File: json.c

package info (click to toggle)
jo 1.3-2
  • links: PTS, VCS
  • area: main
  • in suites: bullseye, sid
  • size: 436 kB
  • sloc: ansic: 1,734; sh: 525; makefile: 41; exp: 18
file content (1390 lines) | stat: -rw-r--r-- 29,514 bytes parent folder | download | duplicates (2)
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
/*
  Copyright (C) 2011 Joseph A. Adams (joeyadams3.14159@gmail.com)
  All rights reserved.

  Permission is hereby granted, free of charge, to any person obtaining a copy
  of this software and associated documentation files (the "Software"), to deal
  in the Software without restriction, including without limitation the rights
  to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
  copies of the Software, and to permit persons to whom the Software is
  furnished to do so, subject to the following conditions:

  The above copyright notice and this permission notice shall be included in
  all copies or substantial portions of the Software.

  THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
  IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
  FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
  AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
  LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
  OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
  THE SOFTWARE.
*/

#include "json.h"

#include <assert.h>
#include <stdint.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>

#define out_of_memory() do {                    \
		fprintf(stderr, "Out of memory.\n");    \
		exit(EXIT_FAILURE);                     \
	} while (0)

/* Sadly, strdup is not portable. */
static char *json_strdup(const char *str)
{
	size_t n = strlen(str) + 1;
	char *ret = (char*) malloc(n);
	if (ret == NULL)
		out_of_memory();
#if HAVE_STRLCPY
	strlcpy(ret, str, n);
#else
	strcpy(ret, str);
#endif
	return ret;
}

/* String buffer */

typedef struct
{
	char *cur;
	char *end;
	char *start;
} SB;

static void sb_init(SB *sb)
{
	sb->start = (char*) malloc(17);
	if (sb->start == NULL)
		out_of_memory();
	sb->cur = sb->start;
	sb->end = sb->start + 16;
}

/* sb and need may be evaluated multiple times. */
#define sb_need(sb, need) do {                  \
		if ((sb)->end - (sb)->cur < (need))     \
			sb_grow(sb, need);                  \
	} while (0)

static void sb_grow(SB *sb, int need)
{
	size_t length = sb->cur - sb->start;
	size_t alloc = sb->end - sb->start;
	
	do {
		alloc *= 2;
	} while (alloc < length + need);
	
	sb->start = (char*) realloc(sb->start, alloc + 1);
	if (sb->start == NULL)
		out_of_memory();
	sb->cur = sb->start + length;
	sb->end = sb->start + alloc;
}

static void sb_put(SB *sb, const char *bytes, int count)
{
	sb_need(sb, count);
	memcpy(sb->cur, bytes, count);
	sb->cur += count;
}

#define sb_putc(sb, c) do {         \
		if ((sb)->cur >= (sb)->end) \
			sb_grow(sb, 1);         \
		*(sb)->cur++ = (c);         \
	} while (0)

static void sb_puts(SB *sb, const char *str)
{
	sb_put(sb, str, strlen(str));
}

static char *sb_finish(SB *sb)
{
	*sb->cur = 0;
	assert(sb->start <= sb->cur && strlen(sb->start) == (size_t)(sb->cur - sb->start));
	return sb->start;
}

static void sb_free(SB *sb)
{
	free(sb->start);
}

/*
 * Unicode helper functions
 *
 * These are taken from the ccan/charset module and customized a bit.
 * Putting them here means the compiler can (choose to) inline them,
 * and it keeps ccan/json from having a dependency.
 */

/*
 * Type for Unicode codepoints.
 * We need our own because wchar_t might be 16 bits.
 */
typedef uint32_t js_uchar_t;

/*
 * Validate a single UTF-8 character starting at @s.
 * The string must be null-terminated.
 *
 * If it's valid, return its length (1 thru 4).
 * If it's invalid or clipped, return 0.
 *
 * This function implements the syntax given in RFC3629, which is
 * the same as that given in The Unicode Standard, Version 6.0.
 *
 * It has the following properties:
 *
 *  * All codepoints U+0000..U+10FFFF may be encoded,
 *    except for U+D800..U+DFFF, which are reserved
 *    for UTF-16 surrogate pair encoding.
 *  * UTF-8 byte sequences longer than 4 bytes are not permitted,
 *    as they exceed the range of Unicode.
 *  * The sixty-six Unicode "non-characters" are permitted
 *    (namely, U+FDD0..U+FDEF, U+xxFFFE, and U+xxFFFF).
 */
static int utf8_validate_cz(const char *s)
{
	unsigned char c = *s++;
	
