summaryrefslogtreecommitdiffstats
path: root/generic/tclParse.c
blob: b863ff276651fe9abe60d81dda81409eed77b120 (plain)
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
1642
1643
1644
1645
1646
1647
1648
1649
1650
1651
1652
1653
1654
1655
1656
1657
1658
1659
1660
1661
1662
1663
1664
1665
1666
1667
1668
1669
1670
1671
1672
1673
1674
1675
1676
1677
1678
1679
1680
1681
1682
1683
1684
1685
1686
1687
1688
1689
1690
1691
1692
1693
1694
1695
1696
1697
1698
1699
1700
1701
1702
1703
1704
1705
1706
1707
1708
1709
1710
1711
1712
1713
1714
1715
1716
1717
1718
1719
1720
1721
1722
1723
1724
1725
1726
1727
1728
1729
1730
1731
1732
1733
1734
1735
1736
1737
1738
1739
1740
1741
1742
1743
1744
1745
1746
1747
1748
1749
1750
1751
1752
1753
1754
1755
1756
1757
1758
1759
1760
1761
1762
1763
1764
1765
1766
1767
1768
1769
1770
1771
1772
1773
1774
1775
1776
1777
1778
1779
1780
1781
1782
1783
1784
1785
1786
1787
1788
1789
1790
1791
1792
1793
1794
1795
1796
1797
1798
1799
1800
1801
1802
1803
1804
1805
1806
1807
1808
1809
1810
1811
1812
1813
1814
1815
1816
1817
1818
1819
1820
1821
1822
1823
1824
1825
1826
1827
1828
1829
1830
1831
1832
1833
1834
1835
1836
1837
1838
1839
1840
1841
1842
1843
1844
1845
1846
1847
1848
1849
1850
1851
1852
1853
1854
1855
1856
1857
1858
1859
1860
1861
1862
1863
1864
1865
1866
1867
1868
1869
1870
1871
1872
1873
1874
1875
1876
1877
1878
1879
1880
1881
1882
1883
1884
1885
1886
1887
1888
1889
1890
1891
1892
1893
1894
1895
1896
1897
1898
1899
1900
1901
1902
1903
1904
1905
1906
1907
1908
1909
1910
1911
1912
1913
1914
1915
1916
1917
1918
1919
1920
1921
1922
1923
1924
1925
1926
1927
1928
1929
1930
1931
1932
1933
1934
1935
1936
1937
1938
1939
1940
1941
1942
1943
1944
1945
1946
1947
1948
1949
1950
1951
1952
1953
1954
1955
1956
1957
1958
1959
1960
1961
1962
1963
1964
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
2036
2037
2038
2039
2040
2041
2042
2043
2044
2045
2046
2047
2048
2049
2050
2051
2052
2053
2054
2055
2056
2057
2058
2059
2060
2061
2062
2063
2064
2065
2066
2067
2068
2069
2070
2071
2072
2073
2074
2075
2076
2077
2078
2079
2080
2081
2082
2083
2084
2085
2086
2087
2088
2089
2090
2091
2092
2093
2094
2095
2096
2097
2098
2099
2100
2101
2102
2103
2104
2105
2106
2107
2108
2109
2110
2111
2112
2113
2114
2115
2116
2117
2118
2119
2120
2121
2122
2123
2124
2125
2126
2127
2128
2129
2130
2131
2132
2133
2134
2135
2136
2137
2138
2139
2140
2141
2142
2143
2144
2145
2146
2147
2148
2149
2150
2151
2152
2153
2154
2155
2156
2157
2158
2159
2160
2161
2162
2163
2164
2165
2166
2167
2168
2169
2170
2171
2172
2173
2174
2175
2176
2177
2178
2179
2180
2181
2182
2183
2184
2185
2186
2187
2188
2189
2190
2191
2192
2193
2194
2195
2196
2197
2198
2199
2200
2201
2202
2203
2204
2205
2206
2207
2208
2209
2210
2211
2212
2213
2214
2215
2216
2217
2218
2219
2220
2221
2222
2223
2224
2225
2226
2227
2228
2229
2230
2231
2232
2233
2234
2235
2236
2237
2238
2239
2240
2241
2242
2243
2244
2245
2246
2247
2248
2249
2250
2251
2252
2253
2254
2255
2256
2257
2258
2259
2260
2261
2262
2263
2264
2265
2266
2267
2268
2269
2270
2271
2272
2273
2274
2275
2276
2277
2278
2279
2280
2281
2282
2283
2284
2285
2286
2287
2288
2289
2290
2291
2292
2293
2294
2295
2296
2297
2298
2299
2300
2301
2302
2303
2304
2305
2306
2307
2308
2309
2310
2311
2312
2313
2314
2315
2316
2317
2318
2319
2320
2321
2322
2323
2324
2325
2326
2327
2328
2329
2330
2331
2332
2333
2334
2335
2336
2337
2338
2339
2340
2341
2342
2343
2344
2345
2346
2347
2348
2349
2350
2351
2352
2353
2354
2355
2356
2357
2358
2359
2360
2361
2362
2363
2364
2365
2366
2367
2368
2369
2370
2371
2372
2373
2374
2375
2376
2377
2378
2379
2380
2381
2382
2383
2384
2385
2386
2387
2388
2389
2390
2391
2392
2393
2394
2395
2396
2397
2398
2399
2400
2401
2402
2403
2404
2405
2406
2407
2408
2409
2410
2411
2412
2413
2414
2415
2416
2417
2418
2419
2420
2421
2422
2423
2424
2425
2426
2427
2428
2429
2430
2431
2432
2433
2434
2435
2436
2437
2438
2439
2440
2441
2442
2443
2444
2445
2446
2447
2448
2449
2450
2451
2452
2453
2454
2455
2456
2457
2458
2459
2460
2461
2462
2463
2464
2465
2466
2467
2468
2469
2470
2471
2472
2473
2474
2475
2476
2477
2478
2479
2480
2481
2482
2483
2484
2485
2486
2487
2488
/*
 * tclParse.c --
 *
 *	This file contains functions that parse Tcl scripts. They do so in a
 *	general-purpose fashion that can be used for many different purposes,
 *	including compilation, direct execution, code analysis, etc.
 *
 * Copyright (c) 1997 Sun Microsystems, Inc.
 * Copyright (c) 1998-2000 Ajuba Solutions.
 * Contributions from Don Porter, NIST, 2002. (not subject to US copyright)
 *
 * See the file "license.terms" for information on usage and redistribution of
 * this file, and for a DISCLAIMER OF ALL WARRANTIES.
 */

#include "tclInt.h"
#include "tclParse.h"
#include <assert.h>

/*
 * The following table provides parsing information about each possible 8-bit
 * character. The table is designed to be referenced with unsigned characters.
 *
 * The macro CHAR_TYPE is used to index into the table and return information
 * about its character argument. The following return values are defined.
 *
 * TYPE_NORMAL -	All characters that don't have special significance to
 *			the Tcl parser.
 * TYPE_SPACE -		The character is a whitespace character other than
 *			newline.
 * TYPE_COMMAND_END -	Character is newline or semicolon.
 * TYPE_SUBS -		Character begins a substitution or has other special
 *			meaning in ParseTokens: backslash, dollar sign, or
 *			open bracket.
 * TYPE_QUOTE -		Character is a double quote.
 * TYPE_CLOSE_PAREN -	Character is a right parenthesis.
 * TYPE_CLOSE_BRACK -	Character is a right square bracket.
 * TYPE_BRACE -		Character is a curly brace (either left or right).
 */

const char tclCharTypeTable[] = {

    /*
     * Positive character values, from 0-127:
     */

    TYPE_SUBS,        TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,
    TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,
    TYPE_NORMAL,      TYPE_SPACE,       TYPE_COMMAND_END, TYPE_SPACE,
    TYPE_SPACE,       TYPE_SPACE,       TYPE_NORMAL,      TYPE_NORMAL,
    TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,
    TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,
    TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,
    TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,
    TYPE_SPACE,       TYPE_NORMAL,      TYPE_QUOTE,       TYPE_NORMAL,
    TYPE_SUBS,        TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,
    TYPE_NORMAL,      TYPE_CLOSE_PAREN, TYPE_NORMAL,      TYPE_NORMAL,
    TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,
    TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,
    TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,
    TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,      TYPE_COMMAND_END,
    TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,
    TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,
    TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,
    TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,
    TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,
    TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,
    TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,
    TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,      TYPE_SUBS,
    TYPE_SUBS,        TYPE_CLOSE_BRACK, TYPE_NORMAL,      TYPE_NORMAL,
    TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,
    TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,
    TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,
    TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,
    TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,
    TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,
    TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,      TYPE_BRACE,
    TYPE_NORMAL,      TYPE_BRACE,       TYPE_NORMAL,      TYPE_NORMAL,

    /*
     * Large unsigned character values, from 128-255:
     */

    TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,
    TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,
    TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,
    TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,
    TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,
    TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,
    TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,
    TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,
    TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,
    TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,
    TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,
    TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,
    TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,
    TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,
    TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,
    TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,
    TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,
    TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,
    TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,
    TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,
    TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,
    TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,
    TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,
    TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,
    TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,
    TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,
    TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,
    TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,
    TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,
    TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,
    TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,
    TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,      TYPE_NORMAL,
};

/*
 * Prototypes for local functions defined in this file:
 */

static int	CommandComplete(const char *script, int numBytes);
static int		ParseComment(const char *src, int numBytes,
			    Tcl_Parse *parsePtr);
static int		ParseTokens(const char *src, int numBytes, int mask,
			    int flags, Tcl_Parse *parsePtr);
static int		ParseWhiteSpace(const char *src, int numBytes,
			    int *incompletePtr, char *typePtr);
static int		ParseAllWhiteSpace(const char *src, int numBytes,
			    int *incompletePtr);
static int		ParseHex(const char *src, int numBytes,
			    int *resultPtr);

