unofficial mirror of bug-gnu-emacs@gnu.org 
 help / color / mirror / code / Atom feed
blob eb446f5c22105c20af9ec7501064c6c25558ecda 52498 bytes (raw)
name: src/json.c 	 # note: path name is non-authoritative(*)

   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
 
/* JSON parsing and serialization.

Copyright (C) 2017-2024 Free Software Foundation, Inc.

This file is part of GNU Emacs.

GNU Emacs is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation, either version 3 of the License, or (at
your option) any later version.

GNU Emacs is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
GNU General Public License for more details.

You should have received a copy of the GNU General Public License
along with GNU Emacs.  If not, see <https://www.gnu.org/licenses/>.  */

#include <config.h>

#include <errno.h>
#include <stddef.h>
#include <stdint.h>
#include <stdlib.h>
#include <math.h>

#include "lisp.h"
#include "buffer.h"
#include "coding.h"
#include "igc.h"

enum json_object_type
  {
    json_object_hashtable,
    json_object_alist,
    json_object_plist,
  };

enum json_array_type
  {
    json_array_array,
    json_array_list,
  };

struct json_configuration
{
  enum json_object_type object_type;
  enum json_array_type array_type;
  Lisp_Object null_object;
  Lisp_Object false_object;
};

static void
json_parse_args (ptrdiff_t nargs, Lisp_Object *args,
		 struct json_configuration *conf,
		 bool parse_object_types)
{
  if ((nargs % 2) != 0)
    wrong_type_argument (Qplistp, Flist (nargs, args));

  /* Start from the back so keyword values appearing first take
     precedence.  */
  for (ptrdiff_t i = nargs; i > 0; i -= 2)
    {
      Lisp_Object key = args[i - 2];
      Lisp_Object value = args[i - 1];
      if (parse_object_types && EQ (key, QCobject_type))
	{
	  if (EQ (value, Qhash_table))
	    conf->object_type = json_object_hashtable;
	  else if (EQ (value, Qalist))
	    conf->object_type = json_object_alist;
	  else if (EQ (value, Qplist))
	    conf->object_type = json_object_plist;
	  else
	    wrong_choice (list3 (Qhash_table, Qalist, Qplist), value);
	}
      else if (parse_object_types && EQ (key, QCarray_type))
	{
	  if (EQ (value, Qarray))
	    conf->array_type = json_array_array;
	  else if (EQ (value, Qlist))
	    conf->array_type = json_array_list;
	  else
	    wrong_choice (list2 (Qarray, Qlist), value);
	}
      else if (EQ (key, QCnull_object))
	conf->null_object = value;
      else if (EQ (key, QCfalse_object))
	conf->false_object = value;
      else if (parse_object_types)
	wrong_choice (list4 (QCobject_type,
			     QCarray_type,
			     QCnull_object,
			     QCfalse_object),
		      value);
      else
	wrong_choice (list2 (QCnull_object,
			     QCfalse_object),
		      value);
    }
}

/* JSON encoding context.  */
typedef struct
{
  char *buf;
  ptrdiff_t size;	      /* number of bytes in buf */
  ptrdiff_t capacity;	      /* allocated size of buf */
  ptrdiff_t chars_delta;      /* size - {number of characters in buf} */

  int maxdepth;
  struct symset_tbl *ss_table;	/* table used by containing object */
  struct json_configuration conf;
} json_out_t;

/* Set of symbols.  */
typedef struct
{
  ptrdiff_t count;		/* symbols in table */
  int bits;			/* log2(table size) */
  struct symset_tbl *table;	/* heap-allocated table */
} symset_t;

struct symset_tbl
{
  /* Table used by the containing object if any, so that we can free all
     tables if an error occurs.  */
  struct symset_tbl *up;
  /* Table of symbols (2**bits elements), Qunbound where unused.  */
  Lisp_Object entries[];
};

static inline ptrdiff_t
symset_size (int bits)
{
  return (ptrdiff_t) 1 << bits;
}

static struct symset_tbl *
make_symset_table (int bits, struct symset_tbl *up)
{
  int maxbits = min (SIZE_WIDTH - 2 - (word_size < 8 ? 2 : 3), 32);
  if (bits > maxbits)
    memory_full (PTRDIFF_MAX);	/* Will never happen in practice.  */
#ifdef HAVE_MPS
  struct symset_tbl *st = igc_xzalloc_ambig (sizeof *st + (sizeof *st->entries << bits));
#else
  struct symset_tbl *st = xmalloc (sizeof *st + (sizeof *st->entries << bits));
#endif
  st->up = up;
  ptrdiff_t size = symset_size (bits);
  for (ptrdiff_t i = 0; i < size; i++)
    st->entries[i] = Qunbound;
  return st;
}

/* Create a new symset to use for a new object.  */
static symset_t
push_symset (json_out_t *jo)
{
  int bits = 4;
  struct symset_tbl *tbl = make_symset_table (bits, jo->ss_table);
  jo->ss_table = tbl;
  return (symset_t){ .count = 0, .bits = bits, .table = tbl };
}

/* Destroy the current symset.  */
static void
pop_symset (json_out_t *jo, symset_t *ss)
{
  jo->ss_table = ss->table->up;
#ifdef HAVE_MPS
  igc_xfree (ss->table);
#else
  xfree (ss->table);
#endif
}

/* Remove all heap-allocated symset tables, in case an error occurred.  */
static void
cleanup_symset_tables (struct symset_tbl *st)
{
  while (st)
    {
      struct symset_tbl *up = st->up;
#ifdef HAVE_MPS
      igc_xfree (st);
#else
      xfree (st);
#endif
      st = up;
    }
}

static inline uint32_t
symset_hash (Lisp_Object sym, int bits)
{
  EMACS_UINT hash;
#ifdef HAVE_MPS
  hash = igc_hash (sym);
#else
  hash = XHASH (sym);
#endif
  return knuth_hash (reduce_emacs_uint_to_hash_hash (hash), bits);
}

/* Enlarge the table used by a symset.  */
static NO_INLINE void
symset_expand (symset_t *ss)
{
  struct symset_tbl *old_table = ss->table;
  int oldbits = ss->bits;
  ptrdiff_t oldsize = symset_size (oldbits);
  int bits = oldbits + 1;
  ss->bits = bits;
  ss->table = make_symset_table (bits, old_table->up);
  /* Move all entries from the old table to the new one.  */
  ptrdiff_t mask = symset_size (bits) - 1;
  struct symset_tbl *tbl = ss->table;
  for (ptrdiff_t i = 0; i < oldsize; i++)
    {
      Lisp_Object sym = old_table->entries[i];
      if (!BASE_EQ (sym, Qunbound))
	{
	  ptrdiff_t j = symset_hash (sym, bits);
	  while (!BASE_EQ (tbl->entries[j], Qunbound))
	    j = (j + 1) & mask;
	  tbl->entries[j] = sym;
	}
    }
#ifdef HAVE_MPS
  igc_xfree (old_table);
#else
  xfree (old_table);
#endif
}

