path: root/mysql-test/t/ctype_ucs.test

-- source include/have_ucs2.inc

--disable_warnings
DROP TABLE IF EXISTS t1;
--enable_warnings

SET NAMES latin1;
SET character_set_connection=ucs2;
-- source include/endspace.inc

SET CHARACTER SET koi8r;

#
# Check that 0x20 is only trimmed when it is 
# a part of real SPACE character, not just a part
# of a multibyte sequence.
# Note, CYRILLIC LETTER ER is used as an example, which
# is stored as 0x0420 in UCS2, thus contains 0x20 in the
# low byte. The second character is THREE-PER-M, U+2004,
# which contains 0x20 in the high byte.
#

CREATE TABLE t1 (word VARCHAR(64) CHARACTER SET ucs2, word2 CHAR(64) CHARACTER SET ucs2);
INSERT INTO t1 VALUES (_koi8r'ò',_koi8r'ò'), (X'2004',X'2004');
SELECT hex(word) FROM t1 ORDER BY word;
SELECT hex(word2) FROM t1 ORDER BY word2;
DELETE FROM t1;

#
# Check that real spaces are correctly trimmed.
#

INSERT INTO t1 VALUES (X'042000200020',X'042000200020'), (X'200400200020', X'200400200020');
SELECT hex(word) FROM t1 ORDER BY word;
SELECT hex(word2) FROM t1 ORDER BY word2;
DROP TABLE t1;

#
# Check LPAD/RPAD
#
SELECT LPAD(_ucs2 X'0420',10,_ucs2 X'0421');
SELECT LPAD(_ucs2 X'0420',10,_ucs2 X'04210422');
SELECT LPAD(_ucs2 X'0420',10,_ucs2 X'042104220423');
SELECT LPAD(_ucs2 X'0420042104220423042404250426042704280429042A042B',10,_ucs2 X'042104220423');

SELECT RPAD(_ucs2 X'0420',10,_ucs2 X'0421');
SELECT RPAD(_ucs2 X'0420',10,_ucs2 X'04210422');
SELECT RPAD(_ucs2 X'0420',10,_ucs2 X'042104220423');
SELECT RPAD(_ucs2 X'0420042104220423042404250426042704280429042A042B',10,_ucs2 X'042104220423');

CREATE TABLE t1 SELECT 
LPAD(_ucs2 X'0420',10,_ucs2 X'0421') l,
RPAD(_ucs2 X'0420',10,_ucs2 X'0421') r;
SHOW CREATE TABLE t1;
DROP TABLE t1;

#
# BUG3946
#

create table t2(f1 Char(30));
insert into t2 values ("103000"), ("22720000"), ("3401200"), ("78000");
select lpad(f1, 12, "-o-/") from t2;
drop table t2;

######################################################
#
# Test of like
#

SET NAMES koi8r;
SET character_set_connection=ucs2;

create table t1 (a varchar(10) character set ucs2, key(a));
insert into t1 values ("a"),("abc"),("abcd"),("hello"),("test");
explain select * from t1 where a like 'abc%';
explain select * from t1 where a like concat('abc','%');
select * from t1 where a like "abc%";
select * from t1 where a like concat("abc","%");
select * from t1 where a like "ABC%";
select * from t1 where a like "test%";
select * from t1 where a like "te_t";
select * from t1 where a like "%a%";
select * from t1 where a like "%abcd%";
select * from t1 where a like "%abc\d%";
drop table t1;

#
# More LIKE test: bug#2619
#
select 'AA' like 'AA'; 
select 'AA' like 'A%A';
select 'AA' like 'A%%A';
select 'AA' like 'AA%';
select 'AA' like '%AA%';
select 'AA' like '%A';
select 'AA' like '%AA';
select 'AA' like 'A%A%';
select 'AA' like '_%_%';
select 'AA' like '%A%A';
select 'AAA'like 'A%A%A';

select 'AZ' like 'AZ'; 
select 'AZ' like 'A%Z';
select 'AZ' like 'A%%Z';
select 'AZ' like 'AZ%';
select 'AZ' like '%AZ%';
select 'AZ' like '%Z';
select 'AZ' like '%AZ';
select 'AZ' like 'A%Z%';
select 'AZ' like '_%_%';
select 'AZ' like '%A%Z';
select 'AZ' like 'A_';
select 'AZ' like '_Z';
select 'AMZ'like 'A%M%Z';