/*
 *----------------------------------------------------------------------
 *
 * TclParseInit --
 *
 *	Initialize the fields of a Tcl_Parse struct.
 *
 * Results:
 *	None.
 *
 * Side effects:
 *	The Tcl_Parse struct pointed to by parsePtr gets initialized.
 *
 *----------------------------------------------------------------------
 */

void
TclParseInit(
    Tcl_Interp *interp,		/* Interpreter to use for error reporting */
    const char *start,		/* Start of string to be parsed. */
    int numBytes,		/* Total number of bytes in string. If < 0,
				 * the script consists of all bytes up to the
				 * first null character. */
    Tcl_Parse *parsePtr)	/* Points to struct to initialize */
{
    parsePtr->numWords = 0;
    parsePtr->tokenPtr = parsePtr->staticTokens;
    parsePtr->numTokens = 0;
    parsePtr->tokensAvailable = NUM_STATIC_TOKENS;
    parsePtr->string = start;
    parsePtr->end = start + numBytes;
    parsePtr->term = parsePtr->end;
    parsePtr->interp = interp;
    parsePtr->incomplete = 0;
    parsePtr->errorType = TCL_PARSE_SUCCESS;
}

/*
 *----------------------------------------------------------------------
 *
 * Tcl_ParseCommand --
 *
 *	Given a string, this function parses the first Tcl command in the
 *	string and returns information about the structure of the command.
 *
 * Results:
 *	The return value is TCL_OK if the command was parsed successfully and
 *	TCL_ERROR otherwise. If an error occurs and interp isn't NULL then an
 *	error message is left in its result. On a successful return, parsePtr
 *	is filled in with information about the command that was parsed.
 *
 * Side effects:
 *	If there is insufficient space in parsePtr to hold all the information
 *	about the command, then additional space is malloc-ed. If the function
 *	returns TCL_OK then the caller must eventually invoke Tcl_FreeParse to
 *	release any additional space that was allocated.
 *
 *----------------------------------------------------------------------
 */

int
Tcl_ParseCommand(
    Tcl_Interp *interp,		/* Interpreter to use for error reporting; if
				 * NULL, then no error message is provided. */
    const char *start,		/* First character of string containing one or
				 * more Tcl commands. */
    int numBytes,	/* Total number of bytes in string. If < 0,
				 * the script consists of all bytes up to the
				 * first null character. */
    int nested,			/* Non-zero means this is a nested command:
				 * close bracket should be considered a
				 * command terminator. If zero, then close
				 * bracket has no special meaning. */
    Tcl_Parse *parsePtr)
				/* Structure to fill in with information about
				 * the parsed command; any previous
				 * information in the structure is ignored. */
{
    const char *src;	/* Points to current character in the
				 * command. */
    char type;			/* Result returned by CHAR_TYPE(*src). */
    Tcl_Token *tokenPtr;	/* Pointer to token being filled in. */
    int wordIndex;		/* Index of word token for current word. */
    int terminators;		/* CHAR_TYPE bits that indicate the end of a
				 * command. */
    const char *termPtr;	/* Set by Tcl_ParseBraces/QuotedString to
				 * point to char after terminating one. */
    int scanned;

    if (numBytes < 0 && start) {
	numBytes = strlen(start);
    }
    TclParseInit(interp, start, numBytes, parsePtr);
    if ((start == NULL) && (numBytes != 0)) {
	if (interp != NULL) {
	    Tcl_SetObjResult(interp, Tcl_NewStringObj(
		    "can't parse a NULL pointer", -1));
	}
	return TCL_ERROR;
    }
    parsePtr->commentStart = NULL;
    parsePtr->commentSize = 0;
    parsePtr->commandStart = NULL;
    parsePtr->commandSize = 0;
    if (nested != 0) {
	terminators = TYPE_COMMAND_END | TYPE_CLOSE_BRACK;
    } else {
	terminators = TYPE_COMMAND_END;
    }

    /*
     * Parse any leading space and comments before the first word of the
     * command.
     */

    scanned = ParseComment(start, numBytes, parsePtr);
    src = (start + scanned);
    numBytes -= scanned;
    if (numBytes == 0) {
	if (nested) {
	    parsePtr->incomplete = nested;
	}
    }

    /*
     * The following loop parses the words of the command, one word in each
     * iteration through the loop.
     */

    parsePtr->commandStart = src;
    type = CHAR_TYPE(*src);
    scanned = 1;	/* Can't have missing whitepsace before first word. */
    while (1) {
	int expandWord = 0;

	/* Are we at command termination? */

	if ((numBytes == 0) || (type & terminators) != 0) {
	    parsePtr->term = src;
	    parsePtr->commandSize = src + (numBytes != 0)
		    - parsePtr->commandStart;
	    return TCL_OK;
	}

	/* Are we missing white space after previous word? */

	if (scanned == 0) {
	    if (src[-1] == '"') {
		if (interp != NULL) {
		    Tcl_SetObjResult(interp, Tcl_NewStringObj(
			    "extra characters after close-quote", -1));
		}
		parsePtr->errorType = TCL_PARSE_QUOTE_EXTRA;
	    } else {
		if (interp != NULL) {
		    Tcl_SetObjResult(interp, Tcl_NewStringObj(
			    "extra characters after close-brace", -1));
		}
		parsePtr->errorType = TCL_PARSE_BRACE_EXTRA;
	    }
	    parsePtr->term = src;
	error:
	    Tcl_FreeParse(parsePtr);
	    parsePtr->commandSize = parsePtr->end - parsePtr->commandStart;
	    return TCL_ERROR;
	}

	/*
	 * Create the token for the word.
	 */

	TclGrowParseTokenArray(parsePtr, 1);
	wordIndex = parsePtr->numTokens;
	tokenPtr = &parsePtr->tokenPtr[wordIndex];
	tokenPtr->type = TCL_TOKEN_WORD;

	tokenPtr->start = src;
	parsePtr->numTokens++;
	parsePtr->numWords++;

	/*
	 * At this point the word can have one of four forms: something
	 * enclosed in quotes, something enclosed in braces, and expanding
	 * word, or an unquoted word (anything else).
	 */

    parseWord:
	if (*src == '"') {
	    if (Tcl_ParseQuotedString(interp, src, numBytes, parsePtr, 1,
		    &termPtr) != TCL_OK) {
		goto error;
	    }
	    src = termPtr;
	    numBytes = parsePtr->end - src;
	} else if (*src == '{') {
	    int expIdx = wordIndex + 1;
	    Tcl_Token *expPtr;

	    if (Tcl_ParseBraces(interp, src, numBytes, parsePtr, 1,
		    &termPtr) != TCL_OK) {
		goto error;
	    }
	    src = termPtr;
	    numBytes = parsePtr->end - src;

	    /*
	     * Check whether the braces contained the word expansion prefix
	     * {*}
	     */

	    expPtr = &parsePtr->tokenPtr[expIdx];
	    if ((0 == expandWord)
		    /* Haven't seen prefix already */
		    && (1 == parsePtr->numTokens - expIdx)
		    /* Only one token */
		    && (((1 == (size_t) expPtr->size)
			    /* Same length as prefix */
			    && (expPtr->start[0] == '*')))
			    /* Is the prefix */
		    && (numBytes > 0) && (0 == ParseWhiteSpace(termPtr,
			    numBytes, &parsePtr->incomplete, &type))
		    && (type != TYPE_COMMAND_END)
		    /* Non-whitespace follows */) {
		expandWord = 1;
		parsePtr->numTokens--;
		goto parseWord;
	    }
	} else {
	    /*
	     * This is an unquoted word. Call ParseTokens and let it do all of
	     * the work.
	     */

	    if (ParseTokens(src, numBytes, TYPE_SPACE|terminators,
		    TCL_SUBST_ALL, parsePtr) != TCL_OK) {
		goto error;
	    }
	    src = parsePtr->term;
	    numBytes = parsePtr->end - src;
	}

	/*
	 * Finish filling in the token for the word and check for the special
	 * case of a word consisting of a single range of literal text.
	 */

	tokenPtr = &parsePtr->tokenPtr[wordIndex];
	tokenPtr->size = src - tokenPtr->start;
	tokenPtr->numComponents = parsePtr->numTokens - (wordIndex + 1);
	if (expandWord) {
	    int i, isLiteral = 1;

	    /*
	     * When a command includes a word that is an expanded literal; for
	     * example, {*}{1 2 3}, the parser performs that expansion
	     * immediately, generating several TCL_TOKEN_SIMPLE_WORDs instead
	     * of a single TCL_TOKEN_EXPAND_WORD that the Tcl_ParseCommand()
	     * caller might have to expand. This notably makes it simpler for
	     * those callers that wish to track line endings, such as those
	     * that implement key parts of TIP 280.
	     *
	     * First check whether the thing to be expanded is a literal,
	     * in the sense of being composed entirely of TCL_TOKEN_TEXT
	     * tokens.
	     */

	    for (i = 1; i <= tokenPtr->numComponents; i++) {
		if (tokenPtr[i].type != TCL_TOKEN_TEXT) {
		    isLiteral = 0;
		    break;
		}
	    }

	    if (isLiteral) {
		int elemCount = 0, code = TCL_OK, literal = 1;
		const char *nextElem, *listEnd, *elemStart;

		/*
		 * The word to be expanded is a literal, so determine the
		 * boundaries of the literal string to be treated as a list
		 * and expanded. That literal string starts at
		 * tokenPtr[1].start, and includes all bytes up to, but not
		 * including (tokenPtr[tokenPtr->numComponents].start +
		 * tokenPtr[tokenPtr->numComponents].size)
		 */

		listEnd = (tokenPtr[tokenPtr->numComponents].start +
			tokenPtr[tokenPtr->numComponents].size);
		nextElem = tokenPtr[1].start;

		/*
		 * Step through the literal string, parsing and counting list
		 * elements.
		 */

		while (nextElem < listEnd) {
		    int size;

		    code = TclFindElement(NULL, nextElem, listEnd - nextElem,
			    &elemStart, &nextElem, &size, &literal);
		    if ((code != TCL_OK) || !literal) {
			break;
		    }
		    if (elemStart < listEnd) {
			elemCount++;
		    }
		}

		if ((code != TCL_OK) || !literal) {
		    /*
		     * Some list element could not be parsed, or is not
		     * present as a literal substring of the script.  The
		     * compiler cannot handle list elements that get generated
		     * by a call to TclCopyAndCollapse(). Defer  the
		     * handling of  this to  compile/eval time, where  code is
		     * already  in place to  report the  "attempt to  expand a
		     * non-list" error or expand lists that require
		     * substitution.
		     */

		    tokenPtr->type = TCL_TOKEN_EXPAND_WORD;
		} else if (elemCount == 0) {
		    /*
		     * We are expanding a literal empty list. This means that
		     * the expanding word completely disappears, leaving no
		     * word generated this pass through the loop. Adjust
		     * accounting appropriately.
		     */

		    parsePtr->numWords--;
		    parsePtr->numTokens = wordIndex;
		} else {
		    /*
		     * Recalculate the number of Tcl_Tokens needed to store
		     * tokens representing the expanded list.
		     */

		    const char *listStart;
		    int growthNeeded = wordIndex + 2*elemCount
			    - parsePtr->numTokens;

		    parsePtr->numWords += elemCount - 1;
		    if (growthNeeded > 0) {
			TclGrowParseTokenArray(parsePtr, growthNeeded);
			tokenPtr = &parsePtr->tokenPtr[wordIndex];
		    }
		    parsePtr->numTokens = wordIndex + 2*elemCount;

