sqlcipher/test/like.test

872 lines
21 KiB
Plaintext

# 2005 August 13
#
# 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 LIKE and GLOB operators and
# in particular the optimizations that occur to help those operators
# run faster.
#
# $Id: like.test,v 1.13 2009/06/07 23:45:11 drh Exp $
set testdir [file dirname $argv0]
source $testdir/tester.tcl
# Create some sample data to work with.
#
do_test like-1.0 {
execsql {
CREATE TABLE t1(x TEXT);
}
foreach str {
a
ab
abc
abcd
acd
abd
bc
bcd
xyz
ABC
CDE
{ABC abc xyz}
} {
db eval {INSERT INTO t1 VALUES(:str)}
}
execsql {
SELECT count(*) FROM t1;
}
} {12}
# Test that both case sensitive and insensitive version of LIKE work.
#
do_test like-1.1 {
execsql {
SELECT x FROM t1 WHERE x LIKE 'abc' ORDER BY 1;
}
} {ABC abc}
do_test like-1.2 {
execsql {
SELECT x FROM t1 WHERE x GLOB 'abc' ORDER BY 1;
}
} {abc}
do_test like-1.3 {
execsql {
SELECT x FROM t1 WHERE x LIKE 'ABC' ORDER BY 1;
}
} {ABC abc}
do_test like-1.4 {
execsql {
SELECT x FROM t1 WHERE x LIKE 'aBc' ORDER BY 1;
}
} {ABC abc}
do_test like-1.5.1 {
execsql {
PRAGMA case_sensitive_like=on;
SELECT x FROM t1 WHERE x LIKE 'abc' ORDER BY 1;
}
} {abc}
do_test like-1.5.2 {
execsql {
PRAGMA case_sensitive_like; -- no argument; does not change setting
SELECT x FROM t1 WHERE x LIKE 'abc' ORDER BY 1;
}
} {abc}
do_test like-1.6 {
execsql {
SELECT x FROM t1 WHERE x GLOB 'abc' ORDER BY 1;
}
} {abc}
do_test like-1.7 {
execsql {
SELECT x FROM t1 WHERE x LIKE 'ABC' ORDER BY 1;
}
} {ABC}
do_test like-1.8 {
execsql {
SELECT x FROM t1 WHERE x LIKE 'aBc' ORDER BY 1;
}
} {}
do_test like-1.9 {
execsql {
PRAGMA case_sensitive_like=off;
SELECT x FROM t1 WHERE x LIKE 'abc' ORDER BY 1;
}
} {ABC abc}
do_test like-1.10 {
execsql {
PRAGMA case_sensitive_like; -- No argument, does not change setting.
SELECT x FROM t1 WHERE x LIKE 'abc' ORDER BY 1;
}
} {ABC abc}
# Tests of the REGEXP operator
#
do_test like-2.1 {
proc test_regexp {a b} {
return [regexp $a $b]
}
db function regexp test_regexp
execsql {
SELECT x FROM t1 WHERE x REGEXP 'abc' ORDER BY 1;
}
} {{ABC abc xyz} abc abcd}
do_test like-2.2 {
execsql {
SELECT x FROM t1 WHERE x REGEXP '^abc' ORDER BY 1;
}
} {abc abcd}
# Tests of the MATCH operator
#
do_test like-2.3 {
proc test_match {a b} {
return [string match $a $b]
}
db function match -argcount 2 test_match
execsql {
SELECT x FROM t1 WHERE x MATCH '*abc*' ORDER BY 1;
}
} {{ABC abc xyz} abc abcd}
do_test like-2.4 {
execsql {
SELECT x FROM t1 WHERE x MATCH 'abc*' ORDER BY 1;
