phrase)
| LexArg(id,pos,lex) -> Xlist.map (transform_ger_subj_pos negation "gerundial" pos) (fun pos -> LexArg(id,pos,lex))
| _ -> failwith "transform_fin_schema"))
else List.flatten (Xlist.map s.morfs (function
Phrase phrase -> Xlist.map (transform_pers_phrase negation "gerundial" phrase) (fun phrase -> Phrase phrase)
| E phrase -> [Phrase Null] (*E(List.flatten (Xlist.map phrases (transform_pers_phrase negation mood)))*) (* FIXME *)
| LexArg(id,pos,lex) -> Xlist.map (transform_pers_pos negation "gerundial" pos) (fun pos -> LexArg(id,pos,lex))
| _ -> failwith "transform_fin_schema"))})
let transform_padv_schema negation mood pro schema =
Xlist.map schema (fun s ->
{s with morfs =
if s.gf = SUBJ then if s.ce = [] then if pro then [Phrase Pro] else [Phrase Null] else [Phrase Null] else
List.flatten (Xlist.map s.morfs (function
Phrase phrase -> Xlist.map (transform_pers_phrase negation mood phrase) (fun phrase -> Phrase phrase)
| E phrase -> [Phrase Null] (*E(List.flatten (Xlist.map phrases (transform_pers_phrase negation mood)))*) (* FIXME *)
| LexArg(id,pos,lex) -> Xlist.map (transform_pers_pos negation mood pos) (fun pos -> LexArg(id,pos,lex))
| _ -> failwith "transform_fin_schema"))})
let transform_pact_schema negation mood schema =
Xlist.map schema (fun s ->
{s with morfs =
if s.gf = SUBJ then [Phrase Null]
else List.flatten (Xlist.map s.morfs (function
Phrase phrase -> Xlist.map (transform_pers_phrase negation mood phrase) (fun phrase -> Phrase phrase)
| E phrase -> [Phrase Null] (*E(List.flatten (Xlist.map phrases (transform_pers_phrase negation mood)))*) (* FIXME *)
| LexArg(id,pos,lex) -> Xlist.map (transform_pers_pos negation mood pos) (fun pos -> LexArg(id,pos,lex))
| _ -> failwith "transform_pact_schema"))})
let transform_ppas_schema negation mood schema =
Xlist.map schema (fun s ->
{s with morfs =
if s.gf = OBJ then [Phrase Null] else
if s.gf = SUBJ then List.flatten (Xlist.map s.morfs (function
Phrase phrase -> Xlist.map (transform_ppas_subj_phrase negation mood (s.cr <> [] || s.ce <> []) phrase) (fun phrase -> Phrase phrase)
| E phrase -> Xlist.map (transform_ppas_subj_phrase negation mood (s.cr <> [] || s.ce <> []) phrase) (fun phrase -> E phrase)
| LexArg(id,SUBST(n,Str),lex) -> raise Not_found (* FIXME!!! *)
| _ -> failwith "transform_ppas_schema"))
else List.flatten (Xlist.map s.morfs (function
Phrase phrase -> Xlist.map (transform_pers_phrase negation mood phrase) (fun phrase -> Phrase phrase)
| E phrase -> [Phrase Null] (*E(List.flatten (Xlist.map phrases (transform_pers_phrase negation mood)))*) (* FIXME *)
| LexArg(id,pos,lex) -> Xlist.map (transform_pers_pos negation mood pos) (fun pos -> LexArg(id,pos,lex))
| _ -> failwith "transform_ppas_schema"))})
let add_padv schema =
List.flatten (Xlist.map schema (fun s ->
if s.gf = SUBJ then
match s.cr with
[] -> [{s with cr=["3"]}; let s = adjunct_schema_field "" Both [Phrase Null;Phrase PadvP] in {s with ce=["3"]}]
| [cr] -> [s; let s = adjunct_schema_field "" Both [Phrase Null;Phrase PadvP] in {s with ce=[cr]}]
| _ -> failwith "add_padv"
else [s]))
let transform_np_schema schema =
Xlist.map schema (fun s ->
{s with morfs=List.flatten (Xlist.map s.morfs (function
Phrase phrase -> Xlist.map (transform_np_phrase phrase) (fun phrase -> Phrase phrase)
