summaryrefslogtreecommitdiffstats
path: root/www/opcode.html
blob: 1e9fa7d1bff71700778ddde6279ff679cc8e7e70 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
1554
1555
1556
1557
1558
1559
1560
1561
1562
1563
1564
1565
1566
1567
1568
1569
1570
1571
1572
1573
1574
1575
1576
1577
1578
1579
1580
1581
1582
1583
1584
1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
1598
1599
1600
1601
1602
1603
1604
1605
1606
1607
1608
1609
1610
1611
1612
1613
1614
1615
1616
1617
1618
1619
1620
1621
1622
1623
1624
1625
1626
1627
1628
1629
1630
1631
1632
1633
1634
1635
1636
1637
1638
1639
1640
1641
1642
1643
1644
1645
1646
1647
1648
1649
1650
1651
1652
1653
1654
1655
1656
1657
1658
1659
1660
1661
1662
1663
1664
1665
1666
1667
1668
1669
1670
1671
1672
1673
1674
1675
1676
1677
1678
1679
1680
1681
1682
1683
1684
1685
1686
1687
1688
1689
1690
1691
1692
1693
1694
1695
1696
1697
1698
1699
1700
1701
1702
1703
1704
1705
1706
1707
1708
1709
1710
1711
1712
1713
1714
1715
1716
1717
1718
1719
1720
1721
1722
1723
1724
1725
1726
1727
1728
1729
1730
1731
1732
1733
1734
1735
1736
1737
1738
1739
1740
1741
1742
1743
1744
1745
1746
1747
1748
1749
1750
1751
1752
1753
1754
1755
1756
1757
1758
1759
1760
1761
1762
1763
1764
1765
1766
1767
1768
1769
1770
1771
1772
1773
1774
1775
1776
1777
1778
1779
1780
1781
1782
1783
1784
1785
1786
1787
1788
1789
1790
1791
1792
1793
1794
1795
1796
1797
1798
1799
1800
1801
1802
1803
1804
1805
1806
1807
1808
1809
1810
1811
1812
1813
1814
1815
1816
1817
1818
1819
1820
1821
1822
1823
1824
1825
1826
1827
1828
1829
1830
1831
1832
1833
1834
1835
1836
1837
1838
1839
1840
1841
1842
1843
1844
1845
1846
1847
1848
1849
1850
1851
1852
1853
1854
1855
1856
1857
1858
1859
1860
1861
1862
1863
1864
1865
1866
1867
1868
1869
1870
1871
1872
1873
1874
1875
1876
1877
1878
1879
1880
1881
1882
1883
1884
1885
1886
1887
1888
1889
1890
1891
1892
1893
1894
1895
1896
1897
1898
1899
1900
1901
1902
1903
1904
1905
1906
1907
1908
1909
1910
1911
1912
1913
1914
1915
1916
1917
1918
1919
1920
1921
1922
1923
1924
1925
1926
1927
1928
1929
1930
1931
1932
1933
1934
1935
1936
1937
1938
1939
1940
1941
1942
1943
1944
1945
1946
1947
1948
1949
1950
1951
1952
1953
1954
1955
1956
1957
1958
1959
1960
1961
1962
1963
1964
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
2036
2037
2038
2039
2040
2041
2042
2043
2044
2045
2046
2047
2048
2049
2050
2051
2052
2053
2054
2055
2056
2057
2058
2059
2060
2061
2062
2063
2064
2065
2066
2067
2068
2069
2070
2071
2072
2073
2074
2075
2076
2077
2078
2079
2080
2081
2082
2083
2084
2085
2086
2087
2088
2089
2090
2091
2092
2093
2094
2095
2096
2097
2098
2099
2100
2101
2102
2103
2104
2105
2106
2107
2108
2109
2110
2111
2112
2113
2114
2115
2116
2117
2118
2119
2120
2121
2122
2123
2124
2125
2126
2127
2128
2129
2130
2131
2132
2133
2134
2135
2136
2137
2138
2139
2140
2141
2142
2143
2144
2145
2146
2147
2148
2149
2150
2151
2152
2153
2154
2155
2156
2157
2158
2159
2160
2161
2162
2163
2164
2165
2166
2167
2168
2169
2170
2171
2172
2173
2174
2175
2176
2177
2178
2179
2180
2181
2182
2183
2184
2185
2186
2187
2188
2189
2190
2191
2192
2193
2194
2195
2196
2197
2198
2199
2200
2201
2202
2203
2204
2205
2206
2207
2208
2209
2210
2211
2212
2213
2214
2215
2216
2217
2218
2219
2220
2221
2222
2223
2224
2225
2226
2227
2228
2229
2230
2231
2232
2233
2234
2235
2236
2237
2238
2239
2240
2241
2242
2243
2244
2245
2246
2247
2248
2249
2250
2251
2252
2253
2254
2255
2256
2257
2258
2259
2260
2261
2262
2263
2264
2265
2266
2267
2268
2269
2270
2271
2272
2273
2274
2275
2276
2277
2278
2279
2280
2281
2282
2283
2284
2285
2286
2287
2288
2289
2290
2291
2292
2293
2294
2295
2296
2297
2298
2299
2300
2301
2302
2303
2304
2305
2306
2307
2308
2309
2310
2311
2312
2313
2314
2315
2316
2317
2318
2319
2320
2321
2322
2323
2324
2325
2326
2327
2328
2329
2330
2331
2332
2333
2334
2335
2336
2337
2338
2339
2340
2341
2342
2343
2344
2345
2346
2347
2348
2349
2350
2351
2352
2353
2354
2355
2356
2357
2358
2359
2360
2361
2362
2363
2364
2365
2366
2367
2368
2369
2370
2371
2372
2373
2374
2375
2376
2377
2378
2379
2380
2381
2382
2383
2384
2385
2386
2387
2388
2389
2390
2391
2392
2393
2394
2395
2396
2397
2398
2399
2400
2401
2402
2403
2404
2405
2406
2407
2408
2409
2410
2411
2412
2413
2414
2415
2416
2417
2418
2419
2420
2421
2422
2423
2424
2425
2426
2427
2428
2429
2430
2431
2432
2433
2434
2435
2436
2437
2438
2439
2440
2441
2442
2443
2444
2445
2446
2447
2448
2449
2450
2451
2452
2453
2454
2455
2456
2457
2458
2459
2460
2461
2462
2463
2464
2465
2466
2467
2468
2469
2470
2471
2472
2473
2474
2475
2476
2477
2478
2479
2480
2481
2482
2483
2484
2485
2486
2487
2488
2489
2490
2491
2492
2493
2494
2495
2496
2497
2498
2499
2500
2501
2502
2503
2504
2505
2506
2507
2508
2509
2510
2511
2512
2513
2514
2515
2516
2517
2518
2519
2520
2521
2522
2523
2524
2525
2526
2527
2528
2529
2530
2531
2532
2533
2534
2535
2536
2537
2538
2539
2540
2541
2542
2543
2544
2545
2546
2547
2548
2549
2550
2551
2552
2553
2554
2555
2556
2557
2558
2559
2560
2561
2562
2563
2564
2565
2566
2567
2568
2569
2570
2571
2572
2573
2574
2575
2576
2577
2578
2579
2580
2581
2582
2583
2584
2585
2586
2587
2588
2589
2590
2591
2592
2593
2594
2595
2596
2597
2598
2599
2600
2601
2602
2603
2604
2605
2606
2607
2608
2609
2610
2611
2612
2613
2614
2615
2616
2617
2618
2619
2620
2621
2622
2623
<!DOCTYPE html>
<html><head>
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<meta http-equiv="content-type" content="text/html; charset=UTF-8">
<link href="sqlite.css" rel="stylesheet">
<title>The SQLite Bytecode Engine</title>
<!-- path= -->
</head>
<body>
<div class=nosearch>
<a href="index.html">
<img class="logo" src="images/sqlite370_banner.gif" alt="SQLite" border="0">
</a>
<div><!-- IE hack to prevent disappearing logo --></div>
<div class="tagline desktoponly">
Small. Fast. Reliable.<br>Choose any three.
</div>
<div class="menu mainmenu">
<ul>
<li><a href="index.html">Home</a>
<li class='mobileonly'><a href="javascript:void(0)" onclick='toggle_div("submenu")'>Menu</a>
<li class='wideonly'><a href='about.html'>About</a>
<li class='desktoponly'><a href="docs.html">Documentation</a>
<li class='desktoponly'><a href="download.html">Download</a>
<li class='wideonly'><a href='copyright.html'>License</a>
<li class='desktoponly'><a href="support.html">Support</a>
<li class='desktoponly'><a href="prosupport.html">Purchase</a>
<li class='search' id='search_menubutton'>
<a href="javascript:void(0)" onclick='toggle_search()'>Search</a>
</ul>
</div>
<div class="menu submenu" id="submenu">
<ul>
<li><a href='about.html'>About</a>
<li><a href='docs.html'>Documentation</a>
<li><a href='download.html'>Download</a>
<li><a href='support.html'>Support</a>
<li><a href='prosupport.html'>Purchase</a>
</ul>
</div>
<div class="searchmenu" id="searchmenu">
<form method="GET" action="search">
<select name="s" id="searchtype">
<option value="d">Search Documentation</option>
<option value="c">Search Changelog</option>
</select>
<input type="text" name="q" id="searchbox" value="">
<input type="submit" value="Go">
</form>
</div>
</div>
<script>
function toggle_div(nm) {
var w = document.getElementById(nm);
if( w.style.display=="block" ){
w.style.display = "none";
}else{
w.style.display = "block";
}
}
function toggle_search() {
var w = document.getElementById("searchmenu");
if( w.style.display=="block" ){
w.style.display = "none";
} else {
w.style.display = "block";
setTimeout(function(){
document.getElementById("searchbox").focus()
}, 30);
}
}
function div_off(nm){document.getElementById(nm).style.display="none";}
window.onbeforeunload = function(e){div_off("submenu");}
/* Disable the Search feature if we are not operating from CGI, since */
/* Search is accomplished using CGI and will not work without it. */
if( !location.origin || !location.origin.match || !location.origin.match(/http/) ){
document.getElementById("search_menubutton").style.display = "none";
}
/* Used by the Hide/Show button beside syntax diagrams, to toggle the */
function hideorshow(btn,obj){
var x = document.getElementById(obj);
var b = document.getElementById(btn);
if( x.style.display!='none' ){
x.style.display = 'none';
b.innerHTML='show';
}else{
x.style.display = '';
b.innerHTML='hide';
}
return false;
}
var antiRobot = 0;
function antiRobotGo(){
if( antiRobot!=3 ) return;
antiRobot = 7;
var j = document.getElementById("mtimelink");
if(j && j.hasAttribute("data-href")) j.href=j.getAttribute("data-href");
}
function antiRobotDefense(){
document.body.onmousedown=function(){
antiRobot |= 2;
antiRobotGo();
document.body.onmousedown=null;
}
document.body.onmousemove=function(){
antiRobot |= 2;
antiRobotGo();
document.body.onmousemove=null;
}
setTimeout(function(){
antiRobot |= 1;
antiRobotGo();
}, 100)
antiRobotGo();
}
antiRobotDefense();
</script>
<div class=fancy>
<div class=nosearch>
<div class="fancy_title">
The SQLite Bytecode Engine
</div>
<div class="fancy_toc">
<a onclick="toggle_toc()">
<span class="fancy_toc_mark" id="toc_mk">&#x25ba;</span>
Table Of Contents
</a>
<div id="toc_sub"><div class="fancy-toc1"><a href="#executive_summary">1. Executive Summary</a></div>
<div class="fancy-toc1"><a href="#introduction">2. Introduction</a></div>
<div class="fancy-toc2"><a href="#vdbe_source_code">2.1. VDBE Source Code</a></div>
<div class="fancy-toc2"><a href="#instruction_format">2.2. Instruction Format</a></div>
<div class="fancy-toc2"><a href="#registers">2.3. Registers</a></div>
<div class="fancy-toc2"><a href="#b_tree_cursors">2.4. B-Tree Cursors</a></div>
<div class="fancy-toc2"><a href="#subroutines_coroutines_and_subprograms">2.5. Subroutines, Coroutines, and Subprograms</a></div>
<div class="fancy-toc2"><a href="#self_altering_code">2.6. Self-Altering Code</a></div>
<div class="fancy-toc1"><a href="#viewing_the_bytecode">3. Viewing The Bytecode</a></div>
<div class="fancy-toc1"><a href="#the_opcodes">4. The Opcodes</a></div>
</div>
</div>
<script>
function toggle_toc(){
var sub = document.getElementById("toc_sub")
var mk = document.getElementById("toc_mk")
if( sub.style.display!="block" ){
sub.style.display = "block";
mk.innerHTML = "&#x25bc;";
} else {
sub.style.display = "none";
mk.innerHTML = "&#x25ba;";
}
}
</script>
</div>





<h1 id="executive_summary"><span>1. </span>Executive Summary</h1>

<p>SQLite works by translating SQL statements into bytecode and
then running that bytecode in a virtual machine.  This document
describes how the bytecode engine works.

</p><p>This document describes SQLite internals.  The information provided
here is not needed for routine application development using SQLite.
This document is intended for people who want to delve more deeply into
the internal operation of SQLite.

</p><p>The bytecode engine is <u>not</u> an API of SQLite.  Details
about the bytecode engine change from one release of SQLite to the next.
Applications that use SQLite should not depend on any of the details
found in this document.



</p><h1 id="introduction"><span>2. </span>Introduction</h1>

<p>SQLite works by translating each SQL statement into bytecode and
then running that bytecode.
A <a href="c3ref/stmt.html">prepared statement</a> in SQLite is mostly just the bytecode needed to
implement the corresponding SQL.  The <a href="c3ref/prepare.html">sqlite3_prepare_v2()</a> interface
is a compiler that translates SQL into bytecode.
The <a href="c3ref/step.html">sqlite3_step()</a> interface is the virtual machine that runs the
bytecode contained within the <a href="c3ref/stmt.html">prepared statement</a>.

</p><p>The bytecode virtual machine is the heart of SQLite.
Programmers who want to understand how SQLite operates internally
must be familiar with the bytecode engine.

</p><p>Historically, the bytecode engine in SQLite is called the
"Virtual DataBase Engine" or "VDBE".  This website uses the terms
"bytecode engine", "VDBE", "virtual machine", and "bytecode virtual
machine" interchangeably, as they all mean the same thing.

</p><p>
This article also uses the terms "bytecode program" and
"prepared statement" interchangeably, as they are mostly the same thing.

</p><h2 id="vdbe_source_code"><span>2.1. </span>VDBE Source Code</h2>

<p>The source code to the bytecode engine is in the 
<a href="http://www.sqlite.org/src/file/src/vdbe.c">vdbe.c</a> source
file.  The <a href="opcode.html#codes">opcode definitions</a> in this document are derived
from comments in that source file. The 
source code comments are the canonical source of information
about the bytecode engine.  When in doubt, refer to the source code.</p>

<p>In addition to the primary vdbe.c source code file, there are 
other helper code files in the source tree, all of whose names
begin with "vdbe" - short for "Virtual DataBase Engine".

</p><p>Remember that the names and meanings of opcodes often change from
one release of SQLite to the next.  So if you are studying the <a href="lang_explain.html">EXPLAIN</a>
output from SQLite, you should reference the version of this document
(or the vdbe.c source code)
that corresponds to the version of SQLite that ran the <a href="lang_explain.html">EXPLAIN</a>.
Otherwise, the description of the opcodes may not be accurate.
This document is derived from SQLite
 version 3.40.1 check-in
<a href='https://www.sqlite.org/src/timeline?c=df5c253c0b3dd'>df5c253c0b3dd</a> dated 2022-12-28.




</p><h2 id="instruction_format"><span>2.2. </span>Instruction Format</h2>


<p>A bytecoded program in SQLite consists of one or more instructions.
Each instruction has an opcode and
five operands named P1, P2  P3, P4, and P5.  The P1, P2, and P3 
operands are 32-bit signed integers.  These operands often refer to 
registers.  For instructions that operate on b-tree cursors,
the P1 operand is usually the cursor number.
For jump instructions, P2 is usually the jump destination.
P4 may be a 32-bit signed integer, a 64-bit signed integer, a
64-bit floating point value, a string literal, a Blob literal,
a pointer to a collating sequence comparison function, or a
pointer to the implementation of an application-defined SQL
function, or various other things.  P5 is an 16-bit unsigned integer
normally used to hold flags.  Bits of the P5 flag can sometimes affect
the opcode in subtle ways.  For example, if the 
SQLITE_NULLEQ (0x0080) bit of the P5 operand
is set on the <a href="opcode.html#Eq">Eq</a> opcode, then the NULL values compare
equal to one another.  Otherwise NULL values compare different
from one another.


<p>Some opcodes use all five operands.  Some opcodes use
one or two.  Some opcodes use none of the operands.</p><p>


<p>The bytecode engine begins execution on instruction number 0.
Execution continues until a <a href="opcode.html#Halt">Halt</a> instruction is seen, or until
the program counter becomes greater than the address of
last instruction, or until there is an error.
When the bytecode engine halts, all memory
that it allocated is released and all database cursors it may
have had open are closed.  If the execution stopped due to an
error, any pending transactions are terminated and changes made
to the database are rolled back.</p>



<p>The <a href="opcode.html#ResultRow">ResultRow</a> opcode causes the
bytecode engine to pause and the corresponding <a href="c3ref/step.html">sqlite3_step()</a>
call to return <a href="rescode.html#row">SQLITE_ROW</a>.  Before invoking
<a href="opcode.html#ResultRow">ResultRow</a>, the bytecoded program will
have loaded the results for a single row of a query into a series
of registers.  C-language APIs such as <a href="c3ref/column_blob.html">sqlite3_column_int()</a>
or <a href="c3ref/column_blob.html">sqlite3_column_text()</a> extract the query results from those
registers.  The bytecode engine resumes with the next instruction
after the <a href="opcode.html#ResultRow">ResultRow</a> on the next call
to <a href="c3ref/step.html">sqlite3_step()</a>.


