AAda Compiler Token Package

8
Jan
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
with aada_character_categories;
with aada_compiler_ucs4_character;
with aada_text_input_package;
with aada_compiler_character_package;
with aada_compiler_token_position;
with aada_vstrings;
 
use aada_character_categories;
use aada_compiler_ucs4_character;
use aada_compiler_character_package;
use aada_compiler_token_position;
use aada_vstrings;
 
package body aada_compiler_token_package is
 
-- open the input text file
procedure open(f: in out compiler_token_input_type; name: in string) is
begin
  aada_text_input_package.open(f.file_input, name);
  -- reset the look ahead stack (mark it empty)
  f.look_ahead.token := token_nil;
end open;
 
procedure close(f: in out compiler_token_input_type) is
begin
  aada_text_input_package.close(f.file_input);
end close;
 
function is_open(f: in compiler_token_input_type) return boolean is
begin
  return aada_text_input_package.is_open(f.file_input);
end is_open;
 
function name(f: in compiler_token_input_type) return string is
begin
  return aada_text_input_package.name(f.file_input);
end name;
 
procedure get_character(f: in out compiler_token_input_type;
                        ct: out compiler_token_character) is
  status: character_status;
  code: ucs4_character;
  position: page_sizes_type;
  last_page: position_type;
begin
  -- called unget_character()?
  if f.stack.pointer /= stack_pointer_type'first then
    -- restore the character including its position
    ct := f.stack.stack(f.stack.pointer);
    return;
  end if;
 
  -- the character being read has this very position
  ct.position := f.current_position;
 
  if f.unget /= 16#0A# then
    set_ucs4_character(ct.c, f.unget);
    f.unget := 16#0A#;
  else
    aada_text_input_package.get(f.file_input, ct.c);
  end if;
  if aada_compiler_character_package.eof(ct.c) then
    -- get_ucs4_code() would raise an error otherwise
    return;
  end if;
 
  -- check whether we reached the end of the line or page
  code := get_ucs4_code(ct.c);
  case code is
  when 16#0D# => -- CR
    -- we do not need to increase char_pos in this case since
    -- we are about to reset it back to 1 anyway
    aada_text_input_package.get(f.file_input, ct.c);
    status := get_status(ct.c);
    if status /= aada_compiler_character_package.eof then
      code := get_ucs4_code(ct.c);
      if code /= 16#0A# then -- not LF? (i.e. for CR LF => 1 line)
        -- save for next call to get_character()
        f.unget := code;
      end if;
    end if;
    get_special_character(ct.c, eol);
    next_line(f.page_sizes, f.current_position);
 
  when 16#000A#    -- LF
     | 16#2028#    -- line separator
     | 16#2029# => -- paragraph separator (this has the same effect as eol)
    get_special_character(ct.c, eol);
    next_line(f.page_sizes, f.current_position);
 
  when 16#0085# | 16#000C# => -- NEL | LF
    -- we should create an EOP but we don't need that distinction
    -- anywhere else so we simplify the code with EOL
    get_special_character(ct.c, eol);
    -- if immediately followed by CR, LF or CR+LF, remove
    aada_text_input_package.get(f.file_input, ct.c);
    status := get_status(ct.c);
    if status /= aada_compiler_character_package.eof then
      code := get_ucs4_code(ct.c);
      if code = 16#0D# then -- CR
        aada_text_input_package.get(f.file_input, ct.c);
        status := get_status(ct.c);
        if status /= aada_compiler_character_package.eof then
          code := get_ucs4_code(ct.c);
          if code /= 16#0A# then -- CR+LF?
            f.unget := code;
          end if;
          -- make sure we enter the LF condition below
          code := 16#0A#;
        end if;
      end if;
      if code = 16#0A# then -- LF
        -- count the line now so we don't start a new page with a new line
        last_page := get_position(f.current_position, page);
        next_line(f.page_sizes, f.current_position);
        if last_page /= get_position(f.current_position, page) then
          -- the next_line() added a page already,
          -- so we don't add it again
          return;
        end if;
      else
        f.unget := code;
      end if;
    end if;
    next_page(f.page_sizes, f.current_position);
 
  when 16#0B# => -- VT
    -- vertical tab
    get_special_character(ct.c, eol);
    next_line(f.page_sizes, f.current_position,
                                      get_size(f.page_sizes, vertical_tab));
 
  when 16#09# => -- HT
    -- horizontal tab
    next_character(f.page_sizes, f.current_position,
                                      get_size(f.page_sizes, horizontal_tab));
    -- transform in a Space so we don't have to handle HT separately
    set_ucs4_character(ct.c, 16#20#);
 
  when others =>
    -- non-formatting characters are counted as one
    -- (of course... next_character() may add a line which in
    -- turn may add a page)
    -- XXX: should characters such as 0xFEFF not be counted?
    next_character(f.page_sizes, f.current_position);
 
  end case;
end get_character;
 
procedure unget_character(f: in out compiler_token_input_type;
                          ct: in compiler_token_character) is
begin
  if f.stack.pointer = stack_pointer_type'last then
    raise unget_stack_is_full_error;
  end if;
  -- save the character including its position
  f.stack.pointer := stack_pointer_type'succ(f.stack.pointer);
  f.stack.stack(f.stack.pointer) := ct;
end unget_character;
 
procedure set_token(f: in out compiler_token_input_type;
                    t: in ada_token;
                    c: in compiler_character) is
begin
  -- f.position := ... already set
  f.token.token := t; -- token
  f.token.operator := token_unknown;
 
  if t = token_eof or t = token_nil then
    f.literal_size := 0; -- this is a special case for EOF, EOL, etc.
    -- the literal is empty
  else
    -- start with a length of 1
    -- use the add_to_token() to add more characters
    f.literal_size := 1;
 
    -- copy the 1 character "string"
    f.literal(integer(f.literal'first)) := c;
  end if;
end set_token;
 