CREATE TABLE t1 (a VARCHAR(10) CHARACTER SET ucs2);
INSERT INTO t1 VALUES ('ÆÙ×Á'),('æÙ×Á'),('Æù×Á'),('ÆÙ÷Á'),('ÆÙ×á'),('æù÷á');
INSERT INTO t1 VALUES ('ÆÙ×ÁÐÒÏÌÄÖ'),('æÙ×ÁÐÒÏÌÄÖ'),('Æù×ÁÐÒÏÌÄÖ'),('ÆÙ÷ÁÐÒÏÌÄÖ');
INSERT INTO t1 VALUES ('ÆÙ×áÐÒÏÌÄÖ'),('ÆÙ×ÁðÒÏÌÄÖ'),('ÆÙ×ÁÐòÏÌÄÖ'),('ÆÙ×ÁÐÒïÌÄÖ');
INSERT INTO t1 VALUES ('ÆÙ×ÁÐÒÏìÄÖ'),('ÆÙ×ÁÐÒÏÌäÖ'),('ÆÙ×ÁÐÒÏÌÄö'),('æù÷áðòïìäö');
SELECT * FROM t1 WHERE a LIKE '%Æù×Á%';
SELECT * FROM t1 WHERE a LIKE '%Æù×%';
SELECT * FROM t1 WHERE a LIKE 'Æù×Á%';
SELECT * FROM t1 WHERE a LIKE 'Æù×Á%' COLLATE ucs2_bin;
DROP TABLE t1;

#
# Bug 1181
#
CREATE TABLE t1 (word varchar(64) NOT NULL, PRIMARY KEY (word))
ENGINE=MyISAM CHARACTER SET ucs2 COLLATE ucs2_general_ci;
INSERT INTO t1 (word) VALUES ("cat");
SELECT * FROM t1 WHERE word LIKE "c%";
SELECT * FROM t1 WHERE word LIKE "ca_";
SELECT * FROM t1 WHERE word LIKE "cat";
SELECT * FROM t1 WHERE word LIKE _ucs2 x'00630025';
SELECT * FROM t1 WHERE word LIKE _ucs2 x'00630061005F';
DROP TABLE t1;

#
# Check that INSERT works fine. 
# This invokes charpos() function.
select insert(_ucs2 0x006100620063,10,2,_ucs2 0x006400650066);
select insert(_ucs2 0x006100620063,1,2,_ucs2 0x006400650066);

######################################################

#
# Bug 1264
#
# Description: 
#
# When using a ucs2 table in MySQL, 
# either with ucs2_general_ci or ucs2_bin collation,
# words are returned in an incorrect order when using ORDER BY
# on an _indexed_ CHAR or VARCHAR column. They are sorted with
# the longest word *first* instead of last. I.E. The word "aardvark"
# is in the results before the word "a".
#
# If there is no index for the column, the problem does not occur.
#
# Interestingly, if there is no second column, the words are returned
# in the correct order. 
#
# According to EXPLAIN, it looks like when the output includes columns that
# are not part of the index sorted on, it does a filesort, which fails. 
# Using a straight index yields correct results.

SET NAMES latin1;

#
# Two fields, index
#

CREATE TABLE t1 (
   word VARCHAR(64),
   bar INT(11) default 0,
   PRIMARY KEY (word))
   ENGINE=MyISAM
   CHARSET ucs2
   COLLATE ucs2_general_ci ;

INSERT INTO t1 (word) VALUES ("aar");
INSERT INTO t1 (word) VALUES ("a");
INSERT INTO t1 (word) VALUES ("aardvar");
INSERT INTO t1 (word) VALUES ("aardvark");
INSERT INTO t1 (word) VALUES ("aardvara");
INSERT INTO t1 (word) VALUES ("aardvarz");
EXPLAIN SELECT * FROM t1 ORDER BY word;
SELECT * FROM t1 ORDER BY word;
EXPLAIN SELECT word FROM t1 ORDER BY word;
SELECT word FROM t1 ORDER by word;
DROP TABLE t1;

