#At file:///home/bar/mysql-bzr/mysql-trunk-bugfixing.w4616v2/ based on revid:bar@strippedwfwi4n9u
3409 Alexander Barkov 2010-12-03
WL#4616 UTF16-LE
added:
@ mysql-test/include/ctype_heap.inc
@ mysql-test/r/ctype_utf16le.result
@ mysql-test/t/ctype_utf16le.test
Adding rests
modified:
@ cmake/character_sets.cmake
Adding utf16le into the list of the possible default character sets
and into the list of complex character sets.
@ include/m_ctype.h
Adding prototype for global variables
@ mysql-test/r/ctype_utf16.result
Conversion from binary to utf16 did not work well.
Recording new results
@ mysys/charset-def.c
Adding initialization for utf16le collations
@ sql/item_func.cc
Error message about bad truncated number in CAST(string AS SIGNED)
did not work well (a missing part of "WL#751 Error message construction").
@ sql/sql_string.cc
Conversion from binary to utf16 and utf16le did not work well.
0x61 was incorrectly extended to 0x00000061 instead of 0x0061.
@ sql/sys_vars.cc
"SET collation_connection=utf16_general_ci" followed by
"SET NAMES 'latin1'" did not work well,
because get_charset_by_csname('latin1') was called with utf16 representation
of 'latin1' string. Fixing val_str() to val_str_ascii() to always call
get_charset_by_csname() with 8-bit representation.
Fixing the similar problem for get_charset_by_name().
@ strings/ctype-ucs2.c
- Changing my_strtoll10_mb2 from accessing memory directly using
s[0] and s[1] to mb_wc() method, to reuse the same code for utf16le.
- Adding useful macro for surrogate high/low first/last values.
- Adding macros MY_UTF16_WC2 and MY_UTF16_WC4, to make some utf16
routines more readable.
- Fixing my_caseup_utf16, my_hash_sort_utf16, my_ismbchar_utf16,
my_strnncoll_utf16_bin, my_hash_sort_utf16_bin to use cs->cset->mb_wc
instead of dirrect call to my_utf16_uni, to reuse the code.
- Fixing my_hash_sort_utf16 and my_hash_sort_utf16_bin to use lengthsp()
instead of direct memory access to reuse the code for utf16le.
- Adding low level utf16le functions my_utf16le_uni, my_uni_utf16le,
my_lengthsp_utf16le. These are the only three new functions.
All other functions reuse utf16/ucs2 code.
- Adding utf16le structures.
added:
mysql-test/include/ctype_heap.inc
mysql-test/r/ctype_utf16le.result
mysql-test/t/ctype_utf16le.test
modified:
cmake/character_sets.cmake
include/m_ctype.h
mysql-test/r/ctype_utf16.result
mysys/charset-def.c
sql/item_func.cc
sql/sql_string.cc
sql/sys_vars.cc
strings/ctype-ucs2.c
=== modified file 'cmake/character_sets.cmake'
--- a/cmake/character_sets.cmake 2010-02-25 08:15:46 +0000
+++ b/cmake/character_sets.cmake 2010-12-03 10:10:05 +0000
@@ -23,14 +23,14 @@ SET(DEFAULT_COLLATION "latin1_swedish_ci
ENDIF()
SET(CHARSETS ${DEFAULT_CHARSET} latin1 utf8 utf8mb4)
-SET(CHARSETS_COMPLEX big5 cp1250 cp932 eucjpms euckr gb2312 gbk latin1 latin2 sjis tis620 ucs2 ujis utf8 utf8mb4 utf16 utf32)
