# 2018-07-26 # # The author disclaims copyright to this source code. In place of # a legal notice, here is a blessing: # # May you do good and not evil. # May you find forgiveness for yourself and forgive others. # May you share freely, never taking more than you give. # #*********************************************************************** # This file implements regression tests for SQLite library. The # focus of this file is testing the WHERE-clause constant propagation # optimization. # set testdir [file dirname $argv0] source $testdir/tester.tcl set ::testprefix whereL do_execsql_test 100 { CREATE TABLE t1(a INT PRIMARY KEY, b, c, d, e); CREATE TABLE t2(a INT PRIMARY KEY, f, g, h, i); CREATE TABLE t3(a INT PRIMARY KEY, j, k, l, m); CREATE VIEW v4 AS SELECT * FROM t2 UNION ALL SELECT * FROM t3; } do_eqp_test 110 { SELECT * FROM t1, v4 WHERE t1.a=?1 AND v4.a=t1.a; } { QUERY PLAN `--COMPOUND QUERY |--LEFT-MOST SUBQUERY | |--SEARCH t1 USING INDEX sqlite_autoindex_t1_1 (a=?) | `--SEARCH t2 USING INDEX sqlite_autoindex_t2_1 (a=?) `--UNION ALL |--SEARCH t1 USING INDEX sqlite_autoindex_t1_1 (a=?) `--SEARCH t3 USING INDEX sqlite_autoindex_t3_1 (a=?) } # The scan of the t1 table goes first since that enables the ORDER BY # sort to be omitted. This would not be possible without constant # propagation because without it the t1 table would depend on t3. # do_eqp_test 120 { SELECT * FROM t1, t2, t3 WHERE t1.a=t2.a AND t2.a=t3.j AND t3.j=5 ORDER BY t1.a; } { QUERY PLAN |--SEARCH t1 USING INDEX sqlite_autoindex_t1_1 (a=?) |--SEARCH t2 USING INDEX sqlite_autoindex_t2_1 (a=?) `--SCAN t3 } do_eqp_test 121 { SELECT * FROM t1, t2, t3 WHERE t1.a=t2.a AND t2.a=t3.j AND t3.j=abs(5) ORDER BY t1.a; } { QUERY PLAN |--SEARCH t1 USING INDEX sqlite_autoindex_t1_1 (a=?) |--SEARCH t2 USING INDEX sqlite_autoindex_t2_1 (a=?) `--SCAN t3 } # The sqlite3ExprIsConstant() routine does not believe that # the expression "coalesce(5,random())" is constant. So the # optimization does not apply in this case. # sqlite3_create_function db do_eqp_test 122 { SELECT * FROM t1, t2, t3 WHERE t1.a=t2.a AND t2.a=t3.j AND t3.j=coalesce(5,random()) ORDER BY t1.a; } { QUERY PLAN |--SCAN t3 |--SEARCH t1 USING INDEX sqlite_autoindex_t1_1 (a=?) |--SEARCH t2 USING INDEX sqlite_autoindex_t2_1 (a=?) `--USE TEMP B-TREE FOR ORDER BY } # Constant propagation in the face of collating sequences: # do_execsql_test 200 { CREATE TABLE c3(x COLLATE binary, y COLLATE nocase, z COLLATE binary); CREATE INDEX c3x ON c3(x); INSERT INTO c3 VALUES('ABC', 'ABC', 'abc'); SELECT * FROM c3 WHERE x=y AND y=z AND z='abc'; } {ABC ABC abc} # If the constants are blindly propagated, as shown in the following # query, the wrong answer results: # do_execsql_test 201 { SELECT * FROM c3 WHERE x='abc' AND y='abc' AND z='abc'; } {} # Constant propagation caused an incorrect answer in the following # query. (Reported by Bentley system on 2018-08-09.) # do_execsql_test 300 { CREATE TABLE A(id INTEGER PRIMARY KEY, label TEXT); CREATE TABLE B(id INTEGER PRIMARY KEY, label TEXT, Aid INTEGER); CREATE TABLE C( id INTEGER PRIMARY KEY, xx INTEGER NOT NULL, yy INTEGER, zz INTEGER ); CREATE UNIQUE INDEX x2 ON C(yy); CREATE UNIQUE INDEX x4 ON C(yy, zz); INSERT INTO A(id) VALUES(1); INSERT INTO B(id) VALUES(2); INSERT INTO C(id,xx,yy,zz) VALUES(99,50,1,2); SELECT 1 FROM A, (SELECT id,xx,yy,zz FROM C) subq, B WHERE A.id='1' AND A.id=subq.yy