(* | LexArg(id,ADV _,lex) as morf -> print_endline (WalStringOf.morf morf); [morf] *)
| LexArg(id,pos,lex) -> Xlist.map (transform_np_pos pos) (fun pos -> LexArg(id,pos,lex))
| Multi[AdjP AllAgr] -> [Multi[AdjP AllAgr]]
| _ -> failwith "transform_np_schema"))})
let transform_num_schema acm schema =
Xlist.map schema (fun s ->
{s with morfs=List.flatten (Xlist.map s.morfs (function (* kierunek argumentu został dodany w expand_lexicalizations_morfs *)
| Phrase Pro -> [Phrase Pro]
| LexArg(id,SUBST(NumberUndef,CaseUndef),lex) ->
(match acm with
Acm "rec" -> [LexArg(id,SUBST(NumberUndef,GenAgr),lex)]
| Acm "congr" -> [LexArg(id,SUBST(NumberUndef,AllAgr),lex)]
| _ -> failwith "transform_num_schema")
| morf -> failwith ("transform_num_schema: " ^ WalStringOf.morf morf)))})
let transform_adj_schema schema =
Xlist.map schema (fun s ->
{s with morfs=List.flatten (Xlist.map s.morfs (function
Phrase phrase -> Xlist.map (transform_adj_phrase phrase) (fun phrase -> Phrase phrase)
| LexArg(id,pos,lex) -> Xlist.map (transform_adj_pos pos) (fun pos -> LexArg(id,pos,lex))
| _ -> failwith "transform_adj_schema"))})
let transform_adv_schema schema =
Xlist.map schema (fun s ->
{s with morfs=List.flatten (Xlist.map s.morfs (function
Phrase phrase -> Xlist.map (transform_adv_phrase phrase) (fun phrase -> Phrase phrase)
| LexArg(id,pos,lex) -> Xlist.map (transform_adv_pos pos) (fun pos -> LexArg(id,pos,lex))
| _ -> failwith "transform_adv_schema"))})
let transform_prep_schema schema =
Xlist.map schema (fun s ->
{s with morfs=List.flatten (Xlist.map s.morfs (function
Phrase(NumP(case)) -> [Phrase(NumP(case))]
| LexArg(id,pos,lex) -> Xlist.map (transform_prep_pos pos) (fun pos -> LexArg(id,pos,lex))
| morf -> failwith ("transform_prep_schema: " ^ WalStringOf.morf morf)))})
let transform_compar_schema schema =
Xlist.map schema (fun s ->
{s with morfs=List.flatten (Xlist.map s.morfs (function
Phrase phrase -> Xlist.map (transform_compar_phrase phrase) (fun phrase -> Phrase phrase)
| LexArg(id,pos,lex) -> Xlist.map (transform_compar_pos pos) (fun pos -> LexArg(id,pos,lex))
| morf -> failwith ("transform_compar_schema: " ^ WalStringOf.morf morf)))})
let transform_comp_schema schema = (* kierunek argumentu został dodany w expand_lexicalizations_morfs *)
Xlist.map schema (fun s ->
{s with morfs=List.flatten (Xlist.map s.morfs (function
| LexArg(_,PERS _,_) as morf -> [morf]
| morf -> failwith ("transform_comp_schema: " ^ WalStringOf.morf morf)))})
let transform_qub_schema schema =
Xlist.map schema (fun s ->
{s with morfs=List.flatten (Xlist.map s.morfs (function
| LexArg(_,PERS _,_) as morf -> [morf]
| morf -> failwith ("transform_qub_schema: " ^ WalStringOf.morf morf)))})
let rec remove_adj_agr = function
[] -> []
| {morfs=[Phrase Null;Phrase(AdjP(CaseAgr))]} :: l -> remove_adj_agr l
| {morfs=[Phrase Null;Phrase(AdjP(Part))]} :: l -> remove_adj_agr l
| s :: l -> (*print_endline (WalStringOf.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]
| NegationNA -> failwith "expand_negation"
let expand_aspect = function
Aspect s -> [Aspect s]
| AspectUndef -> [Aspect "imperf";Aspect "perf"]
| AspectNA -> failwith "expand_aspect"
let load_list filename =
Str.split (Str.regexp "\n") (File.load_file filename)
let subst_uncountable_lexemes = StringSet.of_list (load_list Paths.subst_uncountable_lexemes_filename)