+SET(CHARSETS_COMPLEX big5 cp1250 cp932 eucjpms euckr gb2312 gbk latin1 latin2 sjis tis620 ucs2 ujis utf8 utf8mb4 utf16 utf16le utf32)
SET(CHARSETS_AVAILABLE
binary armscii8 ascii big5 cp1250 cp1251 cp1256 cp1257
cp850 cp852 cp866 cp932 dec8 eucjpms euckr gb2312 gbk geostd8
greek hebrew hp8 keybcs2 koi8r koi8u
latin1 latin2 latin5 latin7 macce macroman
-sjis swe7 tis620 ucs2 ujis utf8 utf8mb4 utf16 utf32)
+sjis swe7 tis620 ucs2 ujis utf8 utf8mb4 utf16 utf16le utf32)
SET (EXTRA_CHARSETS "all")
=== modified file 'include/m_ctype.h'
--- a/include/m_ctype.h 2010-12-02 05:40:58 +0000
+++ b/include/m_ctype.h 2010-12-03 10:10:05 +0000
@@ -427,6 +427,8 @@ extern CHARSET_INFO my_charset_ujis_bin;
extern CHARSET_INFO my_charset_utf16_bin;
extern CHARSET_INFO my_charset_utf16_general_ci;
extern CHARSET_INFO my_charset_utf16_unicode_ci;
+extern CHARSET_INFO my_charset_utf16le_bin;
+extern CHARSET_INFO my_charset_utf16le_general_ci;
extern CHARSET_INFO my_charset_utf32_bin;
extern CHARSET_INFO my_charset_utf32_general_ci;
extern CHARSET_INFO my_charset_utf32_unicode_ci;
=== added file 'mysql-test/include/ctype_heap.inc'
--- a/mysql-test/include/ctype_heap.inc 1970-01-01 00:00:00 +0000
+++ b/mysql-test/include/ctype_heap.inc 2010-12-03 10:10:05 +0000
@@ -0,0 +1,11 @@
+SELECT @@collation_connection;
+CREATE TABLE t1 ENGINE=HEAP AS SELECT REPEAT (' ', 10) AS a LIMIT 0;
+ALTER TABLE t1 ADD KEY (a);
+CREATE TABLE t2 (a VARCHAR(10));
+INSERT INTO t2 VALUES ('0'),('1'),('2'),('3'),('4'),('5'),('6'),('7');
+INSERT INTO t2 VALUES ('8'),('9'),('A'),('B'),('C'),('D'),('E'),('F');
+INSERT INTO t1 SELECT CONCAT('a',t21.a,t22.a) FROM t2 t21, t2 t22 ORDER BY 1;
+DROP TABLE t2;
+INSERT INTO t1 VALUES ('a ');
+SELECT a, HEX(a) FROM t1 WHERE a='a';
+DROP TABLE t1;
=== modified file 'mysql-test/r/ctype_utf16.result'
--- a/mysql-test/r/ctype_utf16.result 2010-12-03 09:24:07 +0000
+++ b/mysql-test/r/ctype_utf16.result 2010-12-03 10:10:05 +0000
@@ -30,13 +30,13 @@ binary 'a a' > 'a' binary 'a \0' > 'a'
1 1 1
select hex(_utf16 0x44);
hex(_utf16 0x44)
-00000044
+0044
select hex(_utf16 0x3344);
hex(_utf16 0x3344)
3344
select hex(_utf16 0x113344);
hex(_utf16 0x113344)
-000000113344
+00113344
CREATE TABLE t1 (word VARCHAR(64), word2 CHAR(64)) CHARACTER SET utf16;
INSERT INTO t1 VALUES (_koi8r 0xF2, _koi8r 0xF2), (X'2004',X'2004');
SELECT hex(word) FROM t1 ORDER BY word;
@@ -434,10 +434,10 @@ aardvarz
DROP TABLE t1;
SELECT hex(cast(0xAA as char character set utf16));
hex(cast(0xAA as char character set utf16))
-000000AA
+00AA
SELECT hex(convert(0xAA using utf16));
hex(convert(0xAA using utf16))
-000000AA
+00AA
CREATE TABLE t1 (a char(10) character set utf16);
INSERT INTO t1 VALUES (0x1),(0x11),(0x111),(0x1111),(0x11111);
SELECT HEX(a) FROM t1;
=== added file 'mysql-test/r/ctype_utf16le.result'
--- a/mysql-test/r/ctype_utf16le.result 1970-01-01 00:00:00 +0000