AND B.id=subq.zz; } {1} do_execsql_test 301 { SELECT 1 FROM A, (SELECT id,xx,yy,zz FROM C) subq, B WHERE A.id=1 AND A.id=subq.yy AND B.id=subq.zz; } {1} do_execsql_test 302 { SELECT 1 FROM A, (SELECT id,yy,zz FROM C) subq, B WHERE A.id='1' AND A.id=subq.yy AND B.id=subq.zz; } {1} # 2018-10-25: Ticket [cf5ed20f] # Incorrect join result with duplicate WHERE clause constraint. # do_execsql_test 400 { CREATE TABLE x(a, b, c); CREATE TABLE y(a, b); INSERT INTO x VALUES (1, 0, 1); INSERT INTO y VALUES (1, 2); SELECT x.a FROM x JOIN y ON x.c = y.a WHERE x.b = 1 AND x.b = 1; } {} # 2020-01-07: ticket 82ac75ba0093e5dc # Incorrect join result due to mishandling of affinity in constant # propagation. # reset_db do_execsql_test 500 { PRAGMA automatic_index=OFF; CREATE TABLE t0(c0); INSERT INTO t0 VALUES('0'); CREATE VIEW v0(c0) AS SELECT CAST(0 AS INT) FROM t0; SELECT 200, * FROM t0, v0 WHERE 0 = t0.c0 AND t0.c0 = v0.c0; } {} do_execsql_test 510 { SELECT 200, * FROM t0, v0 WHERE t0.c0 = 0 AND t0.c0 = v0.c0; } {} do_execsql_test 520 { SELECT 200, * FROM t0, v0 WHERE 0 = t0.c0 AND v0.c0 = t0.c0; } {} do_execsql_test 530 { SELECT 200, * FROM t0, v0 WHERE t0.c0 = 0 AND v0.c0 = t0.c0; } {} # 2020-02-13: ticket 1dcb4d44964846ad # A problem introduced while making optimizations on the fixes above. # reset_db do_execsql_test 600 { CREATE TABLE t1(x TEXT); CREATE TABLE t2(y TEXT); INSERT INTO t1 VALUES('good'),('bad'); INSERT INTO t2 VALUES('good'),('bad'); SELECT * FROM t1 JOIN t2 ON x=y WHERE x='good' AND y='good'; } {good good} # 2020-04-24: Another test case for the previous (1dcb4d44964846ad) # ticket. The test case comes from # https://stackoverflow.com/questions/61399253/sqlite3-different-result-in-console-compared-to-python-script/ # Output verified against postgresql. # do_execsql_test 610 { CREATE TABLE tableA( ID int, RunYearMonth int ); INSERT INTO tableA VALUES(1,202003),(2,202003),(3,202003),(4,202004), (5,202004),(6,202004),(7,202004),(8,202004); CREATE TABLE tableB ( ID int, RunYearMonth int ); INSERT INTO tableB VALUES(1,202004),(2,202004),(3,202004),(4,202004), (5,202004); SELECT * FROM ( SELECT * FROM tableA WHERE RunYearMonth = 202004 ) AS A INNER JOIN ( SELECT * FROM tableB WHERE RunYearMonth = 202004 ) AS B ON A.ID = B.ID AND A.RunYearMonth = B.RunYearMonth; } {4 202004 4 202004 5 202004 5 202004} # 2023-02-10 https://sqlite.org/forum/forumpost/0a539c76db3b9e29 # The original constant propagation implementation caused a performance # regression. Because "abs(v)" was rewritten into "abs(1)" it no longer # matches the indexed column and the index is not used. # reset_db do_execsql_test 700 { CREATE TABLE t1(v INTEGER); WITH RECURSIVE c(x) AS (VALUES(-10) UNION ALL SELECT x+1 FROM c WHERE x<10) INSERT INTO t1(v) SELECT x FROM c; CREATE INDEX idx ON t1( abs(v) ); SELECT v FROM t1 WHERE abs(v)=1 and v=1; } 1 do_eqp_test 710 { SELECT v FROM t1 WHERE abs(v)=1 and v=1; } { QUERY PLAN `--SEARCH t1 USING INDEX idx (=?) } # 2024-03-07 https://sqlite.org/forum/forumpost/ecdfc02339 # A refinement is needed to the enhancements tested by the prior test case # to avoid another problem with indexes on constant expressions. # reset_db db null NULL do_execsql_test 800 { CREATE TABLE t0(c0, c1); CREATE TABLE t1(c2); CREATE INDEX i0 ON t1(NULL); INSERT INTO t1(c2) VALUES (0.2); CREATE VIEW v0(c3) AS SELECT DISTINCT c2 FROM t1; SELECT * FROM v0 LEFT JOIN t0 ON c3