</p><h2 id="registers"><span>2.3. </span>Registers</h2>


<p>Every bytecode program has a fixed (but potentially large) number of
registers.  A single register can hold a variety of objects:
<ul>
<li> A NULL value
<li> A signed 64-bit integer
<li> An IEEE double-precision (64-bit) floating point number
<li> An arbitrary length strings
<li> An arbitrary length BLOB
<li> A RowSet object (See the <a href="opcode.html#RowSetAdd">RowSetAdd</a>, <a href="opcode.html#RowSetRead">RowSetRead</a>, and
                      <a href="opcode.html#RowSetTest">RowSetTest</a> opcodes)
<li> A Frame object (Used by <a href="opcode.html#subprog">subprograms</a> - see <a href="opcode.html#Program">Program</a>)
</ul>


<p>A register can also be "Undefined" meaning that it holds no value
at all.  Undefined is different from NULL.  Depending on compile-time
options, an attempt to read an undefined register will usually cause
a run-time error.  If the code generator (<a href="c3ref/prepare.html">sqlite3_prepare_v2()</a>)
ever generates a <a href="c3ref/stmt.html">prepared statement</a> that reads an Undefined register,
that is a bug in the code generator.

</p><p>
Registers are numbered beginning with 0.
Most opcodes refer to at least one register.

</p><p>The number of registers in a single prepared statement is fixed
at compile-time.  The content of all registers is cleared when
a prepared statement is <a href="c3ref/reset.html">reset</a> or
<a href="c3ref/finalize.html">finalized</a>.

</p><p>The internal Mem object stores the value for a single register.
The abstract <a href="c3ref/value.html">sqlite3_value</a> object that is exposed in the API is really
just a Mem object or register.

</p><h2 id="b_tree_cursors"><span>2.4. </span>B-Tree Cursors</h2>


<p>A prepared statement can have
zero or more open cursors.  Each cursor is identified by a
small integer, which is usually the P1 parameter to the opcode
that uses the cursor.
There can be multiple cursors open on the same index or table.
All cursors operate independently, even cursors pointing to the same
indices or tables.
The only way for the virtual machine to interact with a database
file is through a cursor.
Instructions in the virtual machine can create a new cursor 
(ex: <a href="opcode.html#OpenRead">OpenRead</a> or <a href="opcode.html#OpenWrite">OpenWrite</a>),
read data from a cursor (<a href="opcode.html#Column">Column</a>),
advance the cursor to the next entry in the table
(ex: <a href="opcode.html#Next">Next</a> or <a href="opcode.html#Prev">Prev</a>), and so forth.
All cursors are automatically
closed when the prepared statement is <a href="c3ref/reset.html">reset</a> or
<a href="c3ref/finalize.html">finalized</a>.


<a name="subprog"></a>

<h2 id="subroutines_coroutines_and_subprograms"><span>2.5. </span>Subroutines, Coroutines, and Subprograms</h2>

<p>The bytecode engine has no stack on which to store the return address
of a subroutine.  Return addresses must be stored in registers.
Hence, bytecode subroutines are not reentrant.


<p>The <a href="opcode.html#Gosub">Gosub</a> opcode stores the current program counter into
register P1 then jumps to address P2.  The <a href="opcode.html#Return">Return</a> opcode jumps
to address P1+1.  Hence, every subroutine is associated with two integers:
the address of the entry point in the subroutine and the register number
that is used to hold the return address.

<p>The <a href="opcode.html#Yield">Yield</a> opcode swaps the value of the program counter with
the integer value in register P1.  This opcode is used to implement
coroutines.  Coroutines are often used to implement subqueries from
which content is pulled on an as-needed basis.


</p><p><a href="lang_createtrigger.html">Triggers</a> need to be reentrant.


Since bytecode
subroutines are not reentrant a different mechanism must be used to
implement triggers.  Each trigger is implemented using a separate bytecode
program with its own opcodes, program counter, and register set.  The
<a href="opcode.html#Program">Program</a> opcode invokes the trigger subprogram.  The <a href="opcode.html#Program">Program</a> instruction
allocates and initializes a fresh register set for each invocation of the
subprogram, so subprograms can be reentrant and recursive.  The
<a href="opcode.html#Param">Param</a> opcode is used by subprograms to access content in registers
of the calling bytecode program.


</p><h2 id="self_altering_code"><span>2.6. </span>Self-Altering Code</h2>


<p>Some opcodes are self-altering.
For example, the <a href="opcode.html#Init">Init</a> opcode (which is always the first opcode
in every bytecode program) increments its P1 operand.  Subsequent
<a href="opcode.html#Once">Once</a> opcodes compare their P1 operands to the P1 value for
the <a href="opcode.html#Init">Init</a> opcode in order to determine if the one-time initialization
code that follows should be skipped.
Another example is the <a href="opcode.html#String8">String8</a> opcode which converts its P4
operand from UTF-8 into the correct database string encoding, then
converts itself into a <a href="opcode.html#String">String</a> opcode.



<h1 id="viewing_the_bytecode"><span>3. </span>Viewing The Bytecode</h1>

<p>Every SQL statement that SQLite interprets results in a program
for the virtual machine.  But if the SQL statement begins with
the keyword <a href="lang_explain.html">EXPLAIN</a> the virtual machine will not execute the
program.  Instead, the instructions of the program will be returned,
one instruction per row,
like a query result.  This feature is useful for debugging and
for learning how the virtual machine operates.  For example:
</p>

<blockquote><pre>$&nbsp;<b>sqlite3&nbsp;ex1.db</b>
sqlite&gt;&nbsp;<b>explain&nbsp;delete&nbsp;from&nbsp;tbl1&nbsp;where&nbsp;two&lt;20;</b>
addr&nbsp;&nbsp;opcode&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;p1&nbsp;&nbsp;&nbsp;&nbsp;p2&nbsp;&nbsp;&nbsp;&nbsp;p3&nbsp;&nbsp;&nbsp;&nbsp;p4&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;p5&nbsp;&nbsp;comment&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
----&nbsp;&nbsp;-------------&nbsp;&nbsp;----&nbsp;&nbsp;----&nbsp;&nbsp;----&nbsp;&nbsp;-------------&nbsp;&nbsp;--&nbsp;&nbsp;-------------
0&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;Init&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;0&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;12&nbsp;&nbsp;&nbsp;&nbsp;0&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;00&nbsp;&nbsp;Start&nbsp;at&nbsp;12&nbsp;&nbsp;
1&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;Null&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;0&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;1&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;0&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;00&nbsp;&nbsp;r[1]=NULL&nbsp;&nbsp;&nbsp;&nbsp;
2&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;OpenWrite&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;0&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;2&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;0&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;3&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;00&nbsp;&nbsp;root=2&nbsp;iDb=0;&nbsp;tbl1
3&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;Rewind&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;0&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;10&nbsp;&nbsp;&nbsp;&nbsp;0&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;00&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
4&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;Column&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;0&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;1&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;2&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;00&nbsp;&nbsp;r[2]=tbl1.two
5&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;Ge&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;3&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;9&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;2&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;(BINARY)&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;51&nbsp;&nbsp;if&nbsp;r[2]&gt;=r[3]&nbsp;goto&nbsp;9
6&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;Rowid&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;0&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;4&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;0&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;00&nbsp;&nbsp;r[4]=rowid&nbsp;&nbsp;&nbsp;
7&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;Once&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;0&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;8&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;0&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;00&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
8&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;Delete&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;0&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;1&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;0&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;tbl1&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;02&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
9&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;Next&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;0&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;4&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;0&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;01&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
10&nbsp;&nbsp;&nbsp;&nbsp;Noop&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;0&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;0&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;0&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;00&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
11&nbsp;&nbsp;&nbsp;&nbsp;Halt&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;0&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;0&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;0&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;00&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
12&nbsp;&nbsp;&nbsp;&nbsp;Transaction&nbsp;&nbsp;&nbsp;&nbsp;0&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;1&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;1&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;0&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;01&nbsp;&nbsp;usesStmtJournal=0
13&nbsp;&nbsp;&nbsp;&nbsp;TableLock&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;0&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;2&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;1&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;tbl1&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;00&nbsp;&nbsp;iDb=0&nbsp;root=2&nbsp;write=1
14&nbsp;&nbsp;&nbsp;&nbsp;Integer&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;20&nbsp;&nbsp;&nbsp;&nbsp;3&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;0&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;00&nbsp;&nbsp;r[3]=20&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
15&nbsp;&nbsp;&nbsp;&nbsp;Goto&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;0&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;1&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;0&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;00</pre></blockquote>

<p>Any application can run an <a href="lang_explain.html">EXPLAIN</a> query to get output similar to 
the above.
However, indentation to show the loop structure is not generated
by the SQLite core.  The <a href="cli.html">command-line shell</a> contains extra logic
for indenting loops.
Also, the "comment" column in the <a href="lang_explain.html">EXPLAIN</a> output
is only provided if SQLite is compiled with the
<a href="compile.html#enable_explain_comments">-DSQLITE_ENABLE_EXPLAIN_COMMENTS</a> options.

</p><p>When SQLite is compiled with the <a href="compile.html#debug">SQLITE_DEBUG</a> compile-time option,
extra <a href="pragma.html#syntax">PRAGMA</a> commands are available that are useful for debugging and
for exploring the operation of the VDBE.  For example the <a href="pragma.html#pragma_vdbe_trace">vdbe_trace</a>
pragma can be enabled to cause a disassembly of each VDBE opcode to be
printed on standard output as the opcode is executed.  These debugging
pragmas include:
</p><ul>
<li> <a href="pragma.html#pragma_parser_trace">PRAGMA parser_trace</a>
</li><li> <a href="pragma.html#pragma_vdbe_addoptrace">PRAGMA vdbe_addoptrace</a>
</li><li> <a href="pragma.html#pragma_vdbe_debug">PRAGMA vdbe_debug</a>
</li><li> <a href="pragma.html#pragma_vdbe_listing">PRAGMA vdbe_listing</a>
</li><li> <a href="pragma.html#pragma_vdbe_trace">PRAGMA vdbe_trace</a>
</li></ul>


<h1 id="the_opcodes"><span>4. </span>The Opcodes</h1>

<p>There are currently 186
opcodes defined by the virtual machine.
All currently defined opcodes are described in the table below.
This table was generated automatically by scanning the source code
from the file
<a href="http://www.sqlite.org/src/artifact/0c7cb1b934ad8611e14e7efaf2c3a95df7dd3f7964d63ea07fef42a23df86131">vdbe.c</a>.


</p><p>Remember: The VDBE opcodes are <u>not</u> part of the interface 
definition for SQLite.  The number of opcodes and their names and meanings
change from one release of SQLite to the next.
The opcodes shown in the table below are valid for SQLite
 version 3.40.1 check-in
<a href='https://www.sqlite.org/src/timeline?c=df5c253c0b3dd'>df5c253c0b3dd</a> dated 2022-12-28.

<a name="codes"></a>

  </div>
  <style>.optab td {vertical-align:top; padding: 1ex 1ex;}</style>
  <div class="optab">
  <blockquote><table cellspacing=0 border=1 cellpaddin>
  <tr><th>Opcode Name</th><th>Description</th></tr>
<tr><td valign="top" align="center">
<a name="Abortable"></a>Abortable
<td>Verify that an Abort can happen.  Assert if an Abort at this point
might cause database corruption.  This opcode only appears in debugging
builds.</p>

<p>An Abort is safe if either there have been no writes, or if there is
an active statement journal.</td></tr>
<tr><td valign="top" align="center">
<a name="Add"></a>Add
<td>Add the value in register P1 to the value in register P2
and store the result in register P3.
If either input is NULL, the result is NULL.</td></tr>
<tr><td valign="top" align="center">
<a name="AddImm"></a>AddImm
<td>Add the constant P2 to the value in register P1.
The result is always an integer.</p>

<p>To force any register to be an integer, just add 0.</td></tr>
<tr><td valign="top" align="center">
<a name="Affinity"></a>Affinity
<td>Apply affinities to a range of P2 registers starting with P1.</p>

<p>P4 is a string that is P2 characters long. The N-th character of the
string indicates the column affinity that should be used for the N-th
memory cell in the range.</td></tr>
<tr><td valign="top" align="center">
<a name="AggFinal"></a>AggFinal
<td>P1 is the memory location that is the accumulator for an aggregate
or window function.  Execute the finalizer function
for an aggregate and store the result in P1.</p>

<p>P2 is the number of arguments that the step function takes and
P4 is a pointer to the FuncDef for this function.  The P2
argument is not used by this opcode.  It is only there to disambiguate
functions that can take varying numbers of arguments.  The
P4 argument is only needed for the case where
the step function was not previously called.</td></tr>
<tr><td valign="top" align="center">
<a name="AggInverse"></a>AggInverse
<td>Execute the xInverse function for an aggregate.
The function has P5 arguments.  P4 is a pointer to the
FuncDef structure that specifies the function.  Register P3 is the
accumulator.</p>

<p>The P5 arguments are taken from register P2 and its
successors.</td></tr>
<tr><td valign="top" align="center">
<a name="AggStep"></a>AggStep
<td>Execute the xStep function for an aggregate.
The function has P5 arguments.  P4 is a pointer to the
FuncDef structure that specifies the function.  Register P3 is the
accumulator.</p>

<p>The P5 arguments are taken from register P2 and its
successors.</td></tr>
<tr><td valign="top" align="center">
<a name="AggStep1"></a>AggStep1
<td>Execute the xStep (if P1==0) or xInverse (if P1!=0) function for an
aggregate.  The function has P5 arguments.  P4 is a pointer to the
FuncDef structure that specifies the function.  Register P3 is the
accumulator.</p>

<p>The P5 arguments are taken from register P2 and its
successors.</p>

<p>This opcode is initially coded as OP_AggStep0.  On first evaluation,
the FuncDef stored in P4 is converted into an sqlite3_context and
the opcode is changed.  In this way, the initialization of the
sqlite3_context only happens once, instead of on each call to the
step function.</td></tr>
<tr><td valign="top" align="center">
<a name="AggValue"></a>AggValue
<td>Invoke the xValue() function and store the result in register P3.</p>

<p>P2 is the number of arguments that the step function takes and
P4 is a pointer to the FuncDef for this function.  The P2
argument is not used by this opcode.  It is only there to disambiguate
functions that can take varying numbers of arguments.  The
P4 argument is only needed for the case where
the step function was not previously called.</td></tr>
<tr><td valign="top" align="center">
<a name="And"></a>And
<td>Take the logical AND of the values in registers P1 and P2 and
write the result into register P3.</p>

<p>If either P1 or P2 is 0 (false) then the result is 0 even if
the other input is NULL.  A NULL and true or two NULLs give
a NULL output.</td></tr>
<tr><td valign="top" align="center">
<a name="AutoCommit"></a>AutoCommit
<td>Set the database auto-commit flag to P1 (1 or 0). If P2 is true, roll
back any currently active btree transactions. If there are any active
VMs (apart from this one), then a ROLLBACK fails.  A COMMIT fails if
there are active writing VMs or active VMs that use shared cache.</p>

<p>This instruction causes the VM to halt.</td></tr>
<tr><td valign="top" align="center">
<a name="BeginSubrtn"></a>BeginSubrtn
<td>Mark the beginning of a subroutine that can be entered in-line
or that can be called using <a href="opcode.html#Gosub">Gosub</a>.  The subroutine should
be terminated by an <a href="opcode.html#Return">Return</a> instruction that has a P1 operand that
is the same as the P2 operand to this opcode and that has P3 set to 1.
If the subroutine is entered in-line, then the <a href="opcode.html#Return">Return</a> will simply
fall through.  But if the subroutine is entered using <a href="opcode.html#Gosub">Gosub</a>, then
the <a href="opcode.html#Return">Return</a> will jump back to the first instruction after the <a href="opcode.html#Gosub">Gosub</a>.</p>

<p>This routine works by loading a NULL into the P2 register.  When the
return address register contains a NULL, the <a href="opcode.html#Return">Return</a> instruction is
a no-op that simply falls through to the next instruction (assuming that
the <a href="opcode.html#Return">Return</a> opcode has a P3 value of 1).  Thus if the subroutine is
entered in-line, then the <a href="opcode.html#Return">Return</a> will cause in-line execution to
continue.  But if the subroutine is entered via <a href="opcode.html#Gosub">Gosub</a>, then the
<a href="opcode.html#Return">Return</a> will cause a return to the address following the <a href="opcode.html#Gosub">Gosub</a>.</p>