let subst_uncountable_lexemes2 = StringSet.of_list (load_list Paths.subst_uncountable_lexemes_filename2)
let subst_container_lexemes = StringSet.of_list (load_list Paths.subst_container_lexemes_filename)
let subst_numeral_lexemes = StringSet.of_list (load_list Paths.subst_numeral_lexemes_filename)
let subst_time_lexemes = StringSet.of_list (load_list Paths.subst_time_lexemes_filename)
let subst_pronoun_lexemes = StringSet.of_list ["co"; "kto"; "cokolwiek"; "ktokolwiek"; "nic"; "nikt"; "coś"; "ktoś"; "to"]
let adj_pronoun_lexemes = StringSet.of_list ["czyj"; "jaki"; "który"; "jakiś"; "ten"; "taki"]
(* let adj_quant_lexemes = StringSet.of_list ["każdy"; "wszelki"; "wszystek"; "żaden"; "jakiś"; "pewien"; "niektóry"; "jedyny"; "sam"] *)
let empty_valence_lexemes = StringSet.union subst_pronoun_lexemes adj_pronoun_lexemes
let noun_type lemma pos =
let nsyn =
if pos = "ppron12" || pos = "ppron3" || pos = "siebie" then "pronoun" else
if pos = "psubst" || pos = "pdepr" || pos = "date" then "proper" else
if StringSet.mem subst_pronoun_lexemes lemma then "pronoun" else
"common" in
let nsem =
if pos = "ppron12" || pos = "ppron3" || pos = "siebie" then [Common "count"] else
if StringSet.mem subst_time_lexemes lemma then [Time] else
let l = ["count"] in
let l = if StringSet.mem subst_uncountable_lexemes lemma || StringSet.mem subst_uncountable_lexemes2 lemma then "mass" :: l else l in
let l = if StringSet.mem subst_container_lexemes lemma then "measure" :: l else l in
Xlist.map l (fun s -> Common s) in
nsyn,nsem
let adj_type lemma = (* FIXME: typy przymiotników wymagają zbadania - przejrzenia listy przymiotników *)
let adjsyn = if StringSet.mem adj_pronoun_lexemes lemma then "pronoun" else "common" in (* FIXME: dodać heurystykę uwzględniającą wielkość liter aby wykrywać proper np. Oświęcimski*)
adjsyn
let transform_frame lexeme pos = function (* FIXME: dodać tutaj typy rzeczowników *)
Frame(DefaultAtrs(meanings,refl,opinion,negation,pred,aspect),schema) as frame ->
if pos = "subst" || pos = "depr" || pos = "psubst" || pos = "pdepr" || pos = "ppron12" || pos = "ppron3" || pos = "siebie" then (
if refl <> ReflEmpty || negation <> NegationNA || pred <> PredNA || aspect <> AspectNA then failwith ("transform_frame: " ^ WalStringOf.frame lexeme frame);
let nsyn,nsem(*,typ*) = noun_type lexeme pos in
let schema = if nsyn = "pronoun" then [] else (remove_adj_agr schema) @ noun_adjuncts in (* FIXME: remove_adj_agr jest w słowniku tymczasowo *)
(* List.flatten (Xlist.map typ (fun typ -> *)
Xlist.map nsem (fun nsem -> Frame(NounAtrs(meanings,nsyn,nsem(*,typ*)),transform_np_schema schema)))(* ))*) else
if pos = "symbol" || pos = "date" || pos = "date-interval" || pos = "hour" || pos = "hour-minute" || pos = "hour-interval" || pos = "hour-minute-interval" ||
pos = "year" || pos = "year-interval" || pos = "day" || pos = "day-interval" || pos = "day-month" || pos = "day-month-interval" ||
pos = "match-result" || pos = "month-interval" || pos = "roman" || pos = "roman-interval" || pos = "url" || pos = "email" || pos = "obj-id" then
let nsyn,nsem = "proper",[Common "count"] in
Xlist.map nsem (fun nsem -> Frame(NounAtrs(meanings,nsyn,nsem),transform_np_schema schema)) else
if pos = "adj" || pos = "adjc" || pos = "adjp" || pos = "ordnum" then (