	if (c <= 0x7F) {        /* 00..7F */
		return 1;
	} else if (c <= 0xC1) { /* 80..C1 */
		/* Disallow overlong 2-byte sequence. */
		return 0;
	} else if (c <= 0xDF) { /* C2..DF */
		/* Make sure subsequent byte is in the range 0x80..0xBF. */
		if (((unsigned char)*s++ & 0xC0) != 0x80)
			return 0;
		
		return 2;
	} else if (c <= 0xEF) { /* E0..EF */
		/* Disallow overlong 3-byte sequence. */
		if (c == 0xE0 && (unsigned char)*s < 0xA0)
			return 0;
		
		/* Disallow U+D800..U+DFFF. */
		if (c == 0xED && (unsigned char)*s > 0x9F)
			return 0;
		
		/* Make sure subsequent bytes are in the range 0x80..0xBF. */
		if (((unsigned char)*s++ & 0xC0) != 0x80)
			return 0;
		if (((unsigned char)*s++ & 0xC0) != 0x80)
			return 0;
		
		return 3;
	} else if (c <= 0xF4) { /* F0..F4 */
		/* Disallow overlong 4-byte sequence. */
		if (c == 0xF0 && (unsigned char)*s < 0x90)
			return 0;
		
		/* Disallow codepoints beyond U+10FFFF. */
		if (c == 0xF4 && (unsigned char)*s > 0x8F)
			return 0;
		
		/* Make sure subsequent bytes are in the range 0x80..0xBF. */
		if (((unsigned char)*s++ & 0xC0) != 0x80)
			return 0;
		if (((unsigned char)*s++ & 0xC0) != 0x80)
			return 0;
		if (((unsigned char)*s++ & 0xC0) != 0x80)
			return 0;
		
		return 4;
	} else {                /* F5..FF */
		return 0;
	}
}

/* Validate a null-terminated UTF-8 string. */
static bool utf8_validate(const char *s)
{
	int len;
	
	for (; *s != 0; s += len) {
		len = utf8_validate_cz(s);
		if (len == 0)
			return false;
	}
	
	return true;
}

/*
 * Read a single UTF-8 character starting at @s,
 * returning the length, in bytes, of the character read.
 *
 * This function assumes input is valid UTF-8,
 * and that there are enough characters in front of @s.
 */
static int utf8_read_char(const char *s, js_uchar_t *out)
{
	const unsigned char *c = (const unsigned char*) s;
	
	assert(utf8_validate_cz(s));

	if (c[0] <= 0x7F) {
		/* 00..7F */
		*out = c[0];
		return 1;
	} else if (c[0] <= 0xDF) {
		/* C2..DF (unless input is invalid) */
		*out = ((js_uchar_t)c[0] & 0x1F) << 6 |
		       ((js_uchar_t)c[1] & 0x3F);
		return 2;
	} else if (c[0] <= 0xEF) {
		/* E0..EF */
		*out = ((js_uchar_t)c[0] &  0xF) << 12 |
		       ((js_uchar_t)c[1] & 0x3F) << 6  |
		       ((js_uchar_t)c[2] & 0x3F);
		return 3;
	} else {
		/* F0..F4 (unless input is invalid) */
		*out = ((js_uchar_t)c[0] &  0x7) << 18 |
		       ((js_uchar_t)c[1] & 0x3F) << 12 |
		       ((js_uchar_t)c[2] & 0x3F) << 6  |
		       ((js_uchar_t)c[3] & 0x3F);
		return 4;
	}
}

/*
 * Write a single UTF-8 character to @s,
 * returning the length, in bytes, of the character written.
 *
 * @unicode must be U+0000..U+10FFFF, but not U+D800..U+DFFF.
 *
 * This function will write up to 4 bytes to @out.
 */
static int utf8_write_char(js_uchar_t unicode, char *out)
{
	unsigned char *o = (unsigned char*) out;
	
	assert(unicode <= 0x10FFFF && !(unicode >= 0xD800 && unicode <= 0xDFFF));

	if (unicode <= 0x7F) {
		/* U+0000..U+007F */
		*o++ = unicode;
		return 1;
	} else if (unicode <= 0x7FF) {
		/* U+0080..U+07FF */
		*o++ = 0xC0 | unicode >> 6;
		*o++ = 0x80 | (unicode & 0x3F);
		return 2;
	} else if (unicode <= 0xFFFF) {
		/* U+0800..U+FFFF */
		*o++ = 0xE0 | unicode >> 12;
		*o++ = 0x80 | (unicode >> 6 & 0x3F);
		*o++ = 0x80 | (unicode & 0x3F);
		return 3;
	} else {
		/* U+10000..U+10FFFF */
		*o++ = 0xF0 | unicode >> 18;
		*o++ = 0x80 | (unicode >> 12 & 0x3F);
		*o++ = 0x80 | (unicode >> 6 & 0x3F);
		*o++ = 0x80 | (unicode & 0x3F);
		return 4;
	}
}