+++ b/mysql-test/r/ctype_utf16le.result 2010-12-03 10:10:05 +0000
@@ -0,0 +1,1744 @@
+DROP TABLE IF EXISTS t1;
+#
+# Start of 5.6 tests
+#
+SET NAMES utf8, collation_connection=utf16le_general_ci;
+SELECT HEX('a'), HEX('a ');
+HEX('a') HEX('a ')
+6100 61002000
+select 'a' = 'a', 'a' = 'a ', 'a ' = 'a';
+'a' = 'a' 'a' = 'a ' 'a ' = 'a'
+1 1 1
+select 'a\0' = 'a', 'a\0' < 'a', 'a\0' > 'a';
+'a\0' = 'a' 'a\0' < 'a' 'a\0' > 'a'
+0 1 0
+select 'a' = 'a\0', 'a' < 'a\0', 'a' > 'a\0';
+'a' = 'a\0' 'a' < 'a\0' 'a' > 'a\0'
+0 0 1
+select 'a\0' = 'a ', 'a\0' < 'a ', 'a\0' > 'a ';
+'a\0' = 'a ' 'a\0' < 'a ' 'a\0' > 'a '
+0 1 0
+select 'a ' = 'a\0', 'a ' < 'a\0', 'a ' > 'a\0';
+'a ' = 'a\0' 'a ' < 'a\0' 'a ' > 'a\0'
+0 0 1
+select 'a a' > 'a', 'a \0' < 'a';
+'a a' > 'a' 'a \0' < 'a'
+1 1
+select binary 'a a' > 'a', binary 'a \0' > 'a', binary 'a\0' > 'a';
+binary 'a a' > 'a' binary 'a \0' > 'a' binary 'a\0' > 'a'
+1 1 1
+#
+# Check that incomplete utf16le characters in HEX notation
+# are left-padded with zeros
+#
+SELECT HEX(_utf16le 0x44);
+HEX(_utf16le 0x44)
+0044
+SELECT HEX(_utf16le 0x3344);
+HEX(_utf16le 0x3344)
+3344
+SELECT HEX(_utf16le 0x113344);
+HEX(_utf16le 0x113344)
+00113344
+#
+# 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 utf16le, 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), word2 CHAR(64)) CHARACTER SET utf16le;
+INSERT INTO t1 VALUES (_koi8r 0xF2, _koi8r 0xF2), (_ucs2 X'2004',_ucs2 X'2004');
+SELECT HEX(word) FROM t1 ORDER BY word;
+HEX(word)
+2004
+0420
+SELECT HEX(word2) FROM t1 ORDER BY word2;
+HEX(word2)
+2004
+0420
+DELETE FROM t1;
+#
+# Check that real spaces are correctly trimmed.
+#
+INSERT INTO t1 VALUES (_ucs2 X'042000200020', _ucs2 X'042000200020');
+INSERT INTO t1 VALUES (_ucs2 X'200400200020', _ucs2 X'200400200020');
+SELECT HEX(word) FROM t1 ORDER BY word;
+HEX(word)
+200420002000
+042020002000
+SELECT HEX(word2) FROM t1 ORDER BY word2;
+HEX(word2)
+2004
+0420
+DROP TABLE t1;
+#
+# Check LPAD/RPAD
+#
+CREATE TABLE t1 (a VARCHAR(10), pad INT, b VARCHAR(10)) CHARACTER SET utf16le;
+INSERT INTO t1 VALUES (_ucs2 X'0420', 10, _ucs2 X'0421');
+INSERT INTO t1 VALUES (_ucs2 X'0420', 10, _ucs2 X'04210422');
+INSERT INTO t1 VALUES (_ucs2 X'0420', 10, _ucs2 X'042104220423');
+INSERT INTO t1 VALUES (_ucs2 X'0420042104220423042404250426042704280429042A042B',10,_ucs2 X'042104220423');
+Warnings:
+Warning 1265 Data truncated for column 'a' at row 1
+INSERT INTO t1 VALUES (_utf32 X'010000', 10, _ucs2 X'0421');
+INSERT INTO t1 VALUES (_ucs2 X'0421', 10, _utf32 X'010000');
+SELECT a, pad, b, LPAD(a, pad, b), HEX(LPAD(a, pad, b)) FROM t1;
+a pad b LPAD(a, pad, b) HEX(LPAD(a, pad, b))
+Р 10 С СССССССССР 2104210421042104210421042104210421042004
+Р 10 СТ СТСТСТСТСР 2104220421042204210422042104220421042004
+Р 10 СТУ СТУСТУСТУР 2104220423042104220423042104220423042004
+РСТУФХЦЧШЩ 10 СТУ РСТУФХ ССССССССС? 21042104210421042104210421042104210400D800DC