-- helper function that transforms a UCS4 character in a compiler character
procedure set_token(f: in out compiler_token_input_type;
                    t: in ada_token;
                    u: in ucs4_character) is
  c: compiler_character;
begin
  set_ucs4_character(c, u);
  set_token(f, t, c);
end set_token;
 
procedure add_to_token(f: in out compiler_token_input_type;
                       c: in compiler_character) is
begin
  if f.literal_size >= compiler_literal_index'last then
    -- this literal is too large, we cannot save more characters
    raise literal_too_long_error;
  end if;
  f.literal_size := compiler_literal_index'succ(f.literal_size);
  f.literal(integer(f.literal_size)) := c;
end add_to_token;
 
procedure add_to_token(f: in out compiler_token_input_type;
                       u: in ucs4_character) is
  c: compiler_character;
begin
  set_ucs4_character(c, u);
  add_to_token(f, c);
end add_to_token;
 
function decimal_token_to_integer(f: in compiler_token_input_type)
                                                      return integer is
  r: integer := 0;
  c: compiler_character;
  code: ucs4_character;
begin
  -- this function should never be called if the literal is empty
  if f.literal_size = 0 then
    raise program_error;
  end if;
  -- compute the value, note that the literal is a compiler_character_string
  -- which means we need to get the ucs4_character for each character...
  for i in compiler_literal_index
             range compiler_literal_index(f.literal'first) .. f.literal_size
  loop
    c := f.literal(integer(i));
    code := get_ucs4_code(c);
    r := r * 10 + integer(code - 16#30#);
  end loop;
  return r;
end decimal_token_to_integer;
 
function is_digit(code: in ucs4_character;
                  base: in numeral_base_type) return boolean is
begin
  -- smaller bases means less digits are allowed
  if base <= 10 then
    return code in 16#30# .. ucs4_character(16#30# + base - 1);
  end if;
 
  -- regular digits
  if code in 16#30# .. 16#39# then
    return true;
  end if;
 
  -- extended digits
  return code in 16#41# .. ucs4_character(16#41# + base - 11)   -- uppercase
      or code in 16#61# .. ucs4_character(16#61# + base - 11);  -- lowercase
end is_digit;
 
procedure get_numeral(f: in out compiler_token_input_type;
                      ct: out compiler_token_character;
                      code: out ucs4_character;
                      base: in numeral_base_type) is
  last_was_underscore: boolean;
  status: character_status;
begin
  last_was_underscore := false;
  loop
    get_character(f, ct);
    status := get_status(ct.c);
    case status is
    when aada_compiler_character_package.eol
       | aada_compiler_character_package.eof => -- skip those
      -- there is nothing more we can do here on this one
      code := 16#FFFF#; -- represents EOF/EOL here
 
    when aada_compiler_character_package.valid
       | aada_compiler_character_package.unnormalized =>
      -- check the character
      code := get_ucs4_code(ct.c);
      if code = 16#5F# then -- '_'
        if last_was_underscore then
          -- two underscores in a row is not legal
          aada_compiler_error_package.error(f.compiler_error,
              aada_compiler_error_package.two_underscores_error,
              ct.position,
              "found two underscores in a row in a numeral");
        end if;
        last_was_underscore := true;
      else
        -- check whether code is a valid digit (according to 'base')
        -- '0' .. '9' | 'A' .. 'Z' | 'a' .. 'z'
        exit when not is_digit(code, base);
 
        -- append the next digit
        add_to_token(f, ct.c);
        last_was_underscore := false;
      end if;
 
    when others => -- should never happen (character package raised an error!)
      raise program_error;
 
    end case;
  end loop;
 
  -- verify that we did not end with an underscore
  if last_was_underscore then
    aada_compiler_error_package.error(f.compiler_error,
        aada_compiler_error_package.invalid_underscore_error,
        ct.position,
        "found an illegal underscore at the end of a numeral");
    last_was_underscore := false;
  end if;
end get_numeral;
 
procedure get_number(f: in out compiler_token_input_type;
                     code: in ucs4_character) is
  base: numeral_base_type := 10;
  int_base: integer;
  ct, ct2: compiler_token_character;
  status: character_status;
  subcode: ucs4_character;
begin
  -- case when a period was found first?
  if code = 16#2E# then
    -- prepend a zero before the period
    set_token(f, token_float, 16#30#);
    add_to_token(f, 16#2E#);
    subcode := code;
  else
    -- this is a number, we first read all the digits [0-9_]+
    -- and thus assume it is an integer
    set_token(f, token_integer, code);
    get_numeral(f, ct, subcode, 10);
    if subcode = 16#FFFF# then
      -- found an EOL or an EOF
      return;
    end if;
 
    -- check for a base indicated (#)
    if subcode = 16#23# then -- '#'
      int_base := decimal_token_to_integer(f);
      if int_base not in integer(numeral_base_type'first)
                      .. integer(numeral_base_type'last) then
        aada_compiler_error_package.error(f.compiler_error,
            aada_compiler_error_package.unacceptable_numeric_base_error,
            ct.position,
            "unacceptable base for a numeric");
        -- use a default that should take us through the following
        -- loop albeit with the wrong value
        int_base := 36;
      end if;
      base := numeral_base_type(int_base);
      get_character(f, ct2);
      status := get_status(ct2.c);
      case status is
      when aada_compiler_character_package.eol
         | aada_compiler_character_package.eof => -- skip those
        -- the number is invalid
        aada_compiler_error_package.error(f.compiler_error,
            aada_compiler_error_package.invalid_based_number_error,
            ct2.position,
            "based numeral missing digits after the # character");
        return;
 
      when aada_compiler_character_package.valid
         | aada_compiler_character_package.unnormalized =>
        -- check the character
        subcode := get_ucs4_code(ct2.c);
        if subcode = 16#5F# then -- '_'
          -- start underscores not allowed
          aada_compiler_error_package.error(f.compiler_error,
              aada_compiler_error_package.invalid_underscore_error,
              ct2.position,
              "underscore right at the beginning of a based number");
          get_numeral(f, ct, subcode, base);
        elsif subcode = 16#2E# then
          -- this is a floating point
          f.token.token := token_float;
          -- in case the user missed the 0 before the decimal point
          aada_compiler_error_package.error(f.compiler_error,
              aada_compiler_error_package.invalid_floating_point_error,
              ct2.position,
              "floating point starting with the decimal point");
          add_to_token(f, ct.c); -- '#'
          set_ucs4_character(ct.c, 16#30#); -- '0'
          add_to_token(f, ct.c); -- '0'
          add_to_token(f, ct2.c); -- '.'
          get_numeral(f, ct, subcode, base);
        elsif not is_digit(subcode, base) then
          case subcode is
          when 16#30# .. 16#39# | 16#41# .. 16#5A# | 16#61# .. 16#7A# =>
            -- error about digit being out of bound
            aada_compiler_error_package.error(f.compiler_error,
                aada_compiler_error_package.invalid_floating_point_error,
                ct2.position,
                "digit '" & character'val(subcode) & "' out of range for base "
              & numeral_base_type'image(base));
 
          when others =>
            -- error about invalid character inside the #...# bounds
            aada_compiler_error_package.error(f.compiler_error,
                aada_compiler_error_package.invalid_floating_point_error,
                ct2.position,
                "found a non-digit character inside a based numeral");
 
          end case;
          unget_character(f, ct2);
          return;
        else -- '0' .. '9', 'A' .. 'Z', 'a' .. 'z', limited by base
          add_to_token(f, ct.c);
          add_to_token(f, ct2.c);
          get_numeral(f, ct, subcode, base);
 