/* If sym is in ss, return false; otherwise add it and return true.
   Comparison is done by strict identity.  */
static inline bool
symset_add (json_out_t *jo, symset_t *ss, Lisp_Object sym)
{
  /* Make sure we don't fill more than half of the table.  */
  if (ss->count >= (symset_size (ss->bits) >> 1))
    {
      symset_expand (ss);
      jo->ss_table = ss->table;
    }

  struct symset_tbl *tbl = ss->table;
  ptrdiff_t mask = symset_size (ss->bits) - 1;
  for (ptrdiff_t i = symset_hash (sym, ss->bits); ; i = (i + 1) & mask)
    {
      Lisp_Object s = tbl->entries[i];
      if (BASE_EQ (s, sym))
	return false;		/* Previous occurrence found.  */
      if (BASE_EQ (s, Qunbound))
	{
	  /* Not in set, add it.  */
	  tbl->entries[i] = sym;
	  ss->count++;
	  return true;
	}
    }
}

static NO_INLINE void
json_out_grow_buf (json_out_t *jo, ptrdiff_t bytes)
{
  ptrdiff_t need = jo->size + bytes;
  ptrdiff_t new_size = max (jo->capacity, 512);
  while (new_size < need)
    new_size <<= 1;
  jo->buf = xrealloc (jo->buf, new_size);
  jo->capacity = new_size;
}

static void
cleanup_json_out (void *arg)
{
  json_out_t *jo = arg;
  xfree (jo->buf);
  jo->buf = NULL;
  cleanup_symset_tables (jo->ss_table);
}

/* Make room for `bytes` more bytes in buffer.  */
static void
json_make_room (json_out_t *jo, ptrdiff_t bytes)
{
  if (bytes > jo->capacity - jo->size)
    json_out_grow_buf (jo, bytes);
}

#define JSON_OUT_STR(jo, str) (json_out_str (jo, str, sizeof (str) - 1))

/* Add `bytes` bytes from `str` to the buffer.  */
static void
json_out_str (json_out_t *jo, const char *str, size_t bytes)
{
  json_make_room (jo, bytes);
  memcpy (jo->buf + jo->size, str, bytes);
  jo->size += bytes;
}

static void
json_out_byte (json_out_t *jo, unsigned char c)
{
  json_make_room (jo, 1);
  jo->buf[jo->size++] = c;
}

static void
json_out_fixnum (json_out_t *jo, EMACS_INT x)
{
  char buf[INT_BUFSIZE_BOUND (EMACS_INT)];
  char *end = buf + sizeof buf;
  char *p = fixnum_to_string (x, buf, end);
  json_out_str (jo, p, end - p);
}

static AVOID
string_not_unicode (Lisp_Object obj)
{
  /* FIXME: this is just for compatibility with existing tests, it's not
     a very descriptive error.  */
  wrong_type_argument (Qjson_value_p, obj);
}

static const unsigned char json_plain_char[256] = {
  /* 32 chars/line: 1 for printable ASCII + DEL except " and \, 0 elsewhere */
  0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, /* 00-1f */
  1,1,0,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* 20-3f */
  1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,0,1,1,1, /* 40-5f */
  1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* 60-7f */
  0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, /* 80-9f */
  0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, /* a0-bf */
  0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, /* c0-df */
  0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, /* e0-ff */
};

static void
json_out_string (json_out_t *jo, Lisp_Object str, int skip)
{
  /* FIXME: this code is slow, make faster! */

  static const char hexchar[16] = "0123456789ABCDEF";
  ptrdiff_t len = SBYTES (str);
  json_make_room (jo, len + 2);
  json_out_byte (jo, '"');
  unsigned char *p = SDATA (str);
  unsigned char *end = p + len;
  p += skip;
  while (p < end)
    {
      unsigned char c = *p;
      if (json_plain_char[c])
	{
	  json_out_byte (jo, c);
	  p++;
	}
      else if (c > 0x7f)
	{
	  if (STRING_MULTIBYTE (str))
	    {
	      int n;
	      if (c <= 0xc1)
		string_not_unicode (str);
	      if (c <= 0xdf)
		n = 2;
	      else if (c <= 0xef)
		{
		  int v = (((c & 0x0f) << 12)
			   + ((p[1] & 0x3f) << 6) + (p[2] & 0x3f));
		  if (char_surrogate_p (v))
		    string_not_unicode (str);
		  n = 3;
		}
	      else if (c <= 0xf7)
		{
		  int v = (((c & 0x07) << 18)
			   + ((p[1] & 0x3f) << 12)
			   + ((p[2] & 0x3f) << 6)
			   + (p[3] & 0x3f));
		  if (v > MAX_UNICODE_CHAR)
		    string_not_unicode (str);
		  n = 4;
		}
	      else
		string_not_unicode (str);
	      json_out_str (jo, (const char *)p, n);
	      jo->chars_delta += n - 1;
	      p += n;
	    }
	  else
	    string_not_unicode (str);
	}
      else
	{
	  json_out_byte (jo, '\\');
	  switch (c)
	    {
	    case '"':
	    case '\\': json_out_byte (jo, c); break;
	    case '\b': json_out_byte (jo, 'b'); break;
	    case '\t': json_out_byte (jo, 't'); break;
	    case '\n': json_out_byte (jo, 'n'); break;
	    case '\f': json_out_byte (jo, 'f'); break;
	    case '\r': json_out_byte (jo, 'r'); break;
	    default:
	      {
		char hex[5] = { 'u', '0', '0',
				hexchar[c >> 4], hexchar[c & 0xf] };
		json_out_str (jo, hex, 5);
		break;
	      }
	    }
	  p++;
	}
    }
  json_out_byte (jo, '"');
}

static void
json_out_nest (json_out_t *jo)
{
  --jo->maxdepth;
  if (jo->maxdepth < 0)
    error ("Maximum JSON serialization depth exceeded");
}

static void
json_out_unnest (json_out_t *jo)
{
  ++jo->maxdepth;
}

static void json_out_something (json_out_t *jo, Lisp_Object obj);

static void
json_out_object_cons (json_out_t *jo, Lisp_Object obj)
{
  json_out_nest (jo);
  symset_t ss = push_symset (jo);
  json_out_byte (jo, '{');
  bool is_alist = CONSP (XCAR (obj));
  bool first = true;
  Lisp_Object tail = obj;
  FOR_EACH_TAIL (tail)
    {
      Lisp_Object key;
      Lisp_Object value;
      if (is_alist)
	{
	  Lisp_Object pair = XCAR (tail);
	  CHECK_CONS (pair);
	  key = XCAR (pair);
	  value = XCDR (pair);
	}
      else
	{
	  key = XCAR (tail);
	  tail = XCDR (tail);
	  CHECK_CONS (tail);
	  value = XCAR (tail);
	}
      key = maybe_remove_pos_from_symbol (key);
      CHECK_TYPE (BARE_SYMBOL_P (key), Qsymbolp, key);

      if (symset_add (jo, &ss, key))
	{
	  if (!first)
	    json_out_byte (jo, ',');
	  first = false;

	  Lisp_Object key_str = SYMBOL_NAME (key);
	  const char *str = SSDATA (key_str);
	  /* Skip leading ':' in plist keys.  */
	  int skip = !is_alist && str[0] == ':' && str[1] ? 1 : 0;
	  json_out_string (jo, key_str, skip);
	  json_out_byte (jo, ':');
	  json_out_something (jo, value);
	}
    }
  CHECK_LIST_END (tail, obj);
  json_out_byte (jo, '}');
  pop_symset (jo, &ss);
  json_out_unnest (jo);
}

static void
json_out_object_hash (json_out_t *jo, Lisp_Object obj)
{
  json_out_nest (jo);
  json_out_byte (jo, '{');
  struct Lisp_Hash_Table *h = XHASH_TABLE (obj);
  bool first = true;
  DOHASH (h, k, v)
    {
      if (!first)
	json_out_byte (jo, ',');
      first = false;
      CHECK_STRING (k);
      /* It's the user's responsibility to ensure that hash keys are
	 unique; we don't check for it.  */
      json_out_string (jo, k, 0);
      json_out_byte (jo, ':');
      json_out_something (jo, v);
    }
  json_out_byte (jo, '}');
  json_out_unnest (jo);