+С 10 ? ?????????С 00D800DC00D800DC00D800DC00D800DC00D800DC00D800DC00D800DC00D800DC00D800DC2104
+DROP TABLE t1;
+CREATE TABLE t1 SELECT
+LPAD(_utf16le X'2004',10,_utf16le X'2104') l,
+RPAD(_utf16le X'2004',10,_utf16le X'2104') r;
+SHOW CREATE TABLE t1;
+Table Create Table
+t1 CREATE TABLE `t1` (
+ `l` varchar(10) CHARACTER SET utf16le NOT NULL DEFAULT '',
+ `r` varchar(10) CHARACTER SET utf16le NOT NULL DEFAULT ''
+) ENGINE=MyISAM DEFAULT CHARSET=latin1
+SELECT HEX(l), HEX(r) FROM t1;
+HEX(l) HEX(r)
+2104210421042104210421042104210421042004 2004210421042104210421042104210421042104
+DROP TABLE t1;
+CREATE TABLE t1 (f1 CHAR(30));
+INSERT INTO t1 VALUES ("103000"), ("22720000"), ("3401200"), ("78000");
+SELECT LPAD(f1, 12, "-o-/") FROM t1;
+LPAD(f1, 12, "-o-/")
+-o-/-o103000
+-o-/22720000
+-o-/-3401200
+-o-/-o-78000
+DROP TABLE t1;
+#
+# Testing LIKE
+#
+SET NAMES utf8, collation_connection=utf16le_general_ci;
+select @@collation_connection;
+@@collation_connection
+utf16le_general_ci
+create table t1 as select repeat(' ',10) as a union select null;
+alter table t1 add key(a);
+show create table t1;
+Table Create Table
+t1 CREATE TABLE `t1` (
+ `a` varchar(10) CHARACTER SET utf16le DEFAULT NULL,
+ KEY `a` (`a`)
+) ENGINE=MyISAM DEFAULT CHARSET=latin1
+insert into t1 values ("a"),("abc"),("abcd"),("hello"),("test");
+explain select * from t1 where a like 'abc%';
+id select_type table type possible_keys key key_len ref rows Extra
+1 SIMPLE t1 range a a 43 NULL 1 Using where; Using index
+explain select * from t1 where a like concat('abc','%');
+id select_type table type possible_keys key key_len ref rows Extra
+1 SIMPLE t1 range a a 43 NULL 1 Using where; Using index
+select * from t1 where a like "abc%";
+a
+abc
+abcd
+select * from t1 where a like concat("abc","%");
+a
+abc
+abcd
+select * from t1 where a like "ABC%";
+a
+abc
+abcd
+select * from t1 where a like "test%";
+a
+test
+select * from t1 where a like "te_t";
+a
+test
+select * from t1 where a like "%a%";
+a
+a
+abc
+abcd
+select * from t1 where a like "%abcd%";
+a
+abcd
+select * from t1 where a like "%abc\d%";
+a
+abcd
+drop table t1;
+select 'AA' like 'AA';
+'AA' like 'AA'
+1
+select 'AA' like 'A%A';
+'AA' like 'A%A'
+1
+select 'AA' like 'A%%A';
+'AA' like 'A%%A'
+1
+select 'AA' like 'AA%';
+'AA' like 'AA%'
+1
+select 'AA' like '%AA%';
+'AA' like '%AA%'
+1
+select 'AA' like '%A';
+'AA' like '%A'
+1
+select 'AA' like '%AA';
+'AA' like '%AA'
+1
+select 'AA' like 'A%A%';
+'AA' like 'A%A%'
+1
+select 'AA' like '_%_%';
+'AA' like '_%_%'
+1
+select 'AA' like '%A%A';
+'AA' like '%A%A'
+1
+select 'AAA'like 'A%A%A';
+'AAA'like 'A%A%A'
+1
+select 'AZ' like 'AZ';
+'AZ' like 'AZ'
+1
+select 'AZ' like 'A%Z';
+'AZ' like 'A%Z'
+1
+select 'AZ' like 'A%%Z';
+'AZ' like 'A%%Z'
+1
+select 'AZ' like 'AZ%';
+'AZ' like 'AZ%'
+1
+select 'AZ' like '%AZ%';
+'AZ' like '%AZ%'
+1
+select 'AZ' like '%Z';
+'AZ' like '%Z'