/*
 * Compute the Unicode codepoint of a UTF-16 surrogate pair.
 *
 * @uc should be 0xD800..0xDBFF, and @lc should be 0xDC00..0xDFFF.
 * If they aren't, this function returns false.
 */
static bool from_surrogate_pair(uint16_t uc, uint16_t lc, js_uchar_t *unicode)
{
	if (uc >= 0xD800 && uc <= 0xDBFF && lc >= 0xDC00 && lc <= 0xDFFF) {
		*unicode = 0x10000 + ((((js_uchar_t)uc & 0x3FF) << 10) | (lc & 0x3FF));
		return true;
	} else {
		return false;
	}
}

/*
 * Construct a UTF-16 surrogate pair given a Unicode codepoint.
 *
 * @unicode must be U+10000..U+10FFFF.
 */
static void to_surrogate_pair(js_uchar_t unicode, uint16_t *uc, uint16_t *lc)
{
	js_uchar_t n;
	
	assert(unicode >= 0x10000 && unicode <= 0x10FFFF);
	
	n = unicode - 0x10000;
	*uc = ((n >> 10) & 0x3FF) | 0xD800;
	*lc = (n & 0x3FF) | 0xDC00;
}

#define is_space(c) ((c) == '\t' || (c) == '\n' || (c) == '\r' || (c) == ' ')
#define is_digit(c) ((c) >= '0' && (c) <= '9')

static bool parse_value     (const char **sp, JsonNode        **out);
static bool parse_string    (const char **sp, char            **out);
static bool parse_number    (const char **sp, double           *out);
static bool parse_array     (const char **sp, JsonNode        **out);
static bool parse_object    (const char **sp, JsonNode        **out);
static bool parse_hex16     (const char **sp, uint16_t         *out);

static bool expect_literal  (const char **sp, const char *str);
static void skip_space      (const char **sp);

static void emit_value              (SB *out, const JsonNode *node);
static void emit_value_indented     (SB *out, const JsonNode *node, const char *space, int indent_level);
static void emit_string             (SB *out, const char *str);
static void emit_number             (SB *out, double num);
static void emit_array              (SB *out, const JsonNode *array);
static void emit_array_indented     (SB *out, const JsonNode *array, const char *space, int indent_level);
static void emit_object             (SB *out, const JsonNode *object);
static void emit_object_indented    (SB *out, const JsonNode *object, const char *space, int indent_level);

static int write_hex16(char *out, uint16_t val);

static JsonNode *mknode(JsonTag tag);
static void append_node(JsonNode *parent, JsonNode *child);
static void prepend_node(JsonNode *parent, JsonNode *child);
static void append_member(JsonNode *object, char *key, JsonNode *value);

/* Assertion-friendly validity checks */
static bool tag_is_valid(unsigned int tag);
static bool number_is_valid(const char *num);

JsonNode *json_decode(const char *json)
{
	const char *s = json;
	JsonNode *ret;
	
	skip_space(&s);
	if (!parse_value(&s, &ret))
		return NULL;
	
	skip_space(&s);
	if (*s != 0) {
		json_delete(ret);
		return NULL;
	}
	
	return ret;
}

char *json_encode(const JsonNode *node)
{
	return json_stringify(node, NULL);
}

char *json_encode_string(const char *str)
{
	SB sb;
	sb_init(&sb);
	
	emit_string(&sb, str);
	
	return sb_finish(&sb);
}

char *json_stringify(const JsonNode *node, const char *space)
{
	SB sb;
	sb_init(&sb);
	
	if (space != NULL)
		emit_value_indented(&sb, node, space, 0);
	else
		emit_value(&sb, node);
	
	return sb_finish(&sb);
}

void json_delete(JsonNode *node)
{
	if (node != NULL) {
		json_remove_from_parent(node);
		