<p>This opcode is identical to <a href="opcode.html#Null">Null</a>.  It has a different name
only to make the byte code easier to read and verify.</td></tr>
<tr><td valign="top" align="center">
<a name="BitAnd"></a>BitAnd
<td>Take the bit-wise AND of the values in register P1 and P2 and
store the result in register P3.
If either input is NULL, the result is NULL.</td></tr>
<tr><td valign="top" align="center">
<a name="BitNot"></a>BitNot
<td>Interpret the content of register P1 as an integer.  Store the
ones-complement of the P1 value into register P2.  If P1 holds
a NULL then store a NULL in P2.</td></tr>
<tr><td valign="top" align="center">
<a name="BitOr"></a>BitOr
<td>Take the bit-wise OR of the values in register P1 and P2 and
store the result in register P3.
If either input is NULL, the result is NULL.</td></tr>
<tr><td valign="top" align="center">
<a name="Blob"></a>Blob
<td>P4 points to a blob of data P1 bytes long.  Store this
blob in register P2.  If P4 is a NULL pointer, then construct
a zero-filled blob that is P1 bytes long in P2.</td></tr>
<tr><td valign="top" align="center">
<a name="Cast"></a>Cast
<td>Force the value in register P1 to be the type defined by P2.</p>

<p><ul>
<li> P2=='A' &rarr; BLOB
<li> P2=='B' &rarr; TEXT
<li> P2=='C' &rarr; NUMERIC
<li> P2=='D' &rarr; INTEGER
<li> P2=='E' &rarr; REAL
</ul></p>

<p>A NULL value is not changed by this routine.  It remains NULL.</td></tr>
<tr><td valign="top" align="center">
<a name="Checkpoint"></a>Checkpoint
<td>Checkpoint database P1. This is a no-op if P1 is not currently in
WAL mode. Parameter P2 is one of SQLITE_CHECKPOINT_PASSIVE, FULL,
RESTART, or TRUNCATE.  Write 1 or 0 into mem&#91;P3&#93; if the checkpoint returns
SQLITE_BUSY or not, respectively.  Write the number of pages in the
WAL after the checkpoint into mem&#91;P3+1&#93; and the number of pages
in the WAL that have been checkpointed after the checkpoint
completes into mem&#91;P3+2&#93;.  However on an error, mem&#91;P3+1&#93; and
mem&#91;P3+2&#93; are initialized to -1.</td></tr>
<tr><td valign="top" align="center">
<a name="Clear"></a>Clear
<td>Delete all contents of the database table or index whose root page
in the database file is given by P1.  But, unlike <a href="opcode.html#Destroy">Destroy</a>, do not
remove the table or index from the database file.</p>

<p>The table being clear is in the main database file if P2==0.  If
P2==1 then the table to be clear is in the auxiliary database file
that is used to store tables create using CREATE TEMPORARY TABLE.</p>

<p>If the P3 value is non-zero, then the row change count is incremented
by the number of rows in the table being cleared. If P3 is greater
than zero, then the value stored in register P3 is also incremented
by the number of rows in the table being cleared.</p>

<p>See also: <a href="opcode.html#Destroy">Destroy</a></td></tr>
<tr><td valign="top" align="center">
<a name="Close"></a>Close
<td>Close a cursor previously opened as P1.  If P1 is not
currently open, this instruction is a no-op.</td></tr>
<tr><td valign="top" align="center">
<a name="ClrSubtype"></a>ClrSubtype
<td>Clear the subtype from register P1.</td></tr>
<tr><td valign="top" align="center">
<a name="CollSeq"></a>CollSeq
<td>P4 is a pointer to a CollSeq object. If the next call to a user function
or aggregate calls sqlite3GetFuncCollSeq(), this collation sequence will
be returned. This is used by the built-in min(), max() and nullif()
functions.</p>

<p>If P1 is not zero, then it is a register that a subsequent min() or
max() aggregate will set to 1 if the current row is not the minimum or
maximum.  The P1 register is initialized to 0 by this instruction.</p>

<p>The interface used by the implementation of the aforementioned functions
to retrieve the collation sequence set by this opcode is not available
publicly.  Only built-in functions have access to this feature.</td></tr>
<tr><td valign="top" align="center">
<a name="Column"></a>Column
<td>Interpret the data that cursor P1 points to as a structure built using
the <a href="opcode.html#MakeRecord">MakeRecord</a> instruction.  (See the <a href="opcode.html#MakeRecord">MakeRecord</a> opcode for additional
information about the format of the data.)  Extract the P2-th column
from this record.  If there are less than (P2+1)
values in the record, extract a NULL.</p>

<p>The value extracted is stored in register P3.</p>

<p>If the record contains fewer than P2 fields, then extract a NULL.  Or,
if the P4 argument is a P4_MEM use the value of the P4 argument as
the result.</p>

<p>If the OPFLAG_LENGTHARG bit is set in P5 then the result is guaranteed
to only be used by the length() function or the equivalent.  The content
of large blobs is not loaded, thus saving CPU cycles.  If the
OPFLAG_TYPEOFARG bit is set then the result will only be used by the
typeof() function or the IS NULL or IS NOT NULL operators or the
equivalent.  In this case, all content loading can be omitted.</td></tr>
<tr><td valign="top" align="center">
<a name="ColumnsUsed"></a>ColumnsUsed
<td>This opcode (which only exists if SQLite was compiled with
SQLITE_ENABLE_COLUMN_USED_MASK) identifies which columns of the
table or index for cursor P1 are used.  P4 is a 64-bit integer
(P4_INT64) in which the first 63 bits are one for each of the
first 63 columns of the table or index that are actually used
by the cursor.  The high-order bit is set if any column after
the 64th is used.</td></tr>
<tr><td valign="top" align="center">
<a name="Compare"></a>Compare
<td>Compare two vectors of registers in reg(P1)..reg(P1+P3-1) (call this
vector "A") and in reg(P2)..reg(P2+P3-1) ("B").  Save the result of
the comparison for use by the next <a href="opcode.html#Jump">Jump</a> instruct.</p>

<p>If P5 has the OPFLAG_PERMUTE bit set, then the order of comparison is
determined by the most recent <a href="opcode.html#Permutation">Permutation</a> operator.  If the
OPFLAG_PERMUTE bit is clear, then register are compared in sequential
order.</p>

<p>P4 is a KeyInfo structure that defines collating sequences and sort
orders for the comparison.  The permutation applies to registers
only.  The KeyInfo elements are used sequentially.</p>

<p>The comparison is a sort comparison, so NULLs compare equal,
NULLs are less than numbers, numbers are less than strings,
and strings are less than blobs.</p>

<p>This opcode must be immediately followed by an <a href="opcode.html#Jump">Jump</a> opcode.</td></tr>
<tr><td valign="top" align="center">
<a name="Concat"></a>Concat
<td>Add the text in register P1 onto the end of the text in
register P2 and store the result in register P3.
If either the P1 or P2 text are NULL then store NULL in P3.</p>

<p>P3 = P2 || P1</p>

<p>It is illegal for P1 and P3 to be the same register. Sometimes,
if P3 is the same register as P2, the implementation is able
to avoid a memcpy().</td></tr>
<tr><td valign="top" align="center">
<a name="Copy"></a>Copy
<td>Make a copy of registers P1..P1+P3 into registers P2..P2+P3.</p>

<p>If the 0x0002 bit of P5 is set then also clear the MEM_Subtype flag in the
destination.  The 0x0001 bit of P5 indicates that this <a href="opcode.html#Copy">Copy</a> opcode cannot
be merged.  The 0x0001 bit is used by the query planner and does not
come into play during query execution.</p>

<p>This instruction makes a deep copy of the value.  A duplicate
is made of any string or blob constant.  See also <a href="opcode.html#SCopy">SCopy</a>.</td></tr>
<tr><td valign="top" align="center">
<a name="Count"></a>Count
<td>Store the number of entries (an integer value) in the table or index
opened by cursor P1 in register P2.</p>

<p>If P3==0, then an exact count is obtained, which involves visiting
every btree page of the table.  But if P3 is non-zero, an estimate
is returned based on the current cursor position.</td></tr>
<tr><td valign="top" align="center">
<a name="CreateBtree"></a>CreateBtree
<td>Allocate a new b-tree in the main database file if P1==0 or in the
TEMP database file if P1==1 or in an attached database if
P1&gt;1.  The P3 argument must be 1 (BTREE_INTKEY) for a rowid table
it must be 2 (BTREE_BLOBKEY) for an index or WITHOUT ROWID table.
The root page number of the new b-tree is stored in register P2.</td></tr>
<tr><td valign="top" align="center">
<a name="CursorHint"></a>CursorHint
<td>Provide a hint to cursor P1 that it only needs to return rows that
satisfy the Expr in P4.  TK_REGISTER terms in the P4 expression refer
to values currently held in registers.  TK_COLUMN terms in the P4
expression refer to columns in the b-tree to which cursor P1 is pointing.</td></tr>
<tr><td valign="top" align="center">
<a name="CursorLock"></a>CursorLock
<td>Lock the btree to which cursor P1 is pointing so that the btree cannot be
written by an other cursor.</td></tr>
<tr><td valign="top" align="center">
<a name="CursorUnlock"></a>CursorUnlock
<td>Unlock the btree to which cursor P1 is pointing so that it can be
written by other cursors.</td></tr>
<tr><td valign="top" align="center">
<a name="DecrJumpZero"></a>DecrJumpZero
<td>Register P1 must hold an integer.  Decrement the value in P1
and jump to P2 if the new value is exactly zero.</td></tr>
<tr><td valign="top" align="center">
<a name="DeferredSeek"></a>DeferredSeek
<td>P1 is an open index cursor and P3 is a cursor on the corresponding
table.  This opcode does a deferred seek of the P3 table cursor
to the row that corresponds to the current row of P1.</p>

<p>This is a deferred seek.  Nothing actually happens until
the cursor is used to read a record.  That way, if no reads
occur, no unnecessary I/O happens.</p>

<p>P4 may be an array of integers (type P4_INTARRAY) containing
one entry for each column in the P3 table.  If array entry a(i)
is non-zero, then reading column a(i)-1 from cursor P3 is
equivalent to performing the deferred seek and then reading column i
from P1.  This information is stored in P3 and used to redirect
reads against P3 over to P1, thus possibly avoiding the need to
seek and read cursor P3.</td></tr>
<tr><td valign="top" align="center">
<a name="Delete"></a>Delete
<td>Delete the record at which the P1 cursor is currently pointing.</p>

<p>If the OPFLAG_SAVEPOSITION bit of the P5 parameter is set, then
the cursor will be left pointing at  either the next or the previous
record in the table. If it is left pointing at the next record, then
the next <a href="opcode.html#Next">Next</a> instruction will be a no-op. As a result, in this case
it is ok to delete a record from within a <a href="opcode.html#Next">Next</a> loop. If
OPFLAG_SAVEPOSITION bit of P5 is clear, then the cursor will be
left in an undefined state.</p>

<p>If the OPFLAG_AUXDELETE bit is set on P5, that indicates that this
delete one of several associated with deleting a table row and all its
associated index entries.  Exactly one of those deletes is the "primary"
delete.  The others are all on OPFLAG_FORDELETE cursors or else are
marked with the AUXDELETE flag.</p>

<p>If the OPFLAG_NCHANGE flag of P2 (NB: P2 not P5) is set, then the row
change count is incremented (otherwise not).</p>

<p>P1 must not be pseudo-table.  It has to be a real table with
multiple rows.</p>

<p>If P4 is not NULL then it points to a Table object. In this case either
the update or pre-update hook, or both, may be invoked. The P1 cursor must
have been positioned using <a href="opcode.html#NotFound">NotFound</a> prior to invoking this opcode in
this case. Specifically, if one is configured, the pre-update hook is
invoked if P4 is not NULL. The update-hook is invoked if one is configured,
P4 is not NULL, and the OPFLAG_NCHANGE flag is set in P2.</p>

<p>If the OPFLAG_ISUPDATE flag is set in P2, then P3 contains the address
of the memory cell that contains the value that the rowid of the row will
be set to by the update.</td></tr>
<tr><td valign="top" align="center">
<a name="Destroy"></a>Destroy
<td>Delete an entire database table or index whose root page in the database
file is given by P1.</p>

<p>The table being destroyed is in the main database file if P3==0.  If
P3==1 then the table to be clear is in the auxiliary database file
that is used to store tables create using CREATE TEMPORARY TABLE.</p>

<p>If AUTOVACUUM is enabled then it is possible that another root page
might be moved into the newly deleted root page in order to keep all
root pages contiguous at the beginning of the database.  The former
value of the root page that moved - its value before the move occurred -
is stored in register P2. If no page movement was required (because the
table being dropped was already the last one in the database) then a
zero is stored in register P2.  If AUTOVACUUM is disabled then a zero
is stored in register P2.</p>

<p>This opcode throws an error if there are any active reader VMs when
it is invoked. This is done to avoid the difficulty associated with
updating existing cursors when a root page is moved in an AUTOVACUUM
database. This error is thrown even if the database is not an AUTOVACUUM
db in order to avoid introducing an incompatibility between autovacuum
and non-autovacuum modes.</p>

<p>See also: <a href="opcode.html#Clear">Clear</a></td></tr>
<tr><td valign="top" align="center">
<a name="Divide"></a>Divide
<td>Divide the value in register P1 by the value in register P2
and store the result in register P3 (P3=P2/P1). If the value in
register P1 is zero, then the result is NULL. If either input is
NULL, the result is NULL.</td></tr>
<tr><td valign="top" align="center">
<a name="DropIndex"></a>DropIndex
<td>Remove the internal (in-memory) data structures that describe
the index named P4 in database P1.  This is called after an index
is dropped from disk (using the <a href="opcode.html#Destroy">Destroy</a> opcode)
in order to keep the internal representation of the
schema consistent with what is on disk.</td></tr>
<tr><td valign="top" align="center">
<a name="DropTable"></a>DropTable
<td>Remove the internal (in-memory) data structures that describe
the table named P4 in database P1.  This is called after a table
is dropped from disk (using the <a href="opcode.html#Destroy">Destroy</a> opcode) in order to keep
the internal representation of the
schema consistent with what is on disk.</td></tr>
<tr><td valign="top" align="center">
<a name="DropTrigger"></a>DropTrigger
<td>Remove the internal (in-memory) data structures that describe
the trigger named P4 in database P1.  This is called after a trigger
is dropped from disk (using the <a href="opcode.html#Destroy">Destroy</a> opcode) in order to keep
the internal representation of the
schema consistent with what is on disk.</td></tr>
<tr><td valign="top" align="center">
<a name="ElseEq"></a>ElseEq
<td>This opcode must follow an <a href="opcode.html#Lt">Lt</a> or <a href="opcode.html#Gt">Gt</a> comparison operator.  There
can be zero or more OP_ReleaseReg opcodes intervening, but no other
opcodes are allowed to occur between this instruction and the previous
<a href="opcode.html#Lt">Lt</a> or <a href="opcode.html#Gt">Gt</a>.</p>

<p>If result of an <a href="opcode.html#Eq">Eq</a> comparison on the same two operands as the
prior <a href="opcode.html#Lt">Lt</a> or <a href="opcode.html#Gt">Gt</a> would have been true, then jump to P2.
If the result of an <a href="opcode.html#Eq">Eq</a> comparison on the two previous
operands would have been false or NULL, then fall through.</td></tr>
<tr><td valign="top" align="center">
<a name="EndCoroutine"></a>EndCoroutine
<td>The instruction at the address in register P1 is a <a href="opcode.html#Yield">Yield</a>.
<a href="opcode.html#Jump">Jump</a> to the P2 parameter of that <a href="opcode.html#Yield">Yield</a>.
After the jump, register P1 becomes undefined.</p>

<p>See also: <a href="opcode.html#InitCoroutine">InitCoroutine</a></td></tr>
<tr><td valign="top" align="center">
<a name="Eq"></a>Eq
<td>Compare the values in register P1 and P3.  If reg(P3)==reg(P1) then
jump to address P2.</p>

<p>The SQLITE_AFF_MASK portion of P5 must be an affinity character -
SQLITE_AFF_TEXT, SQLITE_AFF_INTEGER, and so forth. An attempt is made
to coerce both inputs according to this affinity before the
comparison is made. If the SQLITE_AFF_MASK is 0x00, then numeric
affinity is used. Note that the affinity conversions are stored
back into the input registers P1 and P3.  So this opcode can cause
persistent changes to registers P1 and P3.</p>

<p>Once any conversions have taken place, and neither value is NULL,
the values are compared. If both values are blobs then memcmp() is
used to determine the results of the comparison.  If both values
are text, then the appropriate collating function specified in
P4 is used to do the comparison.  If P4 is not specified then
memcmp() is used to compare text string.  If both values are
numeric, then a numeric comparison is used. If the two values
are of different types, then numbers are considered less than
strings and strings are considered less than blobs.</p>

<p>If SQLITE_NULLEQ is set in P5 then the result of comparison is always either
true or false and is never NULL.  If both operands are NULL then the result
of comparison is true.  If either operand is NULL then the result is false.
If neither operand is NULL the result is the same as it would be if
the SQLITE_NULLEQ flag were omitted from P5.</p>

<p>This opcode saves the result of comparison for use by the new
<a href="opcode.html#Jump">Jump</a> opcode.</td></tr>
<tr><td valign="top" align="center">
<a name="Expire"></a>Expire
<td>Cause precompiled statements to expire.  When an expired statement
is executed using sqlite3_step() it will either automatically
reprepare itself (if it was originally created using sqlite3_prepare_v2())
or it will fail with SQLITE_SCHEMA.</p>

<p>If P1 is 0, then all SQL statements become expired. If P1 is non-zero,
then only the currently executing statement is expired.</p>