          -- test whether this is followed by a decimal point
          status := get_status(ct.c);
          case status is
          when aada_compiler_character_package.valid
             | aada_compiler_character_package.unnormalized =>
            -- check the next character for '#'
            subcode := get_ucs4_code(ct.c);
            if subcode = 16#2E# then -- '.'
              -- this is a floating point
              f.token.token := token_float;
              get_character(f, ct);
              status := get_status(ct2.c);
              case status is
              when aada_compiler_character_package.eol
                 | aada_compiler_character_package.eof => -- skip those
                -- the floating point is not valid
                aada_compiler_error_package.error(f.compiler_error,
                    aada_compiler_error_package.invalid_based_number_error,
                    ct2.position,
                    "based floating point must have at least one digit after"
                  & "the decimal point and end with #");
                -- unget_character(f, ct); -- period part of decimal number
                return;
 
              when aada_compiler_character_package.valid
                 | aada_compiler_character_package.unnormalized =>
                -- check the character
                subcode := get_ucs4_code(ct2.c);
                if subcode = 16#2E# then
                  -- okay, then we assume that this is a missing '#'!
                  f.token.token := token_integer;
                  unget_character(f, ct); -- '.'
                  unget_character(f, ct2); -- '.'
                  return;
                elsif is_digit(subcode, base) then
                  add_to_token(f, ct.c); -- '.'
                  add_to_token(f, ct2.c); -- extended digit
                  get_numeral(f, ct, subcode, base);
                else
                  -- invalid digit for this based number
                  case subcode is
                  when 16#30# .. 16#39# | 16#41# .. 16#5A# | 16#61# .. 16#7A# =>
                    -- error about digit being out of bound
                    aada_compiler_error_package.error(f.compiler_error,
                        aada_compiler_error_package.invalid_floating_point_error,
                        ct2.position,
                        "digit '" & character'val(subcode)
                      & "' out of range for base "
                      & numeral_base_type'image(base));
 
                  when others =>
                    -- error about invalid character inside the #...# bounds
                    aada_compiler_error_package.error(f.compiler_error,
                        aada_compiler_error_package.invalid_floating_point_error,
                        ct2.position,
                        "found a non-digit character inside a based numeral");
 
                  end case;
                  unget_character(f, ct2);
                  return;
                end if;
 
              when others => -- should never happen
                raise program_error;
 
              end case;
            else
              aada_compiler_error_package.error(f.compiler_error,
                  aada_compiler_error_package.invalid_based_number_error,
                  ct.position,
                  "based numbers must end with #");
            end if;
 
          when others =>
            -- do nothing here, the following code handles these errors
            null;
 
          end case;
        end if;
 
      when others => -- should never happen
        raise program_error;
 
      end case;
 
      -- test for '#'
      status := get_status(ct.c);
      case status is
      when aada_compiler_character_package.eol
         | aada_compiler_character_package.eof => -- skip those
        -- the number is invalid, missing '#'
        aada_compiler_error_package.error(f.compiler_error,
            aada_compiler_error_package.invalid_based_number_error,
            ct.position,
            "based numbers must end with #");
        return;
 
      when aada_compiler_character_package.valid
         | aada_compiler_character_package.unnormalized =>
        -- check the next character for '#'
        subcode := get_ucs4_code(ct.c);
        if subcode = 16#23# then -- '#'
          -- skip the '#' and see whether we have an 'E'xponent
          -- also avoid taking '.' as a decimal point
          get_character(f, ct);
          status := get_status(ct.c);
          case status is
          when aada_compiler_character_package.eol
             | aada_compiler_character_package.eof => -- skip those
            -- the number is valid, there is no exponent
            return;
 
          when aada_compiler_character_package.valid
             | aada_compiler_character_package.unnormalized =>
            -- check the character
            subcode := get_ucs4_code(ct.c);
            if subcode = 16#2E# then
              -- this number may be valid, but it is followed by a
              -- decimal point which is going to cause problems
              -- unless it is a double dot
              unget_character(f, ct);
              return;
            end if;
            -- FALLTHROUGH -- we allow the exponent now
 
          when others => -- should never happen
            raise program_error;
 
          end case;
        else
          aada_compiler_error_package.error(f.compiler_error,
              aada_compiler_error_package.invalid_based_number_error,
              ct.position,
              "based numbers must end with #");
        end if;
 
      when others => -- should never happen
        raise program_error;
 
      end case;
      -- FALLTHROUGH to get the exponent if there is one
    end if;
  end if;
 
  -- check for a decimal point
  if subcode = 16#2E# then -- '.'
    -- this is a floating point
    f.token.token := token_float;
    -- we don't overwrite ct since we may need it
    get_character(f, ct2);
    status := get_status(ct2.c);
    case status is
    when aada_compiler_character_package.eol
       | aada_compiler_character_package.eof => -- skip those
      -- the number is invalid
      aada_compiler_error_package.error(f.compiler_error,
          aada_compiler_error_package.invalid_floating_point_error,
          ct2.position,
          "floating point missing digits after the decimal point");
      return;
 