		switch (node->tag) {
			case JSON_STRING:
				free(node->string_);
				break;
			case JSON_ARRAY:
			case JSON_OBJECT:
			{
				JsonNode *child, *next;
				for (child = node->children.head; child != NULL; child = next) {
					next = child->next;
					json_delete(child);
				}
				break;
			}
			default:;
		}
		
		free(node);
	}
}

bool json_validate(const char *json)
{
	const char *s = json;
	
	skip_space(&s);
	if (!parse_value(&s, NULL))
		return false;
	
	skip_space(&s);
	if (*s != 0)
		return false;
	
	return true;
}

JsonNode *json_find_element(JsonNode *array, int index)
{
	JsonNode *element;
	int i = 0;
	
	if (array == NULL || array->tag != JSON_ARRAY)
		return NULL;
	
	json_foreach(element, array) {
		if (i == index)
			return element;
		i++;
	}
	
	return NULL;
}

JsonNode *json_find_member(JsonNode *object, const char *name)
{
	JsonNode *member;
	
	if (object == NULL || object->tag != JSON_OBJECT)
		return NULL;
	
	json_foreach(member, object)
		if (strcmp(member->key, name) == 0)
			return member;
	
	return NULL;
}

JsonNode *json_first_child(const JsonNode *node)
{
	if (node != NULL && (node->tag == JSON_ARRAY || node->tag == JSON_OBJECT))
		return node->children.head;
	return NULL;
}

static JsonNode *mknode(JsonTag tag)
{
	JsonNode *ret = (JsonNode*) calloc(1, sizeof(JsonNode));
	if (ret == NULL)
		out_of_memory();
	ret->tag = tag;
	return ret;
}

JsonNode *json_mknull(void)
{
	return mknode(JSON_NULL);
}

JsonNode *json_mkbool(bool b)
{
	JsonNode *ret = mknode(JSON_BOOL);
	ret->bool_ = b;
	return ret;
}

static JsonNode *mkstring(char *s)
{
	JsonNode *ret = mknode(JSON_STRING);
	ret->string_ = s;
	return ret;
}

JsonNode *json_mkstring(const char *s)
{
	return mkstring(json_strdup(s));
}

JsonNode *json_mknumber(double n)
{
	JsonNode *node = mknode(JSON_NUMBER);
	node->number_ = n;
	return node;
}

JsonNode *json_mkarray(void)
{
	return mknode(JSON_ARRAY);
}

JsonNode *json_mkobject(void)
{
	return mknode(JSON_OBJECT);
}

static void append_node(JsonNode *parent, JsonNode *child)
{
	child->parent = parent;
	child->prev = parent->children.tail;
	child->next = NULL;
	
	if (parent->children.tail != NULL)
		parent->children.tail->next = child;
	else
		parent->children.head = child;
	parent->children.tail = child;
}

static void prepend_node(JsonNode *parent, JsonNode *child)
{
	child->parent = parent;
	child->prev = NULL;
	child->next = parent->children.head;
	
	if (parent->children.head != NULL)
		parent->children.head->prev = child;
	else
		parent->children.tail = child;
	parent->children.head = child;
}

static void append_member(JsonNode *object, char *key, JsonNode *value)
{
	value->key = key;
	append_node(object, value);
}

void json_append_element(JsonNode *array, JsonNode *element)
{
	assert(array->tag == JSON_ARRAY);
	assert(element->parent == NULL);
	
	append_node(array, element);
}

void json_prepend_element(JsonNode *array, JsonNode *element)
{
	assert(array->tag == JSON_ARRAY);
	assert(element->parent == NULL);
	
	prepend_node(array, element);
}

void json_append_member(JsonNode *object, const char *key, JsonNode *value)
{
	assert(object->tag == JSON_OBJECT);
	assert(value->parent == NULL);
	
	append_member(object, json_strdup(key), value);
}

void json_prepend_member(JsonNode *object, const char *key, JsonNode *value)
{
	assert(object->tag == JSON_OBJECT);
	assert(value->parent == NULL);
	
	value->key = json_strdup(key);
	prepend_node(object, value);
}

void json_remove_from_parent(JsonNode *node)
{
	JsonNode *parent = node->parent;
	
	if (parent != NULL) {
		if (node->prev != NULL)
			node->prev->next = node->next;
		else
			parent->children.head = node->next;
		if (node->next != NULL)
			node->next->prev = node->prev;
		else
			parent->children.tail = node->prev;
		
		free(node->key);
		
		node->parent = NULL;
		node->prev = node->next = NULL;
		node->key = NULL;
	}
}

static bool parse_value(const char **sp, JsonNode **out)
{
	const char *s = *sp;
	