		    /*
		     * Generate a TCL_TOKEN_SIMPLE_WORD token sequence for
		     * each element of the literal list we are expanding in
		     * place. Take care with the start and size fields of each
		     * token so they point to the right literal characters in
		     * the original script to represent the right expanded
		     * word value.
		     */

		    listStart = nextElem = tokenPtr[1].start;
		    while (nextElem < listEnd) {
			int quoted;

			tokenPtr->type = TCL_TOKEN_SIMPLE_WORD;
			tokenPtr->numComponents = 1;

			tokenPtr++;
			tokenPtr->type = TCL_TOKEN_TEXT;
			tokenPtr->numComponents = 0;
			TclFindElement(NULL, nextElem, listEnd - nextElem,
				&(tokenPtr->start), &nextElem,
				&(tokenPtr->size), NULL);

			quoted = (tokenPtr->start[-1] == '{'
				|| tokenPtr->start[-1] == '"')
				&& tokenPtr->start > listStart;
			tokenPtr[-1].start = tokenPtr->start - quoted;
			tokenPtr[-1].size = tokenPtr->start + tokenPtr->size
				- tokenPtr[-1].start + quoted;

			tokenPtr++;
		    }
		}
	    } else {
		/*
		 * The word to be expanded is not a literal, so defer
		 * expansion to compile/eval time by marking with a
		 * TCL_TOKEN_EXPAND_WORD token.
		 */

		tokenPtr->type = TCL_TOKEN_EXPAND_WORD;
	    }
	} else if ((tokenPtr->numComponents == 1)
		&& (tokenPtr[1].type == TCL_TOKEN_TEXT)) {
	    tokenPtr->type = TCL_TOKEN_SIMPLE_WORD;
	}

	/* Parse the whitespace between words. */

	scanned = ParseWhiteSpace(src,numBytes, &parsePtr->incomplete, &type);
	src += scanned;
	numBytes -= scanned;
    }
}

/*
 *----------------------------------------------------------------------
 *
 * TclIsSpaceProc --
 *
 *	Report whether byte is in the set of whitespace characters used by
 *	Tcl to separate words in scripts or elements in lists.
 *
 * Results:
 *	Returns 1, if byte is in the set, 0 otherwise.
 *
 * Side effects:
 *	None.
 *
 *----------------------------------------------------------------------
 */

int
TclIsSpaceProc(
    int byte)
{
    return CHAR_TYPE(byte) & (TYPE_SPACE) || byte == '\n';
}

/*
 *----------------------------------------------------------------------
 *
 * TclIsBareword--
 *
 *	Report whether byte is one that can be part of a "bareword".
 *	This concept is named in expression parsing, where it determines
 *	what can be a legal function name, but is the same definition used
 *	in determining what variable names can be parsed as variable
 *	substitutions without the benefit of enclosing braces.  The set of
 *	ASCII chars that are accepted are the numeric chars ('0'-'9'),
 *	the alphabetic chars ('a'-'z', 'A'-'Z')	and underscore ('_').
 *
 * Results:
 *	Returns 1, if byte is in the accepted set of chars, 0 otherwise.
 *
 * Side effects:
 *	None.
 *
 *----------------------------------------------------------------------
 */

int
TclIsBareword(
    int byte)
{
    if (byte < '0' || byte > 'z') {
	return 0;
    }
    if (byte <= '9' || byte >= 'a') {
	return 1;
    }
    if (byte == '_') {
	return 1;
    }
    if (byte < 'A' || byte > 'Z') {
	return 0;
    }
    return 1;
}

/*
 *----------------------------------------------------------------------
 *
 * ParseWhiteSpace --
 *
 *	Scans up to numBytes bytes starting at src, consuming white space
 *	between words as defined by Tcl's parsing rules.
 *
 * Results:
 *	Returns the number of bytes recognized as white space. Records at
 *	parsePtr, information about the parse. Records at typePtr the
 *	character type of the non-whitespace character that terminated the
 *	scan.
 *
 * Side effects:
 *	None.
 *
 *----------------------------------------------------------------------
 */

static int
ParseWhiteSpace(
    const char *src,		/* First character to parse. */
    int numBytes,	/* Max number of bytes to scan. */
    int *incompletePtr,		/* Set this boolean memory to true if parsing
				 * indicates an incomplete command. */
    char *typePtr)		/* Points to location to store character type
				 * of character that ends run of whitespace */
{
    char type = TYPE_NORMAL;
    const char *p = src;

    while (1) {
	while (numBytes && ((type = CHAR_TYPE(*p)) & TYPE_SPACE)) {
	    numBytes--;
	    p++;
	}
	if (numBytes && (type & TYPE_SUBS)) {
	    if (*p != '\\') {
		break;
	    }
	    if (--numBytes == 0) {
		break;
	    }
	    if (p[1] != '\n') {
		break;
	    }
	    p += 2;
	    if (--numBytes == 0) {
		*incompletePtr = 1;
		break;
	    }
	    continue;
	}
	break;
    }
    *typePtr = type;
    return (p - src);
}

/*
 *----------------------------------------------------------------------
 *
 * TclParseAllWhiteSpace --
 *
 *	Scans up to numBytes bytes starting at src, consuming all white space
 *	including the command-terminating newline characters.
 *
 * Results:
 *	Returns the number of bytes recognized as white space.
 *
 *----------------------------------------------------------------------
 */

static int
ParseAllWhiteSpace(
    const char *src,		/* First character to parse. */
    int numBytes,		/* Max number of byes to scan */
    int *incompletePtr)		/* Set true if parse is incomplete. */
{
    char type;
    const char *p = src;

    do {
	int scanned = ParseWhiteSpace(p, numBytes, incompletePtr, &type);

	p += scanned;
	numBytes -= scanned;
    } while (numBytes && (*p == '\n') && (p++, --numBytes));
    return (p-src);
}

int
TclParseAllWhiteSpace(
    const char *src,		/* First character to parse. */
    int numBytes)		/* Max number of byes to scan */
{
    int dummy;
    return ParseAllWhiteSpace(src, numBytes, &dummy);
}

/*
 *----------------------------------------------------------------------
 *
 * ParseHex --
 *
 *	Scans a hexadecimal number as a Tcl_UniChar value (e.g., for parsing
 *	\x and \u escape sequences). At most numBytes bytes are scanned.
 *
 * Results:
 *	The numeric value is stored in *resultPtr. Returns the number of bytes
 *	consumed.
 *
 * Notes:
 *	Relies on the following properties of the ASCII character set, with
 *	which UTF-8 is compatible:
 *
 *	The digits '0' .. '9' and the letters 'A' .. 'Z' and 'a' .. 'z' occupy
 *	consecutive code points, and '0' < 'A' < 'a'.
 *
 *----------------------------------------------------------------------
 */

int
ParseHex(
    const char *src,		/* First character to parse. */
    int numBytes,		/* Max number of byes to scan */
    int *resultPtr)	/* Points to storage provided by caller where
				 * the character resulting from the
				 * conversion is to be written. */
{
    int result = 0;
    const char *p = src;

    while (numBytes--) {
	unsigned char digit = UCHAR(*p);

	if (!isxdigit(digit) || (result > 0x10FFF)) {
	    break;
	}

	p++;
	result <<= 4;

	if (digit >= 'a') {
	    result |= (10 + digit - 'a');
	} else if (digit >= 'A') {
	    result |= (10 + digit - 'A');
	} else {
	    result |= (digit - '0');
	}
    }

    *resultPtr = result;
    return (p - src);
}

/*
 *----------------------------------------------------------------------
 *
 * TclParseBackslash --
 *
 *	Scans up to numBytes bytes starting at src, consuming a backslash
 *	sequence as defined by Tcl's parsing rules.
 *
 * Results:
 *	Records at readPtr the number of bytes making up the backslash
 *	sequence. Records at dst the UTF-8 encoded equivalent of that
 *	backslash sequence. Returns the number of bytes written to dst, at
 *	most TCL_UTF_MAX. Either readPtr or dst may be NULL, if the results
 *	are not needed, but the return value is the same either way.
 *
 * Side effects:
 *	None.
 *
 *----------------------------------------------------------------------
 */

int
TclParseBackslash(
    const char *src,		/* Points to the backslash character of a a
				 * backslash sequence. */
    int numBytes,		/* Max number of bytes to scan. */
    int *readPtr,		/* NULL, or points to storage where the number
				 * of bytes scanned should be written. */
    char *dst)			/* NULL, or points to buffer where the UTF-8
				 * encoding of the backslash sequence is to be
				 * written. At most 4 bytes will be written there. */
{
    const char *p = src+1;
    int unichar;
    int result;
    int count;
    char buf[4] = "";

    if (numBytes == 0) {
	if (readPtr != NULL) {
	    *readPtr = 0;
	}
	return 0;
    }

    if (dst == NULL) {
	dst = buf;
    }

    if (numBytes == 1) {
	/*
	 * Can only scan the backslash, so return it.
	 */

	result = '\\';
	count = 1;
	goto done;
    }

    count = 2;
    switch (*p) {
	/*
	 * Note: in the conversions below, use absolute values (e.g., 0xA)
	 * rather than symbolic values (e.g. \n) that get converted by the
	 * compiler. It's possible that compilers on some platforms will do
	 * the symbolic conversions differently, which could result in
	 * non-portable Tcl scripts.
	 */