if refl <> ReflEmpty || negation <> NegationNA || aspect <> AspectNA then failwith ("transform_frame: " ^ WalStringOf.frame lexeme frame);
let adjsyn(*,adjsem,typ*) = adj_type lexeme in
let schema = if pos = "adjp" || pos = "ordnum" then schema else if adjsyn = "pronoun" then [] else schema @ adj_adjuncts in
let case = match pred with Pred -> Case "pred" | PredNA -> CaseUndef in
(* Xlist.map typ (fun typ -> *)
[Frame(AdjAtrs(meanings,case,adjsyn(*,adjsem,typ*)),transform_adj_schema schema)])(* )*) else
if pos = "adv" then (
if refl <> ReflEmpty || negation <> NegationNA || pred <> PredNA || aspect <> AspectNA then failwith ("transform_frame: " ^ WalStringOf.frame lexeme frame); (* FIXME: typy przysłówków *)
[Frame(EmptyAtrs meanings,transform_adv_schema (remove_adj_agr schema))]) else
if pos = "fin" then (
if pred <> PredNA then failwith ("transform_frame: " ^ WalStringOf.frame lexeme frame);
let s,schema = if refl = ReflSie then lexeme ^ " się", nosem_refl_schema_field :: schema else lexeme, schema in
let schema = (add_padv schema) @ verb_adjuncts in
List.flatten (Xlist.map (expand_negation negation) (fun negation ->
Xlist.map (expand_aspect aspect) (function
Aspect "imperf" -> Frame(PersAtrs(meanings,s,negation,"indicative","pres",NoAux,Aspect "imperf"), transform_pers_schema negation "indicative" schema)
| Aspect "perf" -> Frame(PersAtrs(meanings,s,negation,"indicative","fut",NoAux,Aspect "perf"), transform_pers_schema negation "indicative" schema)
| _ -> failwith "transform_frame") @
[Frame(PersAtrs(meanings,s,negation,"imperative","fut",ImpAux,aspect), transform_pers_schema negation "imperative" schema)]))) else
if pos = "bedzie" then (
if pred <> PredNA then failwith ("transform_frame: " ^ WalStringOf.frame lexeme frame);
let s,schema = if refl = ReflSie then lexeme ^ " się", nosem_refl_schema_field :: schema else lexeme, schema in
let schema = (add_padv schema) @ verb_adjuncts in
List.flatten (Xlist.map (expand_negation negation) (fun negation ->
Xlist.map (expand_aspect aspect) (function
Aspect "imperf" -> Frame(PersAtrs(meanings,s,negation,"indicative","fut",NoAux,Aspect "imperf"), transform_pers_schema negation "indicative" schema)
| Aspect "perf" -> Frame(PersAtrs(meanings,s,negation,"indicative","fut",NoAux,Aspect "perf"), transform_pers_schema negation "indicative" schema) (* FIXME: niepotrzebne *)
| _ -> failwith "transform_frame")))) else
if pos = "praet" then (
if pred <> PredNA then failwith ("transform_frame: " ^ WalStringOf.frame lexeme frame);
let s,schema = if refl = ReflSie then lexeme ^ " się", nosem_refl_schema_field :: schema else lexeme, schema in
let schema = (add_padv schema) @ verb_adjuncts in
List.flatten (Xlist.map (expand_negation negation) (fun negation ->
List.flatten (Xlist.map (expand_aspect aspect) (function
Aspect "imperf" ->
[Frame(PersAtrs(meanings,s,negation,"indicative","past",NoAux,Aspect "imperf"), transform_pers_schema negation "indicative" schema);
Frame(PersAtrs(meanings,s,negation,"conditional","past",NoAux,Aspect "imperf"), transform_pers_schema negation "conditional" schema);
Frame(PersAtrs(meanings,s,negation,"indicative","fut",FutAux,Aspect "imperf"), transform_pers_schema negation "indicative" schema)]
| Aspect "perf" ->