	switch (*s) {
		case 'n':
			if (expect_literal(&s, "null")) {
				if (out)
					*out = json_mknull();
				*sp = s;
				return true;
			}
			return false;
		
		case 'f':
			if (expect_literal(&s, "false")) {
				if (out)
					*out = json_mkbool(false);
				*sp = s;
				return true;
			}
			return false;
		
		case 't':
			if (expect_literal(&s, "true")) {
				if (out)
					*out = json_mkbool(true);
				*sp = s;
				return true;
			}
			return false;
		
		case '"': {
			char *str;
			if (parse_string(&s, out ? &str : NULL)) {
				if (out)
					*out = mkstring(str);
				*sp = s;
				return true;
			}
			return false;
		}
		
		case '[':
			if (parse_array(&s, out)) {
				*sp = s;
				return true;
			}
			return false;
		
		case '{':
			if (parse_object(&s, out)) {
				*sp = s;
				return true;
			}
			return false;
		
		default: {
			double num;
			if (parse_number(&s, out ? &num : NULL)) {
				if (out)
					*out = json_mknumber(num);
				*sp = s;
				return true;
			}
			return false;
		}
	}
}

static bool parse_array(const char **sp, JsonNode **out)
{
	const char *s = *sp;
	JsonNode *ret = out ? json_mkarray() : NULL;
	JsonNode *element;
	
	if (*s++ != '[')
		goto failure;
	skip_space(&s);
	
	if (*s == ']') {
		s++;
		goto success;
	}
	
	for (;;) {
		if (!parse_value(&s, out ? &element : NULL))
			goto failure;
		skip_space(&s);
		
		if (out)
			json_append_element(ret, element);
		
		if (*s == ']') {
			s++;
			goto success;
		}
		
		if (*s++ != ',')
			goto failure;
		skip_space(&s);
	}
	
success:
	*sp = s;
	if (out)
		*out = ret;
	return true;

failure:
	json_delete(ret);
	return false;
}

static bool parse_object(const char **sp, JsonNode **out)
{
	const char *s = *sp;
	JsonNode *ret = out ? json_mkobject() : NULL;
	char *key;
	JsonNode *value;
	
	if (*s++ != '{')
		goto failure;
	skip_space(&s);
	
	if (*s == '}') {
		s++;
		goto success;
	}
	
	for (;;) {
		if (!parse_string(&s, out ? &key : NULL))
			goto failure;
		skip_space(&s);
		
		if (*s++ != ':')
			goto failure_free_key;
		skip_space(&s);
		
		if (!parse_value(&s, out ? &value : NULL))
			goto failure_free_key;
		skip_space(&s);
		
		if (out)
			append_member(ret, key, value);
		
		if (*s == '}') {
			s++;
			goto success;
		}
		
		if (*s++ != ',')
			goto failure;
		skip_space(&s);
	}
	
success:
	*sp = s;
	if (out)
		*out = ret;
	return true;

failure_free_key:
	if (out)
		free(key);
failure:
	json_delete(ret);
	return false;
}

bool parse_string(const char **sp, char **out)
{
	const char *s = *sp;
	SB sb;
	char throwaway_buffer[4];
		/* enough space for a UTF-8 character */
	char *b;
	
	if (*s++ != '"')
		return false;
	
	if (out) {
		sb_init(&sb);
		sb_need(&sb, 4);
		b = sb.cur;
	} else {
		b = throwaway_buffer;
	}
	
	while (*s != '"') {
		unsigned char c = *s++;
		
		/* Parse next character, and write it to b. */
		if (c == '\\') {
			c = *s++;
			switch (c) {
				case '"':
				case '\\':
				case '/':
					*b++ = c;
					break;
				case 'b':
					*b++ = '\b';
					break;
				case 'f':
					*b++ = '\f';
					break;
				case 'n':
					*b++ = '\n';
					break;
				case 'r':
					*b++ = '\r';
					break;
				case 't':
					*b++ = '\t';
					break;
				case 'u':
				{
					uint16_t uc, lc;
					js_uchar_t unicode;
					
					if (!parse_hex16(&s, &uc))
						goto failed;
					
					if (uc >= 0xD800 && uc <= 0xDFFF) {
						/* Handle UTF-16 surrogate pair. */
						if (*s++ != '\\' || *s++ != 'u' || !parse_hex16(&s, &lc))
							goto failed; /* Incomplete surrogate pair. */
						if (!from_surrogate_pair(uc, lc, &unicode))
							goto failed; /* Invalid surrogate pair. */
					} else if (uc == 0) {
						/* Disallow "\u0000". */
						goto failed;
					} else {
						unicode = uc;
					}
					