<p>If P2 is 0, then SQL statements are expired immediately.  If P2 is 1,
then running SQL statements are allowed to continue to run to completion.
The P2==1 case occurs when a CREATE INDEX or similar schema change happens
that might help the statement run faster but which does not affect the
correctness of operation.</td></tr>
<tr><td valign="top" align="center">
<a name="Filter"></a>Filter
<td>Compute a hash on the key contained in the P4 registers starting
with r&#91;P3&#93;.  Check to see if that hash is found in the
bloom filter hosted by register P1.  If it is not present then
maybe jump to P2.  Otherwise fall through.</p>

<p>False negatives are harmless.  It is always safe to fall through,
even if the value is in the bloom filter.  A false negative causes
more CPU cycles to be used, but it should still yield the correct
answer.  However, an incorrect answer may well arise from a
false positive - if the jump is taken when it should fall through.</td></tr>
<tr><td valign="top" align="center">
<a name="FilterAdd"></a>FilterAdd
<td>Compute a hash on the P4 registers starting with r&#91;P3&#93; and
add that hash to the bloom filter contained in r&#91;P1&#93;.</td></tr>
<tr><td valign="top" align="center">
<a name="FinishSeek"></a>FinishSeek
<td>If cursor P1 was previously moved via <a href="opcode.html#DeferredSeek">DeferredSeek</a>, complete that
seek operation now, without further delay.  If the cursor seek has
already occurred, this instruction is a no-op.</td></tr>
<tr><td valign="top" align="center">
<a name="FkCheck"></a>FkCheck
<td>Halt with an SQLITE_CONSTRAINT error if there are any unresolved
foreign key constraint violations.  If there are no foreign key
constraint violations, this is a no-op.</p>

<p>FK constraint violations are also checked when the prepared statement
exits.  This opcode is used to raise foreign key constraint errors prior
to returning results such as a row change count or the result of a
RETURNING clause.</td></tr>
<tr><td valign="top" align="center">
<a name="FkCounter"></a>FkCounter
<td>Increment a "constraint counter" by P2 (P2 may be negative or positive).
If P1 is non-zero, the database constraint counter is incremented
(deferred foreign key constraints). Otherwise, if P1 is zero, the
statement counter is incremented (immediate foreign key constraints).</td></tr>
<tr><td valign="top" align="center">
<a name="FkIfZero"></a>FkIfZero
<td>This opcode tests if a foreign key constraint-counter is currently zero.
If so, jump to instruction P2. Otherwise, fall through to the next
instruction.</p>

<p>If P1 is non-zero, then the jump is taken if the database constraint-counter
is zero (the one that counts deferred constraint violations). If P1 is
zero, the jump is taken if the statement constraint-counter is zero
(immediate foreign key constraint violations).</td></tr>
<tr><td valign="top" align="center">
<a name="Found"></a>Found
<td>If P4==0 then register P3 holds a blob constructed by <a href="opcode.html#MakeRecord">MakeRecord</a>.  If
P4&gt;0 then register P3 is the first of P4 registers that form an unpacked
record.</p>

<p>Cursor P1 is on an index btree.  If the record identified by P3 and P4
is a prefix of any entry in P1 then a jump is made to P2 and
P1 is left pointing at the matching entry.</p>

<p>This operation leaves the cursor in a state where it can be
advanced in the forward direction.  The <a href="opcode.html#Next">Next</a> instruction will work,
but not the <a href="opcode.html#Prev">Prev</a> instruction.</p>

<p>See also: <a href="opcode.html#NotFound">NotFound</a>, <a href="opcode.html#NoConflict">NoConflict</a>, <a href="opcode.html#NotExists">NotExists</a>. SeekGe</td></tr>
<tr><td valign="top" align="center">
<a name="Function"></a>Function
<td>Invoke a user function (P4 is a pointer to an sqlite3_context object that
contains a pointer to the function to be run) with arguments taken
from register P2 and successors.  The number of arguments is in
the sqlite3_context object that P4 points to.
The result of the function is stored
in register P3.  Register P3 must not be one of the function inputs.</p>

<p>P1 is a 32-bit bitmask indicating whether or not each argument to the
function was determined to be constant at compile time. If the first
argument was constant then bit 0 of P1 is set. This is used to determine
whether meta data associated with a user function argument using the
sqlite3_set_auxdata() API may be safely retained until the next
invocation of this opcode.</p>

<p>See also: <a href="opcode.html#AggStep">AggStep</a>, <a href="opcode.html#AggFinal">AggFinal</a>, <a href="opcode.html#PureFunc">PureFunc</a></td></tr>
<tr><td valign="top" align="center">
<a name="Ge"></a>Ge
<td>This works just like the Lt opcode except that the jump is taken if
the content of register P3 is greater than or equal to the content of
register P1.  See the Lt opcode for additional information.</td></tr>
<tr><td valign="top" align="center">
<a name="Gosub"></a>Gosub
<td>Write the current address onto register P1
and then jump to address P2.</td></tr>
<tr><td valign="top" align="center">
<a name="Goto"></a>Goto
<td>An unconditional jump to address P2.
The next instruction executed will be
the one at index P2 from the beginning of
the program.</p>

<p>The P1 parameter is not actually used by this opcode.  However, it
is sometimes set to 1 instead of 0 as a hint to the command-line shell
that this <a href="opcode.html#Goto">Goto</a> is the bottom of a loop and that the lines from P2 down
to the current line should be indented for EXPLAIN output.</td></tr>
<tr><td valign="top" align="center">
<a name="Gt"></a>Gt
<td>This works just like the Lt opcode except that the jump is taken if
the content of register P3 is greater than the content of
register P1.  See the Lt opcode for additional information.</td></tr>
<tr><td valign="top" align="center">
<a name="Halt"></a>Halt
<td>Exit immediately.  All open cursors, etc are closed
automatically.</p>

<p>P1 is the result code returned by sqlite3_exec(), sqlite3_reset(),
or sqlite3_finalize().  For a normal halt, this should be SQLITE_OK (0).
For errors, it can be some other value.  If P1!=0 then P2 will determine
whether or not to rollback the current transaction.  Do not rollback
if P2==OE_Fail. Do the rollback if P2==OE_Rollback.  If P2==OE_Abort,
then back out all changes that have occurred during this execution of the
VDBE, but do not rollback the transaction.</p>

<p>If P4 is not null then it is an error message string.</p>

<p>P5 is a value between 0 and 4, inclusive, that modifies the P4 string.</p>

<p>0:  (no change)
1:  NOT NULL contraint failed: P4
2:  UNIQUE constraint failed: P4
3:  CHECK constraint failed: P4
4:  FOREIGN KEY constraint failed: P4</p>

<p>If P5 is not zero and P4 is NULL, then everything after the ":" is
omitted.</p>

<p>There is an implied "<a href="opcode.html#Halt">Halt</a> 0 0 0" instruction inserted at the very end of
every program.  So a jump past the last instruction of the program
is the same as executing <a href="opcode.html#Halt">Halt</a>.</td></tr>
<tr><td valign="top" align="center">
<a name="HaltIfNull"></a>HaltIfNull
<td>Check the value in register P3.  If it is NULL then <a href="opcode.html#Halt">Halt</a> using
parameter P1, P2, and P4 as if this were a <a href="opcode.html#Halt">Halt</a> instruction.  If the
value in register P3 is not NULL, then this routine is a no-op.
The P5 parameter should be 1.</td></tr>
<tr><td valign="top" align="center">
<a name="IdxDelete"></a>IdxDelete
<td>The content of P3 registers starting at register P2 form
an unpacked index key. This opcode removes that entry from the
index opened by cursor P1.</p>

<p>If P5 is not zero, then raise an SQLITE_CORRUPT_INDEX error
if no matching index entry is found.  This happens when running
an UPDATE or DELETE statement and the index entry to be updated
or deleted is not found.  For some uses of <a href="opcode.html#IdxDelete">IdxDelete</a>
(example:  the EXCEPT operator) it does not matter that no matching
entry is found.  For those cases, P5 is zero.  Also, do not raise
this (self-correcting and non-critical) error if in writable_schema mode.</td></tr>
<tr><td valign="top" align="center">
<a name="IdxGE"></a>IdxGE
<td>The P4 register values beginning with P3 form an unpacked index
key that omits the PRIMARY KEY.  <a href="opcode.html#Compare">Compare</a> this key value against the index
that P1 is currently pointing to, ignoring the PRIMARY KEY or ROWID
fields at the end.</p>

<p>If the P1 index entry is greater than or equal to the key value
then jump to P2.  Otherwise fall through to the next instruction.</td></tr>
<tr><td valign="top" align="center">
<a name="IdxGT"></a>IdxGT
<td>The P4 register values beginning with P3 form an unpacked index
key that omits the PRIMARY KEY.  <a href="opcode.html#Compare">Compare</a> this key value against the index
that P1 is currently pointing to, ignoring the PRIMARY KEY or ROWID
fields at the end.</p>

<p>If the P1 index entry is greater than the key value
then jump to P2.  Otherwise fall through to the next instruction.</td></tr>
<tr><td valign="top" align="center">
<a name="IdxInsert"></a>IdxInsert
<td>Register P2 holds an SQL index key made using the
<a href="opcode.html#MakeRecord">MakeRecord</a> instructions.  This opcode writes that key
into the index P1.  Data for the entry is nil.</p>

<p>If P4 is not zero, then it is the number of values in the unpacked
key of reg(P2).  In that case, P3 is the index of the first register
for the unpacked key.  The availability of the unpacked key can sometimes
be an optimization.</p>

<p>If P5 has the OPFLAG_APPEND bit set, that is a hint to the b-tree layer
that this insert is likely to be an append.</p>

<p>If P5 has the OPFLAG_NCHANGE bit set, then the change counter is
incremented by this instruction.  If the OPFLAG_NCHANGE bit is clear,
then the change counter is unchanged.</p>

<p>If the OPFLAG_USESEEKRESULT flag of P5 is set, the implementation might
run faster by avoiding an unnecessary seek on cursor P1.  However,
the OPFLAG_USESEEKRESULT flag must only be set if there have been no prior
seeks on the cursor or if the most recent seek used a key equivalent
to P2.</p>

<p>This instruction only works for indices.  The equivalent instruction
for tables is <a href="opcode.html#Insert">Insert</a>.</td></tr>
<tr><td valign="top" align="center">
<a name="IdxLE"></a>IdxLE
<td>The P4 register values beginning with P3 form an unpacked index
key that omits the PRIMARY KEY or ROWID.  <a href="opcode.html#Compare">Compare</a> this key value against
the index that P1 is currently pointing to, ignoring the PRIMARY KEY or
ROWID on the P1 index.</p>

<p>If the P1 index entry is less than or equal to the key value then jump
to P2. Otherwise fall through to the next instruction.</td></tr>
<tr><td valign="top" align="center">
<a name="IdxLT"></a>IdxLT
<td>The P4 register values beginning with P3 form an unpacked index
key that omits the PRIMARY KEY or ROWID.  <a href="opcode.html#Compare">Compare</a> this key value against
the index that P1 is currently pointing to, ignoring the PRIMARY KEY or
ROWID on the P1 index.</p>

<p>If the P1 index entry is less than the key value then jump to P2.
Otherwise fall through to the next instruction.</td></tr>
<tr><td valign="top" align="center">
<a name="IdxRowid"></a>IdxRowid
<td>Write into register P2 an integer which is the last entry in the record at
the end of the index key pointed to by cursor P1.  This integer should be
the rowid of the table entry to which this index entry points.</p>

<p>See also: <a href="opcode.html#Rowid">Rowid</a>, <a href="opcode.html#MakeRecord">MakeRecord</a>.</td></tr>
<tr><td valign="top" align="center">
<a name="If"></a>If
<td>Jump to P2 if the value in register P1 is true.  The value
is considered true if it is numeric and non-zero.  If the value
in P1 is NULL then take the jump if and only if P3 is non-zero.</td></tr>
<tr><td valign="top" align="center">
<a name="IfNoHope"></a>IfNoHope
<td>Register P3 is the first of P4 registers that form an unpacked
record.  Cursor P1 is an index btree.  P2 is a jump destination.
In other words, the operands to this opcode are the same as the
operands to <a href="opcode.html#NotFound">NotFound</a> and <a href="opcode.html#IdxGT">IdxGT</a>.</p>

<p>This opcode is an optimization attempt only.  If this opcode always
falls through, the correct answer is still obtained, but extra works
is performed.</p>

<p>A value of N in the seekHit flag of cursor P1 means that there exists
a key P3:N that will match some record in the index.  We want to know
if it is possible for a record P3:P4 to match some record in the
index.  If it is not possible, we can skips some work.  So if seekHit
is less than P4, attempt to find out if a match is possible by running
<a href="opcode.html#NotFound">NotFound</a>.</p>

<p>This opcode is used in IN clause processing for a multi-column key.
If an IN clause is attached to an element of the key other than the
left-most element, and if there are no matches on the most recent
seek over the whole key, then it might be that one of the key element
to the left is prohibiting a match, and hence there is "no hope" of
any match regardless of how many IN clause elements are checked.
In such a case, we abandon the IN clause search early, using this
opcode.  The opcode name comes from the fact that the
jump is taken if there is "no hope" of achieving a match.</p>

<p>See also: <a href="opcode.html#NotFound">NotFound</a>, <a href="opcode.html#SeekHit">SeekHit</a></td></tr>
<tr><td valign="top" align="center">
<a name="IfNot"></a>IfNot
<td>Jump to P2 if the value in register P1 is False.  The value
is considered false if it has a numeric value of zero.  If the value
in P1 is NULL then take the jump if and only if P3 is non-zero.</td></tr>
<tr><td valign="top" align="center">
<a name="IfNotOpen"></a>IfNotOpen
<td>If cursor P1 is not open or if P1 is set to a NULL row using the
<a href="opcode.html#NullRow">NullRow</a> opcode, then jump to instruction P2. Otherwise, fall through.</td></tr>
<tr><td valign="top" align="center">
<a name="IfNotZero"></a>IfNotZero
<td>Register P1 must contain an integer.  If the content of register P1 is
initially greater than zero, then decrement the value in register P1.
If it is non-zero (negative or positive) and then also jump to P2.
If register P1 is initially zero, leave it unchanged and fall through.</td></tr>
<tr><td valign="top" align="center">
<a name="IfNullRow"></a>IfNullRow
<td>Check the cursor P1 to see if it is currently pointing at a NULL row.
If it is, then set register P3 to NULL and jump immediately to P2.
If P1 is not on a NULL row, then fall through without making any
changes.</p>

<p>If P1 is not an open cursor, then this opcode is a no-op.</td></tr>
<tr><td valign="top" align="center">
<a name="IfPos"></a>IfPos
<td>Register P1 must contain an integer.
If the value of register P1 is 1 or greater, subtract P3 from the
value in P1 and jump to P2.</p>

<p>If the initial value of register P1 is less than 1, then the
value is unchanged and control passes through to the next instruction.</td></tr>
<tr><td valign="top" align="center">
<a name="IfSmaller"></a>IfSmaller
<td>Estimate the number of rows in the table P1.  <a href="opcode.html#Jump">Jump</a> to P2 if that
estimate is less than approximately 2**(0.1*P3).</td></tr>
<tr><td valign="top" align="center">
<a name="IncrVacuum"></a>IncrVacuum
<td>Perform a single step of the incremental vacuum procedure on
the P1 database. If the vacuum has finished, jump to instruction
P2. Otherwise, fall through to the next instruction.</td></tr>
<tr><td valign="top" align="center">
<a name="Init"></a>Init
<td>Programs contain a single instance of this opcode as the very first
opcode.</p>

<p>If tracing is enabled (by the sqlite3_trace()) interface, then
the UTF-8 string contained in P4 is emitted on the trace callback.
Or if P4 is blank, use the string returned by sqlite3_sql().</p>

<p>If P2 is not zero, jump to instruction P2.</p>

<p>Increment the value of P1 so that <a href="opcode.html#Once">Once</a> opcodes will jump the
first time they are evaluated for this run.</p>

<p>If P3 is not zero, then it is an address to jump to if an SQLITE_CORRUPT
error is encountered.</td></tr>
<tr><td valign="top" align="center">
<a name="InitCoroutine"></a>InitCoroutine
<td>Set up register P1 so that it will <a href="opcode.html#Yield">Yield</a> to the coroutine
located at address P3.</p>

<p>If P2!=0 then the coroutine implementation immediately follows
this opcode.  So jump over the coroutine implementation to
address P2.</p>

<p>See also: <a href="opcode.html#EndCoroutine">EndCoroutine</a></td></tr>
<tr><td valign="top" align="center">
<a name="Insert"></a>Insert
<td>Write an entry into the table of cursor P1.  A new entry is
created if it doesn't already exist or the data for an existing
entry is overwritten.  The data is the value MEM_Blob stored in register
number P2. The key is stored in register P3. The key must
be a MEM_Int.</p>

<p>If the OPFLAG_NCHANGE flag of P5 is set, then the row change count is
incremented (otherwise not).  If the OPFLAG_LASTROWID flag of P5 is set,
then rowid is stored for subsequent return by the
sqlite3_last_insert_rowid() function (otherwise it is unmodified).</p>

<p>If the OPFLAG_USESEEKRESULT flag of P5 is set, the implementation might
run faster by avoiding an unnecessary seek on cursor P1.  However,
the OPFLAG_USESEEKRESULT flag must only be set if there have been no prior
seeks on the cursor or if the most recent seek used a key equal to P3.</p>

<p>If the OPFLAG_ISUPDATE flag is set, then this opcode is part of an
UPDATE operation.  Otherwise (if the flag is clear) then this opcode
is part of an INSERT operation.  The difference is only important to
the update hook.</p>