    case 'a':
	result = 0x7;
	break;
    case 'b':
	result = 0x8;
	break;
    case 'f':
	result = 0xC;
	break;
    case 'n':
	result = 0xA;
	break;
    case 'r':
	result = 0xD;
	break;
    case 't':
	result = 0x9;
	break;
    case 'v':
	result = 0xB;
	break;
    case 'x':
	count += ParseHex(p+1, (numBytes > 3) ? 2 : numBytes-2, &result);
	if (count == 2) {
	    /*
	     * No hexdigits -> This is just "x".
	     */

	    result = 'x';
	} else {
	    /*
	     * Keep only the last byte (2 hex digits).
	     */
	    result = UCHAR(result);
	}
	break;
    case 'u':
	count += ParseHex(p+1, (numBytes > 5) ? 4 : numBytes-2, &result);
	if (count == 2) {
	    /*
	     * No hexdigits -> This is just "u".
	     */
	    result = 'u';
	} else if (((result & 0xFC00) == 0xD800) && (count == 6)
		    && (p[5] == '\\') && (p[6] == 'u') && (numBytes >= 10)) {
	    /* If high surrogate is immediately followed by a low surrogate
	     * escape, combine them into one character. */
	    int low;
	    int count2 = ParseHex(p+7, 4, &low);
	    if ((count2 == 4) && ((low & 0xFC00) == 0xDC00)) {
		result = ((result & 0x3FF)<<10 | (low & 0x3FF)) + 0x10000;
		count += count2 + 2;
	    }
	}
	break;
    case 'U':
	count += ParseHex(p+1, (numBytes > 9) ? 8 : numBytes-2, &result);
	if (count == 2) {
	    /*
	     * No hexdigits -> This is just "U".
	     */
	    result = 'U';
	} else if ((result | 0x7FF) == 0xDFFF) {
	    /* Upper or lower surrogate, not allowed in this syntax. */
	    result = 0xFFFD;
	}
	break;
    case '\n':
	count--;
	do {
	    p++;
	    count++;
	} while ((count < numBytes) && ((*p == ' ') || (*p == '\t')));
	result = ' ';
	break;
    case 0:
	result = '\\';
	count = 1;
	break;
    default:
	/*
	 * Check for an octal number \oo?o?
	 */

	if (isdigit(UCHAR(*p)) && (UCHAR(*p) < '8')) {	/* INTL: digit */
	    result = *p - '0';
	    p++;
	    if ((numBytes == 2) || !isdigit(UCHAR(*p))	/* INTL: digit */
		    || (UCHAR(*p) >= '8')) {
		break;
	    }
	    count = 3;
	    result = (result << 3) + (*p - '0');
	    p++;
	    if ((numBytes == 3) || !isdigit(UCHAR(*p))	/* INTL: digit */
		    || (UCHAR(*p) >= '8') || (result >= 0x20)) {
		break;
	    }
	    count = 4;
	    result = UCHAR((result << 3) + (*p - '0'));
	    break;
	}

	/*
	 * We have to convert here in case the user has put a backslash in
	 * front of a multi-byte utf-8 character. While this means nothing
	 * special, we shouldn't break up a correct utf-8 character. [Bug
	 * #217987] test subst-3.2
	 */

	if (TclUCS4Complete(p, numBytes - 1)) {
	    count = TclUtfToUCS4(p, &unichar) + 1;	/* +1 for '\' */
	} else {
	    char utfBytes[8];

	    memcpy(utfBytes, p, numBytes - 1);
	    utfBytes[numBytes - 1] = '\0';
	    count = TclUtfToUCS4(utfBytes, &unichar) + 1;
	}
	result = unichar;
	break;
    }

  done:
    if (readPtr != NULL) {
	*readPtr = count;
    }
    count = Tcl_UniCharToUtf(result, dst);
    if ((result >= 0xD800) && (count < 3)) {
	/* Special case for handling high surrogates. */
	count += Tcl_UniCharToUtf(-1, dst + count);
    }
    return count;
}

/*
 *----------------------------------------------------------------------
 *
 * ParseComment --
 *
 *	Scans up to numBytes bytes starting at src, consuming a Tcl comment as
 *	defined by Tcl's parsing rules.
 *
 * Results:
 *	Records in parsePtr information about the parse. Returns the number of
 *	bytes consumed.
 *
 * Side effects:
 *	None.
 *
 *----------------------------------------------------------------------
 */

static int
ParseComment(
    const char *src,		/* First character to parse. */
    int numBytes,	/* Max number of bytes to scan. */
    Tcl_Parse *parsePtr)	/* Information about parse in progress.
				 * Updated if parsing indicates an incomplete
				 * command. */
{
    const char *p = src;
    int incomplete = parsePtr->incomplete;

    while (numBytes) {
	int scanned = ParseAllWhiteSpace(p, numBytes, &incomplete);
	p += scanned;
	numBytes -= scanned;

	if ((numBytes == 0) || (*p != '#')) {
	    break;
	}
	if (parsePtr->commentStart == NULL) {
	    parsePtr->commentStart = p;
	}

	p++;
	numBytes--;
	while (numBytes) {
	    if (*p == '\n') {
		p++;
		numBytes--;
		break;
	    }
	    if (*p == '\\') {
		p++;
		numBytes--;
		if (numBytes == 0) {
		    break;
		}
	    }
	    incomplete = (*p == '\n');
	    p++;
	    numBytes--;
	}
	parsePtr->commentSize = p - parsePtr->commentStart;
    }
    parsePtr->incomplete = incomplete;
    return (p - src);
}

/*
 *----------------------------------------------------------------------
 *
 * ParseTokens --
 *
 *	This function forms the heart of the Tcl parser. It parses one or more
 *	tokens from a string, up to a termination point specified by the
 *	caller. This function is used to parse unquoted command words (those
 *	not in quotes or braces), words in quotes, and array indices for
 *	variables. No more than numBytes bytes will be scanned.
 *
 * Results:
 *	Tokens are added to parsePtr and parsePtr->term is filled in with the
 *	address of the character that terminated the parse (the first one
 *	whose CHAR_TYPE matched mask or the character at parsePtr->end). The
 *	return value is TCL_OK if the parse completed successfully and
 *	TCL_ERROR otherwise. If a parse error occurs and parsePtr->interp is
 *	not NULL, then an error message is left in the interpreter's result.
 *
 * Side effects:
 *	None.
 *
 *----------------------------------------------------------------------
 */

static int
ParseTokens(
    const char *src,	/* First character to parse. */
    int numBytes,	/* Max number of bytes to scan. */
    int mask,			/* Specifies when to stop parsing. The parse
				 * stops at the first unquoted character whose
				 * CHAR_TYPE contains any of the bits in
				 * mask. */
    int flags,			/* OR-ed bits indicating what substitutions to
				 * perform: TCL_SUBST_COMMANDS,
				 * TCL_SUBST_VARIABLES, and
				 * TCL_SUBST_BACKSLASHES */
    Tcl_Parse *parsePtr)	/* Information about parse in progress.
				 * Updated with additional tokens and
				 * termination information. */
{
    char type;
    int originalTokens;
    int noSubstCmds = !(flags & TCL_SUBST_COMMANDS);
    int noSubstVars = !(flags & TCL_SUBST_VARIABLES);
    int noSubstBS = !(flags & TCL_SUBST_BACKSLASHES);
    Tcl_Token *tokenPtr;

    /*
     * Each iteration through the following loop adds one token of type
     * TCL_TOKEN_TEXT, TCL_TOKEN_BS, TCL_TOKEN_COMMAND, or TCL_TOKEN_VARIABLE
     * to parsePtr. For TCL_TOKEN_VARIABLE tokens, additional tokens are added
     * for the parsed variable name.
     */

    originalTokens = parsePtr->numTokens;
    while (numBytes && !((type = CHAR_TYPE(*src)) & mask)) {
	TclGrowParseTokenArray(parsePtr, 1);
	tokenPtr = &parsePtr->tokenPtr[parsePtr->numTokens];
	tokenPtr->start = src;
	tokenPtr->numComponents = 0;

	if ((type & TYPE_SUBS) == 0) {
	    /*
	     * This is a simple range of characters. Scan to find the end of
	     * the range.
	     */

	    while ((++src, --numBytes)
		    && !(CHAR_TYPE(*src) & (mask | TYPE_SUBS))) {
		/* empty loop */
	    }
	    tokenPtr->type = TCL_TOKEN_TEXT;
	    tokenPtr->size = src - tokenPtr->start;
	    parsePtr->numTokens++;
	} else if (*src == '$') {
	    int varToken;

	    if (noSubstVars) {
		tokenPtr->type = TCL_TOKEN_TEXT;
		tokenPtr->size = 1;
		parsePtr->numTokens++;
		src++;
		numBytes--;
		continue;
	    }

	    /*
	     * This is a variable reference. Call Tcl_ParseVarName to do all
	     * the dirty work of parsing the name.
	     */

	    varToken = parsePtr->numTokens;
	    if (Tcl_ParseVarName(parsePtr->interp, src, numBytes, parsePtr,
		    1) != TCL_OK) {
		return TCL_ERROR;
	    }
	    src += parsePtr->tokenPtr[varToken].size;
	    numBytes -= parsePtr->tokenPtr[varToken].size;
	} else if (*src == '[') {
	    Tcl_Parse *nestedPtr;

	    if (noSubstCmds) {
		tokenPtr->type = TCL_TOKEN_TEXT;
		tokenPtr->size = 1;
		parsePtr->numTokens++;
		src++;
		numBytes--;
		continue;
	    }

	    /*
	     * Command substitution. Call Tcl_ParseCommand recursively (and
	     * repeatedly) to parse the nested command(s), then throw away the
	     * parse information.
	     */

	    src++;
	    numBytes--;
	    nestedPtr = (Tcl_Parse *)TclStackAlloc(parsePtr->interp, sizeof(Tcl_Parse));
	    while (1) {
		const char *curEnd;

		if (Tcl_ParseCommand(parsePtr->interp, src, numBytes, 1,
			nestedPtr) != TCL_OK) {
		    parsePtr->errorType = nestedPtr->errorType;
		    parsePtr->term = nestedPtr->term;
		    parsePtr->incomplete = nestedPtr->incomplete;
		    TclStackFree(parsePtr->interp, nestedPtr);
		    return TCL_ERROR;
		}
		curEnd = src + numBytes;
		src = nestedPtr->commandStart + nestedPtr->commandSize;
		numBytes = curEnd - src;
		Tcl_FreeParse(nestedPtr);