+1
+select 'AZ' like '%AZ';
+'AZ' like '%AZ'
+1
+select 'AZ' like 'A%Z%';
+'AZ' like 'A%Z%'
+1
+select 'AZ' like '_%_%';
+'AZ' like '_%_%'
+1
+select 'AZ' like '%A%Z';
+'AZ' like '%A%Z'
+1
+select 'AZ' like 'A_';
+'AZ' like 'A_'
+1
+select 'AZ' like '_Z';
+'AZ' like '_Z'
+1
+select 'AMZ'like 'A%M%Z';
+'AMZ'like 'A%M%Z'
+1
+CREATE TABLE t1 (a VARCHAR(10) CHARACTER SET utf16le);
+INSERT INTO t1 VALUES ('фыва'),('Фыва'),('фЫва'),('фыВа'),('фывА'),('ФЫВА');
+INSERT INTO t1 VALUES ('фывапролдж'),('Фывапролдж'),('фO t1 VALUES ('фывАпролдж'),('фываПролдж'),('фывапРолдж'),('фывапрОлдж');
+INSERT INTO t1 VALUES ('фывапроЛдж'),('фывапролДж'),('фывапролдЖ'),('ФЫВАПРОR BY BINARY a;
+a
+ФЫВА
+ФЫВАПРОЛДЖ
+Фыва
+Фывапролдж
+фЫва
+фЫвапролдж
+фыВа
+фыВапролдж
+фывА
+фывАпролдж
+фыва
+фываПролдж
+фывапРолпролДж
+фывапролдЖ
+фывапролдж
+SELECT * FROM t1 WHERE a LIKE '%фЫв%' ORDER BY BINARY a;
+a
+ФЫВА
+ФЫВАПРОЛДЖ
+Фыва
+ФывфыВапролдж
+фывА
+фывАпролдж
+фапрОлдж
+фывапроЛдж
+фывапролДROM t1 WHERE a LIKE 'фЫва%' ORDER BY BINARY a;
+a
+ФЫВА
+ФЫВАПРОЛДЖ
+Фыва
+Фывапролролдж
+фывА
+фывАпролдж
+фыва
+фдж
+фывапроЛдж
+фывапролДж
+фывRE a LIKE 'фЫва%' COLLATE utf16le_bin ORDER BY BINARY a;
+a
+фЫва
+фЫвапролдж
+DROP TABLE t1;
+CREATE TABLE t1 (word VARCHAR(64) NOT NULL, PRIMARY KEY (word))
+ENGINE=MyISAM CHARACTER SET utf16le;
+INSERT INTO t1 (word) VALUES ("cat");
+SELECT * FROM t1 WHERE word LIKE "c%";
+word
+cat
+SELECT * FROM t1 WHERE word LIKE "ca_";
+word
+cat
+SELECT * FROM t1 WHERE word LIKE "cat";
+word
+cat
+SELECT * FROM t1 WHERE word LIKE _ucs2 x'00630025';
+word
+cat
+SELECT * FROM t1 WHERE word LIKE _ucs2 x'00630061005F';
+word
+cat
+DROP TABLE t1;
+#
+# Check that INSERT() works fine.
+# This invokes charpos() function.
+#
+CREATE TABLE t1 (
+a VARCHAR(10) CHARACTER SET utf16le,
+b VARCHAR(10) CHARACTER SET utf16le);
+INSERT INTO t1 VALUES ('abc', 'def');
+SELECT INSERT(a, 10, 2, b) FROM t1;
+INSERT(a, 10, 2, b)
+abc
+SELECT INSERT(a, 1, 2, b) FROM t1;
+INSERT(a, 1, 2, b)
+defc
+DROP TABLE t1;
+#
+# Bug#1264
+#
+SET NAMES utf8, collation_connection=utf16le_general_ci;
+#
+# Two fields, index
+#
+CREATE TABLE t1 (
+word VARCHAR(64),
+bar INT(11) DEFAULT 0,
+PRIMARY KEY (word))
+ENGINE=MyISAM
+CHARSET utf16le
+COLLATE utf16le_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;
+id select_type table type possible_keys key key_len ref rows Extra
+1 SIMPLE t1 ALL NULL NULL NULL NULL 6 Using filesort
+SELECT * FROM t1 ORDER BY word;
+word bar
+a 0
+aar 0
+aardvar 0
+aardvara 0
+aardvark 0
+aardvarz 0
+EXPLAIN SELECT word FROM t1 ORDER BY word;
+id select_type table type possible_keys key key_len ref rows Extra
+1 SIMPLE t1 index NULL PRIMARY 258 NULL 6 Using index
+SELECT word FROM t1 ORDER by word;
+word
+a
+aar
+aardvar
+aardvara
+aardvark
+aardvarz
+DROP TABLE t1;
+#
+# One field, index
+#
+CREATE TABLE t1 (
+word VARCHAR(64) ,
+PRIMARY KEY (word))