    when aada_compiler_character_package.valid
       | aada_compiler_character_package.unnormalized =>
      -- check the character
      subcode := get_ucs4_code(ct2.c);
      if subcode = 16#5F# then -- '_'
        -- an underscores right after the decimal pointer is not legal
        aada_compiler_error_package.error(f.compiler_error,
            aada_compiler_error_package.invalid_underscore_error,
            ct2.position,
            "found an underscore right after a decimal point");
        get_numeral(f, ct, subcode, 10);
      elsif subcode = 16#2E# then
        -- special case of a double dot right after an integer
        unget_character(f, ct2); -- '.'
        unget_character(f, ct); -- '.'
        return;
      elsif subcode = 16#45# or subcode = 16#65# then -- 'E' or 'e'
        -- this is taken as the exponent character although it is not
        -- legal here without at least one digit
        -- (note: Ada does not accept 123.func() so we're good)
        -- XXX: check next character for +, - or a digit?
        aada_compiler_error_package.error(f.compiler_error,
            aada_compiler_error_package.invalid_floating_point_error,
            ct2.position,
            "you need at least one digit after the decimal point"
          & " and before the exponent");
        -- continue to next part
      elsif subcode not in 16#30# .. 16#39# then
        aada_compiler_error_package.error(f.compiler_error,
            aada_compiler_error_package.invalid_floating_point_error,
            ct2.position,
            "floating point without digits after the decimal point");
        return;
      else -- '0' .. '9', we can read a numeral
        add_to_token(f, ct.c); -- '.'
        add_to_token(f, ct2.c); -- '0' .. '9'
        get_numeral(f, ct, subcode, 10);
      end if;
 
    when others => -- should never happen
      raise program_error;
 
    end case;
  end if;
 
  -- see whether there is an exponent
  if subcode = 16#45# or subcode = 16#65# then -- 'E' or 'e'
    -- we don't overwrite ct since we may need it
    get_character(f, ct2);
    status := get_status(ct2.c);
    case status is
    when aada_compiler_character_package.eol
       | aada_compiler_character_package.eof => -- skip those
      -- the number is invalid
      -- as far as I can tell it is not possible to have a number followed by
      -- the letter 'e' and not view that letter as part of the number; however
      -- we may want to restore the 'e'? (in case the user missed typing an
      -- operator) [unless the exponent is on the next line?!]
      -- i.e. 34.0*e was written 34.0e (oops)
      aada_compiler_error_package.error(f.compiler_error,
          aada_compiler_error_package.invalid_exponent_error,
          ct2.position,
          "invalid number followed by an E and no exponent or missing operator");
      -- unget_character(f, ct)?
      return;
 
    when aada_compiler_character_package.valid
       | aada_compiler_character_package.unnormalized =>
      -- check the character
      subcode := get_ucs4_code(ct2.c);
      if subcode = 16#5F# then -- '_'
        -- an underscore right after the decimal pointer is not legal
        aada_compiler_error_package.error(f.compiler_error,
            aada_compiler_error_package.invalid_underscore_error,
            ct2.position,
            "found an underscore right after an exponent mark (E or e)");
        add_to_token(f, ct.c);
        get_numeral(f, ct, subcode, 10);
      elsif subcode = 16#2B# or subcode = 16#2D# then
        -- + or - included, get the first digit
        add_to_token(f, ct.c);
        if subcode = 16#2D# then
          -- keep the -, forget the + altogether
          add_to_token(f, ct2.c);
        end if;
        -- get another character
        get_character(f, ct);
        status := get_status(ct.c);
        case status is
        when aada_compiler_character_package.eol
           | aada_compiler_character_package.eof => -- skip those
          -- the number is definitively invalid
          aada_compiler_error_package.error(f.compiler_error,
              aada_compiler_error_package.invalid_exponent_error,
              ct.position,
              "number followed by an E+ or E- and no exponent");
          -- restore the sign in case it was the operator?
          -- in this case it is more likely that the E was for Exponent
          -- unget_character(f, ct2);
          return;
 
        when aada_compiler_character_package.valid
           | aada_compiler_character_package.unnormalized =>
          -- check the character
          subcode := get_ucs4_code(ct.c);
          if subcode = 16#5F# then -- '_'
            -- an underscore right after the decimal pointer is not legal
            aada_compiler_error_package.error(f.compiler_error,
                aada_compiler_error_package.invalid_underscore_error,
                ct.position,
               "found an underscore right after an exponent sign (E+_ or E-_)");
            get_numeral(f, ct, subcode, 10);
          elsif subcode in 16#30# .. 16#39# then
            -- '0' .. '9', we can read a numeral
            get_numeral(f, ct, subcode, 10);
          else
            -- the number is invalid
            aada_compiler_error_package.error(f.compiler_error,
                aada_compiler_error_package.invalid_exponent_error,
                ct.position,
                "number followed by an E+ or E- and no exponent");
            -- whatever is coming up we restore it and return
            unget_character(f, ct);
            return;
          end if;
 
        when others => -- should never happen
          raise program_error;
 
        end case;
      elsif subcode in 16#30# .. 16#39# then
        -- '0' .. '9', we can read a numeral
        add_to_token(f, ct.c);
        add_to_token(f, ct2.c);
        get_numeral(f, ct, subcode, 10);
      end if;
      -- restore the look ahead, including the 'e' if there wasn't a valid
      -- number after it
      unget_character(f, ct);
    when others => -- should never happen
      raise program_error;
 
    end case;
  else
    unget_character(f, ct);
  end if;
end get_number;
 