}
} {abc abcd}
# For the remaining tests, we need to have the like optimizations
# enabled.
#
ifcapable !like_opt {
finish_test
return
}
# This procedure executes the SQL. Then it appends to the result the
# "sort" or "nosort" keyword (as in the cksort procedure above) then
# it appends the ::sqlite_query_plan variable.
#
proc queryplan {sql} {
set ::sqlite_sort_count 0
set data [execsql $sql]
if {$::sqlite_sort_count} {set x sort} {set x nosort}
lappend data $x
return [concat $data $::sqlite_query_plan]
}
# Perform tests on the like optimization.
#
# With no index on t1.x and with case sensitivity turned off, no optimization
# is performed.
#
do_test like-3.1 {
set sqlite_like_count 0
queryplan {
SELECT x FROM t1 WHERE x LIKE 'abc%' ORDER BY 1;
}
} {ABC {ABC abc xyz} abc abcd sort t1 {}}
do_test like-3.2 {
set sqlite_like_count
} {12}
# With an index on t1.x and case sensitivity on, optimize completely.
#
do_test like-3.3 {
set sqlite_like_count 0
execsql {
PRAGMA case_sensitive_like=on;
CREATE INDEX i1 ON t1(x);
}
queryplan {
SELECT x FROM t1 WHERE x LIKE 'abc%' ORDER BY 1;
}
} {abc abcd nosort {} i1}
do_test like-3.4 {
set sqlite_like_count
} 0
# The LIKE optimization still works when the RHS is a string with no
# wildcard. Ticket [e090183531fc2747]
#
do_test like-3.4.2 {
queryplan {
SELECT x FROM t1 WHERE x LIKE 'a' ORDER BY 1;
}
} {a nosort {} i1}
do_test like-3.4.3 {
queryplan {
SELECT x FROM t1 WHERE x LIKE 'ab' ORDER BY 1;
}
} {ab nosort {} i1}
do_test like-3.4.4 {
queryplan {
SELECT x FROM t1 WHERE x LIKE 'abcd' ORDER BY 1;
}
} {abcd nosort {} i1}
do_test like-3.4.5 {
queryplan {
SELECT x FROM t1 WHERE x LIKE 'abcde' ORDER BY 1;
}
} {nosort {} i1}
# Partial optimization when the pattern does not end in '%'
#
do_test like-3.5 {
set sqlite_like_count 0
queryplan {
SELECT x FROM t1 WHERE x LIKE 'a_c' ORDER BY 1;
}
} {abc nosort {} i1}
do_test like-3.6 {
set sqlite_like_count
} 6
do_test like-3.7 {
set sqlite_like_count 0
queryplan {
SELECT x FROM t1 WHERE x LIKE 'ab%d' ORDER BY 1;
}
} {abcd abd nosort {} i1}
do_test like-3.8 {
set sqlite_like_count
} 4
do_test like-3.9 {
set sqlite_like_count 0
queryplan {
SELECT x FROM t1 WHERE x LIKE 'a_c%' ORDER BY 1;
}
} {abc abcd nosort {} i1}
do_test like-3.10 {
set sqlite_like_count
} 6
# No optimization when the pattern begins with a wildcard.
# Note that the index is still used but only for sorting.
#
do_test like-3.11 {
set sqlite_like_count 0
queryplan {
SELECT x FROM t1 WHERE x LIKE '%bcd' ORDER BY 1;
}
} {abcd bcd nosort {} i1}
do_test like-3.12 {
set sqlite_like_count
} 12
# No optimization for case insensitive LIKE
#
do_test like-3.13 {
set sqlite_like_count 0
queryplan {
PRAGMA case_sensitive_like=off;
SELECT x FROM t1 WHERE x LIKE 'abc%' ORDER BY 1;
}
} {ABC {ABC abc xyz} abc abcd nosort {} i1}
do_test like-3.14 {
set sqlite_like_count
} 12
# No optimization without an index.
#
do_test like-3.15 {
set sqlite_like_count 0
queryplan {
PRAGMA case_sensitive_like=on;
DROP INDEX i1;
SELECT x FROM t1 WHERE x LIKE 'abc%' ORDER BY 1;