}

static void
json_out_array (json_out_t *jo, Lisp_Object obj)
{
  json_out_nest (jo);
  json_out_byte (jo, '[');
  ptrdiff_t n = ASIZE (obj);
  for (ptrdiff_t i = 0; i < n; i++)
    {
      if (i > 0)
	json_out_byte (jo, ',');
      json_out_something (jo, AREF (obj, i));
    }
  json_out_byte (jo, ']');
  json_out_unnest (jo);
}

static void
json_out_float (json_out_t *jo, Lisp_Object f)
{
  double x = XFLOAT_DATA (f);
  if (!isfinite (x))
    signal_error ("JSON does not allow Inf or NaN", f);
  /* As luck has it, float_to_string emits correct JSON float syntax for
     all numbers (because Vfloat_output_format is Qnil).  */
  json_make_room (jo, FLOAT_TO_STRING_BUFSIZE);
  int n = float_to_string (jo->buf + jo->size, x);
  jo->size += n;
}

static void
json_out_bignum (json_out_t *jo, Lisp_Object x)
{
  int base = 10;
  ptrdiff_t size = bignum_bufsize (x, base);
  json_make_room (jo, size);
  int n = bignum_to_c_string (jo->buf + jo->size, size, x, base);
  jo->size += n;
}

static void
json_out_something (json_out_t *jo, Lisp_Object obj)
{
  if (EQ (obj, jo->conf.null_object))
    JSON_OUT_STR (jo, "null");
  else if (EQ (obj, jo->conf.false_object))
    JSON_OUT_STR (jo, "false");
  else if (EQ (obj, Qt))
    JSON_OUT_STR (jo, "true");
  else if (NILP (obj))
    JSON_OUT_STR (jo, "{}");
  else if (FIXNUMP (obj))
    json_out_fixnum (jo, XFIXNUM (obj));
  else if (STRINGP (obj))
    json_out_string (jo, obj, 0);
  else if (CONSP (obj))
    json_out_object_cons (jo, obj);
  else if (FLOATP (obj))
    json_out_float (jo, obj);
  else if (HASH_TABLE_P (obj))
    json_out_object_hash (jo, obj);
  else if (VECTORP (obj))
    json_out_array (jo, obj);
  else if (BIGNUMP (obj))
    json_out_bignum (jo, obj);
  else
    wrong_type_argument (Qjson_value_p, obj);
}

static void
json_serialize (json_out_t *jo, Lisp_Object object,
		ptrdiff_t nargs, Lisp_Object *args)
{
  jo->maxdepth = 50;
  jo->size = 0;
  jo->capacity = 0;
  jo->chars_delta = 0;
  jo->buf = NULL;
  jo->ss_table = NULL;
  jo->conf.object_type = json_object_hashtable;
  jo->conf.array_type = json_array_array;
  jo->conf.null_object = QCnull;
  jo->conf.false_object = QCfalse;

  json_parse_args (nargs, args, &jo->conf, false);
  record_unwind_protect_ptr (cleanup_json_out, jo);

  /* Make float conversion independent of float-output-format.  */
  if (!NILP (Vfloat_output_format))
    specbind (Qfloat_output_format, Qnil);

  json_out_something (jo, object);
}

DEFUN ("json-serialize", Fjson_serialize, Sjson_serialize, 1, MANY,
       NULL,
       doc: /* Return the JSON representation of OBJECT as a unibyte string.

OBJECT is translated as follows:

`t'        -- the JSON `true' value.
number     -- a JSON number.
string     -- a JSON string.
vector     -- a JSON array.
hash-table -- a JSON object.  Keys must be strings.
alist      -- a JSON object.  Keys must be symbols.
plist      -- a JSON object.  Keys must be symbols.
              A leading colon in plist key names is elided.

For duplicate object keys, the first value is used.

The Lisp equivalents to the JSON null and false values are
configurable in the arguments ARGS, a list of keyword/argument pairs:

:null-object OBJ -- use OBJ to represent a JSON null value.
  It defaults to `:null'.

:false-object OBJ -- use OBJ to represent a JSON false value.
  It defaults to `:false'.

In you specify the same value for `:null-object' and `:false-object',
a potentially ambiguous situation, the JSON output will not contain
any JSON false values.
usage: (json-serialize OBJECT &rest ARGS)  */)
  (ptrdiff_t nargs, Lisp_Object *args)
{
  specpdl_ref count = SPECPDL_INDEX ();
  json_out_t jo;
  json_serialize (&jo, args[0], nargs - 1, args + 1);
  return unbind_to (count, make_unibyte_string (jo.buf, jo.size));
}

DEFUN ("json-insert", Fjson_insert, Sjson_insert, 1, MANY,
       NULL,
       doc: /* Insert the JSON representation of OBJECT before point.
This is the same as (insert (json-serialize OBJECT ...)), but potentially
faster, and with the difference that Unicode characters are inserted as
themselves into multibyte buffers, and as UTF-8 byte sequences into
unibyte buffers.
See the function `json-serialize' for allowed values of OBJECT and ARGS.
usage: (json-insert OBJECT &rest ARGS)  */)
  (ptrdiff_t nargs, Lisp_Object *args)
{
  specpdl_ref count = SPECPDL_INDEX ();
  json_out_t jo;
  json_serialize (&jo, args[0], nargs - 1, args + 1);

  prepare_to_modify_buffer (PT, PT, NULL);
  move_gap_both (PT, PT_BYTE);
  if (GAP_SIZE < jo.size)
    make_gap (jo.size - GAP_SIZE);
  memcpy ((char *) BEG_ADDR + PT_BYTE - BEG_BYTE, jo.buf, jo.size);

  /* No need to keep allocation beyond this point.  */
  unbind_to (count, Qnil);

  bool ub_buffer = NILP (BVAR (current_buffer, enable_multibyte_characters));
  ptrdiff_t inserted_bytes = jo.size;
  ptrdiff_t inserted = ub_buffer ? jo.size : jo.size - jo.chars_delta;
  eassert (inserted > 0);

  insert_from_gap_1 (inserted, inserted_bytes, false);
  invalidate_buffer_caches (current_buffer, PT, PT + inserted);
  adjust_after_insert (PT, PT_BYTE, PT + inserted, PT_BYTE + inserted_bytes,
		       inserted);

  /* Call after-change hooks.  */
  signal_after_change (PT, 0, inserted);

  update_compositions (PT, PT, CHECK_BORDER);
  /* Move point to after the inserted text.  */
  SET_PT_BOTH (PT + inserted, PT_BYTE + inserted_bytes);

  return Qnil;
}


#define JSON_PARSER_INTERNAL_OBJECT_WORKSPACE_SIZE 64
#define JSON_PARSER_INTERNAL_BYTE_WORKSPACE_SIZE 512

struct json_parser
{
  /* Because of a possible gap in the input (an emacs buffer can have
     a gap), the input is described by [input_begin;input_end) and
     [secondary_input_begin;secondary_input_end).  If the input is
     continuous, then secondary_input_begin and secondary_input_end
     should be NULL */
  const unsigned char *input_current;
  const unsigned char *input_begin;
  const unsigned char *input_end;

  const unsigned char *secondary_input_begin;
  const unsigned char *secondary_input_end;

  ptrdiff_t current_line;
  ptrdiff_t current_column;
  ptrdiff_t point_of_current_line;