					b += utf8_write_char(unicode, b);
					break;
				}
				default:
					/* Invalid escape */
					goto failed;
			}
		} else if (c <= 0x1F) {
			/* Control characters are not allowed in string literals. */
			goto failed;
		} else {
			/* Validate and echo a UTF-8 character. */
			int len;
			
			s--;
			len = utf8_validate_cz(s);
			if (len == 0)
				goto failed; /* Invalid UTF-8 character. */
			
			while (len--)
				*b++ = *s++;
		}
		
		/*
		 * Update sb to know about the new bytes,
		 * and set up b to write another character.
		 */
		if (out) {
			sb.cur = b;
			sb_need(&sb, 4);
			b = sb.cur;
		} else {
			b = throwaway_buffer;
		}
	}
	s++;
	
	if (out)
		*out = sb_finish(&sb);
	*sp = s;
	return true;

failed:
	if (out)
		sb_free(&sb);
	return false;
}

/*
 * The JSON spec says that a number shall follow this precise pattern
 * (spaces and quotes added for readability):
 *	 '-'? (0 | [1-9][0-9]*) ('.' [0-9]+)? ([Ee] [+-]? [0-9]+)?
 *
 * However, some JSON parsers are more liberal.  For instance, PHP accepts
 * '.5' and '1.'.  JSON.parse accepts '+3'.
 *
 * This function takes the strict approach.
 */
bool parse_number(const char **sp, double *out)
{
	const char *s = *sp;

	/* '-'? */
	if (*s == '-')
		s++;

	/* (0 | [1-9][0-9]*) */
	if (*s == '0') {
		s++;
	} else {
		if (!is_digit(*s))
			return false;
		do {
			s++;
		} while (is_digit(*s));
	}

	/* ('.' [0-9]+)? */
	if (*s == '.') {
		s++;
		if (!is_digit(*s))
			return false;
		do {
			s++;
		} while (is_digit(*s));
	}

	/* ([Ee] [+-]? [0-9]+)? */
	if (*s == 'E' || *s == 'e') {
		s++;
		if (*s == '+' || *s == '-')
			s++;
		if (!is_digit(*s))
			return false;
		do {
			s++;
		} while (is_digit(*s));
	}

	if (out)
		*out = strtod(*sp, NULL);

	*sp = s;
	return true;
}

static void skip_space(const char **sp)
{
	const char *s = *sp;
	while (is_space(*s))
		s++;
	*sp = s;
}

static void emit_value(SB *out, const JsonNode *node)
{
	assert(tag_is_valid(node->tag));
	switch (node->tag) {
		case JSON_NULL:
			sb_puts(out, "null");
			break;
		case JSON_BOOL:
			sb_puts(out, node->bool_ ? "true" : "false");
			break;
		case JSON_STRING:
			emit_string(out, node->string_);
			break;
		case JSON_NUMBER:
			emit_number(out, node->number_);
			break;
		case JSON_ARRAY:
			emit_array(out, node);
			break;
		case JSON_OBJECT:
			emit_object(out, node);
			break;
		default:
			assert(false);
	}
}

void emit_value_indented(SB *out, const JsonNode *node, const char *space, int indent_level)
{
	assert(tag_is_valid(node->tag));
	switch (node->tag) {
		case JSON_NULL:
			sb_puts(out, "null");
			break;
		case JSON_BOOL:
			sb_puts(out, node->bool_ ? "true" : "false");
			break;
		case JSON_STRING:
			emit_string(out, node->string_);
			break;
		case JSON_NUMBER:
			emit_number(out, node->number_);
			break;
		case JSON_ARRAY:
			emit_array_indented(out, node, space, indent_level);
			break;
		case JSON_OBJECT:
			emit_object_indented(out, node, space, indent_level);
			break;
		default:
			assert(false);
	}
}

static void emit_array(SB *out, const JsonNode *array)
{
	const JsonNode *element;
	
	sb_putc(out, '[');
	json_foreach(element, array) {
		emit_value(out, element);
		if (element->next != NULL)
			sb_putc(out, ',');
	}
	sb_putc(out, ']');
}

static void emit_array_indented(SB *out, const JsonNode *array, const char *space, int indent_level)
{
	const JsonNode *element = array->children.head;
	int i;
	