}
} {abc abcd sort t1 {}}
do_test like-3.16 {
set sqlite_like_count
} 12
# No GLOB optimization without an index.
#
do_test like-3.17 {
set sqlite_like_count 0
queryplan {
SELECT x FROM t1 WHERE x GLOB 'abc*' ORDER BY 1;
}
} {abc abcd sort t1 {}}
do_test like-3.18 {
set sqlite_like_count
} 12
# GLOB is optimized regardless of the case_sensitive_like setting.
#
do_test like-3.19 {
set sqlite_like_count 0
queryplan {
CREATE INDEX i1 ON t1(x);
SELECT x FROM t1 WHERE x GLOB 'abc*' ORDER BY 1;
}
} {abc abcd nosort {} i1}
do_test like-3.20 {
set sqlite_like_count
} 0
do_test like-3.21 {
set sqlite_like_count 0
queryplan {
PRAGMA case_sensitive_like=on;
SELECT x FROM t1 WHERE x GLOB 'abc*' ORDER BY 1;
}
} {abc abcd nosort {} i1}
do_test like-3.22 {
set sqlite_like_count
} 0
do_test like-3.23 {
set sqlite_like_count 0
queryplan {
PRAGMA case_sensitive_like=off;
SELECT x FROM t1 WHERE x GLOB 'a[bc]d' ORDER BY 1;
}
} {abd acd nosort {} i1}
do_test like-3.24 {
set sqlite_like_count
} 6
# GLOB optimization when there is no wildcard. Ticket [e090183531fc2747]
#
do_test like-3.25 {
queryplan {
SELECT x FROM t1 WHERE x GLOB 'a' ORDER BY 1;
}
} {a nosort {} i1}
do_test like-3.26 {
queryplan {
SELECT x FROM t1 WHERE x GLOB 'abcd' ORDER BY 1;
}
} {abcd nosort {} i1}
do_test like-3.27 {
queryplan {
SELECT x FROM t1 WHERE x GLOB 'abcde' ORDER BY 1;