		/*
		 * Check for the closing ']' that ends the command
		 * substitution. It must have been the last character of the
		 * parsed command.
		 */

		if ((nestedPtr->term < parsePtr->end)
			&& (*(nestedPtr->term) == ']')
			&& !(nestedPtr->incomplete)) {
		    break;
		}
		if (numBytes == 0) {
		    if (parsePtr->interp != NULL) {
			Tcl_SetObjResult(parsePtr->interp, Tcl_NewStringObj(
				"missing close-bracket", -1));
		    }
		    parsePtr->errorType = TCL_PARSE_MISSING_BRACKET;
		    parsePtr->term = tokenPtr->start;
		    parsePtr->incomplete = 1;
		    TclStackFree(parsePtr->interp, nestedPtr);
		    return TCL_ERROR;
		}
	    }
	    TclStackFree(parsePtr->interp, nestedPtr);
	    tokenPtr->type = TCL_TOKEN_COMMAND;
	    tokenPtr->size = src - tokenPtr->start;
	    parsePtr->numTokens++;
	} else if (*src == '\\') {
	    if (noSubstBS) {
		tokenPtr->type = TCL_TOKEN_TEXT;
		tokenPtr->size = 1;
		parsePtr->numTokens++;
		src++;
		numBytes--;
		continue;
	    }

	    /*
	     * Backslash substitution.
	     */

	    TclParseBackslash(src, numBytes, &tokenPtr->size, NULL);

	    if (tokenPtr->size == 1) {
		/*
		 * Just a backslash, due to end of string.
		 */

		tokenPtr->type = TCL_TOKEN_TEXT;
		parsePtr->numTokens++;
		src++;
		numBytes--;
		continue;
	    }

	    if (src[1] == '\n') {
		if (numBytes == 2) {
		    parsePtr->incomplete = 1;
		}

		/*
		 * Note: backslash-newline is special in that it is treated
		 * the same as a space character would be. This means that it
		 * could terminate the token.
		 */

		if (mask & TYPE_SPACE) {
		    if (parsePtr->numTokens == originalTokens) {
			goto finishToken;
		    }
		    break;
		}
	    }

	    tokenPtr->type = TCL_TOKEN_BS;
	    parsePtr->numTokens++;
	    src += tokenPtr->size;
	    numBytes -= tokenPtr->size;
	} else if (*src == 0) {
	    tokenPtr->type = TCL_TOKEN_TEXT;
	    tokenPtr->size = 1;
	    parsePtr->numTokens++;
	    src++;
	    numBytes--;
	} else {
	    Tcl_Panic("ParseTokens encountered unknown character");
	}
    }
    if (parsePtr->numTokens == originalTokens) {
	/*
	 * There was nothing in this range of text. Add an empty token for the
	 * empty range, so that there is always at least one token added.
	 */

	TclGrowParseTokenArray(parsePtr, 1);
	tokenPtr = &parsePtr->tokenPtr[parsePtr->numTokens];
	tokenPtr->start = src;
	tokenPtr->numComponents = 0;

    finishToken:
	tokenPtr->type = TCL_TOKEN_TEXT;
	tokenPtr->size = 0;
	parsePtr->numTokens++;
    }
    parsePtr->term = src;
    return TCL_OK;
}

/*
 *----------------------------------------------------------------------
 *
 * Tcl_FreeParse --
 *
 *	This function is invoked to free any dynamic storage that may have
 *	been allocated by a previous call to Tcl_ParseCommand.
 *
 * Results:
 *	None.
 *
 * Side effects:
 *	If there is any dynamically allocated memory in *parsePtr, it is
 *	freed.
 *
 *----------------------------------------------------------------------
 */

void
Tcl_FreeParse(
    Tcl_Parse *parsePtr)	/* Structure that was filled in by a previous
				 * call to Tcl_ParseCommand. */
{
    if (parsePtr->tokenPtr != parsePtr->staticTokens) {
	ckfree(parsePtr->tokenPtr);
	parsePtr->tokenPtr = parsePtr->staticTokens;
    }
}

/*
 *----------------------------------------------------------------------
 *
 * Tcl_ParseVarName --
 *
 *	Given a string starting with a $ sign, parse off a variable name and
 *	return information about the parse. No more than numBytes bytes will
 *	be scanned.
 *
 * Results:
 *	The return value is TCL_OK if the command was parsed successfully and
 *	TCL_ERROR otherwise. If an error occurs and interp isn't NULL then an
 *	error message is left in its result. On a successful return, tokenPtr
 *	and numTokens fields of parsePtr are filled in with information about
 *	the variable name that was parsed. The "size" field of the first new
 *	token gives the total number of bytes in the variable name. Other
 *	fields in parsePtr are undefined.
 *
 * Side effects:
 *	If there is insufficient space in parsePtr to hold all the information
 *	about the command, then additional space is malloc-ed. If the function
 *	returns TCL_OK then the caller must eventually invoke Tcl_FreeParse to
 *	release any additional space that was allocated.
 *
 *----------------------------------------------------------------------
 */

int
Tcl_ParseVarName(
    Tcl_Interp *interp,		/* Interpreter to use for error reporting; if
				 * NULL, then no error message is provided. */
    const char *start,		/* Start of variable substitution string.
				 * First character must be "$". */
    int numBytes,	/* Total number of bytes in string. If < 0,
				 * the string consists of all bytes up to the
				 * first null character. */
    Tcl_Parse *parsePtr,	/* Structure to fill in with information about
				 * the variable name. */
    int append)			/* Non-zero means append tokens to existing
				 * information in parsePtr; zero means ignore
				 * existing tokens in parsePtr and
				 * reinitialize it. */
{
    Tcl_Token *tokenPtr;
    const char *src;
    int varIndex;
    unsigned array;

    if (numBytes < 0 && start) {
	numBytes = strlen(start);
    }
    if (!append) {
	TclParseInit(interp, start, numBytes, parsePtr);
    }
    if ((numBytes == 0) || (start == NULL)) {
	return TCL_ERROR;
    }

    /*
     * Generate one token for the variable, an additional token for the name,
     * plus any number of additional tokens for the index, if there is one.
     */

    src = start;
    TclGrowParseTokenArray(parsePtr, 2);
    tokenPtr = &parsePtr->tokenPtr[parsePtr->numTokens];
    tokenPtr->type = TCL_TOKEN_VARIABLE;
    tokenPtr->start = src;
    varIndex = parsePtr->numTokens;
    parsePtr->numTokens++;
    tokenPtr++;
    src++;
    numBytes--;
    if (numBytes == 0) {
	goto justADollarSign;
    }
    tokenPtr->type = TCL_TOKEN_TEXT;
    tokenPtr->start = src;
    tokenPtr->numComponents = 0;

    /*
     * The name of the variable can have three forms:
     * 1. The $ sign is followed by an open curly brace. Then the variable
     *	  name is everything up to the next close curly brace, and the
     *	  variable is a scalar variable.
     * 2. The $ sign is not followed by an open curly brace. Then the variable
     *	  name is everything up to the next character that isn't a letter,
     *	  digit, or underscore. :: sequences are also considered part of the
     *	  variable name, in order to support namespaces. If the following
     *	  character is an open parenthesis, then the information between
     *	  parentheses is the array element name.
     * 3. The $ sign is followed by something that isn't a letter, digit, or
     *	  underscore: in this case, there is no variable name and the token is
     *	  just "$".
     */

    if (*src == '{') {
	src++;
	numBytes--;
	tokenPtr->type = TCL_TOKEN_TEXT;
	tokenPtr->start = src;
	tokenPtr->numComponents = 0;

	while (numBytes && (*src != '}')) {
	    numBytes--;
	    src++;
	}
	if (numBytes == 0) {
	    if (parsePtr->interp != NULL) {
		Tcl_SetObjResult(parsePtr->interp, Tcl_NewStringObj(
			"missing close-brace for variable name", -1));
	    }
	    parsePtr->errorType = TCL_PARSE_MISSING_VAR_BRACE;
	    parsePtr->term = tokenPtr->start-1;
	    parsePtr->incomplete = 1;
	    goto error;
	}
	tokenPtr->size = src - tokenPtr->start;
	tokenPtr[-1].size = src - tokenPtr[-1].start;
	parsePtr->numTokens++;
	src++;
    } else {
	tokenPtr->type = TCL_TOKEN_TEXT;
	tokenPtr->start = src;
	tokenPtr->numComponents = 0;

	while (numBytes) {
	    if (TclIsBareword(*src)) {
		src += 1;
		numBytes -= 1;
		continue;
	    }
	    if ((src[0] == ':') && (numBytes != 1) && (src[1] == ':')) {
		src += 2;
		numBytes -= 2;
		while (numBytes && (*src == ':')) {
		    src++;
		    numBytes--;
		}
		continue;
	    }
	    break;
	}

	/*
	 * Support for empty array names here.
	 */

	array = (numBytes && (*src == '('));
	tokenPtr->size = src - tokenPtr->start;
	if ((tokenPtr->size == 0) && !array) {
	    goto justADollarSign;
	}
	parsePtr->numTokens++;
	if (array) {
	    /*
	     * This is a reference to an array element. Call ParseTokens
	     * recursively to parse the element name, since it could contain
	     * any number of substitutions.
	     */

	    if (TCL_OK != ParseTokens(src+1, numBytes-1, TYPE_CLOSE_PAREN,
		    TCL_SUBST_ALL, parsePtr)) {
		goto error;
	    }
	    if ((parsePtr->term == src+numBytes) || (*parsePtr->term != ')')){
		if (parsePtr->interp != NULL) {
		    Tcl_SetObjResult(parsePtr->interp, Tcl_NewStringObj(
			    "missing )", -1));
		}
		parsePtr->errorType = TCL_PARSE_MISSING_PAREN;
		parsePtr->term = src;
		parsePtr->incomplete = 1;
		goto error;
	    }
	    src = parsePtr->term + 1;
	}
    }
    tokenPtr = &parsePtr->tokenPtr[varIndex];
    tokenPtr->size = src - tokenPtr->start;
    tokenPtr->numComponents = parsePtr->numTokens - (varIndex + 1);
    return TCL_OK;