  /* The parser has a maximum allowed depth.  available_depth
     decreases at each object/array begin.  If reaches zero, then an
     error is generated */
  int available_depth;

  struct json_configuration conf;

  size_t additional_bytes_count;

  /* Lisp_Objects are collected in this area during object/array
     parsing.  To avoid allocations, initially
     internal_object_workspace is used.  If it runs out of space then
     we switch to allocated space.  Important note: with this design,
     GC must not run during JSON parsing, otherwise Lisp_Objects in
     the workspace may get incorrectly collected. */
  Lisp_Object internal_object_workspace
  [JSON_PARSER_INTERNAL_OBJECT_WORKSPACE_SIZE];
  Lisp_Object *object_workspace;
  size_t object_workspace_size;
  size_t object_workspace_current;

  /* String and number parsing uses this workspace.  The idea behind
     internal_byte_workspace is the same as the idea behind
     internal_object_workspace */
  unsigned char
  internal_byte_workspace[JSON_PARSER_INTERNAL_BYTE_WORKSPACE_SIZE];
  unsigned char *byte_workspace;
  unsigned char *byte_workspace_end;
  unsigned char *byte_workspace_current;
};

static AVOID
json_signal_error (struct json_parser *parser, Lisp_Object error)
{
  xsignal3 (error, INT_TO_INTEGER (parser->current_line),
	    INT_TO_INTEGER (parser->current_column),
	    INT_TO_INTEGER (parser->point_of_current_line
			    + parser->current_column));
}

static void
json_parser_init (struct json_parser *parser,
		  struct json_configuration conf,
		  const unsigned char *input,
		  const unsigned char *input_end,
		  const unsigned char *secondary_input,
		  const unsigned char *secondary_input_end)
{
  if (secondary_input >= secondary_input_end)
    {
      secondary_input = NULL;
      secondary_input_end = NULL;
    }

  if (input < input_end)
    {
      parser->input_begin = input;
      parser->input_end = input_end;

      parser->secondary_input_begin = secondary_input;
      parser->secondary_input_end = secondary_input_end;
    }
  else
    {
      parser->input_begin = secondary_input;
      parser->input_end = secondary_input_end;

      parser->secondary_input_begin = NULL;
      parser->secondary_input_end = NULL;
    }

  parser->input_current = parser->input_begin;

  parser->current_line = 1;
  parser->current_column = 0;
  parser->point_of_current_line = 0;
  parser->available_depth = 10000;
  parser->conf = conf;

  parser->additional_bytes_count = 0;

  parser->object_workspace = parser->internal_object_workspace;
  parser->object_workspace_size
    = JSON_PARSER_INTERNAL_OBJECT_WORKSPACE_SIZE;
  parser->object_workspace_current = 0;

  parser->byte_workspace = parser->internal_byte_workspace;
  parser->byte_workspace_end = (parser->byte_workspace
				+ JSON_PARSER_INTERNAL_BYTE_WORKSPACE_SIZE);
}

static void
json_parser_done (void *parser)
{
  struct json_parser *p = (struct json_parser *) parser;
  if (p->object_workspace != p->internal_object_workspace)
    xfree (p->object_workspace);
  if (p->byte_workspace != p->internal_byte_workspace)
    xfree (p->byte_workspace);
}

/* Makes sure that the object_workspace has 'size' available space for
   Lisp_Objects */
NO_INLINE static void
json_make_object_workspace_for_slow_path (struct json_parser *parser,
					  size_t size)
{
  size_t needed_workspace_size
    = (parser->object_workspace_current + size);
  size_t new_workspace_size = parser->object_workspace_size;
  while (new_workspace_size < needed_workspace_size)
    {
      if (ckd_mul (&new_workspace_size, new_workspace_size, 2))
	{
	  json_signal_error (parser, Qjson_out_of_memory);
	}
    }

  Lisp_Object *new_workspace_ptr;
  if (parser->object_workspace_size
      == JSON_PARSER_INTERNAL_OBJECT_WORKSPACE_SIZE)
    {
      new_workspace_ptr
	= xnmalloc (new_workspace_size, sizeof (Lisp_Object));
      memcpy (new_workspace_ptr, parser->object_workspace,
	      (sizeof (Lisp_Object)
	       * parser->object_workspace_current));
    }
  else
    {
      new_workspace_ptr
	= xnrealloc (parser->object_workspace, new_workspace_size,
		     sizeof (Lisp_Object));
    }

  parser->object_workspace = new_workspace_ptr;
  parser->object_workspace_size = new_workspace_size;
}

INLINE void
json_make_object_workspace_for (struct json_parser *parser,
				size_t size)
{
  if (parser->object_workspace_size - parser->object_workspace_current
      < size)
    {
      json_make_object_workspace_for_slow_path (parser, size);
    }
}

static void
json_byte_workspace_reset (struct json_parser *parser)
{
  parser->byte_workspace_current = parser->byte_workspace;
}

/* Puts 'value' into the byte_workspace.  If there is no space
   available, it allocates space */
NO_INLINE static void
json_byte_workspace_put_slow_path (struct json_parser *parser,
				   unsigned char value)
{
  size_t new_workspace_size
    = parser->byte_workspace_end - parser->byte_workspace;
  if (ckd_mul (&new_workspace_size, new_workspace_size, 2))
    {
      json_signal_error (parser, Qjson_out_of_memory);
    }

  size_t offset
    = parser->byte_workspace_current - parser->byte_workspace;

  if (parser->byte_workspace == parser->internal_byte_workspace)
    {
      parser->byte_workspace = xmalloc (new_workspace_size);
      memcpy (parser->byte_workspace, parser->internal_byte_workspace,
	      offset);
    }
  else
    {
      parser->byte_workspace
	= xrealloc (parser->byte_workspace, new_workspace_size);
    }
  parser->byte_workspace_end
    = parser->byte_workspace + new_workspace_size;
  parser->byte_workspace_current = parser->byte_workspace + offset;
  *parser->byte_workspace_current++ = value;
}

INLINE void
json_byte_workspace_put (struct json_parser *parser,
			 unsigned char value)
{
  if (parser->byte_workspace_current < parser->byte_workspace_end)
    {
      *parser->byte_workspace_current++ = value;
    }
  else
    {
      json_byte_workspace_put_slow_path (parser, value);
    }
}

static bool
json_input_at_eof (struct json_parser *parser)
{
  if (parser->input_current < parser->input_end)
    return false;
  return parser->secondary_input_end == NULL;
}

/* If there is a secondary buffer, this switches to it */
static int
json_input_switch_to_secondary (struct json_parser *parser)
{
  if (parser->secondary_input_begin < parser->secondary_input_end)
    {
      parser->additional_bytes_count
	= parser->input_end - parser->input_begin;
      parser->input_begin = parser->secondary_input_begin;
      parser->input_end = parser->secondary_input_end;
      parser->input_current = parser->secondary_input_begin;
      parser->secondary_input_begin = NULL;
      parser->secondary_input_end = NULL;
      return 0;
    }
  else
    return -1;
}

/* Reads a byte from the JSON input stream */
NO_INLINE static unsigned char
json_input_get_slow_path (struct json_parser *parser)
{
  if (json_input_switch_to_secondary (parser) < 0)
    json_signal_error (parser, Qjson_end_of_file);
  return *parser->input_current++;
}

static unsigned char
json_input_get (struct json_parser *parser)
{
  if (parser->input_current < parser->input_end)
    return *parser->input_current++;
  return json_input_get_slow_path (parser);
}