	if (element == NULL) {
		sb_puts(out, "[]");
		return;
	}
	
	sb_puts(out, "[\n");
	while (element != NULL) {
		for (i = 0; i < indent_level + 1; i++)
			sb_puts(out, space);
		emit_value_indented(out, element, space, indent_level + 1);
		
		element = element->next;
		sb_puts(out, element != NULL ? ",\n" : "\n");
	}
	for (i = 0; i < indent_level; i++)
		sb_puts(out, space);
	sb_putc(out, ']');
}

static void emit_object(SB *out, const JsonNode *object)
{
	const JsonNode *member;
	
	sb_putc(out, '{');
	json_foreach(member, object) {
		emit_string(out, member->key);
		sb_putc(out, ':');
		emit_value(out, member);
		if (member->next != NULL)
			sb_putc(out, ',');
	}
	sb_putc(out, '}');
}

static void emit_object_indented(SB *out, const JsonNode *object, const char *space, int indent_level)
{
	const JsonNode *member = object->children.head;
	int i;
	
	if (member == NULL) {
		sb_puts(out, "{}");
		return;
	}
	
	sb_puts(out, "{\n");
	while (member != NULL) {
		for (i = 0; i < indent_level + 1; i++)
			sb_puts(out, space);
		emit_string(out, member->key);
		sb_puts(out, ": ");
		emit_value_indented(out, member, space, indent_level + 1);
		
		member = member->next;
		sb_puts(out, member != NULL ? ",\n" : "\n");
	}
	for (i = 0; i < indent_level; i++)
		sb_puts(out, space);
	sb_putc(out, '}');
}

void emit_string(SB *out, const char *str)
{
	bool escape_unicode = false;
	const char *s = str;
	char *b;
	
	assert(utf8_validate(str));
	
	/*
	 * 14 bytes is enough space to write up to two
	 * \uXXXX escapes and two quotation marks.
	 */
	sb_need(out, 14);
	b = out->cur;
	
	*b++ = '"';
	while (*s != 0) {
		unsigned char c = *s++;
		
		/* Encode the next character, and write it to b. */
		switch (c) {
			case '"':
				*b++ = '\\';
				*b++ = '"';
				break;
			case '\\':
				*b++ = '\\';
				*b++ = '\\';
				break;
			case '\b':
				*b++ = '\\';
				*b++ = 'b';
				break;
			case '\f':
				*b++ = '\\';
				*b++ = 'f';
				break;
			case '\n':
				*b++ = '\\';
				*b++ = 'n';
				break;
			case '\r':
				*b++ = '\\';
				*b++ = 'r';
				break;
			case '\t':
				*b++ = '\\';
				*b++ = 't';
				break;
			default: {
				int len;
				
				s--;
				len = utf8_validate_cz(s);
				
				if (len == 0) {
					/*
					 * Handle invalid UTF-8 character gracefully in production
					 * by writing a replacement character (U+FFFD)
					 * and skipping a single byte.
					 *
					 * This should never happen when assertions are enabled
					 * due to the assertion at the beginning of this function.
					 */
					assert(false);
					if (escape_unicode) {
#if HAVE_STRLCPY
						strlcpy(b, "\\uFFFD", out->end - out->start );
#else
						strcpy(b, "\\uFFFD");
#endif
						b += 6;
					} else {
						*b++ = 0xEF;
						*b++ = 0xBF;
						*b++ = 0xBD;
					}
					s++;
				} else if (c < 0x1F || (c >= 0x80 && escape_unicode)) {
					/* Encode using \u.... */
					uint32_t unicode;
					
					s += utf8_read_char(s, &unicode);
					
					if (unicode <= 0xFFFF) {
						*b++ = '\\';
						*b++ = 'u';
						b += write_hex16(b, unicode);
					} else {
						/* Produce a surrogate pair. */
						uint16_t uc, lc;
						assert(unicode <= 0x10FFFF);
						to_surrogate_pair(unicode, &uc, &lc);
						*b++ = '\\';
						*b++ = 'u';
						b += write_hex16(b, uc);
						*b++ = '\\';
						*b++ = 'u';
						b += write_hex16(b, lc);
					}
				} else {
					/* Write the character directly. */
					while (len--)
						*b++ = *s++;
				}
				
				break;
			}
		}
	
		/*
		 * Update *out to know about the new bytes,
		 * and set up b to write another encoded character.
		 */
		out->cur = b;
		sb_need(out, 14);
		b = out->cur;
	}
	*b++ = '"';
	