+ENGINE=MyISAM
+CHARSET utf16le
+COLLATE utf16le_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;
+id select_type table type possible_keys key key_len ref rows Extra
+1 SIMPLE t1 index NULL PRIMARY 258 NULL 6 Using index
+SELECT * FROM t1 ORDER BY word;
+word
+a
+aar
+aardvar
+aardvara
+aardvark
+aardvarz
+DROP TABLE t1;
+#
+# Two fields, no index
+#
+CREATE TABLE t1 (
+word TEXT,
+bar INT(11) AUTO_INCREMENT,
+PRIMARY KEY (bar))
+ENGINE=MyISAM
+CHARSET utf16le
+COLLATE utf16le_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;
+id select_type table type possible_keys key key_len ref rows Extra
+1 SIMPLE t1 ALL NULL NULL NULL NULL 6 Using filesort
+SELECT * FROM t1 ORDER BY word;
+word bar
+a 2
+aar 1
+aardvar 3
+aardvara 5
+aardvark 4
+aardvarz 6
+EXPLAIN SELECT word FROM t1 ORDER BY word;
+id select_type table type possible_keys key key_len ref rows Extra
+1 SIMPLE t1 ALL NULL NULL NULL NULL 6 Using filesort
+SELECT word FROM t1 ORDER BY word;
+word
+a
+aar
+aardvar
+aardvara
+aardvark
+aardvarz
+DROP TABLE t1;
+#
+# END OF Bug 1264 test
+#
+#
+# Check alignment for from-binary-conversion with CAST and CONVERT
+#
+SELECT HEX(CAST(0xAA as char CHARACTER SET utf16le));
+HEX(CAST(0xAA as char CHARACTER SET utf16le))
+00AA
+SELECT HEX(CONVERT(0xAA USING utf16le));
+HEX(CONVERT(0xAA USING utf16le))
+00AA
+#
+# Check alignment for string types
+#
+CREATE TABLE t1 (a CHAR(10) CHARACTER SET utf16le);
+INSERT INTO t1 VALUES (0x1),(0x11),(0x111),(0x1111),(0x11111);
+SELECT HEX(a) FROM t1;
+HEX(a)
+0001
+0011
+0111
+1111
+00011111
+DROP TABLE t1;
+CREATE TABLE t1 (a VARCHAR(10) CHARACTER SET utf16le);
+INSERT INTO t1 VALUES (0x1),(0x11),(0x111),(0x1111),(0x11111);
+SELECT HEX(a) FROM t1;
+HEX(a)
+0001
+0011
+0111
+1111
+00011111
+DROP TABLE t1;
+CREATE TABLE t1 (a TEXT CHARACTER SET utf16le);
+INSERT INTO t1 VALUES (0x1),(0x11),(0x111),(0x1111),(0x11111);
+SELECT HEX(a) FROM t1;
+HEX(a)
+0001
+0011
+0111
+1111
+00011111
+DROP TABLE t1;
+CREATE TABLE t1 (a MEDIUMTEXT CHARACTER SET utf16le);
+INSERT INTO t1 VALUES (0x1),(0x11),(0x111),(0x1111),(0x11111);
+SELECT HEX(a) FROM t1;
+HEX(a)
+0001
+0011
+0111
+1111
+00011111
+DROP TABLE t1;
+CREATE TABLE t1 (a LONGTEXT CHARACTER SET utf16le);
+INSERT INTO t1 VALUES (0x1),(0x11),(0x111),(0x1111),(0x11111);
+SELECT HEX(a) FROM t1;
+HEX(a)
+0001
+0011
+0111
+1111
+00011111
+DROP TABLE t1;
+#
+# Bug#5081 : UCS2 fields are filled with '0x2020'
+# after extending field length
+#
+CREATE TABLE t1(a CHAR(1)) DEFAULT CHARSET utf16le;
+INSERT INTO t1 VALUES ('a'),('b'),('c');
+ALTER TABLE t1 MODIFY a CHAR(5);
+SELECT a, HEX(a) FROM t1;
+a HEX(a)
+a 6100
+b 6200
+c 6300
+DROP TABLE t1;
+#
+# Check prepare statement from an UTF16 string
+#
+SET NAMES latin1;
+SET @ivar= 1234;
+SET @str1 = 'SELECT ?';
+SET @str2 = CONVERT(@str1 USING utf16le);
+PREPARE stmt1 FROM @str2;
+EXECUTE stmt1 USING @ivar;
+?