/* Reads a byte from the JSON input stream, if the stream is not at
 * eof.  At eof, returns -1 */
static int
json_input_get_if_possible (struct json_parser *parser)
{
  if (parser->input_current >= parser->input_end
      && json_input_switch_to_secondary (parser) < 0)
    return -1;
  return *parser->input_current++;
}

/* Puts back the last read input byte.  Only one byte can be put back,
   because otherwise this code would need to handle switching from
   the secondary buffer to the initial */
static void
json_input_put_back (struct json_parser *parser)
{
  parser->input_current--;
}

static bool
json_skip_whitespace_internal (struct json_parser *parser, int c)
{
  parser->current_column++;
  if (c == 0x20 || c == 0x09 || c == 0x0d)
    return false;
  else if (c == 0x0a)
    {
      parser->current_line++;
      parser->point_of_current_line += parser->current_column;
      parser->current_column = 0;
      return false;
    }
  else
    return true;
}

/* Skips JSON whitespace, and returns with the first non-whitespace
 * character */
static int
json_skip_whitespace (struct json_parser *parser)
{
  for (;;)
    {
      int c = json_input_get (parser);
      if (json_skip_whitespace_internal (parser, c))
	return c;
    }
}

/* Skips JSON whitespace, and returns with the first non-whitespace
 * character, if possible.  If there is no non-whitespace character
 * (because we reached the end), it returns -1 */
static int
json_skip_whitespace_if_possible (struct json_parser *parser)
{
  for (;;)
    {
      int c = json_input_get_if_possible (parser);
      if (c < 0)
	return c;
      if (json_skip_whitespace_internal (parser, c))
	return c;
    }
}

static int
json_hex_value (int c)
{
  if (c >= '0' && c <= '9')
    return c - '0';
  else if (c >= 'A' && c <= 'F')
    return c - 'A' + 10;
  else if (c >= 'a' && c <= 'f')
    return c - 'a' + 10;
  else
    return -1;
}

/* Parses the CCCC part of the unicode escape sequence \uCCCC */
static int
json_parse_unicode (struct json_parser *parser)
{
  unsigned char v[4];
  for (int i = 0; i < 4; i++)
    {
      int c = json_hex_value (json_input_get (parser));
      parser->current_column++;
      if (c < 0)
	json_signal_error (parser, Qjson_escape_sequence_error);
      v[i] = c;
    }

  return v[0] << 12 | v[1] << 8 | v[2] << 4 | v[3];
}

static AVOID
utf8_error (struct json_parser *parser)
{
  json_signal_error (parser, Qjson_utf8_decode_error);
}

/* Parse a string literal.  Optionally prepend a ':'.
   Return the string or an interned symbol.  */
static Lisp_Object
json_parse_string (struct json_parser *parser, bool intern, bool leading_colon)
{
  json_byte_workspace_reset (parser);
  if (leading_colon)
    json_byte_workspace_put (parser, ':');
  ptrdiff_t chars_delta = 0;	/* nbytes - nchars */
  for (;;)
    {
      /* This if is only here for a possible speedup.  If there are 4
	 bytes available, and all of them are single_uninteresting,
	 then we can just copy these 4 bytes to output */
      if (parser->input_end - parser->input_current >= 4)
	{
	  int c0 = parser->input_current[0];
	  int c1 = parser->input_current[1];
	  int c2 = parser->input_current[2];
	  int c3 = parser->input_current[3];
	  bool v0 = json_plain_char[c0];
	  bool v1 = json_plain_char[c1];
	  bool v2 = json_plain_char[c2];
	  bool v3 = json_plain_char[c3];
	  if (v0 && v1 && v2 && v3)
	    {
	      json_byte_workspace_put (parser, c0);
	      json_byte_workspace_put (parser, c1);
	      json_byte_workspace_put (parser, c2);
	      json_byte_workspace_put (parser, c3);
	      parser->input_current += 4;
	      parser->current_column += 4;
	      continue;
	    }
	}

      int c = json_input_get (parser);
      parser->current_column++;
      if (json_plain_char[c])
	{
	  json_byte_workspace_put (parser, c);
	  continue;
	}

      if (c == '"')
	{
	  ptrdiff_t nbytes
	    = parser->byte_workspace_current - parser->byte_workspace;
	  ptrdiff_t nchars = nbytes - chars_delta;
	  const char *str = (const char *) parser->byte_workspace;
	  return (intern
		  ? intern_c_multibyte (str, nchars, nbytes)
		  : make_multibyte_string (str, nchars, nbytes));
	}

      if (c & 0x80)
	{
	  /* Parse UTF-8, strictly.  This is the correct thing to do
	     whether the input is a unibyte or multibyte string.  */
	  json_byte_workspace_put (parser, c);
	  unsigned char c1 = json_input_get (parser);
	  if ((c1 & 0xc0) != 0x80)
	    utf8_error (parser);
	  json_byte_workspace_put (parser, c1);
	  if (c <= 0xc1)
	    utf8_error (parser);
	  else if (c <= 0xdf)
	    chars_delta += 1;
	  else if (c <= 0xef)
	    {
	      unsigned char c2 = json_input_get (parser);
	      if ((c2 & 0xc0) != 0x80)
		utf8_error (parser);
	      int v = ((c & 0x0f) << 12) + ((c1 & 0x3f) << 6) + (c2 & 0x3f);
	      if (v < 0x800 || (v >= 0xd800 && v <= 0xdfff))
		utf8_error (parser);
	      json_byte_workspace_put (parser, c2);
	      chars_delta += 2;
	    }
	  else if (c <= 0xf7)
	    {
	      unsigned char c2 = json_input_get (parser);
	      unsigned char c3 = json_input_get (parser);
	      if ((c2 & 0xc0) != 0x80 || (c3 & 0xc0) != 0x80)
		utf8_error (parser);
	      int v = (((c & 0x07) << 18) + ((c1 & 0x3f) << 12)
		       + ((c2 & 0x3f) << 6) + (c3 & 0x3f));
	      if (v < 0x10000 || v > 0x10ffff)
		utf8_error (parser);
	      json_byte_workspace_put (parser, c2);
	      json_byte_workspace_put (parser, c3);
	      chars_delta += 3;
	    }
	  else
	    utf8_error (parser);
	}
      else if (c == '\\')
	{
	  /* Handle escape sequences */
	  c = json_input_get (parser);
	  parser->current_column++;
	  if (c == '"')
	    json_byte_workspace_put (parser, '"');
	  else if (c == '\\')
	    json_byte_workspace_put (parser, '\\');
	  else if (c == '/')
	    json_byte_workspace_put (parser, '/');
	  else if (c == 'b')
	    json_byte_workspace_put (parser, '\b');
	  else if (c == 'f')
	    json_byte_workspace_put (parser, '\f');
	  else if (c == 'n')
	    json_byte_workspace_put (parser, '\n');
	  else if (c == 'r')
	    json_byte_workspace_put (parser, '\r');
	  else if (c == 't')
	    json_byte_workspace_put (parser, '\t');
	  else if (c == 'u')
	    {
	      int num = json_parse_unicode (parser);
	      /* is the first half of the surrogate pair */
	      if (num >= 0xd800 && num < 0xdc00)
		{
		  parser->current_column++;
		  if (json_input_get (parser) != '\\')
		    json_signal_error (parser,
				       Qjson_invalid_surrogate_error);
		  parser->current_column++;
		  if (json_input_get (parser) != 'u')
		    json_signal_error (parser,
				       Qjson_invalid_surrogate_error);
		  int num2 = json_parse_unicode (parser);
		  if (num2 < 0xdc00 || num2 >= 0xe000)
		    json_signal_error (parser,
				       Qjson_invalid_surrogate_error);
		  num = (0x10000 + ((num - 0xd800) << 10 | (num2 - 0xdc00)));
		}
	      else if (num >= 0xdc00 && num < 0xe000)
		/* is the second half of the surrogate pair without
		   the first half */
		json_signal_error (parser,
				   Qjson_invalid_surrogate_error);