procedure identifier_to_reserved_word(f: in out compiler_token_input_type;
                                      ascii_token: in boolean) is
  type token_list_item is
    record
      length: positive;
      word: string(1 .. 12);
    end record;
  type token_list is array(token_abort .. token_xor) of token_list_item;
  -- only latin letter currently not used in a token is J
  tokens: constant token_list := (
    token_abort        => ( 5, "ABORT       "),
    token_abs          => ( 3, "ABS         "),
    token_abstract     => ( 8, "ABSTRACT    "),
    token_accept       => ( 6, "ACCEPT      "),
    token_access       => ( 6, "ACCESS      "),
    token_aliased      => ( 7, "ALIASED     "),
    token_all          => ( 3, "ALL         "),
    token_and          => ( 3, "AND         "),
    token_array        => ( 5, "ARRAY       "),
    token_at           => ( 2, "AT          "),
    token_begin        => ( 5, "BEGIN       "),
    token_body         => ( 4, "BODY        "),
    token_case         => ( 4, "CASE        "),
    token_constant     => ( 8, "CONSTANT    "),
    token_declare      => ( 7, "DECLARE     "),
    token_delay        => ( 5, "DELAY       "),
    token_delta        => ( 5, "DELTA       "),
    token_digits       => ( 6, "DIGITS      "),
    token_do           => ( 2, "DO          "),
    token_else         => ( 4, "ELSE        "),
    token_elsif        => ( 5, "ELSIF       "),
    token_end          => ( 3, "END         "),
    token_entry        => ( 5, "ENTRY       "),
    token_exception    => ( 9, "EXCEPTION   "),
    token_exit         => ( 4, "EXIT        "),
    token_for          => ( 3, "FOR         "),
    token_function     => ( 8, "FUNCTION    "),
    token_generic      => ( 7, "GENERIC     "),
    token_goto         => ( 4, "GOTO        "),
    token_if           => ( 2, "IF          "),
    token_in           => ( 2, "IN          "),
    token_interface    => ( 9, "INTERFACE   "),
    token_is           => ( 2, "IS          "),
    token_limited      => ( 7, "LIMITED     "),
    token_loop         => ( 4, "LOOP        "),
    token_mod          => ( 3, "MOD         "),
    token_new          => ( 3, "NEW         "),
    token_not          => ( 3, "NOT         "),
    token_null         => ( 4, "NULL        "),
    token_of           => ( 2, "OF          "),
    token_or           => ( 2, "OR          "),
    token_others       => ( 6, "OTHERS      "),
    token_out          => ( 3, "OUT         "),
    token_overriding   => (10, "OVERRIDING  "),
    token_package      => ( 6, "PACKAGE     "),
    token_pragma       => ( 6, "PRAGMA      "),
    token_private      => ( 7, "PRIVATE     "),
    token_procedure    => ( 9, "PROCEDURE   "),
    token_protected    => ( 9, "PROTECTED   "),
    token_raise        => ( 5, "RAISE       "),
    token_range        => ( 5, "RANGE       "),
    token_record       => ( 6, "RECORD      "),
    token_rem          => ( 3, "REM         "),
    token_renames      => ( 7, "RENAMES     "),
    token_requeue      => ( 7, "REQUEUE     "),
    token_return       => ( 6, "RETURN      "),
    token_reverse      => ( 7, "REVERSE     "),
    token_select       => ( 6, "SELECT      "),
    token_separate     => ( 8, "SEPARATE    "),
    token_subtype      => ( 7, "SUBTYPE     "),
    token_synchronized => (12, "SYNCHRONIZED"),
    token_tagged       => ( 6, "TAGGED      "),
    token_task         => ( 4, "TASK        "),
    token_terminate    => ( 9, "TERMINATE   "),
    token_then         => ( 4, "THEN        "),
    token_type         => ( 4, "TYPE        "),
    token_until        => ( 5, "UNTIL       "),
    token_use          => ( 3, "USE         "),
    token_when         => ( 4, "WHEN        "),
    token_while        => ( 5, "WHILE       "),
    token_with         => ( 4, "WITH        "),
    token_xor          => ( 3, "XOR         ")
  );
  ch: character;
  l: natural;
begin
  -- get the uppercase version (necessary to compare two identifiers
  -- so we cache it in the token)
  get_uppercase(f.literal(integer(f.literal'first) .. integer(f.literal_size)),
                f.token.uppercase);
 
  -- get the string and check the size
  l := aada_vstrings.length(f.token.uppercase);
  if l < 2 or else l > 12 then
    -- longest reserved word is SYNCHRONIZED (12 chars)
    -- smallests (AT, DO, IF, IN, IS, OF, OR) are 2
    -- too small or large to be a token, return immediately
    return;
  end if;
 
  -- check each character; if not ASCII, return immediately
  -- (this is partially a left over since before I would copy the string)
  for i in positive range 1 .. l
  loop
    ch := aada_vstrings.get(f.token.uppercase, i);
    if ch not in 'A' .. 'Z' or else ch = 'J' then
      -- 'J' is not currently used by any reserved keyword!
      --
      -- ignore any identifier that is not exclusively composed
      -- of ASCII letters as expected in a keyword
      return;
    end if;
  end loop;
 
  -- check whether this is a reserved word
  -- TODO: use a binary search! (we can easily check just the 1st char.)
  for j in token_list'range
  loop
    if l = tokens(j).length
    and then aada_vstrings.str(f.token.uppercase) = tokens(j).word then
      if not ascii_token then
        -- this is not a valid reserved keyword
        -- (i.e. ACCESS with a lowercase German Sharp S or "Szet")
        aada_compiler_error_package.error(f.compiler_error,
            aada_compiler_error_package.illegal_identifier_error,
            f.token.position,
            "identifier """ & aada_vstrings.str(f.token.uppercase)
          & """ matches a reserved word but is not exclusively composed"
          & " of the letters A to Z, it is considered illegal.");
        -- keep it as an identifier and return
      else
        -- valid reserved word!
        f.token.token := j;
      end if;
      return;
    end if;
  end loop;
end identifier_to_reserved_word;
 
procedure get_token_literal(f: in out compiler_token_input_type) is
  status: character_status;
  ct, ct2: compiler_token_character;
  code: ucs4_character;
  category: character_category;
  last_character_category: character_category;
  ascii_token: boolean;
begin
  loop
    f.literal_size := 0;
    get_character(f, ct);
    f.token.position := ct.position;
    status := get_status(ct.c);
    case status is
    when aada_compiler_character_package.eol => -- skip those
      null;
 
    when aada_compiler_character_package.eof => -- we're done here
      -- the character is ignored in this call
      set_token(f, token_eof, ct.c);
      return;
 
    when aada_compiler_character_package.valid
       | aada_compiler_character_package.unnormalized =>
      code := get_ucs4_code(ct.c);
      case code is
      when 16#01# .. 16#1F# -- SOH to US
         | 16#80# .. 16#9F# -- Reserved_128 to APC
        =>
        -- generate an error about controls and go on
        -- (note that controls that we must handle were
        -- handled in the get_character() call)
        aada_compiler_error_package.error(f.compiler_error,
            aada_compiler_error_package.unexpected_character_error,
            ct.position,
            "illegal control character in input stream");
 