<p>Parameter P4 may point to a Table structure, or may be NULL. If it is
not NULL, then the update-hook (sqlite3.xUpdateCallback) is invoked
following a successful insert.</p>

<p>(WARNING/TODO: If P1 is a pseudo-cursor and P2 is dynamically
allocated, then ownership of P2 is transferred to the pseudo-cursor
and register P2 becomes ephemeral.  If the cursor is changed, the
value of register P2 will then change.  Make sure this does not
cause any problems.)</p>

<p>This instruction only works on tables.  The equivalent instruction
for indices is <a href="opcode.html#IdxInsert">IdxInsert</a>.</td></tr>
<tr><td valign="top" align="center">
<a name="Int64"></a>Int64
<td>P4 is a pointer to a 64-bit integer value.
Write that value into register P2.</td></tr>
<tr><td valign="top" align="center">
<a name="IntCopy"></a>IntCopy
<td>Transfer the integer value held in register P1 into register P2.</p>

<p>This is an optimized version of <a href="opcode.html#SCopy">SCopy</a> that works only for integer
values.</td></tr>
<tr><td valign="top" align="center">
<a name="Integer"></a>Integer
<td>The 32-bit integer value P1 is written into register P2.</td></tr>
<tr><td valign="top" align="center">
<a name="IntegrityCk"></a>IntegrityCk
<td>Do an analysis of the currently open database.  Store in
register P1 the text of an error message describing any problems.
If no problems are found, store a NULL in register P1.</p>

<p>The register P3 contains one less than the maximum number of allowed errors.
At most reg(P3) errors will be reported.
In other words, the analysis stops as soon as reg(P1) errors are
seen.  Reg(P1) is updated with the number of errors remaining.</p>

<p>The root page numbers of all tables in the database are integers
stored in P4_INTARRAY argument.</p>

<p>If P5 is not zero, the check is done on the auxiliary database
file, not the main database file.</p>

<p>This opcode is used to implement the integrity_check pragma.</td></tr>
<tr><td valign="top" align="center">
<a name="IsNull"></a>IsNull
<td>Jump to P2 if the value in register P1 is NULL.</td></tr>
<tr><td valign="top" align="center">
<a name="IsTrue"></a>IsTrue
<td>This opcode implements the IS TRUE, IS FALSE, IS NOT TRUE, and
IS NOT FALSE operators.</p>

<p>Interpret the value in register P1 as a boolean value.  Store that
boolean (a 0 or 1) in register P2.  Or if the value in register P1 is
NULL, then the P3 is stored in register P2.  Invert the answer if P4
is 1.</p>

<p>The logic is summarized like this:</p>

<p><ul>
<li> If P3==0 and P4==0  then  r&#91;P2&#93; := r&#91;P1&#93; IS TRUE
<li> If P3==1 and P4==1  then  r&#91;P2&#93; := r&#91;P1&#93; IS FALSE
<li> If P3==0 and P4==1  then  r&#91;P2&#93; := r&#91;P1&#93; IS NOT TRUE
<li> If P3==1 and P4==0  then  r&#91;P2&#93; := r&#91;P1&#93; IS NOT FALSE
</ul></td></tr>
<tr><td valign="top" align="center">
<a name="IsType"></a>IsType
<td>Jump to P2 if the type of a column in a btree is one of the types specified
by the P5 bitmask.</p>

<p>P1 is normally a cursor on a btree for which the row decode cache is
valid through at least column P3.  In other words, there should have been
a prior <a href="opcode.html#Column">Column</a> for column P3 or greater.  If the cursor is not valid,
then this opcode might give spurious results.
The the btree row has fewer than P3 columns, then use P4 as the
datatype.</p>

<p>If P1 is -1, then P3 is a register number and the datatype is taken
from the value in that register.</p>

<p>P5 is a bitmask of data types.  SQLITE_INTEGER is the least significant
(0x01) bit. SQLITE_FLOAT is the 0x02 bit. SQLITE_TEXT is 0x04.
SQLITE_BLOB is 0x08.  SQLITE_NULL is 0x10.</p>

<p>Take the jump to address P2 if and only if the datatype of the
value determined by P1 and P3 corresponds to one of the bits in the
P5 bitmask.</p></td></tr>
<tr><td valign="top" align="center">
<a name="JournalMode"></a>JournalMode
<td>Change the journal mode of database P1 to P3. P3 must be one of the
PAGER_JOURNALMODE_XXX values. If changing between the various rollback
modes (delete, truncate, persist, off and memory), this is a simple
operation. No IO is required.</p>

<p>If changing into or out of WAL mode the procedure is more complicated.</p>

<p>Write a string containing the final journal-mode to register P2.</td></tr>
<tr><td valign="top" align="center">
<a name="Jump"></a>Jump
<td>Jump to the instruction at address P1, P2, or P3 depending on whether
in the most recent <a href="opcode.html#Compare">Compare</a> instruction the P1 vector was less than
equal to, or greater than the P2 vector, respectively.</p>

<p>This opcode must immediately follow an <a href="opcode.html#Compare">Compare</a> opcode.</td></tr>
<tr><td valign="top" align="center">
<a name="Last"></a>Last
<td>The next use of the <a href="opcode.html#Rowid">Rowid</a> or <a href="opcode.html#Column">Column</a> or <a href="opcode.html#Prev">Prev</a> instruction for P1
will refer to the last entry in the database table or index.
If the table or index is empty and P2&gt;0, then jump immediately to P2.
If P2 is 0 or if the table or index is not empty, fall through
to the following instruction.</p>

<p>This opcode leaves the cursor configured to move in reverse order,
from the end toward the beginning.  In other words, the cursor is
configured to use <a href="opcode.html#Prev">Prev</a>, not <a href="opcode.html#Next">Next</a>.</td></tr>
<tr><td valign="top" align="center">
<a name="Le"></a>Le
<td>This works just like the Lt opcode except that the jump is taken if
the content of register P3 is less than or equal to the content of
register P1.  See the Lt opcode for additional information.</td></tr>
<tr><td valign="top" align="center">
<a name="LoadAnalysis"></a>LoadAnalysis
<td>Read the sqlite_stat1 table for database P1 and load the content
of that table into the internal index hash table.  This will cause
the analysis to be used when preparing all subsequent queries.</td></tr>
<tr><td valign="top" align="center">
<a name="Lt"></a>Lt
<td>Compare the values in register P1 and P3.  If reg(P3)&lt;reg(P1) then
jump to address P2.</p>

<p>If the SQLITE_JUMPIFNULL bit of P5 is set and either reg(P1) or
reg(P3) is NULL then the take the jump.  If the SQLITE_JUMPIFNULL
bit is clear then fall through if either operand is NULL.</p>

<p>The SQLITE_AFF_MASK portion of P5 must be an affinity character -
SQLITE_AFF_TEXT, SQLITE_AFF_INTEGER, and so forth. An attempt is made
to coerce both inputs according to this affinity before the
comparison is made. If the SQLITE_AFF_MASK is 0x00, then numeric
affinity is used. Note that the affinity conversions are stored
back into the input registers P1 and P3.  So this opcode can cause
persistent changes to registers P1 and P3.</p>

<p>Once any conversions have taken place, and neither value is NULL,
the values are compared. If both values are blobs then memcmp() is
used to determine the results of the comparison.  If both values
are text, then the appropriate collating function specified in
P4 is  used to do the comparison.  If P4 is not specified then
memcmp() is used to compare text string.  If both values are
numeric, then a numeric comparison is used. If the two values
are of different types, then numbers are considered less than
strings and strings are considered less than blobs.</p>

<p>This opcode saves the result of comparison for use by the new
<a href="opcode.html#Jump">Jump</a> opcode.</td></tr>
<tr><td valign="top" align="center">
<a name="MakeRecord"></a>MakeRecord
<td>Convert P2 registers beginning with P1 into the <a href="fileformat2.html#record_format">record format</a>
use as a data record in a database table or as a key
in an index.  The <a href="opcode.html#Column">Column</a> opcode can decode the record later.</p>

<p>P4 may be a string that is P2 characters long.  The N-th character of the
string indicates the column affinity that should be used for the N-th
field of the index key.</p>

<p>The mapping from character to affinity is given by the SQLITE_AFF_
macros defined in sqliteInt.h.</p>

<p>If P4 is NULL then all index fields have the affinity BLOB.</p>

<p>The meaning of P5 depends on whether or not the SQLITE_ENABLE_NULL_TRIM
compile-time option is enabled:</p>

<p>* If SQLITE_ENABLE_NULL_TRIM is enabled, then the P5 is the index
of the right-most table that can be null-trimmed.</p>

<p>* If SQLITE_ENABLE_NULL_TRIM is omitted, then P5 has the value
OPFLAG_NOCHNG_MAGIC if the <a href="opcode.html#MakeRecord">MakeRecord</a> opcode is allowed to
accept no-change records with serial_type 10.  This value is
only used inside an assert() and does not affect the end result.</td></tr>
<tr><td valign="top" align="center">
<a name="MaxPgcnt"></a>MaxPgcnt
<td>Try to set the maximum page count for database P1 to the value in P3.
Do not let the maximum page count fall below the current page count and
do not change the maximum page count value if P3==0.</p>

<p>Store the maximum page count after the change in register P2.</td></tr>
<tr><td valign="top" align="center">
<a name="MemMax"></a>MemMax
<td>P1 is a register in the root frame of this VM (the root frame is
different from the current frame if this instruction is being executed
within a sub-program). Set the value of register P1 to the maximum of
its current value and the value in register P2.</p>

<p>This instruction throws an error if the memory cell is not initially
an integer.</td></tr>
<tr><td valign="top" align="center">
<a name="Move"></a>Move
<td>Move the P3 values in register P1..P1+P3-1 over into
registers P2..P2+P3-1.  Registers P1..P1+P3-1 are
left holding a NULL.  It is an error for register ranges
P1..P1+P3-1 and P2..P2+P3-1 to overlap.  It is an error
for P3 to be less than 1.</td></tr>
<tr><td valign="top" align="center">
<a name="Multiply"></a>Multiply
<td>Multiply the value in register P1 by the value in register P2
and store the result in register P3.
If either input is NULL, the result is NULL.</td></tr>
<tr><td valign="top" align="center">
<a name="MustBeInt"></a>MustBeInt
<td>Force the value in register P1 to be an integer.  If the value
in P1 is not an integer and cannot be converted into an integer
without data loss, then jump immediately to P2, or if P2==0
raise an SQLITE_MISMATCH exception.</td></tr>
<tr><td valign="top" align="center">
<a name="Ne"></a>Ne
<td>This works just like the Eq opcode except that the jump is taken if
the operands in registers P1 and P3 are not equal.  See the Eq opcode for
additional information.</td></tr>
<tr><td valign="top" align="center">
<a name="NewRowid"></a>NewRowid
<td>Get a new integer record number (a.k.a "rowid") used as the key to a table.
The record number is not previously used as a key in the database
table that cursor P1 points to.  The new record number is written
written to register P2.</p>

<p>If P3&gt;0 then P3 is a register in the root frame of this VDBE that holds
the largest previously generated record number. No new record numbers are
allowed to be less than this value. When this value reaches its maximum,
an SQLITE_FULL error is generated. The P3 register is updated with the '
generated record number. This P3 mechanism is used to help implement the
AUTOINCREMENT feature.</td></tr>
<tr><td valign="top" align="center">
<a name="Next"></a>Next
<td>Advance cursor P1 so that it points to the next key/data pair in its
table or index.  If there are no more key/value pairs then fall through
to the following instruction.  But if the cursor advance was successful,
jump immediately to P2.</p>

<p>The <a href="opcode.html#Next">Next</a> opcode is only valid following an <a href="opcode.html#SeekGT">SeekGT</a>, <a href="opcode.html#SeekGE">SeekGE</a>, or
<a href="opcode.html#Rewind">Rewind</a> opcode used to position the cursor.  <a href="opcode.html#Next">Next</a> is not allowed
to follow <a href="opcode.html#SeekLT">SeekLT</a>, <a href="opcode.html#SeekLE">SeekLE</a>, or <a href="opcode.html#Last">Last</a>.</p>

<p>The P1 cursor must be for a real table, not a pseudo-table.  P1 must have
been opened prior to this opcode or the program will segfault.</p>

<p>The P3 value is a hint to the btree implementation. If P3==1, that
means P1 is an SQL index and that this instruction could have been
omitted if that index had been unique.  P3 is usually 0.  P3 is
always either 0 or 1.</p>

<p>If P5 is positive and the jump is taken, then event counter
number P5-1 in the prepared statement is incremented.</p>

<p>See also: <a href="opcode.html#Prev">Prev</a></td></tr>
<tr><td valign="top" align="center">
<a name="NoConflict"></a>NoConflict
<td>If P4==0 then register P3 holds a blob constructed by <a href="opcode.html#MakeRecord">MakeRecord</a>.  If
P4&gt;0 then register P3 is the first of P4 registers that form an unpacked
record.</p>

<p>Cursor P1 is on an index btree.  If the record identified by P3 and P4
contains any NULL value, jump immediately to P2.  If all terms of the
record are not-NULL then a check is done to determine if any row in the
P1 index btree has a matching key prefix.  If there are no matches, jump
immediately to P2.  If there is a match, fall through and leave the P1
cursor pointing to the matching row.</p>

<p>This opcode is similar to <a href="opcode.html#NotFound">NotFound</a> with the exceptions that the
branch is always taken if any part of the search key input is NULL.</p>

<p>This operation leaves the cursor in a state where it cannot be
advanced in either direction.  In other words, the <a href="opcode.html#Next">Next</a> and <a href="opcode.html#Prev">Prev</a>
opcodes do not work after this operation.</p>

<p>See also: <a href="opcode.html#NotFound">NotFound</a>, <a href="opcode.html#Found">Found</a>, <a href="opcode.html#NotExists">NotExists</a></td></tr>
<tr><td valign="top" align="center">
<a name="Noop"></a>Noop
<td>Do nothing.  This instruction is often useful as a jump
destination.</td></tr>
<tr><td valign="top" align="center">
<a name="Not"></a>Not
<td>Interpret the value in register P1 as a boolean value.  Store the
boolean complement in register P2.  If the value in register P1 is
NULL, then a NULL is stored in P2.</td></tr>
<tr><td valign="top" align="center">
<a name="NotExists"></a>NotExists
<td>P1 is the index of a cursor open on an SQL table btree (with integer
keys).  P3 is an integer rowid.  If P1 does not contain a record with
rowid P3 then jump immediately to P2.  Or, if P2 is 0, raise an
SQLITE_CORRUPT error. If P1 does contain a record with rowid P3 then
leave the cursor pointing at that record and fall through to the next
instruction.</p>

<p>The <a href="opcode.html#SeekRowid">SeekRowid</a> opcode performs the same operation but also allows the
P3 register to contain a non-integer value, in which case the jump is
always taken.  This opcode requires that P3 always contain an integer.</p>

<p>The <a href="opcode.html#NotFound">NotFound</a> opcode performs the same operation on index btrees
(with arbitrary multi-value keys).</p>

<p>This opcode leaves the cursor in a state where it cannot be advanced
in either direction.  In other words, the <a href="opcode.html#Next">Next</a> and <a href="opcode.html#Prev">Prev</a> opcodes will
not work following this opcode.</p>

<p>See also: <a href="opcode.html#Found">Found</a>, <a href="opcode.html#NotFound">NotFound</a>, <a href="opcode.html#NoConflict">NoConflict</a>, <a href="opcode.html#SeekRowid">SeekRowid</a></td></tr>
<tr><td valign="top" align="center">
<a name="NotFound"></a>NotFound
<td>If P4==0 then register P3 holds a blob constructed by <a href="opcode.html#MakeRecord">MakeRecord</a>.  If
P4&gt;0 then register P3 is the first of P4 registers that form an unpacked
record.</p>

<p>Cursor P1 is on an index btree.  If the record identified by P3 and P4
is not the prefix of any entry in P1 then a jump is made to P2.  If P1
does contain an entry whose prefix matches the P3/P4 record then control
falls through to the next instruction and P1 is left pointing at the
matching entry.</p>

<p>This operation leaves the cursor in a state where it cannot be
advanced in either direction.  In other words, the <a href="opcode.html#Next">Next</a> and <a href="opcode.html#Prev">Prev</a>
opcodes do not work after this operation.</p>

<p>See also: <a href="opcode.html#Found">Found</a>, <a href="opcode.html#NotExists">NotExists</a>, <a href="opcode.html#NoConflict">NoConflict</a>, <a href="opcode.html#IfNoHope">IfNoHope</a></td></tr>
<tr><td valign="top" align="center">
<a name="NotNull"></a>NotNull
<td>Jump to P2 if the value in register P1 is not NULL.</td></tr>
<tr><td valign="top" align="center">
<a name="Null"></a>Null
<td>Write a NULL into registers P2.  If P3 greater than P2, then also write
NULL into register P3 and every register in between P2 and P3.  If P3
is less than P2 (typically P3 is zero) then only register P2 is
set to NULL.</p>

<p>If the P1 value is non-zero, then also set the MEM_Cleared flag so that
NULL values will not compare equal even if SQLITE_NULLEQ is set on
<a href="opcode.html#Ne">Ne</a> or <a href="opcode.html#Eq">Eq</a>.</td></tr>
<tr><td valign="top" align="center">
<a name="NullRow"></a>NullRow
<td>Move the cursor P1 to a null row.  Any <a href="opcode.html#Column">Column</a> operations
that occur while the cursor is on the null row will always
write a NULL.</p>