	      /* utf-8 encode the code-point */
	      if (num < 0x80)
		json_byte_workspace_put (parser, num);
	      else if (num < 0x800)
		{
		  json_byte_workspace_put (parser, 0xc0 | num >> 6);
		  json_byte_workspace_put (parser,
					   0x80 | (num & 0x3f));
		  chars_delta += 1;
		}
	      else if (num < 0x10000)
		{
		  json_byte_workspace_put (parser, 0xe0 | num >> 12);
		  json_byte_workspace_put (parser,
					   (0x80
					    | ((num >> 6) & 0x3f)));
		  json_byte_workspace_put (parser,
					   0x80 | (num & 0x3f));
		  chars_delta += 2;
		}
	      else
		{
		  json_byte_workspace_put (parser, 0xf0 | num >> 18);
		  json_byte_workspace_put (parser,
					   (0x80
					    | ((num >> 12) & 0x3f)));
		  json_byte_workspace_put (parser,
					   (0x80
					    | ((num >> 6) & 0x3f)));
		  json_byte_workspace_put (parser,
					   0x80 | (num & 0x3f));
		  chars_delta += 3;
		}
	    }
	  else
	    json_signal_error (parser, Qjson_escape_sequence_error);
	}
      else
	json_signal_error (parser, Qjson_parse_error);
    }
}

/* If there was no integer overflow during parsing the integer, this
   puts 'value' to the output. Otherwise this calls string_to_number
   to parse integer on the byte workspace.  This could just always
   call string_to_number, but for performance reasons, during parsing
   the code tries to calculate the value, so in most cases, we can
   save call of string_to_number */
static Lisp_Object
json_create_integer (struct json_parser *parser,
		     bool integer_overflow, bool negative,
		     EMACS_UINT value)
{
  if (!integer_overflow)
    {
      if (negative)
	{
	  uintmax_t v = value;
	  if (v <= (uintmax_t) INTMAX_MAX + 1)
	    return INT_TO_INTEGER ((intmax_t) -v);
	}
      else
	return INT_TO_INTEGER (value);
    }

  json_byte_workspace_put (parser, 0);
  ptrdiff_t len;
  Lisp_Object result
    = string_to_number ((const char *) parser->byte_workspace, 10, &len);
  if (len != parser->byte_workspace_current - parser->byte_workspace - 1)
    json_signal_error (parser, Qjson_error);
  return result;
}

/* Parses a float using the byte workspace */
static Lisp_Object
json_create_float (struct json_parser *parser)
{
  json_byte_workspace_put (parser, 0);
  errno = 0;
  char *e;
  double value = strtod ((const char *) parser->byte_workspace, &e);
  bool out_of_range = (errno != 0 && (value == HUGE_VAL || value == -HUGE_VAL));
  if (out_of_range)
    json_signal_error (parser, Qjson_number_out_of_range);
  else if ((const unsigned char *) e != parser->byte_workspace_current - 1)
    json_signal_error (parser, Qjson_error);
  else
    return make_float (value);
}

/* Parses a number.  The first character is the input parameter 'c'.
 */
static Lisp_Object
json_parse_number (struct json_parser *parser, int c)
{
  json_byte_workspace_reset (parser);
  json_byte_workspace_put (parser, c);

  bool negative = false;
  if (c == '-')
    {
      negative = true;
      c = json_input_get (parser);
      json_byte_workspace_put (parser, c);
      parser->current_column++;
    }
  if (c < '0' || c > '9')
    json_signal_error (parser, Qjson_parse_error);

  /* The idea is that during finding the last character of the
     number, the for loop below also tries to calculate the value.  If
     the parsed number is an integer which fits into unsigned long,
     then the parser can use the value of 'integer' right away,
     instead of having to re-parse the byte workspace later.
     Ideally, this integer should have the same size as a CPU general
     purpose register. */
  EMACS_UINT integer = c - '0';
  bool integer_overflow = false;

  if (integer == 0)
    {
      if (json_input_at_eof (parser))
	return INT_TO_INTEGER (0);
      c = json_input_get (parser);
    }
  else
    {
      for (;;)
	{
	  if (json_input_at_eof (parser))
	    return json_create_integer (parser, integer_overflow,
					negative, integer);
	  c = json_input_get (parser);
	  if (c < '0' || c > '9')
	    break;
	  json_byte_workspace_put (parser, c);
	  parser->current_column++;

	  integer_overflow |= ckd_mul (&integer, integer, 10);
	  integer_overflow |= ckd_add (&integer, integer, c - '0');
	}
    }

  bool is_float = false;
  if (c == '.')
    {
      json_byte_workspace_put (parser, c);
      parser->current_column++;

      is_float = true;
      c = json_input_get (parser);
      json_byte_workspace_put (parser, c);
      parser->current_column++;
      if (c < '0' || c > '9')
	json_signal_error (parser, Qjson_parse_error);
      for (;;)
	{
	  if (json_input_at_eof (parser))
	    return json_create_float (parser);
	  c = json_input_get (parser);
	  if (c < '0' || c > '9')
	    break;
	  json_byte_workspace_put (parser, c);
	  parser->current_column++;
	}
    }
  if (c == 'e' || c == 'E')
    {
      json_byte_workspace_put (parser, c);
      parser->current_column++;

      is_float = true;
      c = json_input_get (parser);
      json_byte_workspace_put (parser, c);
      parser->current_column++;
      if (c == '-' || c == '+')
	{
	  c = json_input_get (parser);
	  json_byte_workspace_put (parser, c);
	  parser->current_column++;
	}
      if (c < '0' || c > '9')
	json_signal_error (parser, Qjson_parse_error);
      for (;;)
	{
	  if (json_input_at_eof (parser))
	    return json_create_float (parser);
	  c = json_input_get (parser);
	  if (c < '0' || c > '9')
	    break;
	  json_byte_workspace_put (parser, c);
	  parser->current_column++;
	}
    }

  /* 'c' contains a character which is not part of the number,
     so it is need to be put back */
  json_input_put_back (parser);

  if (is_float)
    return json_create_float (parser);
  else
    return json_create_integer (parser, integer_overflow, negative,
				integer);
}

static Lisp_Object json_parse_value (struct json_parser *parser,
				     int c);

/* Parses a JSON array. */
static Lisp_Object
json_parse_array (struct json_parser *parser)
{
  int c = json_skip_whitespace (parser);

  const size_t first = parser->object_workspace_current;
  Lisp_Object result = Qnil;

  if (c != ']')
    {
      parser->available_depth--;
      if (parser->available_depth < 0)
	json_signal_error (parser, Qjson_object_too_deep);