      when 16#0020# -- space
         | 16#00A0# -- no-break space
         | 16#1680# -- ogham space mark
         | 16#180E# -- mongolian vowel separator
         | 16#2000# -- en quad
         | 16#2001# -- em quad
         | 16#2002# -- en space
         | 16#2003# -- em space
         | 16#2004# -- three-per-em space
         | 16#2005# -- four-per-em space
         | 16#2006# -- six-per-em space
         | 16#2007# -- figure space
         | 16#2008# -- punctuation space
         | 16#2009# -- thin space
         | 16#200A# -- hair space
         | 16#202F# -- narrow no-break space
         | 16#205F# -- medium mathematical space
         | 16#3000# -- ideographic space
         =>
        -- ignore all blanks
        null;
 
      when 16#23# => -- '#'
        -- check for a digit, if there is one right after the '#'
        -- then we assume the base is missing; otherwise view the
        -- '#' as an invalid comment introducer
        get_character(f, ct2);
        status := get_status(ct2.c);
        if status /= aada_compiler_character_package.eol
        and then status /= aada_compiler_character_package.eof then
          if is_digit(get_ucs4_code(ct2.c), 36) then
            -- okay, so this is a digit, assume the base is missing
            aada_compiler_error_package.error(f.compiler_error,
                aada_compiler_error_package.unexpected_character_error,
                ct.position,
                "found a lone # character, assuming this is a number"
              & " and the base is missing");
            unget_character(f, ct2);
            unget_character(f, ct);
            set_ucs4_character(ct.c, 16#36#);
            unget_character(f, ct);
            get_number(f, 16#33#);
            return;
          end if;
        end if;
        -- '# ...': we assume we found a comment, skip it
        aada_compiler_error_package.error(f.compiler_error,
            aada_compiler_error_package.unexpected_character_error,
            ct.position,
            "Ada does not support comments introduced with a # character;"
          & " skipping assuming it is a comment");
        status := get_status(ct2.c);
        loop
          exit when status = aada_compiler_character_package.eol
            or else status = aada_compiler_character_package.eof;
          get_character(f, ct);
          status := get_status(ct.c);
        end loop;
        -- comments are completely ignored like blanks
        -- FALLTHROUGH
 
      when 16#26# => -- '&'
        set_token(f, token_and, ct.c);
        return;
 
      when 16#27# => -- '''
        set_token(f, token_apostrophe, ct.c);
        return;
 
      when 16#28# => -- '('
        set_token(f, token_open_parenthesis, ct.c);
 
        -- check for '(*'
        get_character(f, ct2);
        status := get_status(ct2.c);
        if status /= aada_compiler_character_package.eol
        and then status /= aada_compiler_character_package.eof then
          if get_ucs4_code(ct2.c) = 16#2A# then
            aada_compiler_error_package.error(f.compiler_error,
                aada_compiler_error_package.unexpected_character_error,
                ct.position,
                "Pascal comments are not allowed in Ada");
          end if;
        else
          unget_character(f, ct2);
        end if;
        return;
 
      when 16#29# => -- ')'
        set_token(f, token_close_parenthesis, ct.c);
        return;
 
      when 16#2A# => -- '*'
        get_character(f, ct2);
        status := get_status(ct2.c);
        if status = aada_compiler_character_package.eol
        or else status = aada_compiler_character_package.eof then
          -- '*': just before the end of a line or end of the file
          set_token(f, token_multiply, ct.c);
        elsif get_ucs4_code(ct.c) = 16#2A# then
          -- '**': exponential
          set_token(f, token_exponential, ct.c);
          add_to_token(f, ct2.c);
        else
          -- '*': this is not exponential
          unget_character(f, ct2);
          set_token(f, token_multiply, ct.c);
        end if;
        return;
 
      when 16#2B# => -- '+'
        set_token(f, token_plus, ct.c);
        return;
 
      when 16#2C# => -- ','
        set_token(f, token_comma, ct.c);
        return;
 
      when 16#2D# => -- '-'
        get_character(f, ct2);
        status := get_status(ct2.c);
        if status = aada_compiler_character_package.eol
        or else status = aada_compiler_character_package.eof then
          -- '-': just before the end of a line or end of the file
          set_token(f, token_minus, ct.c);
          return;
        end if;
        if get_ucs4_code(ct2.c) = 16#2D# then
          -- '--': we found a comment, skip it
          loop
            get_character(f, ct);
            status := get_status(ct.c);
            exit when status = aada_compiler_character_package.eol
              or else status = aada_compiler_character_package.eof;
          end loop;
          -- comments are completely ignored like blanks
          -- FALLTHROUGH
        else
          -- '-': this is not a comment, restore the last get_character()
          unget_character(f, ct2);
          set_token(f, token_minus, ct.c);
          return;
        end if;
 
      when 16#2E# => -- '.'
        get_character(f, ct2);
        status := get_status(ct2.c);
        if status = aada_compiler_character_package.eol
        or else status = aada_compiler_character_package.eof then
          -- '.': just before the end of a line or end of the file
          set_token(f, token_divide, ct.c);
        elsif get_ucs4_code(ct2.c) = 16#2E# then
          -- '..': the double dot
          set_token(f, token_double_dot, ct.c);
          add_to_token(f, ct2.c);
        elsif get_ucs4_code(ct2.c) in 16#30# .. 16#39# then -- '0' .. '9'
          -- this is a digit! a period followed by a digit is most certainly
          -- a floating point with a missing 0 before the period
          aada_compiler_error_package.error(f.compiler_error,
              aada_compiler_error_package.invalid_floating_point_error,
              ct.position,
              "floating point numerals must start with a digit,"
            & " not a decimal point");
          -- transform into a valid floating point
          unget_character(f, ct2); -- put the digit back on the stack
          get_number(f, code);
        else
          -- '.': this is not the double dot
          unget_character(f, ct2);
          set_token(f, token_period, ct.c);
        end if;
        return;
 