+1234
+#
+# Check that utf16le works with ENUM and SET type
+#
+SET NAMES utf8, collation_connection=utf16le_general_ci;
+CREATE TABLE t1 (a ENUM('x','y','z') CHARACTER SET utf16le);
+SHOW CREATE TABLE t1;
+Table Create Table
+t1 CREATE TABLE `t1` (
+ `a` enum('x','y','z') CHARACTER SET utf16le DEFAULT NULL
+) ENGINE=MyISAM DEFAULT CHARSET=latin1
+INSERT INTO t1 VALUES ('x');
+INSERT INTO t1 VALUES ('y');
+INSERT INTO t1 VALUES ('z');
+SELECT a, HEX(a) FROM t1 ORDER BY a;
+a HEX(a)
+x 7800
+y 7900
+z 7A00
+ALTER TABLE t1 CHANGE a a ENUM('x','y','z','d','e','ä','able Create Table
+t1 CREATE TABLE `t1` (
+ `a` enum('x','y','z','d','e','ä','ö','ü') CHARACTER SET utf16le DEFAULT NULL
+) ENGINE=MyISAM DEFAULT CHARSET=latin1
+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;
+a HEX(a)
+x 7800
+y 7900
+z 7A00
+d 6400
+e 6500
+ä E400
+ö F600
+ü FC00
+DROP TABLE t1;
+CREATE TABLE t1 (a set ('x','y','z','ä','ö','ü') CHARACTER SET utf16le);
+SHOW CREATE TABLE t1;
+Table Create Table
+t1 CREATE TABLE `t1` (
+ `a` set('x','y','z','ä','ö','ü') CHARACTER SET utf16le DEFAULT NULL
+) ENGINE=MyISAM DEFAULT CHARSET=latin1
+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;
+a HEX(a)
+x 7800
+y 7900
+x,y 78002C007900
+z 7A00
+x,y,z,ä,ö,ü 78002C0079002C007A002C00E4002C00F6002C00FC00
+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 utf16le;
+INSERT INTO t1 VALUES('a'),('b'),('c');
+ALTER TABLE t1 ADD b CHAR(1);
+SHOW WARNINGS;
+Level Code Message
+SELECT * FROM t1 ORDER BY a;
+a b
+a NULL
+b NULL
+c NULL
+DROP TABLE t1;
+SET NAMES utf8, collation_connection='utf16le_general_ci';
+create table t1 select repeat('a',4000) a;
+delete from t1;
+insert into t1 values ('a'), ('a '), ('a\t');
+select collation(a),hex(a) from t1 order by a;
+collation(a) hex(a)
+utf16le_general_ci 61000900
+utf16le_general_ci 6100
+utf16le_general_ci 61002000
+drop table t1;
+select @@collation_connection;
+@@collation_connection
+utf16le_general_ci
+create table t1 ROW_FORMAT=DYNAMIC select repeat('a',50) as c1 ;
+insert into t1 values('abcdef');
+insert into t1 values('_bcdef');
+insert into t1 values('a_cdef');
+insert into t1 values('ab_def');
+insert into t1 values('abc_ef');
+insert into t1 values('abcd_f');
+insert into t1 values('abcde_');
+select c1 as c1u from t1 where c1 like 'ab\_def';
+c1u
+ab_def
+select c1 as c2h from t1 where c1 like 'ab#_def' escape '#';
+c2h
+ab_def
+drop table t1;
+SET NAMES utf8, collation_connection='utf16le_bin';
+create table t1 select repeat('a',4000) a;
+delete from t1;
+insert into t1 values ('a'), ('a '), ('a\t');
+select collation(a),hex(a) from t1 order by a;
+collation(a) hex(a)
+utf16le_bin 61000900
+utf16le_bin 6100
+utf16le_bin 61002000
+drop table t1;
+#
+# Bug#55980 Character sets: supplementary character _bin ordering is wrong
+#
+CREATE TABLE t1 AS SELECT REPEAT('a',1) AS a LIMIT 0;
+SHOW CREATE TABLE t1;
+Table Create Table
+t1 CREATE TABLE `t1` (
+ `a` varchar(1) CHARACTER SET utf16le COLLATE utf16le_bin NOT NULL DEFAULT ''
+) ENGINE=MyISAM DEFAULT CHARSET=latin1
+INSERT INTO t1 VALUES (_utf8mb4 0xEFBE9D),(_utf8mb4 0xF0908E84);
+INSERT INTO t1 VALUES (_utf8mb4 0xCE85),(_utf8mb4 0xF4808080);
+SELECT HEX(a), HEX(CONVERT(a USING utf8mb4)) FROM t1 ORDER BY a;
+HEX(a) HEX(CONVERT(a USING utf8mb4))
+8503 CE85
+9DFF EFBE9D
+00D884DF F0908E84
+C0DB00DC F4808080
+ALTER TABLE t1 ADD KEY(a);
+SELECT HEX(a), HEX(CONVERT(a USING utf8mb4)) FROM t1 ORDER BY a;
+HEX(a) HEX(CONVERT(a USING utf8mb4))
+8503 CE85
+9DFF EFBE9D