      Lisp_Object *cdr = &result;
      /* This loop collects the array elements in the object workspace
       */
      for (;;)
	{
	  Lisp_Object element = json_parse_value (parser, c);
	  switch (parser->conf.array_type)
	    {
	    case json_array_array:
	      json_make_object_workspace_for (parser, 1);
	      parser->object_workspace[parser->object_workspace_current]
		= element;
	      parser->object_workspace_current++;
	      break;
	    case json_array_list:
	      {
		Lisp_Object nc = Fcons (element, Qnil);
		*cdr = nc;
		cdr = xcdr_addr (nc);
		break;
	      }
	    default:
	      emacs_abort ();
	    }

	  c = json_skip_whitespace (parser);
	  if (c == ']')
	    {
	      parser->available_depth++;
	      break;
	    }

	  if (c != ',')
	    json_signal_error (parser, Qjson_parse_error);

	  c = json_skip_whitespace (parser);
	}
    }

  switch (parser->conf.array_type)
    {
    case json_array_array:
      {
	size_t number_of_elements
	  = parser->object_workspace_current - first;
	result = make_vector (number_of_elements, Qnil);
	for (size_t i = 0; i < number_of_elements; i++)
	  {
	    rarely_quit (i);
	    ASET (result, i, parser->object_workspace[first + i]);
	  }
	parser->object_workspace_current = first;
	break;
      }
    case json_array_list:
      break;
    default:
      emacs_abort ();
    }

  return result;
}

/* Parses the ": value" part of a JSON object member. */
static Lisp_Object
json_parse_object_member_value (struct json_parser *parser)
{
  int c = json_skip_whitespace (parser);
  if (c != ':')
    json_signal_error (parser, Qjson_parse_error);

  c = json_skip_whitespace (parser);

  return json_parse_value (parser, c);
}

/* Parses a JSON object. */
static Lisp_Object
json_parse_object (struct json_parser *parser)
{
  int c = json_skip_whitespace (parser);

  const size_t first = parser->object_workspace_current;
  Lisp_Object result = Qnil;

  if (c != '}')
    {
      parser->available_depth--;
      if (parser->available_depth < 0)
	json_signal_error (parser, Qjson_object_too_deep);

      Lisp_Object *cdr = &result;

      /* This loop collects the object members (key/value pairs) in
       * the object workspace */
      for (;;)
	{
	  if (c != '"')
	    json_signal_error (parser, Qjson_parse_error);

	  switch (parser->conf.object_type)
	    {
	    case json_object_hashtable:
	      {
		Lisp_Object key = json_parse_string (parser, false, false);
		Lisp_Object value = json_parse_object_member_value (parser);
		json_make_object_workspace_for (parser, 2);
		parser->object_workspace[parser->object_workspace_current] = key;
		parser->object_workspace_current++;
		parser->object_workspace[parser->object_workspace_current] = value;
		parser->object_workspace_current++;
		break;
	      }
	    case json_object_alist:
	      {
		Lisp_Object key = json_parse_string (parser, true, false);
		Lisp_Object value = json_parse_object_member_value (parser);
		Lisp_Object nc = Fcons (Fcons (key, value), Qnil);
		*cdr = nc;
		cdr = xcdr_addr (nc);
		break;
	      }
	    case json_object_plist:
	      {
		Lisp_Object key = json_parse_string (parser, true, true);
		Lisp_Object value = json_parse_object_member_value (parser);
		Lisp_Object nc = Fcons (key, Qnil);
		*cdr = nc;
		cdr = xcdr_addr (nc);

		nc = Fcons (value, Qnil);
		*cdr = nc;
		cdr = xcdr_addr (nc);
		break;
	      }
	    default:
	      emacs_abort ();
	    }

	  c = json_skip_whitespace (parser);

	  if (c == '}')
	    {
	      parser->available_depth++;
	      break;
	    }

	  if (c != ',')
	    json_signal_error (parser, Qjson_parse_error);

	  c = json_skip_whitespace (parser);
	}
    }

  switch (parser->conf.object_type)
    {
    case json_object_hashtable:
      {
	EMACS_INT value = (parser->object_workspace_current - first) / 2;
	result = make_hash_table (&hashtest_equal, value, Weak_None, false);
	struct Lisp_Hash_Table *h = XHASH_TABLE (result);
	for (size_t i = first; i < parser->object_workspace_current; i += 2)
	  {
	    hash_hash_t hash;
	    Lisp_Object key = parser->object_workspace[i];
	    Lisp_Object value = parser->object_workspace[i + 1];
	    ptrdiff_t i = hash_lookup_get_hash (h, key, &hash);
	    if (i < 0)
	      hash_put (h, key, value, hash);
	    else
	      set_hash_value_slot (h, i, value);
	  }
	parser->object_workspace_current = first;
	break;
      }
    case json_object_alist:
    case json_object_plist:
      break;
    default:
      emacs_abort ();
    }

  return result;
}

/* Token-char is not a JSON terminology.  When parsing
   null/false/true, this function tells the character set that is need
   to be considered as part of a token.  For example, if the input is
   "truesomething", then the parser shouldn't consider it as "true",
   and an additional later "something" token. An additional example:
   if the input is "truetrue", then calling (json-parse-buffer) twice
   shouldn't produce two successful calls which return t, but a
   parsing error */
static bool
json_is_token_char (int c)
{
  return ((c >= 'A' && c <= 'Z') || (c >= 'a' && c <= 'z')
	  || (c >= '0' && c <= '9') || (c == '-'));
}

/* This is the entry point to the value parser, this parses a JSON
 * value */
Lisp_Object
json_parse_value (struct json_parser *parser, int c)
{
  if (c == '{')
    return json_parse_object (parser);
  else if (c == '[')
    return json_parse_array (parser);
  else if (c == '"')
    return json_parse_string (parser, false, false);
  else if ((c >= '0' && c <= '9') || (c == '-'))
    return json_parse_number (parser, c);
  else
    {
      int c2 = json_input_get_if_possible (parser);
      int c3 = json_input_get_if_possible (parser);
      int c4 = json_input_get_if_possible (parser);
      int c5 = json_input_get_if_possible (parser);

      if (c == 't' && c2 == 'r' && c3 == 'u' && c4 == 'e'
	  && (c5 < 0 || !json_is_token_char (c5)))
	{
	  if (c5 >= 0)
	    json_input_put_back (parser);
	  parser->current_column += 3;
	  return Qt;
	}
      if (c == 'n' && c2 == 'u' && c3 == 'l' && c4 == 'l'
	  && (c5 < 0 || !json_is_token_char (c5)))
	{
	  if (c5 >= 0)
	    json_input_put_back (parser);
	  parser->current_column += 3;
	  return parser->conf.null_object;
	}
      if (c == 'f' && c2 == 'a' && c3 == 'l' && c4 == 's'
	  && c5 == 'e')
	{
	  int c6 = json_input_get_if_possible (parser);
	  if (c6 < 0 || !json_is_token_char (c6))
	    {
	      if (c6 >= 0)
		json_input_put_back (parser);
	      parser->current_column += 4;
	      return parser->conf.false_object;
	    }
	}

      json_signal_error (parser, Qjson_parse_error);
    }
}

static Lisp_Object
json_parse (struct json_parser *parser)
{
  return json_parse_value (parser, json_skip_whitespace (parser));
}

DEFUN ("json-parse-string", Fjson_parse_string, Sjson_parse_string, 1, MANY,
       NULL,
       doc: /* Parse the JSON STRING into a Lisp value.
This is essentially the reverse operation of `json-serialize', which
see.  The returned value will be the JSON null value, the JSON false
value, t, a number, a string, a vector, a list, a hash-table, an alist,
or a plist.  Its elements will be further values of these types.
If STRING doesn't contain a valid JSON value, this function
signals an error of type `json-parse-error'.