      when 16#2F# => -- '/'
        get_character(f, ct2);
        status := get_status(ct2.c);
        if status = aada_compiler_character_package.eol
        or else status = aada_compiler_character_package.eof then
          -- '/': just before the end of a line or end of the file
          set_token(f, token_divide, ct.c);
          return;
        elsif get_ucs4_code(ct2.c) = 16#3D# then
          -- '/=': the inequality operator
          set_token(f, token_inequality, ct.c);
          add_to_token(f, ct2.c);
          return;
        elsif get_ucs4_code(ct2.c) = 16#2F# then
          -- '//': a C/C++ comment?
          aada_compiler_error_package.error(f.compiler_error,
              aada_compiler_error_package.invalid_floating_point_error,
              ct.position,
              "C/C++ comments are not allowed in ada,"
            & " skipping up to the end of the line");
          loop
            get_character(f, ct);
            status := get_status(ct.c);
            exit when status = aada_compiler_character_package.eol
              or else status = aada_compiler_character_package.eof;
          end loop;
          -- comments are completely ignored like blanks
          -- FALLTHROUGH
        elsif get_ucs4_code(ct2.c) = 16#2A# then
          -- '/* ...': a C/C++ comment introducer?
          aada_compiler_error_package.error(f.compiler_error,
              aada_compiler_error_package.invalid_floating_point_error,
              ct.position,
              """/*"" looks like a C/C++ comments introducer,"
            & " these are not allowed in Ada; ignored");
          -- if it was a comment introducer, we're in "trouble"
          -- anyway... (but since Ada does not support those,
          -- skipping to the next */ is not sensible)
          -- FALLTHROUGH
        else
          -- '/': this is not the inequality operator
          unget_character(f, ct2);
          set_token(f, token_divide, ct.c);
          return;
        end if;
 
      when 16#30# .. 16#39# => -- '0' .. '9'
        get_number(f, code);
        return;
 
      when 16#3A# => -- ':'
        get_character(f, ct2);
        status := get_status(ct2.c);
        if status = aada_compiler_character_package.eol
        or else status = aada_compiler_character_package.eof then
          -- ':': just before the end of a line or end of the file
          set_token(f, token_colon, ct.c);
        elsif get_ucs4_code(ct2.c) = 16#3D# then
          -- ':=': the inequality operator
          set_token(f, token_assignment, ct.c);
          add_to_token(f, ct2.c);
        else
          -- ':': this is not the assignment operator
          unget_character(f, ct2);
          set_token(f, token_colon, ct.c);
        end if;
        return;
 
      when 16#3B# => -- ';'
        set_token(f, token_semi_colon, ct.c);
        return;
 
      when 16#3C# => -- '<'
        get_character(f, ct2);
        status := get_status(ct2.c);
        if status = aada_compiler_character_package.eol
        or else status = aada_compiler_character_package.eof then
          -- '<': just before the end of a line or end of the file
          set_token(f, token_less, ct.c);
        else
          case get_ucs4_code(ct2.c) is
          when 16#3C# =>
            -- '<<': the left label mark
            set_token(f, token_left_label, ct.c);
            add_to_token(f, ct2.c);
          when 16#3D# =>
            -- '<=': the inequality operator
            set_token(f, token_less_equal, ct.c);
            add_to_token(f, ct2.c);
          when 16#3E# =>
            -- '<>': the box
            set_token(f, token_box, ct.c);
            add_to_token(f, ct2.c);
          when others =>
            -- '<': this is just less than
            unget_character(f, ct2);
            set_token(f, token_less, ct.c);
          end case;
        end if;
        return;
 
      when 16#3D# => -- '='
        get_character(f, ct2);
        status := get_status(ct2.c);
        if status = aada_compiler_character_package.eol
        or else status = aada_compiler_character_package.eof then
          -- '=': just before the end of a line or end of the file
          set_token(f, token_equal, ct.c);
        elsif get_ucs4_code(ct2.c) = 16#3E# then
          -- '=>': the arrow
          set_token(f, token_arrow, ct.c);
          add_to_token(f, ct2.c);
        else
          -- '=': this is not the assignment operator
          unget_character(f, ct2);
          set_token(f, token_equal, ct.c);
        end if;
        return;
 
      when 16#3E# => -- '>'
        get_character(f, ct2);
        status := get_status(ct2.c);
        if status = aada_compiler_character_package.eol
        or else status = aada_compiler_character_package.eof then
          -- '>': just before the end of a line or end of the file
          set_token(f, token_greater, ct.c);
        else
          case get_ucs4_code(ct2.c) is
          when 16#3E# =>
            -- '>>': the right label mark
            set_token(f, token_right_label, ct.c);
            add_to_token(f, ct2.c);
          when 16#3D# =>
            -- '>=': the greater than or equal
            set_token(f, token_greater_equal, ct.c);
            add_to_token(f, ct2.c);
          when others =>
            -- '>': this is just greater than
            unget_character(f, ct2);
            set_token(f, token_greater, ct.c);
          end case;
        end if;
        return;
 
      when 16#5F# =>
        aada_compiler_error_package.error(f.compiler_error,
            aada_compiler_error_package.unexpected_underscore_error,
            ct.position,
            "underscores cannot appear as the first character of a numeral,"
          & "an identifier, or by itself");
        -- FALLTHROUGH -- we just skip this character!
 
      when 16#7B# => -- '{'
        aada_compiler_error_package.error(f.compiler_error,
            aada_compiler_error_package.unexpected_character_error,
            ct.position,
            "to open a block use ""begin"" instead of {");
        return;
 
      when 16#7C# => -- '|'
        set_token(f, token_or, ct.c);
        return;
 
      when 16#D800# .. 16#DFFF# -- surrogates
        | 16#E000# .. 16#F8FF# -- private use
        | 16#F0000# .. 16#FFFFD#
        | 16#100000# .. 16#10FFFD# =>
        -- Other surrogate characters are forbidden
        aada_compiler_error_package.error(f.compiler_error,
            aada_compiler_error_package.unexpected_character_error,
            ct.position,
            "an unexpected character was"
          & " encountered outside of a comment");
 
      when others =>
        get_category(ct.c, category);
        case category is
        when unknown | invalid =>
          -- TODO: determine whether 'unknown' is a program error or input error
          -- 'invalid' should all be caught outside of this
          -- case so if we detect it here we have a problem
          raise program_error;
 
        when other | punctuation | format =>
          -- TODO: generate an error
          aada_compiler_error_package.error(f.compiler_error,
              aada_compiler_error_package.unexpected_character_error,
              ct.position,
              "an unexpected (formatting, punctuation) character was"
            & " encountered outside of an identifier or literal");
          -- in this case we continue
 