<p>If cursor P1 is not previously opened, open it now to a special
pseudo-cursor that always returns NULL for every column.</td></tr>
<tr><td valign="top" align="center">
<a name="Offset"></a>Offset
<td>Store in register r&#91;P3&#93; the byte offset into the database file that is the
start of the payload for the record at which that cursor P1 is currently
pointing.</p>

<p>P2 is the column number for the argument to the sqlite_offset() function.
This opcode does not use P2 itself, but the P2 value is used by the
code generator.  The P1, P2, and P3 operands to this opcode are the
same as for <a href="opcode.html#Column">Column</a>.</p>

<p>This opcode is only available if SQLite is compiled with the
-DSQLITE_ENABLE_OFFSET_SQL_FUNC option.</td></tr>
<tr><td valign="top" align="center">
<a name="OffsetLimit"></a>OffsetLimit
<td>This opcode performs a commonly used computation associated with
LIMIT and OFFSET processing.  r&#91;P1&#93; holds the limit counter.  r&#91;P3&#93;
holds the offset counter.  The opcode computes the combined value
of the LIMIT and OFFSET and stores that value in r&#91;P2&#93;.  The r&#91;P2&#93;
value computed is the total number of rows that will need to be
visited in order to complete the query.</p>

<p>If r&#91;P3&#93; is zero or negative, that means there is no OFFSET
and r&#91;P2&#93; is set to be the value of the LIMIT, r&#91;P1&#93;.</p>

<p>if r&#91;P1&#93; is zero or negative, that means there is no LIMIT
and r&#91;P2&#93; is set to -1.</p>

<p>Otherwise, r&#91;P2&#93; is set to the sum of r&#91;P1&#93; and r&#91;P3&#93;.</td></tr>
<tr><td valign="top" align="center">
<a name="Once"></a>Once
<td>Fall through to the next instruction the first time this opcode is
encountered on each invocation of the byte-code program.  <a href="opcode.html#Jump">Jump</a> to P2
on the second and all subsequent encounters during the same invocation.</p>

<p>Top-level programs determine first invocation by comparing the P1
operand against the P1 operand on the <a href="opcode.html#Init">Init</a> opcode at the beginning
of the program.  If the P1 values differ, then fall through and make
the P1 of this opcode equal to the P1 of <a href="opcode.html#Init">Init</a>.  If P1 values are
the same then take the jump.</p>

<p>For subprograms, there is a bitmask in the VdbeFrame that determines
whether or not the jump should be taken.  The bitmask is necessary
because the self-altering code trick does not work for recursive
triggers.</td></tr>
<tr><td valign="top" align="center">
<a name="OpenAutoindex"></a>OpenAutoindex
<td>This opcode works the same as <a href="opcode.html#OpenEphemeral">OpenEphemeral</a>.  It has a
different name to distinguish its use.  Tables created using
by this opcode will be used for automatically created transient
indices in joins.</td></tr>
<tr><td valign="top" align="center">
<a name="OpenDup"></a>OpenDup
<td>Open a new cursor P1 that points to the same ephemeral table as
cursor P2.  The P2 cursor must have been opened by a prior <a href="opcode.html#OpenEphemeral">OpenEphemeral</a>
opcode.  Only ephemeral cursors may be duplicated.</p>

<p>Duplicate ephemeral cursors are used for self-joins of materialized views.</td></tr>
<tr><td valign="top" align="center">
<a name="OpenEphemeral"></a>OpenEphemeral
<td>Open a new cursor P1 to a transient table.
The cursor is always opened read/write even if
the main database is read-only.  The ephemeral
table is deleted automatically when the cursor is closed.</p>

<p>If the cursor P1 is already opened on an ephemeral table, the table
is cleared (all content is erased).</p>

<p>P2 is the number of columns in the ephemeral table.
The cursor points to a BTree table if P4==0 and to a BTree index
if P4 is not 0.  If P4 is not NULL, it points to a KeyInfo structure
that defines the format of keys in the index.</p>

<p>The P5 parameter can be a mask of the BTREE_* flags defined
in btree.h.  These flags control aspects of the operation of
the btree.  The BTREE_OMIT_JOURNAL and BTREE_SINGLE flags are
added automatically.</p>

<p>If P3 is positive, then reg&#91;P3&#93; is modified slightly so that it
can be used as zero-length data for <a href="opcode.html#Insert">Insert</a>.  This is an optimization
that avoids an extra <a href="opcode.html#Blob">Blob</a> opcode to initialize that register.</td></tr>
<tr><td valign="top" align="center">
<a name="OpenPseudo"></a>OpenPseudo
<td>Open a new cursor that points to a fake table that contains a single
row of data.  The content of that one row is the content of memory
register P2.  In other words, cursor P1 becomes an alias for the
MEM_Blob content contained in register P2.</p>

<p>A pseudo-table created by this opcode is used to hold a single
row output from the sorter so that the row can be decomposed into
individual columns using the <a href="opcode.html#Column">Column</a> opcode.  The <a href="opcode.html#Column">Column</a> opcode
is the only cursor opcode that works with a pseudo-table.</p>

<p>P3 is the number of fields in the records that will be stored by
the pseudo-table.</td></tr>
<tr><td valign="top" align="center">
<a name="OpenRead"></a>OpenRead
<td>Open a read-only cursor for the database table whose root page is
P2 in a database file.  The database file is determined by P3.
P3==0 means the main database, P3==1 means the database used for
temporary tables, and P3&gt;1 means used the corresponding attached
database.  Give the new cursor an identifier of P1.  The P1
values need not be contiguous but all P1 values should be small integers.
It is an error for P1 to be negative.</p>

<p>Allowed P5 bits:
<ul>
<li>  <b>0x02 OPFLAG_SEEKEQ</b>: This cursor will only be used for
equality lookups (implemented as a pair of opcodes <a href="opcode.html#SeekGE">SeekGE</a>/<a href="opcode.html#IdxGT">IdxGT</a>
of <a href="opcode.html#SeekLE">SeekLE</a>/<a href="opcode.html#IdxLT">IdxLT</a>)
</ul></p>

<p>The P4 value may be either an integer (P4_INT32) or a pointer to
a KeyInfo structure (P4_KEYINFO). If it is a pointer to a KeyInfo
object, then table being opened must be an <a href="fileformat2.html#btypes">index b-tree</a> where the
KeyInfo object defines the content and collating
sequence of that index b-tree. Otherwise, if P4 is an integer
value, then the table being opened must be a <a href="fileformat2.html#btypes">table b-tree</a> with a
number of columns no less than the value of P4.</p>

<p>See also: <a href="opcode.html#OpenWrite">OpenWrite</a>, <a href="opcode.html#ReopenIdx">ReopenIdx</a></td></tr>
<tr><td valign="top" align="center">
<a name="OpenWrite"></a>OpenWrite
<td>Open a read/write cursor named P1 on the table or index whose root
page is P2 (or whose root page is held in register P2 if the
OPFLAG_P2ISREG bit is set in P5 - see below).</p>

<p>The P4 value may be either an integer (P4_INT32) or a pointer to
a KeyInfo structure (P4_KEYINFO). If it is a pointer to a KeyInfo
object, then table being opened must be an <a href="fileformat2.html#btypes">index b-tree</a> where the
KeyInfo object defines the content and collating
sequence of that index b-tree. Otherwise, if P4 is an integer
value, then the table being opened must be a <a href="fileformat2.html#btypes">table b-tree</a> with a
number of columns no less than the value of P4.</p>

<p>Allowed P5 bits:
<ul>
<li>  <b>0x02 OPFLAG_SEEKEQ</b>: This cursor will only be used for
equality lookups (implemented as a pair of opcodes <a href="opcode.html#SeekGE">SeekGE</a>/<a href="opcode.html#IdxGT">IdxGT</a>
of <a href="opcode.html#SeekLE">SeekLE</a>/<a href="opcode.html#IdxLT">IdxLT</a>)
<li>  <b>0x08 OPFLAG_FORDELETE</b>: This cursor is used only to seek
and subsequently delete entries in an index btree.  This is a
hint to the storage engine that the storage engine is allowed to
ignore.  The hint is not used by the official SQLite b*tree storage
engine, but is used by COMDB2.
<li>  <b>0x10 OPFLAG_P2ISREG</b>: Use the content of register P2
as the root page, not the value of P2 itself.
</ul></p>

<p>This instruction works like <a href="opcode.html#OpenRead">OpenRead</a> except that it opens the cursor
in read/write mode.</p>

<p>See also: <a href="opcode.html#OpenRead">OpenRead</a>, <a href="opcode.html#ReopenIdx">ReopenIdx</a></td></tr>
<tr><td valign="top" align="center">
<a name="Or"></a>Or
<td>Take the logical OR of the values in register P1 and P2 and
store the answer in register P3.</p>

<p>If either P1 or P2 is nonzero (true) then the result is 1 (true)
even if the other input is NULL.  A NULL and false or two NULLs
give a NULL output.</td></tr>
<tr><td valign="top" align="center">
<a name="Pagecount"></a>Pagecount
<td>Write the current number of pages in database P1 to memory cell P2.</td></tr>
<tr><td valign="top" align="center">
<a name="Param"></a>Param
<td>This opcode is only ever present in sub-programs called via the
<a href="opcode.html#Program">Program</a> instruction. <a href="opcode.html#Copy">Copy</a> a value currently stored in a memory
cell of the calling (parent) frame to cell P2 in the current frames
address space. This is used by trigger programs to access the new.*
and old.* values.</p>

<p>The address of the cell in the parent frame is determined by adding
the value of the P1 argument to the value of the P1 argument to the
calling <a href="opcode.html#Program">Program</a> instruction.</td></tr>
<tr><td valign="top" align="center">
<a name="ParseSchema"></a>ParseSchema
<td>Read and parse all entries from the schema table of database P1
that match the WHERE clause P4.  If P4 is a NULL pointer, then the
entire schema for P1 is reparsed.</p>

<p>This opcode invokes the parser to create a new virtual machine,
then runs the new virtual machine.  It is thus a re-entrant opcode.</td></tr>
<tr><td valign="top" align="center">
<a name="Permutation"></a>Permutation
<td>Set the permutation used by the <a href="opcode.html#Compare">Compare</a> operator in the next
instruction.  The permutation is stored in the P4 operand.</p>

<p>The permutation is only valid for the next opcode which must be
an <a href="opcode.html#Compare">Compare</a> that has the OPFLAG_PERMUTE bit set in P5.</p>

<p>The first integer in the P4 integer array is the length of the array
and does not become part of the permutation.</td></tr>
<tr><td valign="top" align="center">
<a name="Prev"></a>Prev
<td>Back up cursor P1 so that it points to the previous key/data pair in its
table or index.  If there is no previous key/value pairs then fall through
to the following instruction.  But if the cursor backup was successful,
jump immediately to P2.</p>

<p>The <a href="opcode.html#Prev">Prev</a> opcode is only valid following an <a href="opcode.html#SeekLT">SeekLT</a>, <a href="opcode.html#SeekLE">SeekLE</a>, or
<a href="opcode.html#Last">Last</a> opcode used to position the cursor.  <a href="opcode.html#Prev">Prev</a> is not allowed
to follow <a href="opcode.html#SeekGT">SeekGT</a>, <a href="opcode.html#SeekGE">SeekGE</a>, or <a href="opcode.html#Rewind">Rewind</a>.</p>

<p>The P1 cursor must be for a real table, not a pseudo-table.  If P1 is
not open then the behavior is undefined.</p>

<p>The P3 value is a hint to the btree implementation. If P3==1, that
means P1 is an SQL index and that this instruction could have been
omitted if that index had been unique.  P3 is usually 0.  P3 is
always either 0 or 1.</p>

<p>If P5 is positive and the jump is taken, then event counter
number P5-1 in the prepared statement is incremented.</td></tr>
<tr><td valign="top" align="center">
<a name="Program"></a>Program
<td>Execute the trigger program passed as P4 (type P4_SUBPROGRAM).</p>

<p>P1 contains the address of the memory cell that contains the first memory
cell in an array of values used as arguments to the sub-program. P2
contains the address to jump to if the sub-program throws an IGNORE
exception using the RAISE() function. Register P3 contains the address
of a memory cell in this (the parent) VM that is used to allocate the
memory required by the sub-vdbe at runtime.</p>

<p>P4 is a pointer to the VM containing the trigger program.</p>

<p>If P5 is non-zero, then recursive program invocation is enabled.</td></tr>
<tr><td valign="top" align="center">
<a name="PureFunc"></a>PureFunc
<td>Invoke a user function (P4 is a pointer to an sqlite3_context object that
contains a pointer to the function to be run) with arguments taken
from register P2 and successors.  The number of arguments is in
the sqlite3_context object that P4 points to.
The result of the function is stored
in register P3.  Register P3 must not be one of the function inputs.</p>

<p>P1 is a 32-bit bitmask indicating whether or not each argument to the
function was determined to be constant at compile time. If the first
argument was constant then bit 0 of P1 is set. This is used to determine
whether meta data associated with a user function argument using the
sqlite3_set_auxdata() API may be safely retained until the next
invocation of this opcode.</p>

<p>This opcode works exactly like <a href="opcode.html#Function">Function</a>.  The only difference is in
its name.  This opcode is used in places where the function must be
purely non-deterministic.  Some built-in date/time functions can be
either determinitic of non-deterministic, depending on their arguments.
When those function are used in a non-deterministic way, they will check
to see if they were called using <a href="opcode.html#PureFunc">PureFunc</a> instead of <a href="opcode.html#Function">Function</a>, and
if they were, they throw an error.</p>

<p>See also: <a href="opcode.html#AggStep">AggStep</a>, <a href="opcode.html#AggFinal">AggFinal</a>, <a href="opcode.html#Function">Function</a></td></tr>
<tr><td valign="top" align="center">
<a name="ReadCookie"></a>ReadCookie
<td>Read cookie number P3 from database P1 and write it into register P2.
P3==1 is the schema version.  P3==2 is the database format.
P3==3 is the recommended pager cache size, and so forth.  P1==0 is
the main database file and P1==1 is the database file used to store
temporary tables.</p>

<p>There must be a read-lock on the database (either a transaction
must be started or there must be an open cursor) before
executing this instruction.</td></tr>
<tr><td valign="top" align="center">
<a name="Real"></a>Real
<td>P4 is a pointer to a 64-bit floating point value.
Write that value into register P2.</td></tr>
<tr><td valign="top" align="center">
<a name="RealAffinity"></a>RealAffinity
<td>If register P1 holds an integer convert it to a real value.</p>

<p>This opcode is used when extracting information from a column that
has REAL affinity.  Such column values may still be stored as
integers, for space efficiency, but after extraction we want them
to have only a real value.</td></tr>
<tr><td valign="top" align="center">
<a name="ReleaseReg"></a>ReleaseReg
<td>Release registers from service.  Any content that was in the
the registers is unreliable after this opcode completes.</p>

<p>The registers released will be the P2 registers starting at P1,
except if bit ii of P3 set, then do not release register P1+ii.
In other words, P3 is a mask of registers to preserve.</p>

<p>Releasing a register clears the Mem.pScopyFrom pointer.  That means
that if the content of the released register was set using <a href="opcode.html#SCopy">SCopy</a>,
a change to the value of the source register for the <a href="opcode.html#SCopy">SCopy</a> will no longer
generate an assertion fault in sqlite3VdbeMemAboutToChange().</p>

<p>If P5 is set, then all released registers have their type set
to MEM_Undefined so that any subsequent attempt to read the released
register (before it is reinitialized) will generate an assertion fault.</p>

<p>P5 ought to be set on every call to this opcode.
However, there are places in the code generator will release registers
before their are used, under the (valid) assumption that the registers
will not be reallocated for some other purpose before they are used and
hence are safe to release.</p>

<p>This opcode is only available in testing and debugging builds.  It is
not generated for release builds.  The purpose of this opcode is to help
validate the generated bytecode.  This opcode does not actually contribute
to computing an answer.</td></tr>
<tr><td valign="top" align="center">
<a name="Remainder"></a>Remainder
<td>Compute the remainder after integer register P2 is divided by
register P1 and store the result in register P3.
If the value in register P1 is zero the result is NULL.
If either operand is NULL, the result is NULL.</td></tr>
<tr><td valign="top" align="center">
<a name="ReopenIdx"></a>ReopenIdx
<td>The <a href="opcode.html#ReopenIdx">ReopenIdx</a> opcode works like <a href="opcode.html#OpenRead">OpenRead</a> except that it first
checks to see if the cursor on P1 is already open on the same
b-tree and if it is this opcode becomes a no-op.  In other words,
if the cursor is already open, do not reopen it.</p>

<p>The <a href="opcode.html#ReopenIdx">ReopenIdx</a> opcode may only be used with P5==0 or P5==OPFLAG_SEEKEQ
and with P4 being a P4_KEYINFO object.  Furthermore, the P3 value must
be the same as every other <a href="opcode.html#ReopenIdx">ReopenIdx</a> or <a href="opcode.html#OpenRead">OpenRead</a> for the same cursor
number.</p>

<p>Allowed P5 bits:
<ul>
<li>  <b>0x02 OPFLAG_SEEKEQ</b>: This cursor will only be used for
equality lookups (implemented as a pair of opcodes <a href="opcode.html#SeekGE">SeekGE</a>/<a href="opcode.html#IdxGT">IdxGT</a>
of <a href="opcode.html#SeekLE">SeekLE</a>/<a href="opcode.html#IdxLT">IdxLT</a>)
</ul></p>