[Frame(PersAtrs(meanings,s,negation,"indicative","past",NoAux,Aspect "perf"), transform_pers_schema negation "indicative" schema);
Frame(PersAtrs(meanings,s,negation,"conditional","past",NoAux,Aspect "perf"), transform_pers_schema negation "conditional" schema)]
| _ -> failwith "transform_frame"))))) else
if pos = "winien" then (
if pred <> PredNA then failwith ("transform_frame: " ^ WalStringOf.frame lexeme frame);
let s,schema = if refl = ReflSie then lexeme ^ " się", nosem_refl_schema_field :: schema else lexeme, schema in
let schema = (add_padv schema) @ verb_adjuncts in
List.flatten (Xlist.map (expand_negation negation) (fun negation ->
List.flatten (Xlist.map (expand_aspect aspect) (fun aspect ->
[Frame(PersAtrs(meanings,s,negation,"indicative","pres",NoAux,aspect), transform_pers_schema negation "indicative" schema);
Frame(PersAtrs(meanings,s,negation,"conditional","past",NoAux,aspect), transform_pers_schema negation "conditional" schema);
Frame(PersAtrs(meanings,s,negation,"indicative","past",PastAux,aspect), transform_pers_schema negation "indicative" schema)]))))) else
if pos = "impt" then (
if pred <> PredNA then failwith ("transform_frame: " ^ WalStringOf.frame lexeme frame);
let s,schema = if refl = ReflSie then lexeme ^ " się", nosem_refl_schema_field :: schema else lexeme, schema in
let schema = (add_padv schema) @ verb_adjuncts in
Xlist.map (expand_negation negation) (fun negation ->
Frame(PersAtrs(meanings,s,negation,"imperative","fut",NoAux,aspect),transform_impt_schema negation "imperative" schema))) else
if pos = "imps" then (
if pred <> PredNA then failwith ("transform_frame: " ^ WalStringOf.frame lexeme frame);
let s,schema = if refl = ReflSie then lexeme ^ " się", nosem_refl_schema_field :: schema else lexeme, schema in
let schema = (add_padv schema) @ verb_adjuncts in
Xlist.map (expand_negation negation) (fun negation ->
Frame(PersAtrs(meanings,s,negation,"indicative","past",NoAux,aspect),transform_imps_schema negation "indicative" schema))) else
if pos = "pred" then (
if pred <> PredNA then failwith ("transform_frame: " ^ WalStringOf.frame lexeme frame);
let s,schema = if refl = ReflSie then lexeme ^ " się", nosem_refl_schema_field :: schema else lexeme, schema in
let schema = (add_padv schema) @ verb_adjuncts in
List.flatten (Xlist.map (expand_negation negation) (fun negation ->
[Frame(PersAtrs(meanings,s,negation,"indicative","pres",NoAux,aspect), transform_pers_schema negation "indicative" schema);
Frame(PersAtrs(meanings,s,negation,"indicative","fut",FutAux,aspect), transform_pers_schema negation "indicative" schema);
Frame(PersAtrs(meanings,s,negation,"indicative","past",PastAux,aspect), transform_pers_schema negation "indicative" schema)]))) else
if pos = "pcon" || pos = "pant" || pos = "inf" then (
if pred <> PredNA then failwith ("transform_frame: " ^ WalStringOf.frame lexeme frame);
let role,role_attr = try get_role SUBJ schema with Not_found -> "Initiator","" in
let s,schema = if refl = ReflSie then lexeme ^ " się", nosem_refl_schema_field :: schema else lexeme, schema in
let schema = schema @ verb_adjuncts in
Xlist.map (expand_negation negation) (fun negation ->
Frame(NonPersAtrs(meanings,s,role,role_attr,negation,aspect),transform_padv_schema negation "indicative" true schema))) else
if pos = "pact" then (