The arguments ARGS are a list of keyword/argument pairs:

:object-type TYPE -- use TYPE to represent JSON objects.
  TYPE can be `hash-table' (the default), `alist' or `plist'.
  If an object has members with the same key, `hash-table' keeps only
  the last value of such keys, while `alist' and `plist' keep all the
  members.

:array-type TYPE -- use TYPE to represent JSON arrays.
  TYPE can be `array' (the default) or `list'.

:null-object OBJ -- use OBJ to represent a JSON null value.
  It defaults to `:null'.

:false-object OBJ -- use OBJ to represent a JSON false value.
  It defaults to `:false'.
usage: (json-parse-string STRING &rest ARGS) */)
(ptrdiff_t nargs, Lisp_Object *args)
{
  specpdl_ref count = SPECPDL_INDEX ();

  Lisp_Object string = args[0];
  CHECK_STRING (string);
  struct json_configuration conf
    = { json_object_hashtable, json_array_array, QCnull, QCfalse };
  json_parse_args (nargs - 1, args + 1, &conf, true);

  struct json_parser p;
  const unsigned char *begin = SDATA (string);
  json_parser_init (&p, conf, begin, begin + SBYTES (string), NULL, NULL);
  record_unwind_protect_ptr (json_parser_done, &p);
  Lisp_Object result = json_parse (&p);

  if (json_skip_whitespace_if_possible (&p) >= 0)
    json_signal_error (&p, Qjson_trailing_content);

  return unbind_to (count, result);
}

DEFUN ("json-parse-buffer", Fjson_parse_buffer, Sjson_parse_buffer,
       0, MANY, NULL,
       doc: /* Read a JSON value from current buffer starting at point.
Move point after the end of the value if parsing was successful.
On error, don't move point.

The returned value will be a vector, list, hashtable, alist, or
plist.  Its elements will be the JSON null value, the JSON false
value, t, numbers, strings, or further vectors, lists, hashtables,
alists, or plists.

If the current buffer doesn't contain a valid JSON value, the
function signals an error of type `json-parse-error'.

The arguments ARGS are a list of keyword/argument pairs:

:object-type TYPE -- use TYPE to represent JSON objects.
  TYPE can be `hash-table' (the default), `alist' or `plist'.
  If an object has members with the same key, `hash-table' keeps only
  the last value of such keys, while `alist' and `plist' keep all the
  members.

:array-type TYPE -- use TYPE to represent JSON arrays.
  TYPE can be `array' (the default) or `list'.

:null-object OBJ -- use OBJ to represent a JSON null value.
  It defaults to `:null'.

:false-object OBJ -- use OBJ to represent a JSON false value.
  It defaults to `:false'.
usage: (json-parse-buffer &rest args) */)
(ptrdiff_t nargs, Lisp_Object *args)
{
  specpdl_ref count = SPECPDL_INDEX ();

  struct json_configuration conf
    = { json_object_hashtable, json_array_array, QCnull, QCfalse };
  json_parse_args (nargs, args, &conf, true);

  struct json_parser p;
  unsigned char *begin = PT_ADDR;
  unsigned char *end = GPT_ADDR;
  unsigned char *secondary_begin = NULL;
  unsigned char *secondary_end = NULL;
  if (GPT_ADDR < Z_ADDR)
    {
      secondary_begin = GAP_END_ADDR;
      if (secondary_begin < PT_ADDR)
	secondary_begin = PT_ADDR;
      secondary_end = Z_ADDR;
    }

  json_parser_init (&p, conf, begin, end, secondary_begin,
		    secondary_end);
  record_unwind_protect_ptr (json_parser_done, &p);
  Lisp_Object result = json_parse (&p);

  ptrdiff_t byte = (PT_BYTE + p.input_current - p.input_begin
		    + p.additional_bytes_count);
  ptrdiff_t position = (NILP (BVAR (current_buffer,
				    enable_multibyte_characters))
			? byte
			: PT + p.point_of_current_line + p.current_column);
  SET_PT_BOTH (position, byte);

  return unbind_to (count, result);
}

void
syms_of_json (void)
{
  DEFSYM (QCnull, ":null");
  DEFSYM (QCfalse, ":false");

  DEFSYM (Qjson_value_p, "json-value-p");

  DEFSYM (Qjson_error, "json-error");
  DEFSYM (Qjson_out_of_memory, "json-out-of-memory");
  DEFSYM (Qjson_parse_error, "json-parse-error");
  DEFSYM (Qjson_end_of_file, "json-end-of-file");
  DEFSYM (Qjson_trailing_content, "json-trailing-content");
  DEFSYM (Qjson_object_too_deep, "json-object-too-deep");
  DEFSYM (Qjson_utf8_decode_error, "json-utf8-decode-error");
  DEFSYM (Qjson_invalid_surrogate_error, "json-invalid-surrogate-error");
  DEFSYM (Qjson_number_out_of_range, "json-number-out-of-range-error");
  DEFSYM (Qjson_escape_sequence_error, "json-escape-sequence-error");
  define_error (Qjson_error, "generic JSON error", Qerror);
  define_error (Qjson_out_of_memory,
		"not enough memory for creating JSON object", Qjson_error);
  define_error (Qjson_parse_error, "could not parse JSON stream",
		Qjson_error);
  define_error (Qjson_end_of_file, "end of JSON stream", Qjson_parse_error);
  define_error (Qjson_trailing_content, "trailing content after JSON stream",
		Qjson_parse_error);
  define_error (Qjson_object_too_deep,
		"object cyclic or Lisp evaluation too deep", Qjson_error);
  define_error (Qjson_utf8_decode_error,
		"invalid utf-8 encoding", Qjson_error);
  define_error (Qjson_invalid_surrogate_error,
		"invalid surrogate pair", Qjson_error);
  define_error (Qjson_number_out_of_range,
		"number out of range", Qjson_error);
  define_error (Qjson_escape_sequence_error,
		"invalid escape sequence", Qjson_parse_error);

  DEFSYM (Qpure, "pure");
  DEFSYM (Qside_effect_free, "side-effect-free");

  DEFSYM (Qjson_serialize, "json-serialize");
  DEFSYM (Qjson_parse_string, "json-parse-string");
  Fput (Qjson_serialize, Qpure, Qt);
  Fput (Qjson_serialize, Qside_effect_free, Qt);
  Fput (Qjson_parse_string, Qpure, Qt);
  Fput (Qjson_parse_string, Qside_effect_free, Qt);

  DEFSYM (QCobject_type, ":object-type");
  DEFSYM (QCarray_type, ":array-type");
  DEFSYM (QCnull_object, ":null-object");
  DEFSYM (QCfalse_object, ":false-object");
  DEFSYM (Qalist, "alist");
  DEFSYM (Qplist, "plist");
  DEFSYM (Qarray, "array");

  defsubr (&Sjson_serialize);
  defsubr (&Sjson_insert);
  defsubr (&Sjson_parse_string);
  defsubr (&Sjson_parse_buffer);
}

debug log:

solving eb446f5c221 ...
found eb446f5c221 in https://git.savannah.gnu.org/cgit/emacs.git

(*) Git path names are given by the tree(s) the blob belongs to.
    Blobs themselves have no identifier aside from the hash of its contents.^

Code repositories for project(s) associated with this public inbox

	https://git.savannah.gnu.org/cgit/emacs.git

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).