}
} {nosort {} i1}
# No optimization if the LHS of the LIKE is not a column name or
# if the RHS is not a string.
#
do_test like-4.1 {
execsql {PRAGMA case_sensitive_like=on}
set sqlite_like_count 0
queryplan {
SELECT x FROM t1 WHERE x LIKE 'abc%' ORDER BY 1
}
} {abc abcd nosort {} i1}
do_test like-4.2 {
set sqlite_like_count
} 0
do_test like-4.3 {
set sqlite_like_count 0
queryplan {
SELECT x FROM t1 WHERE +x LIKE 'abc%' ORDER BY 1
}
} {abc abcd nosort {} i1}
do_test like-4.4 {
set sqlite_like_count
} 12
do_test like-4.5 {
set sqlite_like_count 0
queryplan {
SELECT x FROM t1 WHERE x LIKE ('ab' || 'c%') ORDER BY 1
}
} {abc abcd nosort {} i1}
do_test like-4.6 {
set sqlite_like_count
} 12
# Collating sequences on the index disable the LIKE optimization.
# Or if the NOCASE collating sequence is used, the LIKE optimization
# is enabled when case_sensitive_like is OFF.
#
do_test like-5.1 {
execsql {PRAGMA case_sensitive_like=off}
set sqlite_like_count 0
queryplan {
SELECT x FROM t1 WHERE x LIKE 'abc%' ORDER BY 1
}
} {ABC {ABC abc xyz} abc abcd nosort {} i1}
do_test like-5.2 {
set sqlite_like_count
} 12
do_test like-5.3 {
execsql {
CREATE TABLE t2(x TEXT COLLATE NOCASE);
INSERT INTO t2 SELECT * FROM t1;
CREATE INDEX i2 ON t2(x COLLATE NOCASE);
}
set sqlite_like_count 0
queryplan {
SELECT x FROM t2 WHERE x LIKE 'abc%' ORDER BY 1
}
} {abc ABC {ABC abc xyz} abcd nosort {} i2}
do_test like-5.4 {
set sqlite_like_count
} 0
do_test like-5.5 {
execsql {
PRAGMA case_sensitive_like=on;
}
set sqlite_like_count 0
queryplan {
SELECT x FROM t2 WHERE x LIKE 'abc%' ORDER BY 1
}
} {abc abcd nosort {} i2}
do_test like-5.6 {
set sqlite_like_count
} 12
do_test like-5.7 {
execsql {
PRAGMA case_sensitive_like=off;
}
set sqlite_like_count 0
queryplan {
SELECT x FROM t2 WHERE x GLOB 'abc*' ORDER BY 1
}
} {abc abcd nosort {} i2}
do_test like-5.8 {
set sqlite_like_count
} 12
do_test like-5.11 {
execsql {PRAGMA case_sensitive_like=off}
set sqlite_like_count 0
queryplan {
SELECT x FROM t1 WHERE x LIKE 'ABC%' ORDER BY 1
}
} {ABC {ABC abc xyz} abc abcd nosort {} i1}
do_test like-5.12 {
set sqlite_like_count
} 12
do_test like-5.13 {
set sqlite_like_count 0
queryplan {
SELECT x FROM t2 WHERE x LIKE 'ABC%' ORDER BY 1
}
} {abc ABC {ABC abc xyz} abcd nosort {} i2}
do_test like-5.14 {
set sqlite_like_count
} 0
do_test like-5.15 {
execsql {
PRAGMA case_sensitive_like=on;
}
set sqlite_like_count 0
queryplan {
SELECT x FROM t2 WHERE x LIKE 'ABC%' ORDER BY 1
}
} {ABC {ABC abc xyz} nosort {} i2}
do_test like-5.16 {
set sqlite_like_count
} 12
do_test like-5.17 {
execsql {
PRAGMA case_sensitive_like=off;
}
set sqlite_like_count 0
queryplan {
SELECT x FROM t2 WHERE x GLOB 'ABC*' ORDER BY 1
}
} {ABC {ABC abc xyz} nosort {} i2}
do_test like-5.18 {
set sqlite_like_count
} 12
# Boundary case. The prefix for a LIKE comparison is rounded up
# when constructing the comparison. Example: "ab" becomes "ac".
# In other words, the last character is increased by one.
#
# Make sure this happens correctly when the last character is a
# "z" and we are doing case-insensitive comparisons.
#
# Ticket #2959
#
do_test like-5.21 {
execsql {
PRAGMA case_sensitive_like=off;
INSERT INTO t2 VALUES('ZZ-upper-upper');
INSERT INTO t2 VALUES('zZ-lower-upper');
INSERT INTO t2 VALUES('Zz-upper-lower');
INSERT INTO t2 VALUES('zz-lower-lower');
}
queryplan {
SELECT x FROM t2 WHERE x LIKE 'zz%';
}
} {zz-lower-lower zZ-lower-upper Zz-upper-lower ZZ-upper-upper nosort {} i2}
do_test like-5.22 {
queryplan {
SELECT x FROM t2 WHERE x LIKE 'zZ%';
}
} {zz-lower-lower zZ-lower-upper Zz-upper-lower ZZ-upper-upper nosort {} i2}
do_test like-5.23 {
queryplan {
SELECT x FROM t2 WHERE x LIKE 'Zz%';
}
} {zz-lower-lower zZ-lower-upper Zz-upper-lower ZZ-upper-upper nosort {} i2}
do_test like-5.24 {
queryplan {
SELECT x FROM t2 WHERE x LIKE 'ZZ%';
}
} {zz-lower-lower zZ-lower-upper Zz-upper-lower ZZ-upper-upper nosort {} i2}
do_test like-5.25 {
queryplan {
PRAGMA case_sensitive_like=on;
CREATE TABLE t3(x TEXT);
CREATE INDEX i3 ON t3(x);
INSERT INTO t3 VALUES('ZZ-upper-upper');
INSERT INTO t3 VALUES('zZ-lower-upper');
INSERT INTO t3 VALUES('Zz-upper-lower');
INSERT INTO t3 VALUES('zz-lower-lower');
SELECT x FROM t3 WHERE x LIKE 'zz%';
}
} {zz-lower-lower nosort {} i3}
do_test like-5.26 {
queryplan {
SELECT x FROM t3 WHERE x LIKE 'zZ%';
}
} {zZ-lower-upper nosort {} i3}
do_test like-5.27 {
queryplan {
SELECT x FROM t3 WHERE x LIKE 'Zz%';
}
} {Zz-upper-lower nosort {} i3}
do_test like-5.28 {
queryplan {
SELECT x FROM t3 WHERE x LIKE 'ZZ%';