<p>See also: <a href="opcode.html#OpenRead">OpenRead</a>, <a href="opcode.html#OpenWrite">OpenWrite</a></td></tr>
<tr><td valign="top" align="center">
<a name="ResetCount"></a>ResetCount
<td>The value of the change counter is copied to the database handle
change counter (returned by subsequent calls to sqlite3_changes()).
Then the VMs internal change counter resets to 0.
This is used by trigger programs.</td></tr>
<tr><td valign="top" align="center">
<a name="ResetSorter"></a>ResetSorter
<td>Delete all contents from the ephemeral table or sorter
that is open on cursor P1.</p>

<p>This opcode only works for cursors used for sorting and
opened with <a href="opcode.html#OpenEphemeral">OpenEphemeral</a> or <a href="opcode.html#SorterOpen">SorterOpen</a>.</td></tr>
<tr><td valign="top" align="center">
<a name="ResultRow"></a>ResultRow
<td>The registers P1 through P1+P2-1 contain a single row of
results. This opcode causes the sqlite3_step() call to terminate
with an SQLITE_ROW return code and it sets up the sqlite3_stmt
structure to provide access to the r(P1)..r(P1+P2-1) values as
the result row.</td></tr>
<tr><td valign="top" align="center">
<a name="Return"></a>Return
<td>Jump to the address stored in register P1.  If P1 is a return address
register, then this accomplishes a return from a subroutine.</p>

<p>If P3 is 1, then the jump is only taken if register P1 holds an integer
values, otherwise execution falls through to the next opcode, and the
<a href="opcode.html#Return">Return</a> becomes a no-op. If P3 is 0, then register P1 must hold an
integer or else an assert() is raised.  P3 should be set to 1 when
this opcode is used in combination with <a href="opcode.html#BeginSubrtn">BeginSubrtn</a>, and set to 0
otherwise.</p>

<p>The value in register P1 is unchanged by this opcode.</p>

<p>P2 is not used by the byte-code engine.  However, if P2 is positive
and also less than the current address, then the "EXPLAIN" output
formatter in the CLI will indent all opcodes from the P2 opcode up
to be not including the current <a href="opcode.html#Return">Return</a>.   P2 should be the first opcode
in the subroutine from which this opcode is returning.  Thus the P2
value is a byte-code indentation hint.  See tag-20220407a in
wherecode.c and shell.c.</td></tr>
<tr><td valign="top" align="center">
<a name="Rewind"></a>Rewind
<td>The next use of the <a href="opcode.html#Rowid">Rowid</a> or <a href="opcode.html#Column">Column</a> or <a href="opcode.html#Next">Next</a> instruction for P1
will refer to the first entry in the database table or index.
If the table or index is empty, jump immediately to P2.
If the table or index is not empty, fall through to the following
instruction.</p>

<p>This opcode leaves the cursor configured to move in forward order,
from the beginning toward the end.  In other words, the cursor is
configured to use <a href="opcode.html#Next">Next</a>, not <a href="opcode.html#Prev">Prev</a>.</td></tr>
<tr><td valign="top" align="center">
<a name="RowCell"></a>RowCell
<td>P1 and P2 are both open cursors. Both must be opened on the same type
of table - intkey or index. This opcode is used as part of copying
the current row from P2 into P1. If the cursors are opened on intkey
tables, register P3 contains the rowid to use with the new record in
P1. If they are opened on index tables, P3 is not used.</p>

<p>This opcode must be followed by either an <a href="opcode.html#Insert">Insert</a> or InsertIdx opcode
with the OPFLAG_PREFORMAT flag set to complete the insert operation.</td></tr>
<tr><td valign="top" align="center">
<a name="RowData"></a>RowData
<td>Write into register P2 the complete row content for the row at
which cursor P1 is currently pointing.
There is no interpretation of the data.
It is just copied onto the P2 register exactly as
it is found in the database file.</p>

<p>If cursor P1 is an index, then the content is the key of the row.
If cursor P2 is a table, then the content extracted is the data.</p>

<p>If the P1 cursor must be pointing to a valid row (not a NULL row)
of a real table, not a pseudo-table.</p>

<p>If P3!=0 then this opcode is allowed to make an ephemeral pointer
into the database page.  That means that the content of the output
register will be invalidated as soon as the cursor moves - including
moves caused by other cursors that "save" the current cursors
position in order that they can write to the same table.  If P3==0
then a copy of the data is made into memory.  P3!=0 is faster, but
P3==0 is safer.</p>

<p>If P3!=0 then the content of the P2 register is unsuitable for use
in OP_Result and any OP_Result will invalidate the P2 register content.
The P2 register content is invalidated by opcodes like <a href="opcode.html#Function">Function</a> or
by any use of another cursor pointing to the same table.</td></tr>
<tr><td valign="top" align="center">
<a name="Rowid"></a>Rowid
<td>Store in register P2 an integer which is the key of the table entry that
P1 is currently point to.</p>

<p>P1 can be either an ordinary table or a virtual table.  There used to
be a separate OP_VRowid opcode for use with virtual tables, but this
one opcode now works for both table types.</td></tr>
<tr><td valign="top" align="center">
<a name="RowSetAdd"></a>RowSetAdd
<td>Insert the integer value held by register P2 into a RowSet object
held in register P1.</p>

<p>An assertion fails if P2 is not an integer.</td></tr>
<tr><td valign="top" align="center">
<a name="RowSetRead"></a>RowSetRead
<td>Extract the smallest value from the RowSet object in P1
and put that value into register P3.
Or, if RowSet object P1 is initially empty, leave P3
unchanged and jump to instruction P2.</td></tr>
<tr><td valign="top" align="center">
<a name="RowSetTest"></a>RowSetTest
<td>Register P3 is assumed to hold a 64-bit integer value. If register P1
contains a RowSet object and that RowSet object contains
the value held in P3, jump to register P2. Otherwise, insert the
integer in P3 into the RowSet and continue on to the
next opcode.</p>

<p>The RowSet object is optimized for the case where sets of integers
are inserted in distinct phases, which each set contains no duplicates.
Each set is identified by a unique P4 value. The first set
must have P4==0, the final set must have P4==-1, and for all other sets
must have P4&gt;0.</p>

<p>This allows optimizations: (a) when P4==0 there is no need to test
the RowSet object for P3, as it is guaranteed not to contain it,
(b) when P4==-1 there is no need to insert the value, as it will
never be tested for, and (c) when a value that is part of set X is
inserted, there is no need to search to see if the same value was
previously inserted as part of set X (only if it was previously
inserted as part of some other set).</td></tr>
<tr><td valign="top" align="center">
<a name="Savepoint"></a>Savepoint
<td>Open, release or rollback the savepoint named by parameter P4, depending
on the value of P1. To open a new savepoint set P1==0 (SAVEPOINT_BEGIN).
To release (commit) an existing savepoint set P1==1 (SAVEPOINT_RELEASE).
To rollback an existing savepoint set P1==2 (SAVEPOINT_ROLLBACK).</td></tr>
<tr><td valign="top" align="center">
<a name="SCopy"></a>SCopy
<td>Make a shallow copy of register P1 into register P2.</p>

<p>This instruction makes a shallow copy of the value.  If the value
is a string or blob, then the copy is only a pointer to the
original and hence if the original changes so will the copy.
Worse, if the original is deallocated, the copy becomes invalid.
Thus the program must guarantee that the original will not change
during the lifetime of the copy.  Use <a href="opcode.html#Copy">Copy</a> to make a complete
copy.</td></tr>
<tr><td valign="top" align="center">
<a name="SeekEnd"></a>SeekEnd
<td>Position cursor P1 at the end of the btree for the purpose of
appending a new entry onto the btree.</p>

<p>It is assumed that the cursor is used only for appending and so
if the cursor is valid, then the cursor must already be pointing
at the end of the btree and so no changes are made to
the cursor.</td></tr>
<tr><td valign="top" align="center">
<a name="SeekGE"></a>SeekGE
<td>If cursor P1 refers to an SQL table (B-Tree that uses integer keys),
use the value in register P3 as the key.  If cursor P1 refers
to an SQL index, then P3 is the first in an array of P4 registers
that are used as an unpacked index key.</p>

<p>Reposition cursor P1 so that  it points to the smallest entry that
is greater than or equal to the key value. If there are no records
greater than or equal to the key and P2 is not zero, then jump to P2.</p>

<p>If the cursor P1 was opened using the OPFLAG_SEEKEQ flag, then this
opcode will either land on a record that exactly matches the key, or
else it will cause a jump to P2.  When the cursor is OPFLAG_SEEKEQ,
this opcode must be followed by an <a href="opcode.html#IdxLE">IdxLE</a> opcode with the same arguments.
The <a href="opcode.html#IdxGT">IdxGT</a> opcode will be skipped if this opcode succeeds, but the
<a href="opcode.html#IdxGT">IdxGT</a> opcode will be used on subsequent loop iterations.  The
OPFLAG_SEEKEQ flags is a hint to the btree layer to say that this
is an equality search.</p>

<p>This opcode leaves the cursor configured to move in forward order,
from the beginning toward the end.  In other words, the cursor is
configured to use <a href="opcode.html#Next">Next</a>, not <a href="opcode.html#Prev">Prev</a>.</p>

<p>See also: <a href="opcode.html#Found">Found</a>, <a href="opcode.html#NotFound">NotFound</a>, SeekLt, SeekGt, SeekLe</td></tr>
<tr><td valign="top" align="center">
<a name="SeekGT"></a>SeekGT
<td>If cursor P1 refers to an SQL table (B-Tree that uses integer keys),
use the value in register P3 as a key. If cursor P1 refers
to an SQL index, then P3 is the first in an array of P4 registers
that are used as an unpacked index key.</p>

<p>Reposition cursor P1 so that it points to the smallest entry that
is greater than the key value. If there are no records greater than
the key and P2 is not zero, then jump to P2.</p>

<p>This opcode leaves the cursor configured to move in forward order,
from the beginning toward the end.  In other words, the cursor is
configured to use <a href="opcode.html#Next">Next</a>, not <a href="opcode.html#Prev">Prev</a>.</p>

<p>See also: <a href="opcode.html#Found">Found</a>, <a href="opcode.html#NotFound">NotFound</a>, SeekLt, SeekGe, SeekLe</td></tr>
<tr><td valign="top" align="center">
<a name="SeekHit"></a>SeekHit
<td>Increase or decrease the seekHit value for cursor P1, if necessary,
so that it is no less than P2 and no greater than P3.</p>

<p>The seekHit integer represents the maximum of terms in an index for which
there is known to be at least one match.  If the seekHit value is smaller
than the total number of equality terms in an index lookup, then the
<a href="opcode.html#IfNoHope">IfNoHope</a> opcode might run to see if the IN loop can be abandoned
early, thus saving work.  This is part of the IN-early-out optimization.</p>

<p>P1 must be a valid b-tree cursor.</td></tr>
<tr><td valign="top" align="center">
<a name="SeekLE"></a>SeekLE
<td>If cursor P1 refers to an SQL table (B-Tree that uses integer keys),
use the value in register P3 as a key. If cursor P1 refers
to an SQL index, then P3 is the first in an array of P4 registers
that are used as an unpacked index key.</p>

<p>Reposition cursor P1 so that it points to the largest entry that
is less than or equal to the key value. If there are no records
less than or equal to the key and P2 is not zero, then jump to P2.</p>

<p>This opcode leaves the cursor configured to move in reverse order,
from the end toward the beginning.  In other words, the cursor is
configured to use <a href="opcode.html#Prev">Prev</a>, not <a href="opcode.html#Next">Next</a>.</p>

<p>If the cursor P1 was opened using the OPFLAG_SEEKEQ flag, then this
opcode will either land on a record that exactly matches the key, or
else it will cause a jump to P2.  When the cursor is OPFLAG_SEEKEQ,
this opcode must be followed by an <a href="opcode.html#IdxLE">IdxLE</a> opcode with the same arguments.
The <a href="opcode.html#IdxGE">IdxGE</a> opcode will be skipped if this opcode succeeds, but the
<a href="opcode.html#IdxGE">IdxGE</a> opcode will be used on subsequent loop iterations.  The
OPFLAG_SEEKEQ flags is a hint to the btree layer to say that this
is an equality search.</p>

<p>See also: <a href="opcode.html#Found">Found</a>, <a href="opcode.html#NotFound">NotFound</a>, SeekGt, SeekGe, SeekLt</td></tr>
<tr><td valign="top" align="center">
<a name="SeekLT"></a>SeekLT
<td>If cursor P1 refers to an SQL table (B-Tree that uses integer keys),
use the value in register P3 as a key. If cursor P1 refers
to an SQL index, then P3 is the first in an array of P4 registers
that are used as an unpacked index key.</p>

<p>Reposition cursor P1 so that  it points to the largest entry that
is less than the key value. If there are no records less than
the key and P2 is not zero, then jump to P2.</p>

<p>This opcode leaves the cursor configured to move in reverse order,
from the end toward the beginning.  In other words, the cursor is
configured to use <a href="opcode.html#Prev">Prev</a>, not <a href="opcode.html#Next">Next</a>.</p>

<p>See also: <a href="opcode.html#Found">Found</a>, <a href="opcode.html#NotFound">NotFound</a>, SeekGt, SeekGe, SeekLe</td></tr>
<tr><td valign="top" align="center">
<a name="SeekRowid"></a>SeekRowid
<td>P1 is the index of a cursor open on an SQL table btree (with integer
keys).  If register P3 does not contain an integer or if P1 does not
contain a record with rowid P3 then jump immediately to P2.
Or, if P2 is 0, raise an SQLITE_CORRUPT error. If P1 does contain
a record with rowid P3 then
leave the cursor pointing at that record and fall through to the next
instruction.</p>

<p>The <a href="opcode.html#NotExists">NotExists</a> opcode performs the same operation, but with <a href="opcode.html#NotExists">NotExists</a>
the P3 register must be guaranteed to contain an integer value.  With this
opcode, register P3 might not contain an integer.</p>

<p>The <a href="opcode.html#NotFound">NotFound</a> opcode performs the same operation on index btrees
(with arbitrary multi-value keys).</p>

<p>This opcode leaves the cursor in a state where it cannot be advanced
in either direction.  In other words, the <a href="opcode.html#Next">Next</a> and <a href="opcode.html#Prev">Prev</a> opcodes will
not work following this opcode.</p>

<p>See also: <a href="opcode.html#Found">Found</a>, <a href="opcode.html#NotFound">NotFound</a>, <a href="opcode.html#NoConflict">NoConflict</a>, <a href="opcode.html#SeekRowid">SeekRowid</a></td></tr>
<tr><td valign="top" align="center">
<a name="SeekScan"></a>SeekScan
<td>This opcode is a prefix opcode to <a href="opcode.html#SeekGE">SeekGE</a>.  In other words, this
opcode must be immediately followed by <a href="opcode.html#SeekGE">SeekGE</a>. This constraint is
checked by assert() statements.</p>

<p>This opcode uses the P1 through P4 operands of the subsequent
<a href="opcode.html#SeekGE">SeekGE</a>.  In the text that follows, the operands of the subsequent
<a href="opcode.html#SeekGE">SeekGE</a> opcode are denoted as SeekOP.P1 through SeekOP.P4.   Only
the P1, P2 and P5 operands of this opcode are also used, and  are called
This.P1, This.P2 and This.P5.</p>

<p>This opcode helps to optimize IN operators on a multi-column index
where the IN operator is on the later terms of the index by avoiding
unnecessary seeks on the btree, substituting steps to the next row
of the b-tree instead.  A correct answer is obtained if this opcode
is omitted or is a no-op.</p>

<p>The <a href="opcode.html#SeekGE">SeekGE</a>.P3 and <a href="opcode.html#SeekGE">SeekGE</a>.P4 operands identify an unpacked key which
is the desired entry that we want the cursor <a href="opcode.html#SeekGE">SeekGE</a>.P1 to be pointing
to.  Call this <a href="opcode.html#SeekGE">SeekGE</a>.P3/P4 row the "target".</p>

<p>If the <a href="opcode.html#SeekGE">SeekGE</a>.P1 cursor is not currently pointing to a valid row,
then this opcode is a no-op and control passes through into the <a href="opcode.html#SeekGE">SeekGE</a>.</p>

<p>If the <a href="opcode.html#SeekGE">SeekGE</a>.P1 cursor is pointing to a valid row, then that row
might be the target row, or it might be near and slightly before the
target row, or it might be after the target row.  If the cursor is
currently before the target row, then this opcode attempts to position
the cursor on or after the target row by invoking sqlite3BtreeStep()
on the cursor between 1 and This.P1 times.</p>

<p>The This.P5 parameter is a flag that indicates what to do if the
cursor ends up pointing at a valid row that is past the target
row.  If This.P5 is false (0) then a jump is made to <a href="opcode.html#SeekGE">SeekGE</a>.P2.  If
This.P5 is true (non-zero) then a jump is made to This.P2.  The P5==0
case occurs when there are no inequality constraints to the right of
the IN constraing.  The jump to <a href="opcode.html#SeekGE">SeekGE</a>.P2 ends the loop.  The P5!=0 case
occurs when there are inequality constraints to the right of the IN
operator.  In that case, the This.P2 will point either directly to or
to setup code prior to the <a href="opcode.html#IdxGT">IdxGT</a> or <a href="opcode.html#IdxGE">IdxGE</a> opcode that checks for
loop terminate.</p>

<p>Possible outcomes from this opcode:<ol></p>

<p><li> If the cursor is initally not pointed to any valid row, then
fall through into the subsequent <a href="opcode.html#SeekGE">SeekGE</a> opcode.</p>