    /*
     * The dollar sign isn't followed by a variable name. Replace the
     * TCL_TOKEN_VARIABLE token with a TCL_TOKEN_TEXT token for the dollar
     * sign.
     */

  justADollarSign:
    tokenPtr = &parsePtr->tokenPtr[varIndex];
    tokenPtr->type = TCL_TOKEN_TEXT;
    tokenPtr->size = 1;
    tokenPtr->numComponents = 0;
    return TCL_OK;

  error:
    Tcl_FreeParse(parsePtr);
    return TCL_ERROR;
}

/*
 *----------------------------------------------------------------------
 *
 * Tcl_ParseVar --
 *
 *	Given a string starting with a $ sign, parse off a variable name and
 *	return its value.
 *
 * Results:
 *	The return value is the contents of the variable given by the leading
 *	characters of string. If termPtr isn't NULL, *termPtr gets filled in
 *	with the address of the character just after the last one in the
 *	variable specifier. If the variable doesn't exist, then the return
 *	value is NULL and an error message will be left in interp's result.
 *
 * Side effects:
 *	None.
 *
 *----------------------------------------------------------------------
 */

const char *
Tcl_ParseVar(
    Tcl_Interp *interp,		/* Context for looking up variable. */
    const char *start,	/* Start of variable substitution. First
				 * character must be "$". */
    const char **termPtr)	/* If non-NULL, points to word to fill in with
				 * character just after last one in the
				 * variable specifier. */
{
    Tcl_Obj *objPtr;
    int code;
    Tcl_Parse *parsePtr = (Tcl_Parse *)TclStackAlloc(interp, sizeof(Tcl_Parse));

    if (Tcl_ParseVarName(interp, start, -1, parsePtr, 0) != TCL_OK) {
	TclStackFree(interp, parsePtr);
	return NULL;
    }

    if (termPtr != NULL) {
	*termPtr = start + parsePtr->tokenPtr->size;
    }
    if (parsePtr->numTokens == 1) {
	/*
	 * There isn't a variable name after all: the $ is just a $.
	 */

	TclStackFree(interp, parsePtr);
	return "$";
    }

    code = TclSubstTokens(interp, parsePtr->tokenPtr, parsePtr->numTokens,
	    NULL, 1, NULL, NULL);
    Tcl_FreeParse(parsePtr);
    TclStackFree(interp, parsePtr);
    if (code != TCL_OK) {
	return NULL;
    }
    objPtr = Tcl_GetObjResult(interp);

    /*
     * At this point we should have an object containing the value of a
     * variable. Just return the string from that object.
     *
     * Since TclSubstTokens above returned TCL_OK, we know that objPtr
     * is shared.  It is in both the interp result and the value of the
     * variable.  Returning the string relies on that to be true.
     */

    assert( Tcl_IsShared(objPtr) );

    Tcl_ResetResult(interp);
    return TclGetString(objPtr);
}

/*
 *----------------------------------------------------------------------
 *
 * Tcl_ParseBraces --
 *
 *	Given a string in braces such as a Tcl command argument or a string
 *	value in a Tcl expression, this function parses the string and returns
 *	information about the parse. No more than numBytes bytes will be
 *	scanned.
 *
 * Results:
 *	The return value is TCL_OK if the string was parsed successfully and
 *	TCL_ERROR otherwise. If an error occurs and interp isn't NULL then an
 *	error message is left in its result. On a successful return, tokenPtr
 *	and numTokens fields of parsePtr are filled in with information about
 *	the string that was parsed. Other fields in parsePtr are undefined.
 *	termPtr is set to point to the character just after the last one in
 *	the braced string.
 *
 * Side effects:
 *	If there is insufficient space in parsePtr to hold all the information
 *	about the command, then additional space is malloc-ed. If the function
 *	returns TCL_OK then the caller must eventually invoke Tcl_FreeParse to
 *	release any additional space that was allocated.
 *
 *----------------------------------------------------------------------
 */

int
Tcl_ParseBraces(
    Tcl_Interp *interp,		/* Interpreter to use for error reporting; if
				 * NULL, then no error message is provided. */
    const char *start,		/* Start of string enclosed in braces. The
				 * first character must be {'. */
    int numBytes,	/* Total number of bytes in string. If < 0,
				 * the string consists of all bytes up to the
				 * first null character. */
    Tcl_Parse *parsePtr,
				/* Structure to fill in with information about
				 * the string. */
    int append,			/* Non-zero means append tokens to existing
				 * information in parsePtr; zero means ignore
				 * existing tokens in parsePtr and
				 * reinitialize it. */
    const char **termPtr)	/* If non-NULL, points to word in which to
				 * store a pointer to the character just after
				 * the terminating '}' if the parse was
				 * successful. */
{
    Tcl_Token *tokenPtr;
    const char *src;
    int startIndex, level, length;

    if (numBytes < 0 && start) {
	numBytes = strlen(start);
    }
    if (!append) {
	TclParseInit(interp, start, numBytes, parsePtr);
    }
    if ((numBytes == 0) || (start == NULL)) {
	return TCL_ERROR;
    }

    src = start;
    startIndex = parsePtr->numTokens;

    TclGrowParseTokenArray(parsePtr, 1);
    tokenPtr = &parsePtr->tokenPtr[startIndex];
    tokenPtr->type = TCL_TOKEN_TEXT;
    tokenPtr->start = src+1;
    tokenPtr->numComponents = 0;
    level = 1;
    while (1) {
	while (++src, --numBytes) {
	    if (CHAR_TYPE(*src) != TYPE_NORMAL) {
		break;
	    }
	}
	if (numBytes == 0) {
	    goto missingBraceError;
	}

	switch (*src) {
	case '{':
	    level++;
	    break;
	case '}':
	    if (--level == 0) {
		/*
		 * Decide if we need to finish emitting a partially-finished
		 * token. There are 3 cases:
		 *     {abc \newline xyz} or {xyz}
		 *		- finish emitting "xyz" token
		 *     {abc \newline}
		 *		- don't emit token after \newline
		 *     {}	- finish emitting zero-sized token
		 *
		 * The last case ensures that there is a token (even if empty)
		 * that describes the braced string.
		 */

		if ((src != tokenPtr->start)
			|| (parsePtr->numTokens == startIndex)) {
		    tokenPtr->size = (src - tokenPtr->start);
		    parsePtr->numTokens++;
		}
		if (termPtr != NULL) {
		    *termPtr = src+1;
		}
		return TCL_OK;
	    }
	    break;
	case '\\':
	    TclParseBackslash(src, numBytes, &length, NULL);
	    if ((length > 1) && (src[1] == '\n')) {
		/*
		 * A backslash-newline sequence must be collapsed, even inside
		 * braces, so we have to split the word into multiple tokens
		 * so that the backslash-newline can be represented
		 * explicitly.
		 */

		if (numBytes == 2) {
		    parsePtr->incomplete = 1;
		}
		tokenPtr->size = (src - tokenPtr->start);
		if (tokenPtr->size != 0) {
		    parsePtr->numTokens++;
		}
		TclGrowParseTokenArray(parsePtr, 2);
		tokenPtr = &parsePtr->tokenPtr[parsePtr->numTokens];
		tokenPtr->type = TCL_TOKEN_BS;
		tokenPtr->start = src;
		tokenPtr->size = length;
		tokenPtr->numComponents = 0;
		parsePtr->numTokens++;

		src += length - 1;
		numBytes -= length - 1;
		tokenPtr++;
		tokenPtr->type = TCL_TOKEN_TEXT;
		tokenPtr->start = src + 1;
		tokenPtr->numComponents = 0;
	    } else {
		src += length - 1;
		numBytes -= length - 1;
	    }
	    break;
	}
    }

  missingBraceError:
    parsePtr->errorType = TCL_PARSE_MISSING_BRACE;
    parsePtr->term = start;
    parsePtr->incomplete = 1;
    if (parsePtr->interp == NULL) {
	/*
	 * Skip straight to the exit code since we have no interpreter to put
	 * error message in.
	 */

	goto error;
    }

    Tcl_SetObjResult(parsePtr->interp, Tcl_NewStringObj(
	    "missing close-brace", -1));

    /*
     * Guess if the problem is due to comments by searching the source string
     * for a possible open brace within the context of a comment. Since we
     * aren't performing a full Tcl parse, just look for an open brace
     * preceded by a '<whitespace>#' on the same line.
     */

    {
	int openBrace = 0;

	while (--src > start) {
	    switch (*src) {
	    case '{':
		openBrace = 1;
		break;
	    case '\n':
		openBrace = 0;
		break;
	    case '#' :
		if (openBrace && TclIsSpaceProcM(src[-1])) {
		    Tcl_AppendToObj(Tcl_GetObjResult(parsePtr->interp),
			    ": possible unbalanced brace in comment", -1);
		    goto error;
		}
		break;
	    }
	}
    }

  error:
    Tcl_FreeParse(parsePtr);
    return TCL_ERROR;
}

/*
 *----------------------------------------------------------------------
 *
 * Tcl_ParseQuotedString --
 *
 *	Given a double-quoted string such as a quoted Tcl command argument or
 *	a quoted value in a Tcl expression, this function parses the string
 *	and returns information about the parse. No more than numBytes bytes
 *	will be scanned.
 *
 * Results:
 *	The return value is TCL_OK if the string was parsed successfully and
 *	TCL_ERROR otherwise. If an error occurs and interp isn't NULL then an
 *	error message is left in its result. On a successful return, tokenPtr
 *	and numTokens fields of parsePtr are filled in with information about
 *	the string that was parsed. Other fields in parsePtr are undefined.
 *	termPtr is set to point to the character just after the quoted
 *	string's terminating close-quote.
 *
 * Side effects:
 *	If there is insufficient space in parsePtr to hold all the information
 *	about the command, then additional space is malloc-ed. If the function
 *	returns TCL_OK then the caller must eventually invoke Tcl_FreeParse to
 *	release any additional space that was allocated.
 *
 *----------------------------------------------------------------------
 */