        when letter =>
          -- found an identifier
          set_token(f, token_identifier, ct.c);
          ascii_token := is_ascii(ct.c);
          last_character_category := letter;
          loop
            get_character(f, ct);
            status := get_status(ct.c);
            exit when status = aada_compiler_character_package.eol
              or else status = aada_compiler_character_package.eof;
            get_category(ct.c, category);
            if category not in letter .. format then
              -- restore this character for next round
              unget_character(f, ct);
              exit;
            end if;
            if ascii_token and then not is_ascii(ct.c) then
              ascii_token := false;
            end if;
            -- for this test, ignore formatting characters
            if category /= format then
              if last_character_category = punctuation
              and then category = punctuation then
                aada_compiler_error_package.error(f.compiler_error,
                    aada_compiler_error_package.two_punctuation_error,
                    ct.position,
                    "found two punctuation characters"
                  & " in a row in an identifier");
              end if;
              last_character_category := category;
            end if;
            add_to_token(f, ct.c);
          end loop;
          if last_character_category = punctuation then
            aada_compiler_error_package.error(f.compiler_error,
                aada_compiler_error_package.illegal_punctuation_error,
                ct.position,
                "found an illegal punctuation character"
              & " at the end of an identifier");
          end if;
          -- check whether this identifier is a reserved word
          identifier_to_reserved_word(f, ascii_token);
          return;
 
        end case;
      end case;
 
    when others =>
      -- should not happen as the character package should raise an error
      raise program_error;
 
    end case;
  end loop;
end get_token_literal;
 
procedure get_token(f: in out compiler_token_input_type) is
begin
  get_token_literal(f);
  if f.literal_size > 0 then
    -- copy the literal in the token if any
    -- (this is dynamically copied to reduce the size as much as possible)
    get_utf8_string(f.literal(integer(f.literal'first)
                           .. integer(f.literal_size)),
                f.token.literal);
  else
    -- reset to empty on eof (necessary?!)
    aada_vstrings.vstr("", f.token.literal);
  end if;
end get_token;
 
procedure next_token(f: in out compiler_token_input_type;
                     t: out compiler_token_type) is
begin
  -- is there a look-ahead token? if so, return it instead
  if f.look_ahead.token /= token_nil then
    t := f.look_ahead;
    f.look_ahead.token := token_nil;
    return;
  end if;
  loop
    get_token(f);
    t := f.token;
    -- although at this time we should never return a NIL token
    -- we do the test here, just in case we change our mind
    exit when t.token /= token_nil;
  end loop;
end next_token;
 
procedure unget_token(f: in out compiler_token_input_type;
                      t: in compiler_token_type) is
begin
  -- TBD: we may want to have a stack instead of just one item
  if f.look_ahead.token /= token_nil then
    raise token_look_ahead_buffer_full;
  end if;
  f.look_ahead := t;
end unget_token;
 
function get_token_token(t: in compiler_token_type) return ada_token is
begin
  return t.token;
end get_token_token;
 
procedure set_token_token(t: in out compiler_token_type; token: ada_token) is
begin
  -- to make the change secure we avoid changing tokens at will
  case token is
  when token_null_record =>
    if t.token = token_record then
      t.token := token;
      return;
    end if;
 
  when others => null;
  end case;
 
  -- any other case we refuse to change the token
  raise token_change_not_allowed;
end set_token_token;
 
function get_token_position(t: in compiler_token_type)
                    return aada_compiler_token_position.token_position_type is
begin
  return t.position;
end get_token_position;
 
procedure get_token_operator(--f: in out compiler_token_input_type;
                             t: in out compiler_token_type;
                             token: out ada_token) is
begin
  --
  -- a string token can represent an operator; this is true
  -- in a function declaration:
  --    function "**" (a, b: this_type) return that_type;
  -- and also in an expression as in:
  --    a := "**" (b, c);
  -- plus it is possible to qualify the operator in an expression
  --    a := b."**" (c, d);
  --
  if t.operator = token_unknown then
    -- check whether this token can represent an operator
    if t.token /= token_string then
      -- only strings can be used to this effect
      -- we should never be called otherwise!!!
      raise program_error;
    elsif t.uppercase = "&" then
      t.operator := token_and;
    elsif t.uppercase = "*" then
      t.operator := token_multiply;
    elsif t.uppercase = "**" then
      t.operator := token_exponential;
    elsif t.uppercase = "+" then
      t.operator := token_plus;
    elsif t.uppercase = "-" then
      t.operator := token_minus;
    elsif t.uppercase = "/" then
      t.operator := token_divide;
    elsif t.uppercase = "/=" then
      t.operator := token_inequality;
    elsif t.uppercase = "<" then
      t.operator := token_less;
    elsif t.uppercase = "<=" then
      t.operator := token_less_equal;
    elsif t.uppercase = "=" then
      t.operator := token_equal;
    elsif t.uppercase = ">" then
      t.operator := token_greater;
    elsif t.uppercase = ">=" then
      t.operator := token_greater_equal;
    elsif t.uppercase = "ABS" then
      t.operator := token_abs;
    elsif t.uppercase = "AND" then
      t.operator := token_and;
    elsif t.uppercase = "MOD" then
      t.operator := token_mod;
    elsif t.uppercase = "NOT" then
      t.operator := token_not;
    elsif t.uppercase = "OR" then
      t.operator := token_or;
    elsif t.uppercase = "REM" then
      t.operator := token_rem;
    elsif t.uppercase = "XOR" then
      t.operator := token_xor;
    else
      -- none of the 19 allowed operators
      -- TODO: try a "trim()" of the keyword and check again
      --       so we can err intelligently in the event the
      --       user added spaces
      t.operator := token_nil;
      --aada_compiler_error_package.error(f.compiler_error,
      --    aada_compiler_error_package.unknown_operator_error,
      --    t.position,
      --    "an operator was expected; """ & aada_vstrings.str(t.literal)
      --                                      & """ was found instead");
    end if;
  end if;
  token := t.operator;
end get_token_operator;
 
function get_token_literal(t: in compiler_token_type)
                      return aada_vstrings.vstring is
begin
  return t.literal;
end get_token_literal;
 
function get_token_uppercase(t: in compiler_token_type)
                      return aada_vstrings.vstring is
begin
  return t.uppercase;
end get_token_uppercase;
 
 
end aada_compiler_token_package;
 
-- vim: ts=2 sw=2 et syntax=ada
Project aada v1.0-338 (Project id #3)
Process Done (Last compiled on 2012/01/13 01:21:26)
Description Alexis Ada Compiler written in Ada (my first attempt was in C++ which is not correct for an Ada compiler.)
Syndicate content