<p><li> If the cursor is left pointing to a row that is before the target
row, even after making as many as This.P1 calls to
sqlite3BtreeNext(), then also fall through into <a href="opcode.html#SeekGE">SeekGE</a>.</p>

<p><li> If the cursor is left pointing at the target row, either because it
was at the target row to begin with or because one or more
sqlite3BtreeNext() calls moved the cursor to the target row,
then jump to This.P2..,</p>

<p><li> If the cursor started out before the target row and a call to
to sqlite3BtreeNext() moved the cursor off the end of the index
(indicating that the target row definitely does not exist in the
btree) then jump to <a href="opcode.html#SeekGE">SeekGE</a>.P2, ending the loop.</p>

<p><li> If the cursor ends up on a valid row that is past the target row
(indicating that the target row does not exist in the btree) then
jump to SeekOP.P2 if This.P5==0 or to This.P2 if This.P5&gt;0.
</ol></td></tr>
<tr><td valign="top" align="center">
<a name="Sequence"></a>Sequence
<td>Find the next available sequence number for cursor P1.
Write the sequence number into register P2.
The sequence number on the cursor is incremented after this
instruction.</td></tr>
<tr><td valign="top" align="center">
<a name="SequenceTest"></a>SequenceTest
<td>P1 is a sorter cursor. If the sequence counter is currently zero, jump
to P2. Regardless of whether or not the jump is taken, increment the
the sequence value.</td></tr>
<tr><td valign="top" align="center">
<a name="SetCookie"></a>SetCookie
<td>Write the integer value P3 into cookie number P2 of database P1.
P2==1 is the schema version.  P2==2 is the database format.
P2==3 is the recommended pager cache
size, and so forth.  P1==0 is the main database file and P1==1 is the
database file used to store temporary tables.</p>

<p>A transaction must be started before executing this opcode.</p>

<p>If P2 is the SCHEMA_VERSION cookie (cookie number 1) then the internal
schema version is set to P3-P5.  The "PRAGMA schema_version=N" statement
has P5 set to 1, so that the internal schema version will be different
from the database schema version, resulting in a schema reset.</td></tr>
<tr><td valign="top" align="center">
<a name="ShiftLeft"></a>ShiftLeft
<td>Shift the integer value in register P2 to the left by the
number of bits specified by the integer in register P1.
Store the result in register P3.
If either input is NULL, the result is NULL.</td></tr>
<tr><td valign="top" align="center">
<a name="ShiftRight"></a>ShiftRight
<td>Shift the integer value in register P2 to the right by the
number of bits specified by the integer in register P1.
Store the result in register P3.
If either input is NULL, the result is NULL.</td></tr>
<tr><td valign="top" align="center">
<a name="SoftNull"></a>SoftNull
<td>Set register P1 to have the value NULL as seen by the <a href="opcode.html#MakeRecord">MakeRecord</a>
instruction, but do not free any string or blob memory associated with
the register, so that if the value was a string or blob that was
previously copied using <a href="opcode.html#SCopy">SCopy</a>, the copies will continue to be valid.</td></tr>
<tr><td valign="top" align="center">
<a name="Sort"></a>Sort
<td>This opcode does exactly the same thing as <a href="opcode.html#Rewind">Rewind</a> except that
it increments an undocumented global variable used for testing.</p>

<p>Sorting is accomplished by writing records into a sorting index,
then rewinding that index and playing it back from beginning to
end.  We use the <a href="opcode.html#Sort">Sort</a> opcode instead of <a href="opcode.html#Rewind">Rewind</a> to do the
rewinding so that the global variable will be incremented and
regression tests can determine whether or not the optimizer is
correctly optimizing out sorts.</td></tr>
<tr><td valign="top" align="center">
<a name="SorterCompare"></a>SorterCompare
<td>P1 is a sorter cursor. This instruction compares a prefix of the
record blob in register P3 against a prefix of the entry that
the sorter cursor currently points to.  Only the first P4 fields
of r&#91;P3&#93; and the sorter record are compared.</p>

<p>If either P3 or the sorter contains a NULL in one of their significant
fields (not counting the P4 fields at the end which are ignored) then
the comparison is assumed to be equal.</p>

<p>Fall through to next instruction if the two records compare equal to
each other.  <a href="opcode.html#Jump">Jump</a> to P2 if they are different.</td></tr>
<tr><td valign="top" align="center">
<a name="SorterData"></a>SorterData
<td>Write into register P2 the current sorter data for sorter cursor P1.
Then clear the column header cache on cursor P3.</p>

<p>This opcode is normally use to move a record out of the sorter and into
a register that is the source for a pseudo-table cursor created using
<a href="opcode.html#OpenPseudo">OpenPseudo</a>.  That pseudo-table cursor is the one that is identified by
parameter P3.  Clearing the P3 column cache as part of this opcode saves
us from having to issue a separate <a href="opcode.html#NullRow">NullRow</a> instruction to clear that cache.</td></tr>
<tr><td valign="top" align="center">
<a name="SorterInsert"></a>SorterInsert
<td>Register P2 holds an SQL index key made using the
<a href="opcode.html#MakeRecord">MakeRecord</a> instructions.  This opcode writes that key
into the sorter P1.  Data for the entry is nil.</td></tr>
<tr><td valign="top" align="center">
<a name="SorterNext"></a>SorterNext
<td>This opcode works just like <a href="opcode.html#Next">Next</a> except that P1 must be a
sorter object for which the <a href="opcode.html#SorterSort">SorterSort</a> opcode has been
invoked.  This opcode advances the cursor to the next sorted
record, or jumps to P2 if there are no more sorted records.</td></tr>
<tr><td valign="top" align="center">
<a name="SorterOpen"></a>SorterOpen
<td>This opcode works like <a href="opcode.html#OpenEphemeral">OpenEphemeral</a> except that it opens
a transient index that is specifically designed to sort large
tables using an external merge-sort algorithm.</p>

<p>If argument P3 is non-zero, then it indicates that the sorter may
assume that a stable sort considering the first P3 fields of each
key is sufficient to produce the required results.</td></tr>
<tr><td valign="top" align="center">
<a name="SorterSort"></a>SorterSort
<td>After all records have been inserted into the Sorter object
identified by P1, invoke this opcode to actually do the sorting.
<a href="opcode.html#Jump">Jump</a> to P2 if there are no records to be sorted.</p>

<p>This opcode is an alias for <a href="opcode.html#Sort">Sort</a> and <a href="opcode.html#Rewind">Rewind</a> that is used
for Sorter objects.</td></tr>
<tr><td valign="top" align="center">
<a name="SqlExec"></a>SqlExec
<td>Run the SQL statement or statements specified in the P4 string.</td></tr>
<tr><td valign="top" align="center">
<a name="String"></a>String
<td>The string value P4 of length P1 (bytes) is stored in register P2.</p>

<p>If P3 is not zero and the content of register P3 is equal to P5, then
the datatype of the register P2 is converted to BLOB.  The content is
the same sequence of bytes, it is merely interpreted as a BLOB instead
of a string, as if it had been CAST.  In other words:</p>

<p>if( P3!=0 and reg&#91;P3&#93;==P5 ) reg&#91;P2&#93; := CAST(reg&#91;P2&#93; as BLOB)</td></tr>
<tr><td valign="top" align="center">
<a name="String8"></a>String8
<td>P4 points to a nul terminated UTF-8 string. This opcode is transformed
into a <a href="opcode.html#String">String</a> opcode before it is executed for the first time.  During
this transformation, the length of string P4 is computed and stored
as the P1 parameter.</td></tr>
<tr><td valign="top" align="center">
<a name="Subtract"></a>Subtract
<td>Subtract the value in register P1 from the value in register P2
and store the result in register P3.
If either input is NULL, the result is NULL.</td></tr>
<tr><td valign="top" align="center">
<a name="TableLock"></a>TableLock
<td>Obtain a lock on a particular table. This instruction is only used when
the shared-cache feature is enabled.</p>

<p>P1 is the index of the database in sqlite3.aDb[] of the database
on which the lock is acquired.  A readlock is obtained if P3==0 or
a write lock if P3==1.</p>

<p>P2 contains the root-page of the table to lock.</p>

<p>P4 contains a pointer to the name of the table being locked. This is only
used to generate an error message if the lock cannot be obtained.</td></tr>
<tr><td valign="top" align="center">
<a name="Trace"></a>Trace
<td>Write P4 on the statement trace output if statement tracing is
enabled.</p>

<p>Operand P1 must be 0x7fffffff and P2 must positive.</td></tr>
<tr><td valign="top" align="center">
<a name="Transaction"></a>Transaction
<td>Begin a transaction on database P1 if a transaction is not already
active.
If P2 is non-zero, then a write-transaction is started, or if a
read-transaction is already active, it is upgraded to a write-transaction.
If P2 is zero, then a read-transaction is started.  If P2 is 2 or more
then an exclusive transaction is started.</p>

<p>P1 is the index of the database file on which the transaction is
started.  Index 0 is the main database file and index 1 is the
file used for temporary tables.  Indices of 2 or more are used for
attached databases.</p>

<p>If a write-transaction is started and the Vdbe.usesStmtJournal flag is
true (this flag is set if the Vdbe may modify more than one row and may
throw an ABORT exception), a statement transaction may also be opened.
More specifically, a statement transaction is opened iff the database
connection is currently not in autocommit mode, or if there are other
active statements. A statement transaction allows the changes made by this
VDBE to be rolled back after an error without having to roll back the
entire transaction. If no error is encountered, the statement transaction
will automatically commit when the VDBE halts.</p>

<p>If P5!=0 then this opcode also checks the schema cookie against P3
and the schema generation counter against P4.
The cookie changes its value whenever the database schema changes.
This operation is used to detect when that the cookie has changed
and that the current process needs to reread the schema.  If the schema
cookie in P3 differs from the schema cookie in the database header or
if the schema generation counter in P4 differs from the current
generation counter, then an SQLITE_SCHEMA error is raised and execution
halts.  The sqlite3_step() wrapper function might then reprepare the
statement and rerun it from the beginning.</td></tr>
<tr><td valign="top" align="center">
<a name="TypeCheck"></a>TypeCheck
<td>Apply affinities to the range of P2 registers beginning with P1.
Take the affinities from the Table object in P4.  If any value
cannot be coerced into the correct type, then raise an error.</p>

<p>This opcode is similar to <a href="opcode.html#Affinity">Affinity</a> except that this opcode
forces the register type to the Table column type.  This is used
to implement "strict affinity".</p>

<p>GENERATED ALWAYS AS ... STATIC columns are only checked if P3
is zero.  When P3 is non-zero, no type checking occurs for
static generated columns.  Virtual columns are computed at query time
and so they are never checked.</p>

<p>Preconditions:</p>

<p><ul>
<li> P2 should be the number of non-virtual columns in the
table of P4.
<li> Table P4 should be a STRICT table.
</ul></p>

<p>If any precondition is false, an assertion fault occurs.</td></tr>
<tr><td valign="top" align="center">
<a name="Vacuum"></a>Vacuum
<td>Vacuum the entire database P1.  P1 is 0 for "main", and 2 or more
for an attached database.  The "temp" database may not be vacuumed.</p>

<p>If P2 is not zero, then it is a register holding a string which is
the file into which the result of vacuum should be written.  When
P2 is zero, the vacuum overwrites the original database.</td></tr>
<tr><td valign="top" align="center">
<a name="Variable"></a>Variable
<td>Transfer the values of bound parameter P1 into register P2</p>

<p>If the parameter is named, then its name appears in P4.
The P4 value is used by sqlite3_bind_parameter_name().</td></tr>
<tr><td valign="top" align="center">
<a name="VBegin"></a>VBegin
<td>P4 may be a pointer to an sqlite3_vtab structure. If so, call the
xBegin method for that table.</p>

<p>Also, whether or not P4 is set, check that this is not being called from
within a callback to a virtual table xSync() method. If it is, the error
code will be set to SQLITE_LOCKED.</td></tr>
<tr><td valign="top" align="center">
<a name="VColumn"></a>VColumn
<td>Store in register P3 the value of the P2-th column of
the current row of the virtual-table of cursor P1.</p>

<p>If the <a href="opcode.html#VColumn">VColumn</a> opcode is being used to fetch the value of
an unchanging column during an UPDATE operation, then the P5
value is OPFLAG_NOCHNG.  This will cause the sqlite3_vtab_nochange()
function to return true inside the xColumn method of the virtual
table implementation.  The P5 column might also contain other
bits (OPFLAG_LENGTHARG or OPFLAG_TYPEOFARG) but those bits are
unused by <a href="opcode.html#VColumn">VColumn</a>.</td></tr>
<tr><td valign="top" align="center">
<a name="VCreate"></a>VCreate
<td>P2 is a register that holds the name of a virtual table in database
P1. Call the xCreate method for that table.</td></tr>
<tr><td valign="top" align="center">
<a name="VDestroy"></a>VDestroy
<td>P4 is the name of a virtual table in database P1.  Call the xDestroy method
of that table.</td></tr>
<tr><td valign="top" align="center">
<a name="VFilter"></a>VFilter
<td>P1 is a cursor opened using <a href="opcode.html#VOpen">VOpen</a>.  P2 is an address to jump to if
the filtered result set is empty.</p>

<p>P4 is either NULL or a string that was generated by the xBestIndex
method of the module.  The interpretation of the P4 string is left
to the module implementation.</p>

<p>This opcode invokes the xFilter method on the virtual table specified
by P1.  The integer query plan parameter to xFilter is stored in register
P3. Register P3+1 stores the argc parameter to be passed to the
xFilter method. Registers P3+2..P3+1+argc are the argc
additional parameters which are passed to
xFilter as argv. Register P3+2 becomes argv[0] when passed to xFilter.</p>

<p>A jump is made to P2 if the result set after filtering would be empty.</td></tr>
<tr><td valign="top" align="center">
<a name="VInitIn"></a>VInitIn
<td>Set register P2 to be a pointer to a ValueList object for cursor P1
with cache register P3 and output register P3+1.  This ValueList object
can be used as the first argument to sqlite3_vtab_in_first() and
sqlite3_vtab_in_next() to extract all of the values stored in the P1
cursor.  Register P3 is used to hold the values returned by
sqlite3_vtab_in_first() and sqlite3_vtab_in_next().</td></tr>
<tr><td valign="top" align="center">
<a name="VNext"></a>VNext
<td>Advance virtual table P1 to the next row in its result set and
jump to instruction P2.  Or, if the virtual table has reached
the end of its result set, then fall through to the next instruction.</td></tr>
<tr><td valign="top" align="center">
<a name="VOpen"></a>VOpen
<td>P4 is a pointer to a virtual table object, an sqlite3_vtab structure.
P1 is a cursor number.  This opcode opens a cursor to the virtual
table and stores that cursor in P1.</td></tr>
<tr><td valign="top" align="center">
<a name="VRename"></a>VRename
<td>P4 is a pointer to a virtual table object, an sqlite3_vtab structure.
This opcode invokes the corresponding xRename method. The value
in register P1 is passed as the zName argument to the xRename method.</td></tr>
<tr><td valign="top" align="center">
<a name="VUpdate"></a>VUpdate
<td>P4 is a pointer to a virtual table object, an sqlite3_vtab structure.
This opcode invokes the corresponding xUpdate method. P2 values
are contiguous memory cells starting at P3 to pass to the xUpdate
invocation. The value in register (P3+P2-1) corresponds to the
p2th element of the argv array passed to xUpdate.</p>

<p>The xUpdate method will do a DELETE or an INSERT or both.
The argv[0] element (which corresponds to memory cell P3)
is the rowid of a row to delete.  If argv[0] is NULL then no
deletion occurs.  The argv[1] element is the rowid of the new
row.  This can be NULL to have the virtual table select the new
rowid for itself.  The subsequent elements in the array are
the values of columns in the new row.</p>

<p>If P2==1 then no insert is performed.  argv[0] is the rowid of
a row to delete.</p>

<p>P1 is a boolean flag. If it is set to true and the xUpdate call
is successful, then the value returned by sqlite3_last_insert_rowid()
is set to the value of the rowid for the row just inserted.</p>

<p>P5 is the error actions (OE_Replace, OE_Fail, OE_Ignore, etc) to
apply in the case of a constraint failure on an insert or update.</td></tr>
<tr><td valign="top" align="center">
<a name="Yield"></a>Yield
<td>Swap the program counter with the value in register P1.  This
has the effect of yielding to a coroutine.</p>

<p>If the coroutine that is launched by this instruction ends with
<a href="opcode.html#Yield">Yield</a> or <a href="opcode.html#Return">Return</a> then continue to the next instruction.  But if
the coroutine launched by this instruction ends with
<a href="opcode.html#EndCoroutine">EndCoroutine</a>, then jump to P2 rather than continuing with the
next instruction.</p>

<p>See also: <a href="opcode.html#InitCoroutine">InitCoroutine</a></td></tr>
<tr><td valign="top" align="center">
<a name="ZeroOrNull"></a>ZeroOrNull
<td>If all both registers P1 and P3 are NOT NULL, then store a zero in
register P2.  If either registers P1 or P3 are NULL then put
a NULL in register P2.</td></tr>

  </table></blockquote>
  </div>

</p><p align="center"><small><i>This page last modified on  <a href="https://sqlite.org/docsrc/honeypot" id="mtimelink"  data-href="https://sqlite.org/docsrc/finfo/pages/opcode.in?m=c96f076c3fd941bcc">2022-08-10 18:45:48</a> UTC </small></i></p>