}
} {ZZ-upper-upper nosort {} i3}
# ticket #2407
#
# Make sure the LIKE prefix optimization does not strip off leading
# characters of the like pattern that happen to be quote characters.
#
do_test like-6.1 {
foreach x { 'abc 'bcd 'def 'ax } {
set x2 '[string map {' ''} $x]'
db eval "INSERT INTO t2 VALUES($x2)"
}
execsql {
SELECT * FROM t2 WHERE x LIKE '''a%'
}
} {'abc 'ax}
do_test like-7.1 {
execsql {
SELECT rowid, * FROM t1 WHERE rowid GLOB '1*' ORDER BY rowid;
}
} {1 a 10 ABC 11 CDE 12 {ABC abc xyz}}
# ticket #3345.
#
# Overloading the LIKE function with -1 for the number of arguments
# will overload both the 2-argument and the 3-argument LIKE.
#
do_test like-8.1 {
db eval {
CREATE TABLE t8(x);
INSERT INTO t8 VALUES('abcdef');
INSERT INTO t8 VALUES('ghijkl');
INSERT INTO t8 VALUES('mnopqr');
SELECT 1, x FROM t8 WHERE x LIKE '%h%';
SELECT 2, x FROM t8 WHERE x LIKE '%h%' ESCAPE 'x';
}
} {1 ghijkl 2 ghijkl}
do_test like-8.2 {
proc newlike {args} {return 1} ;# Alternative LIKE is always return TRUE
db function like newlike ;# Uses -1 for nArg in sqlite3_create_function
db cache flush
db eval {
SELECT 1, x FROM t8 WHERE x LIKE '%h%';
SELECT 2, x FROM t8 WHERE x LIKE '%h%' ESCAPE 'x';
}
} {1 ghijkl 2 ghijkl}
do_test like-8.3 {
db function like -argcount 2 newlike
db eval {
SELECT 1, x FROM t8 WHERE x LIKE '%h%';
SELECT 2, x FROM t8 WHERE x LIKE '%h%' ESCAPE 'x';
}
} {1 abcdef 1 ghijkl 1 mnopqr 2 ghijkl}
do_test like-8.4 {
db function like -argcount 3 newlike
db eval {
SELECT 1, x FROM t8 WHERE x LIKE '%h%';
SELECT 2, x FROM t8 WHERE x LIKE '%h%' ESCAPE 'x';
}
} {1 abcdef 1 ghijkl 1 mnopqr 2 abcdef 2 ghijkl 2 mnopqr}
ifcapable like_opt {
# Evaluate SQL. Return the result set followed by the
# and the number of full-scan steps.
#
db close
sqlite3 db test.db
proc count_steps {sql} {
set r [db eval $sql]
lappend r scan [db status step] sort [db status sort]
}
do_test like-9.1 {
count_steps {
SELECT x FROM t2 WHERE x LIKE 'x%'
}
} {xyz scan 0 sort 0}
do_test like-9.2 {
count_steps {
SELECT x FROM t2 WHERE x LIKE '_y%'
}
} {xyz scan 19 sort 0}
do_test like-9.3.1 {
set res [sqlite3_exec_hex db {
SELECT x FROM t2 WHERE x LIKE '%78%25'
}]
} {0 {x xyz}}
ifcapable explain {
do_test like-9.3.2 {
set res [sqlite3_exec_hex db {
EXPLAIN QUERY PLAN SELECT x FROM t2 WHERE x LIKE '%78%25'
}]
regexp {INDEX i2} $res
} {1}
}
do_test like-9.4.1 {
sqlite3_exec_hex db {INSERT INTO t2 VALUES('%ffhello')}
set res [sqlite3_exec_hex db {
SELECT substr(x,2) AS x FROM t2 WHERE +x LIKE '%ff%25'
}]
} {0 {x hello}}
do_test like-9.4.2 {
set res [sqlite3_exec_hex db {
SELECT substr(x,2) AS x FROM t2 WHERE x LIKE '%ff%25'
}]
} {0 {x hello}}
ifcapable explain {
do_test like-9.4.3 {
set res [sqlite3_exec_hex db {
EXPLAIN QUERY PLAN SELECT x FROM t2 WHERE x LIKE '%ff%25'
}]
regexp {INDEX i2} $res
} {0}
}
do_test like-9.5.1 {
set res [sqlite3_exec_hex db {
SELECT x FROM t2 WHERE x LIKE '%fe%25'
}]
} {0 {}}
ifcapable explain {
do_test like-9.5.2 {
set res [sqlite3_exec_hex db {
EXPLAIN QUERY PLAN SELECT x FROM t2 WHERE x LIKE '%fe%25'
}]