if pred <> PredNA then failwith ("transform_frame: " ^ WalStringOf.frame lexeme frame);
try
let role,role_attr = try get_role SUBJ schema with Not_found -> "Initiator","" in
let s,schema = if refl = ReflSie then lexeme ^ " się", nosem_refl_schema_field :: schema else lexeme, schema in
let schema = schema @ verb_adjuncts in
Xlist.map (expand_negation negation) (fun negation ->
Frame(NonPersAtrs(meanings,s,role,role_attr,negation,aspect),transform_pact_schema negation "indicative" schema))
with Not_found -> []) else
if pos = "ppas" then (
if pred <> PredNA then failwith ("transform_frame: " ^ WalStringOf.frame lexeme frame);
try
let role,role_attr = try get_role OBJ schema with Not_found -> "Theme","" in
let s,schema = if refl = ReflSie then raise Not_found else lexeme, schema in
let schema = schema @ verb_adjuncts in
Xlist.map (expand_negation negation) (fun negation ->
Frame(NonPersAtrs(meanings,s,role,role_attr,negation,aspect),transform_ppas_schema negation "indicative" schema))
with Not_found -> []) else
if pos = "ger" then (
if pred <> PredNA then failwith ("transform_frame: " ^ WalStringOf.frame lexeme frame);
let s,schema = if refl = ReflSie then lexeme ^ " się", nosem_refl_schema_field :: schema else lexeme, schema in (* FIXME: czy ger może mieć niesemantyczne się? *)
let schema = schema @ verb_adjuncts in
Xlist.map (expand_negation negation) (fun negation ->
Frame(GerAtrs(meanings,s,negation,aspect),transform_ger_schema negation schema))) else
failwith ("transform_frame: " ^ pos)
| LexFrame(id,pos,NoRestr,schema) as frame ->
(match pos with
SUBST _ -> [LexFrame(id,pos,NoRestr,transform_np_schema schema)]
| PREP _ -> [LexFrame(id,pos,NoRestr,transform_prep_schema schema)]
| NUM(c,g,AcmUndef) ->
Xlist.map [Acm "congr";Acm "rec"] (fun acm ->
LexFrame(id,NUM(c,g,acm),NoRestr,transform_num_schema acm schema))
| ADJ(n,c,g,gr) -> [LexFrame(id,pos,NoRestr,transform_adj_schema schema)]
| ADV(gr) -> [LexFrame(id,pos,NoRestr,transform_adv_schema schema)]
| GER(n,c,g,a,negation,ReflEmpty) ->
Xlist.map (expand_negation negation) (fun negation ->
LexFrame(id,GER(n,c,g,a,negation,ReflEmpty),NoRestr,transform_ger_schema negation schema))
| PACT(n,c,g,a,negation,ReflEmpty) ->
Xlist.map (expand_negation negation) (fun negation ->
LexFrame(id,PACT(n,c,g,a,negation,ReflEmpty),NoRestr,transform_pact_schema negation "indicative" schema))
| PPAS(n,c,g,a,negation) ->
Xlist.map (expand_negation negation) (fun negation ->
LexFrame(id,PPAS(n,c,g,a,negation),NoRestr,transform_ppas_schema negation "indicative" schema))
| INF(a,negation,r) ->
Xlist.map (expand_negation negation) (fun negation ->
LexFrame(id,INF(a,negation,r),NoRestr,transform_padv_schema negation "indicative" false schema))
| QUB -> [LexFrame(id,pos,NoRestr,transform_qub_schema schema)]
| COMPAR -> [LexFrame(id,pos,NoRestr,transform_compar_schema schema)]
| COMP _ -> [LexFrame(id,pos,NoRestr,transform_comp_schema schema)]
| PERS(negation,r) ->
Xlist.map (expand_negation negation) (fun negation ->
LexFrame(id,PERS(negation,r),NoRestr,transform_pers_schema negation "indicative" schema))
| _ -> failwith ("transform_frame:" ^ WalStringOf.frame lexeme frame))
| ComprepFrame(s,pos,NoRestr,schema) as frame ->
(match pos with
PREP _ -> [ComprepFrame(s,pos,NoRestr,transform_prep_schema schema)]