int
Tcl_ParseQuotedString(
    Tcl_Interp *interp,		/* Interpreter to use for error reporting; if
				 * NULL, then no error message is provided. */
    const char *start,		/* Start of the quoted string. The first
				 * character must be '"'. */
    int numBytes,	/* Total number of bytes in string. If < 0,
				 * the string consists of all bytes up to the
				 * first null character. */
    Tcl_Parse *parsePtr,
				/* Structure to fill in with information about
				 * the string. */
    int append,			/* Non-zero means append tokens to existing
				 * information in parsePtr; zero means ignore
				 * existing tokens in parsePtr and
				 * reinitialize it. */
    const char **termPtr)	/* If non-NULL, points to word in which to
				 * store a pointer to the character just after
				 * the quoted string's terminating close-quote
				 * if the parse succeeds. */
{
    if (numBytes < 0 && start) {
	numBytes = strlen(start);
    }
    if (!append) {
	TclParseInit(interp, start, numBytes, parsePtr);
    }
    if ((numBytes == 0) || (start == NULL)) {
	return TCL_ERROR;
    }

    if (TCL_OK != ParseTokens(start+1, numBytes-1, TYPE_QUOTE, TCL_SUBST_ALL,
	    parsePtr)) {
	goto error;
    }
    if (*parsePtr->term != '"') {
	if (parsePtr->interp != NULL) {
	    Tcl_SetObjResult(parsePtr->interp, Tcl_NewStringObj(
		    "missing \"", -1));
	}
	parsePtr->errorType = TCL_PARSE_MISSING_QUOTE;
	parsePtr->term = start;
	parsePtr->incomplete = 1;
	goto error;
    }
    if (termPtr != NULL) {
	*termPtr = (parsePtr->term + 1);
    }
    return TCL_OK;

  error:
    Tcl_FreeParse(parsePtr);
    return TCL_ERROR;
}

/*
 *----------------------------------------------------------------------
 *
 * TclSubstParse --
 *
 *	Token parser used by the [subst] command. Parses the string made up of
 *	'numBytes' bytes starting at 'bytes'. Parsing is controlled by the
 *	flags argument to provide support for the -nobackslashes, -nocommands,
 *	and -novariables options, as represented by the flag values
 *	TCL_SUBST_BACKSLASHES, TCL_SUBST_COMMANDS, TCL_SUBST_VARIABLES.
 *
 * Results:
 *	None.
 *
 * Side effects:
 *	The Tcl_Parse struct '*parsePtr' is filled with parse results.
 *	The caller is expected to eventually call Tcl_FreeParse() to properly
 *	cleanup the value written there.
 *
 *	If a parse error occurs, the Tcl_InterpState value '*statePtr' is
 *	filled with the state created by that error. When *statePtr is written
 *	to, the caller is expected to make the required calls to either
 *	Tcl_RestoreInterpState() or Tcl_DiscardInterpState() to dispose of the
 *	value written there.
 *
 *----------------------------------------------------------------------
 */

void
TclSubstParse(
    Tcl_Interp *interp,
    const char *bytes,
    int numBytes,
    int flags,
    Tcl_Parse *parsePtr,
    Tcl_InterpState *statePtr)
{
    int length = numBytes;
    const char *p = bytes;

    TclParseInit(interp, p, length, parsePtr);

    /*
     * First parse the string rep of objPtr, as if it were enclosed as a
     * "-quoted word in a normal Tcl command. Honor flags that selectively
     * inhibit types of substitution.
     */

    if (TCL_OK != ParseTokens(p, length, /* mask */ 0, flags, parsePtr)) {
	/*
	 * There was a parse error. Save the interpreter state for possible
	 * error reporting later.
	 */

	*statePtr = Tcl_SaveInterpState(interp, TCL_ERROR);

	/*
	 * We need to re-parse to get the portion of the string we can [subst]
	 * before the parse error. Sadly, all the Tcl_Token's created by the
	 * first parse attempt are gone, freed according to the public spec
	 * for the Tcl_Parse* routines. The only clue we have is parse.term,
	 * which points to either the unmatched opener, or to characters that
	 * follow a close brace or close quote.
	 *
	 * Call ParseTokens again, working on the string up to parse.term.
	 * Keep repeating until we get a good parse on a prefix.
	 */

	do {
	    parsePtr->numTokens = 0;
	    parsePtr->tokensAvailable = NUM_STATIC_TOKENS;
	    parsePtr->end = parsePtr->term;
	    parsePtr->incomplete = 0;
	    parsePtr->errorType = TCL_PARSE_SUCCESS;
	} while (TCL_OK !=
		ParseTokens(p, parsePtr->end - p, 0, flags, parsePtr));

	/*
	 * The good parse will have to be followed by {, (, or [.
	 */

	switch (*(parsePtr->term)) {
	case '{':
	    /*
	     * Parse error was a missing } in a ${varname} variable
	     * substitution at the toplevel. We will subst everything up to
	     * that broken variable substitution before reporting the parse
	     * error. Substituting the leftover '$' will have no side-effects,
	     * so the current token stream is fine.
	     */
	    break;

	case '(':
	    /*
	     * Parse error was during the parsing of the index part of an
	     * array variable substitution at the toplevel.
	     */

	    if (*(parsePtr->term - 1) == '$') {
		/*
		 * Special case where removing the array index left us with
		 * just a dollar sign (array variable with name the empty
		 * string as its name), instead of with a scalar variable
		 * reference.
		 *
		 * As in the previous case, existing token stream is OK.
		 */
	    } else {
		/*
		 * The current parse includes a successful parse of a scalar
		 * variable substitution where there should have been an array
		 * variable substitution. We remove that mistaken part of the
		 * parse before moving on. A scalar variable substitution is
		 * two tokens.
		 */

		Tcl_Token *varTokenPtr =
			parsePtr->tokenPtr + parsePtr->numTokens - 2;

		if (varTokenPtr->type != TCL_TOKEN_VARIABLE) {
		    Tcl_Panic("TclSubstParse: programming error");
		}
		if (varTokenPtr[1].type != TCL_TOKEN_TEXT) {
		    Tcl_Panic("TclSubstParse: programming error");
		}
		parsePtr->numTokens -= 2;
	    }
	    break;
	case '[':
	    /*
	     * Parse error occurred during parsing of a toplevel command
	     * substitution.
	     */

	    parsePtr->end = p + length;
	    p = parsePtr->term + 1;
	    length = parsePtr->end - p;
	    if (length == 0) {
		/*
		 * No commands, just an unmatched [. As in previous cases,
		 * existing token stream is OK.
		 */
	    } else {
		/*
		 * We want to add the parsing of as many commands as we can
		 * within that substitution until we reach the actual parse
		 * error. We'll do additional parsing to determine what length
		 * to claim for the final TCL_TOKEN_COMMAND token.
		 */

		Tcl_Token *tokenPtr;
		const char *lastTerm = parsePtr->term;
		Tcl_Parse *nestedPtr = (Tcl_Parse *)
			TclStackAlloc(interp, sizeof(Tcl_Parse));

		while (TCL_OK ==
			Tcl_ParseCommand(NULL, p, length, 0, nestedPtr)) {
		    Tcl_FreeParse(nestedPtr);
		    p = nestedPtr->term + (nestedPtr->term < nestedPtr->end);
		    length = nestedPtr->end - p;
		    if ((length == 0) && (nestedPtr->term == nestedPtr->end)) {
			/*
			 * If we run out of string, blame the missing close
			 * bracket on the last command, and do not evaluate it
			 * during substitution.
			 */

			break;
		    }
		    lastTerm = nestedPtr->term;
		}
		TclStackFree(interp, nestedPtr);

		if (lastTerm == parsePtr->term) {
		    /*
		     * Parse error in first command. No commands to subst, add
		     * no more tokens.
		     */
		    break;
		}

		/*
		 * Create a command substitution token for whatever commands
		 * got parsed.
		 */

		TclGrowParseTokenArray(parsePtr, 1);
		tokenPtr = &(parsePtr->tokenPtr[parsePtr->numTokens]);
		tokenPtr->start = parsePtr->term;
		tokenPtr->numComponents = 0;
		tokenPtr->type = TCL_TOKEN_COMMAND;
		tokenPtr->size = lastTerm - tokenPtr->start + 1;
		parsePtr->numTokens++;
	    }
	    break;

	default:
	    Tcl_Panic("bad parse in TclSubstParse: %c", p[length]);
	}
    }
}

/*
 *----------------------------------------------------------------------
 *
 * TclSubstTokens --
 *
 *	Accepts an array of count Tcl_Token's, and creates a result value in
 *	the interp from concatenating the results of performing Tcl
 *	substitution on each Tcl_Token. Substitution is interrupted if any
 *	non-TCL_OK completion code arises.
 *
 * Results:
 *	The return value is a standard Tcl completion code. The result in
 *	interp is the substituted value, or an error message if TCL_ERROR is
 *	returned. If tokensLeftPtr is not NULL, then it points to an int where
 *	the number of tokens remaining to be processed is written.
 *
 * Side effects:
 *	Can be anything, depending on the types of substitution done.
 *
 *----------------------------------------------------------------------
 */

int
TclSubstTokens(
    Tcl_Interp *interp,		/* Interpreter in which to lookup variables,
				 * execute nested commands, and report
				 * errors. */
    Tcl_Token *tokenPtr,	/* Pointer to first in an array of tokens to
				 * evaluate and concatenate. */
    int count,			/* Number of tokens to consider at tokenPtr.
				 * Must be at least 1. */
    int *tokensLeftPtr,		/* If not NULL, points to memory where an
				 * integer representing the number of tokens
				 * left to be substituted will be written */
    int line,			/* The line the script starts on. */
    int *clNextOuter,		/* Information about an outer context for */
    const char *outerScript)	/* continuation line data. This is set by
				 * EvalEx() to properly handle [...]-nested
				 * commands. The 'outerScript' refers to the
				 * most-outer script containing the embedded
				 * command, which is refered to by 'script'.
				 * The 'clNextOuter' refers to the current
				 * entry in the table of continuation lines in
				 * this "main script", and the character
				 * offsets are relative to the 'outerScript'
				 * as well.
				 *
				 * If outerScript == script, then this call is
				 * for words in the outer-most script or
				 * command. See Tcl_EvalEx and TclEvalObjEx
				 * for the places generating arguments for
				 * which this is true. */
{
    Tcl_Obj *result;
    int code = TCL_OK;
#define NUM_STATIC_POS 20
    int isLiteral, maxNumCL, numCL, i, adjust;
    int *clPosition = NULL;
    Interp *iPtr = (Interp *) interp;
    int inFile = iPtr->evalFlags & TCL_EVAL_FILE;