regexp {INDEX i2} $res
} {1}
}
}
# Do an SQL statement. Append the search count to the end of the result.
#
proc count sql {
set ::sqlite_search_count 0
set ::sqlite_like_count 0
return [concat [execsql $sql] scan $::sqlite_search_count \
like $::sqlite_like_count]
}
# The LIKE and GLOB optimizations do not work on columns with
# affinity other than TEXT.
# Ticket #3901
#
do_test like-10.1 {
db close
sqlite3 db test.db
execsql {
CREATE TABLE t10(
a INTEGER PRIMARY KEY,
b INTEGER COLLATE nocase UNIQUE,
c NUMBER COLLATE nocase UNIQUE,
d BLOB COLLATE nocase UNIQUE,
e COLLATE nocase UNIQUE,
f TEXT COLLATE nocase UNIQUE
);
INSERT INTO t10 VALUES(1,1,1,1,1,1);
INSERT INTO t10 VALUES(12,12,12,12,12,12);
INSERT INTO t10 VALUES(123,123,123,123,123,123);
INSERT INTO t10 VALUES(234,234,234,234,234,234);
INSERT INTO t10 VALUES(345,345,345,345,345,345);
INSERT INTO t10 VALUES(45,45,45,45,45,45);
}
count {
SELECT a FROM t10 WHERE b LIKE '12%' ORDER BY a;
}
} {12 123 scan 5 like 6}
do_test like-10.2 {
count {
SELECT a FROM t10 WHERE c LIKE '12%' ORDER BY a;
}
} {12 123 scan 5 like 6}
do_test like-10.3 {
count {
SELECT a FROM t10 WHERE d LIKE '12%' ORDER BY a;
}
} {12 123 scan 5 like 6}
do_test like-10.4 {
count {
SELECT a FROM t10 WHERE e LIKE '12%' ORDER BY a;
}
} {12 123 scan 5 like 6}
do_test like-10.5 {
count {
SELECT a FROM t10 WHERE f LIKE '12%' ORDER BY a;
}
} {12 123 scan 3 like 0}
do_test like-10.6 {
count {
SELECT a FROM t10 WHERE a LIKE '12%' ORDER BY a;
}
} {12 123 scan 5 like 6}
do_test like-10.10 {
execsql {
CREATE TABLE t10b(
a INTEGER PRIMARY KEY,
b INTEGER UNIQUE,
c NUMBER UNIQUE,
d BLOB UNIQUE,
e UNIQUE,
f TEXT UNIQUE
);
INSERT INTO t10b SELECT * FROM t10;
}
count {
SELECT a FROM t10b WHERE b GLOB '12*' ORDER BY a;
}
} {12 123 scan 5 like 6}
do_test like-10.11 {
count {
SELECT a FROM t10b WHERE c GLOB '12*' ORDER BY a;
}
} {12 123 scan 5 like 6}
do_test like-10.12 {
count {
SELECT a FROM t10b WHERE d GLOB '12*' ORDER BY a;
}
} {12 123 scan 5 like 6}
do_test like-10.13 {
count {
SELECT a FROM t10b WHERE e GLOB '12*' ORDER BY a;
}
} {12 123 scan 5 like 6}
do_test like-10.14 {
count {
SELECT a FROM t10b WHERE f GLOB '12*' ORDER BY a;
}
} {12 123 scan 3 like 0}
do_test like-10.15 {
count {
SELECT a FROM t10b WHERE a GLOB '12*' ORDER BY a;