	out->cur = b;
}

static void emit_number(SB *out, double num)
{
	/*
	 * This isn't exactly how JavaScript renders numbers,
	 * but it should produce valid JSON for reasonable numbers
	 * preserve precision well enough, and avoid some oddities
	 * like 0.3 -> 0.299999999999999988898 .
	 */
	char buf[64];
	snprintf(buf, sizeof(buf), "%.16g", num);
	
	if (number_is_valid(buf))
		sb_puts(out, buf);
	else
		sb_puts(out, "null");
}

static bool tag_is_valid(unsigned int tag)
{
	return (/* tag >= JSON_NULL && */ tag <= JSON_OBJECT);
}

static bool number_is_valid(const char *num)
{
	return (parse_number(&num, NULL) && *num == '\0');
}

static bool expect_literal(const char **sp, const char *str)
{
	const char *s = *sp;
	
	while (*str != '\0')
		if (*s++ != *str++)
			return false;
	
	*sp = s;
	return true;
}

/*
 * Parses exactly 4 hex characters (capital or lowercase).
 * Fails if any input chars are not [0-9A-Fa-f].
 */
static bool parse_hex16(const char **sp, uint16_t *out)
{
	const char *s = *sp;
	uint16_t ret = 0;
	uint16_t i;
	uint16_t tmp;
	char c;

	for (i = 0; i < 4; i++) {
		c = *s++;
		if (c >= '0' && c <= '9')
			tmp = c - '0';
		else if (c >= 'A' && c <= 'F')
			tmp = c - 'A' + 10;
		else if (c >= 'a' && c <= 'f')
			tmp = c - 'a' + 10;
		else
			return false;

		ret <<= 4;
		ret += tmp;
	}
	
	if (out)
		*out = ret;
	*sp = s;
	return true;
}

/*
 * Encodes a 16-bit number into hexadecimal,
 * writing exactly 4 hex chars.
 */
static int write_hex16(char *out, uint16_t val)
{
	const char *hex = "0123456789ABCDEF";
	
	*out++ = hex[(val >> 12) & 0xF];
	*out++ = hex[(val >> 8)  & 0xF];
	*out++ = hex[(val >> 4)  & 0xF];
	*out++ = hex[ val        & 0xF];
	
	return 4;
}

bool json_check(const JsonNode *node, char errmsg[256])
{
	#define problem(...) do { \
			if (errmsg != NULL) \
				snprintf(errmsg, 256, __VA_ARGS__); \
			return false; \
		} while (0)
	
	if (node->key != NULL && !utf8_validate(node->key))
		problem("key contains invalid UTF-8");
	
	if (!tag_is_valid(node->tag))
		problem("tag is invalid (%u)", node->tag);
	
	if (node->tag == JSON_BOOL) {
		if (node->bool_ != false && node->bool_ != true)
			problem("bool_ is neither false (%d) nor true (%d)", (int)false, (int)true);
	} else if (node->tag == JSON_STRING) {
		if (node->string_ == NULL)
			problem("string_ is NULL");
		if (!utf8_validate(node->string_))
			problem("string_ contains invalid UTF-8");
	} else if (node->tag == JSON_ARRAY || node->tag == JSON_OBJECT) {
		JsonNode *head = node->children.head;
		JsonNode *tail = node->children.tail;
		
		if (head == NULL || tail == NULL) {
			if (head != NULL)
				problem("tail is NULL, but head is not");
			if (tail != NULL)
				problem("head is NULL, but tail is not");
		} else {
			JsonNode *child;
			JsonNode *last = NULL;
			
			if (head->prev != NULL)
				problem("First child's prev pointer is not NULL");
			
			for (child = head; child != NULL; last = child, child = child->next) {
				if (child == node)
					problem("node is its own child");
				if (child->next == child)
					problem("child->next == child (cycle)");
				if (child->next == head)
					problem("child->next == head (cycle)");
				
				if (child->parent != node)
					problem("child does not point back to parent");
				if (child->next != NULL && child->next->prev != child)
					problem("child->next does not point back to child");
				
				if (node->tag == JSON_ARRAY && child->key != NULL)
					problem("Array element's key is not NULL");
				if (node->tag == JSON_OBJECT && child->key == NULL)
					problem("Object member's key is NULL");
				
				if (!json_check(child, errmsg))
					return false;
			}
			
			if (last != tail)
				problem("tail does not match pointer found by starting at head and following next links");
		}
	}
	
	return true;
	
	#undef problem
}