#
# One field, index
# 

CREATE TABLE t1 (
   word VARCHAR(64) ,
   PRIMARY KEY (word))
   ENGINE=MyISAM
   CHARSET ucs2
   COLLATE ucs2_general_ci;

INSERT INTO t1 (word) VALUES ("aar");
INSERT INTO t1 (word) VALUES ("a");
INSERT INTO t1 (word) VALUES ("aardvar");
INSERT INTO t1 (word) VALUES ("aardvark");
INSERT INTO t1 (word) VALUES ("aardvara");
INSERT INTO t1 (word) VALUES ("aardvarz");
EXPLAIN SELECT * FROM t1 ORDER BY WORD;
SELECT * FROM t1 ORDER BY word;
DROP TABLE t1;


#
# Two fields, no index
#

CREATE TABLE t1 (
   word TEXT,
   bar INT(11) AUTO_INCREMENT,
   PRIMARY KEY (bar))
   ENGINE=MyISAM
   CHARSET ucs2
   COLLATE ucs2_general_ci ;
INSERT INTO t1 (word) VALUES ("aar");
INSERT INTO t1 (word) VALUES ("a" );
INSERT INTO t1 (word) VALUES ("aardvar");
INSERT INTO t1 (word) VALUES ("aardvark");
INSERT INTO t1 (word) VALUES ("aardvara");
INSERT INTO t1 (word) VALUES ("aardvarz");
EXPLAIN SELECT * FROM t1 ORDER BY word;
SELECT * FROM t1 ORDER BY word;
EXPLAIN SELECT word FROM t1 ORDER BY word;
SELECT word FROM t1 ORDER BY word;
DROP TABLE t1;

#
# END OF Bug 1264 test
#
########################################################


# Bug #2390
# Check alignment for constants
#
SELECT HEX(_ucs2 0x0);
SELECT HEX(_ucs2 0x01);
SELECT HEX(_ucs2 0x012);
SELECT HEX(_ucs2 0x0123);
SELECT HEX(_ucs2 0x01234);
SELECT HEX(_ucs2 0x012345);
SELECT HEX(_ucs2 0x0123456);
SELECT HEX(_ucs2 0x01234567);
SELECT HEX(_ucs2 0x012345678);
SELECT HEX(_ucs2 0x0123456789);
SELECT HEX(_ucs2 0x0123456789A);
SELECT HEX(_ucs2 0x0123456789AB);
SELECT HEX(_ucs2 0x0123456789ABC);
SELECT HEX(_ucs2 0x0123456789ABCD);
SELECT HEX(_ucs2 0x0123456789ABCDE);
SELECT HEX(_ucs2 0x0123456789ABCDEF);

#
# Check alignment for from-binary-conversion with CAST and CONVERT
#
SELECT hex(cast(0xAA as char character set ucs2));
SELECT hex(convert(0xAA using ucs2));

#
# Check alignment for string types
#
CREATE TABLE t1 (a char(10) character set ucs2);
INSERT INTO t1 VALUES (0xA),(0xAA),(0xAAA),(0xAAAA),(0xAAAAA);
SELECT HEX(a) FROM t1;
DROP TABLE t1;

CREATE TABLE t1 (a varchar(10) character set ucs2);
INSERT INTO t1 VALUES (0xA),(0xAA),(0xAAA),(0xAAAA),(0xAAAAA);
SELECT HEX(a) FROM t1;
DROP TABLE t1;

CREATE TABLE t1 (a text character set ucs2);
INSERT INTO t1 VALUES (0xA),(0xAA),(0xAAA),(0xAAAA),(0xAAAAA);
SELECT HEX(a) FROM t1;
DROP TABLE t1;

CREATE TABLE t1 (a mediumtext character set ucs2);
INSERT INTO t1 VALUES (0xA),(0xAA),(0xAAA),(0xAAAA),(0xAAAAA);
SELECT HEX(a) FROM t1;
DROP TABLE t1;