| ADV _ -> [ComprepFrame(s,pos,NoRestr,transform_adv_schema schema)]
| _ -> failwith ("transform_frame:" ^ WalStringOf.frame lexeme frame))
| frame -> failwith ("transform_frame:" ^ WalStringOf.frame lexeme frame)
let reduce_frame_negation lexemes = function
Negation -> StringMap.mem lexemes "nie"
| _ -> true
let reduce_frame_mood lexemes = function
"conditional" -> StringMap.mem lexemes "by"
| _ -> true
let reduce_frame_aux lexemes = function
NoAux -> true
| PastAux -> (try let poss = StringMap.find lexemes "być" in StringSet.mem poss "praet" with Not_found -> false)
| FutAux -> (try let poss = StringMap.find lexemes "być" in StringSet.mem poss "bedzie" with Not_found -> false)
| ImpAux -> StringMap.mem lexemes "niech" || StringMap.mem lexemes "niechaj" || StringMap.mem lexemes "niechże" || StringMap.mem lexemes "niechajże"
let reduce_frame_atrs pos lexemes = function
Frame(NounAtrs _,_) -> true
| Frame(AdjAtrs _,_) -> true
| Frame(EmptyAtrs _,_) -> true
| Frame(PersAtrs(_,_,negation,mood,_,aux,_),_) -> reduce_frame_negation lexemes negation && reduce_frame_mood lexemes mood && reduce_frame_aux lexemes aux
| Frame(NonPersAtrs(_,_,_,_,negation,_),_) -> if pos = "pact" || pos = "ppas" then true else reduce_frame_negation lexemes negation
| Frame(GerAtrs(_,_,negation,_),_) -> reduce_frame_negation lexemes negation
| Frame(_,_) as frame -> failwith ("reduce_frame_atrs: " ^ WalStringOf.frame "" frame)
| LexFrame _ -> true
| ComprepFrame _ -> true
let rec reduce_frame_atrs_list pos lexemes = function
[] -> []
| frame :: l -> (if reduce_frame_atrs pos lexemes frame then [frame] else []) @ reduce_frame_atrs_list pos lexemes l
let find_frames lexemes =
(* print_endline "find_frames 1"; *)
let valence = StringMap.fold lexemes StringMap.empty (fun valence lexeme poss ->
(* let poss = StringSet.fold poss StringSet.empty (fun poss pos -> StringSet.add poss (simplify_pos pos)) in *)
(* Printf.printf "find_frame: %s |%s|\n" s (String.concat " " (StringSet.to_list lexemes)); *)
StringSet.fold poss valence (fun valence pos ->
let valence =
let frames_sem = try StringMap.find (StringMap.find walenty (simplify_pos pos)) lexeme with Not_found -> [] in
(* if frames_sem <> [] then Printf.printf "%s %s in TEI\n%!" lexeme pos; *)
if frames_sem <> [] then
Xlist.fold frames_sem valence (fun valence frame ->
convert_frame_sem expands subtypes equivs lexemes valence lexeme pos frame)
else
let frames = match simplify_pos pos with
"verb" -> ((*try StringMap.find verb_frames lexeme with Not_found ->*) ["verb","","","","",""])
| "noun" -> ((*try StringMap.find noun_frames lexeme with Not_found ->*) if StringSet.mem empty_valence_lexemes lexeme then ["empty","","","","",""] else ["noun","","","","",""])
| "adj" -> ((*try StringMap.find adj_frames lexeme with Not_found ->*) if StringSet.mem empty_valence_lexemes lexeme then ["empty","","","","",""] else ["adj","","","","",""])
| "adv" -> ((*try StringMap.find adv_frames lexeme with Not_found ->*) ["adv","","","","",""])
| "pron" -> ["empty","","","","",""]
| "adjp" -> ["empty","","","","",""]
| "ordnum" -> ["empty","","","","",""]
| "symbol" -> ["empty","","","","",""]
| "date" -> ["date","","","","",""]
| "date-interval" -> ["empty","","","","",""]
| "hour" -> ["hour","","","","",""]
| "hour-minute" -> ["hour","","","","",""]