}
} {12 123 scan 5 like 6}
# LIKE and GLOB where the default collating sequence is not appropriate
# but an index with the appropriate collating sequence exists.
#
do_test like-11.0 {
execsql {
CREATE TABLE t11(
a INTEGER PRIMARY KEY,
b TEXT COLLATE nocase,
c TEXT COLLATE binary
);
INSERT INTO t11 VALUES(1, 'a','a');
INSERT INTO t11 VALUES(2, 'ab','ab');
INSERT INTO t11 VALUES(3, 'abc','abc');
INSERT INTO t11 VALUES(4, 'abcd','abcd');
INSERT INTO t11 VALUES(5, 'A','A');
INSERT INTO t11 VALUES(6, 'AB','AB');
INSERT INTO t11 VALUES(7, 'ABC','ABC');
INSERT INTO t11 VALUES(8, 'ABCD','ABCD');
INSERT INTO t11 VALUES(9, 'x','x');
INSERT INTO t11 VALUES(10, 'yz','yz');
INSERT INTO t11 VALUES(11, 'X','X');
INSERT INTO t11 VALUES(12, 'YZ','YZ');
SELECT count(*) FROM t11;
}
} {12}
do_test like-11.1 {
queryplan {
PRAGMA case_sensitive_like=OFF;
SELECT b FROM t11 WHERE b LIKE 'abc%' ORDER BY a;
}
} {abc abcd ABC ABCD nosort t11 *}
do_test like-11.2 {
queryplan {
PRAGMA case_sensitive_like=ON;
SELECT b FROM t11 WHERE b LIKE 'abc%' ORDER BY a;
}
} {abc abcd nosort t11 *}
do_test like-11.3 {
queryplan {
PRAGMA case_sensitive_like=OFF;
CREATE INDEX t11b ON t11(b);
SELECT b FROM t11 WHERE b LIKE 'abc%' ORDER BY a;
}
} {abc abcd ABC ABCD sort {} t11b}
do_test like-11.4 {
queryplan {
PRAGMA case_sensitive_like=ON;
SELECT b FROM t11 WHERE b LIKE 'abc%' ORDER BY a;
}
} {abc abcd nosort t11 *}
do_test like-11.5 {
queryplan {
PRAGMA case_sensitive_like=OFF;
DROP INDEX t11b;
CREATE INDEX t11bnc ON t11(b COLLATE nocase);
SELECT b FROM t11 WHERE b LIKE 'abc%' ORDER BY a;
}
} {abc abcd ABC ABCD sort {} t11bnc}
do_test like-11.6 {
queryplan {
CREATE INDEX t11bb ON t11(b COLLATE binary);
SELECT b FROM t11 WHERE b LIKE 'abc%' ORDER BY a;
}
} {abc abcd ABC ABCD sort {} t11bnc}
do_test like-11.7 {
queryplan {
PRAGMA case_sensitive_like=ON;
SELECT b FROM t11 WHERE b LIKE 'abc%' ORDER BY a;
}
} {abc abcd sort {} t11bb}
do_test like-11.8 {
queryplan {
PRAGMA case_sensitive_like=OFF;
SELECT b FROM t11 WHERE b GLOB 'abc*' ORDER BY a;
}
} {abc abcd sort {} t11bb}
do_test like-11.9 {
queryplan {
CREATE INDEX t11cnc ON t11(c COLLATE nocase);
CREATE INDEX t11cb ON t11(c COLLATE binary);
SELECT c FROM t11 WHERE c LIKE 'abc%' ORDER BY a;
}
} {abc abcd ABC ABCD sort {} t11cnc}
do_test like-11.10 {
queryplan {
SELECT c FROM t11 WHERE c GLOB 'abc*' ORDER BY a;
}
} {abc abcd sort {} t11cb}
finish_test