CREATE TABLE t1 (a longtext character set ucs2);
INSERT INTO t1 VALUES (0xA),(0xAA),(0xAAA),(0xAAAA),(0xAAAAA);
SELECT HEX(a) FROM t1;
DROP TABLE t1;

-- the same should be also done with enum and set


#
# Bug #5024 Server crashes with queries on fields
#  with certain charset/collation settings
#

create table t1 (s1 char character set `ucs2` collate `ucs2_czech_ci`);
insert into t1 values ('0'),('1'),('2'),('a'),('b'),('c');
select s1 from t1 where s1 > 'a' order by s1;
drop table t1;

#
# Bug #5081 : UCS2 fields are filled with '0x2020'
# after extending field length
#
create table t1(a char(1)) default charset = ucs2;
insert into t1 values ('a'),('b'),('c');
alter table t1 modify a char(5);
select a, hex(a) from t1;
drop table t1;

#
# Check prepare statement from an UCS2 string
#
set @ivar= 1234;
set @str1 = 'select ?';
set @str2 = convert(@str1 using ucs2);
prepare stmt1 from @str2;
execute stmt1 using @ivar;

#
# Check correct binlogging of UCS2 user variables (BUG#3875)
#
SET TIMESTAMP=10000;
create table t2 (c char(30)) charset=ucs2;
set @v=convert('abc' using ucs2);
reset master;
insert into t2 values (@v);
show binlog events from 96;
# more important than SHOW BINLOG EVENTS, mysqlbinlog (where we
# absolutely need variables names to be quoted and strings to be
# escaped).
--replace_result $MYSQL_TEST_DIR MYSQL_TEST_DIR
--exec $MYSQL_BINLOG --short-form $MYSQL_TEST_DIR/var/log/master-bin.000001 
drop table t2;


#
# Check that ucs2 works with ENUM and SET type
#
set names latin1;
create table t1 (a enum('x','y','z') character set ucs2);
show create table t1;
insert into t1 values ('x');
insert into t1 values ('y');
insert into t1 values ('z');
select a, hex(a) from t1 order by a;
alter table t1 change a a enum('x','y','z','d','e','ä','ö','ü') character set ucs2;
show create table t1;
insert into t1 values ('D');
insert into t1 values ('E ');
insert into t1 values ('Ä');
insert into t1 values ('Ö');
insert into t1 values ('Ü');
select a, hex(a) from t1 order by a;
drop table t1;

create table t1 (a set ('x','y','z','ä','ö','ü') character set ucs2);
show create table t1;
insert into t1 values ('x');
insert into t1 values ('y');
insert into t1 values ('z');
insert into t1 values ('x,y');
insert into t1 values ('x,y,z,Ä,Ö,Ü');
select a, hex(a) from t1 order by a;
drop table t1;

#
# Bug#7302 UCS2 data in ENUM fields get truncated when new column is added
#
create table t1(a enum('a','b','c')) default character set ucs2;
insert into t1 values('a'),('b'),('c');
alter table t1 add b char(1);
show warnings;
select * from t1 order by a;
drop table t1;

SET collation_connection='ucs2_general_ci';
-- source include/ctype_filesort.inc
SET NAMES latin1;
SET collation_connection='ucs2_bin';
-- source include/ctype_filesort.inc

SET NAMES latin1;
#
# Bug#8235
#
# This bug also helped to find another problem that
# INSERT of a UCS2 string containing a negative number
# into a unsigned int column didn't produce warnings.
# This test covers both problems.
#
SET collation_connection='ucs2_swedish_ci';
CREATE TABLE t1 (Field1 int(10) default '0');
# no warnings, negative numbers are allowed
INSERT INTO t1 VALUES ('-1');
SELECT * FROM t1;
DROP TABLE t1;
CREATE TABLE t1 (Field1 int(10) unsigned default '0');
# this should generate a "Data truncated" warning
INSERT INTO t1 VALUES ('-1');
DROP TABLE t1;
SET NAMES latin1;