| "hour-interval" -> ["empty","","","","",""]
| "hour-minute-interval" -> ["empty","","","","",""]
| "year" -> ["empty","","","","",""]
| "year-interval" -> ["empty","","","","",""]
| "day" -> ["day","","","","",""]
| "day-interval" -> ["day","","","","",""]
| "day-month" -> ["date2","","","","",""]
| "day-month-interval" -> ["empty","","","","",""]
| "match-result" -> ["empty","","","","",""]
| "month-interval" -> ["empty","","","","",""]
| "roman" -> ["empty","","","","",""]
| "roman-interval" -> ["empty","","","","",""]
| "url" -> ["empty","","","","",""]
| "email" -> ["empty","","","","",""]
| "obj-id" -> ["empty","","","","",""]
| _ -> [] in
(* if frames = [] then valence else
Printf.printf "find_frame: %s |l|=%d\n" s (Xlist.size l); *)
Xlist.fold frames valence (fun valence frame ->
convert_frame expands subtypes equivs lexemes valence lexeme pos frame) in
Xlist.fold ((*try StringMap.find compreps lexeme with Not_found ->*) []) valence (fun valence (cpos,frame) -> (* FIXME: na razie przyimki złożone są wyłączone *)
if cpos = pos then convert_comprep_frame expands subtypes equivs lexemes valence lexeme pos frame else valence))) in
(* print_endline "find_frames 2"; *)
let valence = StringMap.mapi valence (fun lexeme poss ->
StringMap.mapi poss (fun pos frames ->
List.flatten (Xlist.map frames (fun frame ->
(* print_endline ("find_frames: " ^ WalStringOf.frame lexeme frame); *)
expand_restr valence lexeme pos frame)))) in
(* print_endline "find_frames 3"; *)
let valence = StringMap.mapi valence (fun lexeme poss ->
StringMap.mapi poss (fun pos frames ->
reduce_frame_atrs_list pos lexemes (List.flatten (Xlist.map frames (transform_frame lexeme pos))))) in
(* let valence = StringMap.mapi valence (fun lexeme poss ->
StringMap.mapi poss (fun pos frames ->
Xlist.map frames (assign_thematic_role pos))) in*)
(* StringMap.iter valence (fun lexeme poss ->
StringMap.iter poss (fun pos frames ->
Xlist.iter frames (fun frame -> print_endline (WalStringOf.frame lexeme frame))));*)
(* print_endline "find_frames 4"; *)
valence
(*let _ =
let valence = Xlist.fold all_frames StringMap.empty (fun valence (pos,frame_map) ->
print_endline pos;
StringMap.fold frame_map valence (fun valence lexeme frames ->
Xlist.fold frames valence (fun valence frame ->
(* print_endline (WalStringOf.unparsed_frame lexeme frame); *)
convert_frame expands subtypes equivs StringMap.empty valence lexeme pos frame))) in
print_endline "comprepnp";
let valence = StringMap.fold compreps valence (fun valence lexeme frames ->
Xlist.fold frames valence (fun valence (pos,frame) ->
convert_comprep_frame expands subtypes equivs StringMap.empty valence lexeme pos frame)) in
print_endline "expand_restr";
let valence = StringMap.mapi valence (fun lexeme poss ->
StringMap.mapi poss (fun pos frames ->
List.flatten (Xlist.map frames (expand_restr valence lexeme pos)))) in
print_endline "transform_frame";
let _ = StringMap.mapi valence (fun lexeme poss ->
StringMap.mapi poss (fun pos frames ->
(* print_endline lexeme; *)
List.flatten (Xlist.map frames (transform_frame lexeme pos)))) in
print_endline "done";
()*)
(* StringMap.iter valence (fun lexeme poss ->
StringMap.iter poss (fun pos frames ->
Xlist.iter frames (fun frame -> print_endline (WalStringOf.frame lexeme frame))))*)