    /*
     * Each pass through this loop will substitute one token, and its
     * components, if any. The only thing tricky here is that we go to some
     * effort to pass Tcl_Obj's through untouched, to avoid string copying and
     * Tcl_Obj creation if possible, to aid performance and limit shimmering.
     *
     * Further optimization opportunities might be to check for the equivalent
     * of Tcl_SetObjResult(interp, Tcl_GetObjResult(interp)) and omit them.
     */

    /*
     * For the handling of continuation lines in literals we first check if
     * this is actually a literal. For if not we can forego the additional
     * processing. Otherwise we pre-allocate a small table to store the
     * locations of all continuation lines we find in this literal, if any.
     * The table is extended if needed.
     */

    numCL = 0;
    maxNumCL = 0;
    isLiteral = 1;
    for (i=0 ; i < count; i++) {
	if ((tokenPtr[i].type != TCL_TOKEN_TEXT)
		&& (tokenPtr[i].type != TCL_TOKEN_BS)) {
	    isLiteral = 0;
	    break;
	}
    }

    if (isLiteral) {
	maxNumCL = NUM_STATIC_POS;
	clPosition = (int *)ckalloc(maxNumCL * sizeof(int));
    }

    adjust = 0;
    result = NULL;
    for (; count>0 && code==TCL_OK ; count--, tokenPtr++) {
	Tcl_Obj *appendObj = NULL;
	const char *append = NULL;
	int appendByteLength = 0;
	char utfCharBytes[4] = "";

	switch (tokenPtr->type) {
	case TCL_TOKEN_TEXT:
	    append = tokenPtr->start;
	    appendByteLength = tokenPtr->size;
	    break;

	case TCL_TOKEN_BS:
	    appendByteLength = TclParseBackslash(tokenPtr->start,
		    tokenPtr->size, NULL, utfCharBytes);
	    append = utfCharBytes;

	    /*
	     * If the backslash sequence we found is in a literal, and
	     * represented a continuation line, we compute and store its
	     * location (as char offset to the beginning of the _result_
	     * script). We may have to extend the table of locations.
	     *
	     * Note that the continuation line information is relevant even if
	     * the word we are processing is not a literal, as it can affect
	     * nested commands. See the branch for TCL_TOKEN_COMMAND below,
	     * where the adjustment we are tracking here is taken into
	     * account. The good thing is that we do not need a table of
	     * everything, just the number of lines we have to add as
	     * correction.
	     */

	    if ((appendByteLength == 1) && (utfCharBytes[0] == ' ')
		    && (tokenPtr->start[1] == '\n')) {
		if (isLiteral) {
		    int clPos;

		    if (result == 0) {
			clPos = 0;
		    } else {
			TclGetStringFromObj(result, &clPos);
		    }

		    if (numCL >= maxNumCL) {
			maxNumCL *= 2;
			clPosition = (int *)ckrealloc(clPosition,
				maxNumCL * sizeof(int));
		    }
		    clPosition[numCL] = clPos;
		    numCL++;
		}
		adjust++;
	    }
	    break;

	case TCL_TOKEN_COMMAND: {
	    /* TIP #280: Transfer line information to nested command */
	    iPtr->numLevels++;
	    code = TclInterpReady(interp);
	    if (code == TCL_OK) {
		/*
		 * Test cases: info-30.{6,8,9}
		 */

		int theline;

		TclAdvanceContinuations(&line, &clNextOuter,
			tokenPtr->start - outerScript);
		theline = line + adjust;
		code = TclEvalEx(interp, tokenPtr->start+1, tokenPtr->size-2,
			0, theline, clNextOuter, outerScript);

		TclAdvanceLines(&line, tokenPtr->start+1,
			tokenPtr->start + tokenPtr->size - 1);

		/*
		 * Restore flag reset by nested eval for future bracketed
		 * commands and their cmdframe setup
		 */

		if (inFile) {
		    iPtr->evalFlags |= TCL_EVAL_FILE;
		}
	    }
	    iPtr->numLevels--;
	    TclResetCancellation(interp, 0);
	    appendObj = Tcl_GetObjResult(interp);
	    break;
	}

	case TCL_TOKEN_VARIABLE: {
	    Tcl_Obj *arrayIndex = NULL;
	    Tcl_Obj *varName = NULL;

	    if (tokenPtr->numComponents > 1) {
		/*
		 * Subst the index part of an array variable reference.
		 */

		code = TclSubstTokens(interp, tokenPtr+2,
			tokenPtr->numComponents - 1, NULL, line, NULL, NULL);
		arrayIndex = Tcl_GetObjResult(interp);
		Tcl_IncrRefCount(arrayIndex);
	    }

	    if (code == TCL_OK) {
		varName = Tcl_NewStringObj(tokenPtr[1].start,
			tokenPtr[1].size);
		appendObj = Tcl_ObjGetVar2(interp, varName, arrayIndex,
			TCL_LEAVE_ERR_MSG);
		Tcl_DecrRefCount(varName);
		if (appendObj == NULL) {
		    code = TCL_ERROR;
		}
	    }

	    switch (code) {
	    case TCL_OK:	/* Got value */
	    case TCL_ERROR:	/* Already have error message */
	    case TCL_BREAK:	/* Will not substitute anyway */
	    case TCL_CONTINUE:	/* Will not substitute anyway */
		break;
	    default:
		/*
		 * All other return codes, we will subst the result from the
		 * code-throwing evaluation.
		 */

		appendObj = Tcl_GetObjResult(interp);
	    }

	    if (arrayIndex != NULL) {
		Tcl_DecrRefCount(arrayIndex);
	    }
	    count -= tokenPtr->numComponents;
	    tokenPtr += tokenPtr->numComponents;
	    break;
	}

	default:
	    Tcl_Panic("unexpected token type in TclSubstTokens: %d",
		    tokenPtr->type);
	}

	if ((code == TCL_BREAK) || (code == TCL_CONTINUE)) {
	    /*
	     * Inhibit substitution.
	     */
	    continue;
	}

	if (result == NULL) {
	    /*
	     * First pass through. If we have a Tcl_Obj, just use it. If not,
	     * create one from our string.
	     */

	    if (appendObj != NULL) {
		result = appendObj;
	    } else {
		result = Tcl_NewStringObj(append, appendByteLength);
	    }
	    Tcl_IncrRefCount(result);
	} else {
	    /*
	     * Subsequent passes. Append to result.
	     */

	    if (Tcl_IsShared(result)) {
		Tcl_DecrRefCount(result);
		result = Tcl_DuplicateObj(result);
		Tcl_IncrRefCount(result);
	    }
	    if (appendObj != NULL) {
		Tcl_AppendObjToObj(result, appendObj);
	    } else {
		Tcl_AppendToObj(result, append, appendByteLength);
	    }
	}
    }

    if (code != TCL_ERROR) {		/* Keep error message in result! */
	if (result != NULL) {
	    Tcl_SetObjResult(interp, result);

	    /*
	     * If the code found continuation lines (which implies that this
	     * word is a literal), then we store the accumulated table of
	     * locations in the thread-global data structure for the bytecode
	     * compiler to find later, assuming that the literal is a script
	     * which will be compiled.
	     */

	    if (numCL) {
		TclContinuationsEnter(result, numCL, clPosition);
	    }

	    /*
	     * Release the temp table we used to collect the locations of
	     * continuation lines, if any.
	     */

	    if (maxNumCL) {
		ckfree(clPosition);
	    }
	} else {
	    Tcl_ResetResult(interp);
	}
    }
    if (tokensLeftPtr != NULL) {
	*tokensLeftPtr = count;
    }
    if (result != NULL) {
	Tcl_DecrRefCount(result);
    }
    return code;
}

/*
 *----------------------------------------------------------------------
 *
 * CommandComplete --
 *
 *	This function is shared by TclCommandComplete and
 *	Tcl_ObjCommandComplete; it does all the real work of seeing whether a
 *	script is complete
 *
 * Results:
 *	1 is returned if the script is complete, 0 if there are open
 *	delimiters such as " or (. 1 is also returned if there is a parse
 *	error in the script other than unmatched delimiters.
 *
 * Side effects:
 *	None.
 *
 *----------------------------------------------------------------------
 */

static int
CommandComplete(
    const char *script,		/* Script to check. */
    int numBytes)		/* Number of bytes in script. */
{
    Tcl_Parse parse;
    const char *p, *end;
    int result;

    p = script;
    end = p + numBytes;
    while (Tcl_ParseCommand(NULL, p, end - p, 0, &parse) == TCL_OK) {
	p = parse.commandStart + parse.commandSize;
	if (p >= end) {
	    break;
	}
	Tcl_FreeParse(&parse);
    }
    if (parse.incomplete) {
	result = 0;
    } else {
	result = 1;
    }
    Tcl_FreeParse(&parse);
    return result;
}

/*
 *----------------------------------------------------------------------
 *
 * Tcl_CommandComplete --
 *
 *	Given a partial or complete Tcl script, this function determines
 *	whether the script is complete in the sense of having matched braces
 *	and quotes and brackets.
 *
 * Results:
 *	1 is returned if the script is complete, 0 otherwise. 1 is also
 *	returned if there is a parse error in the script other than unmatched
 *	delimiters.
 *
 * Side effects:
 *	None.
 *
 *----------------------------------------------------------------------
 */

int
Tcl_CommandComplete(
    const char *script)		/* Script to check. */
{
    return CommandComplete(script, (int) strlen(script));
}

/*
 *----------------------------------------------------------------------
 *
 * TclObjCommandComplete --
 *
 *	Given a partial or complete Tcl command in a Tcl object, this function
 *	determines whether the command is complete in the sense of having
 *	matched braces and quotes and brackets.
 *
 * Results:
 *	1 is returned if the command is complete, 0 otherwise.
 *
 * Side effects:
 *	None.
 *
 *----------------------------------------------------------------------
 */

int
TclObjCommandComplete(
    Tcl_Obj *objPtr)		/* Points to object holding script to
				 * check. */
{
    int length;
    const char *script = TclGetStringFromObj(objPtr, &length);

    return CommandComplete(script, length);
}

/*
 * Local Variables:
 * mode: c
 * c-basic-offset: 4
 * fill-column: 78
 * End:
 */