+00D884DF F0908E84
+C0DB00DC F4808080
+# Additional test for bug#37244 Character sets: short utf8_bin weight_string value
+SELECT HEX(a), HEX(WEIGHT_STRING(a)) FROM t1 ORDER BY a;
+HEX(a) HEX(WEIGHT_STRING(a))
+8503 000385
+9DFF 00FF9D
+00D884DF 010384
+C0DB00DC 100000
+DROP TABLE IF EXISTS t1;
+select @@collation_connection;
+@@collation_connection
+utf16le_bin
+create table t1 ROW_FORMAT=DYNAMIC select repeat('a',50) as c1 ;
+insert into t1 values('abcdef');
+insert into t1 values('_bcdef');
+insert into t1 values('a_cdef');
+insert into t1 values('ab_def');
+insert into t1 values('abc_ef');
+insert into t1 values('abcd_f');
+insert into t1 values('abcde_');
+select c1 as c1u from t1 where c1 like 'ab\_def';
+c1u
+ab_def
+select c1 as c2h from t1 where c1 like 'ab#_def' escape '#';
+c2h
+ab_def
+drop table t1;
+#
+# Bug#10344 Some string functions fail for UCS2
+#
+CREATE TABLE t1 (a VARCHAR(10) CHARACTER SET utf16le, pos INT);
+INSERT INTO t1 VALUES (_ucs2 0x00e400e50068,1);
+INSERT INTO t1 VALUES (_ucs2 0x00e400e50068,2);
+INSERT INTO t1 VALUES (_ucs2 0x00e400e50068,3);
+INSERT INTO t1 VALUES (_ucs2 0x00e400e50068,-1);
+INSERT INTO t1 VALUES (_ucs2 0x00e400e50068,-2);
+INSERT INTO t1 VALUES (_ucs2 0x00e400e50068,-3);
+INSERT INTO t1 VALUES (_utf32 0x000000e4000000e500010000, 1);
+INSERT INTO t1 VALUES (_utf32 0x000000e4000000e500010000, 2);
+INSERT INTO t1 VALUES (_utf32 0x000000e4000000e500010000, 3);
+INSERT INTO t1 VALUES (_utf32 0x000000e4000000e500010000, -1);
+INSERT INTO t1 VALUES (_utf32 0x000000e4000000e500010000, -2);
+INSERT INTO t1 VALUES (_utf32 0x000000e4000000e500010000, -3);
+SELECT HEX(SUBSTR(a, pos)), SUBSTR(a, pos) FROM t1;
+HEX(SUBSTR(a, pos)) SUBSTR(a, pos)
+E400E5006800 äåh
+E5006800 åh
+6800 h
+6800 h
+E5006800 åh
+E400E5006800 äåh
+E400E50000D800DC äå?
+E50000D800DC å?
+00D800DC ?
+00D800DC ?
+E50000D800DC å?
+E400E50000D800DC äå?
+DROP TABLE t1;
+SET NAMES utf8, collation_connection=utf16le_general_ci;
+#
+# Bug#9442 Set parameter make query fail if column CHARACTER SET is UCS2
+#
+CREATE TABLE t1 (utext VARCHAR(20) CHARACTER SET utf16le);
+INSERT INTO t1 VALUES ("lily");
+INSERT INTO t1 VALUES ("river");
+PREPARE stmt FROM 'SELECT utext FROM t1 where utext like ?';
+SET @param1='%%';
+EXECUTE stmt USING @param1;
+utext
+lily
+river
+EXECUTE stmt USING @param1;
+utext
+lily
+river
+SELECT utext FROM t1 where utext like '%%';
+utext
+lily
+river
+DROP TABLE t1;
+DEALLOCATE PREPARE stmt;
+#
+# Bug#22052 Trailing spaces are not removed FROM UNICODE fields in an index
+#
+CREATE TABLE t1 (
+a CHAR(10) CHARACTER SET utf16le NOT NULL,
+INDEX a (a)
+) engine=myisam;
+INSERT INTO t1 VALUES (REPEAT(_ucs2 0x201f, 10));
+INSERT INTO t1 VALUES (REPEAT(_ucs2 0x2020, 10));
+INSERT INTO t1 VALUES (REPEAT(_ucs2 0x2021, 10));
+# make sure "index read" is used
+explain SELECT HEX(a) FROM t1 ORDER BY a;
+id select_type table type possible_keys key key_len ref rows Extra
+1 SIMPLE t1 index NULL a 40 NULL 3 Using index
+SELECT HEX(a) FROM t1 ORDER BY a;
+HEX(a)
+1F201F201F201F201F201F201F201F201F201F20
+2020202020202020202020202020202020202020
+2120212021202120212021202120212021202120
+ALTER TABLE t1 DROP INDEX a;
+SELECT HEX(a) FROM t1 ORDER BY a;
+HEX(a)
+1F201F201F201F201F201F201F201F201F201F20
+2020202020202020202020202020202020202020
+2120212021202120212021202120212021202120
+DROP TABLE t1;
+#
+# Bug #20108: corrupted default enum value for a ucs2 field
+#
+CREATE TABLE t1 (
+status ENUM('active','passive') CHARACTER SET utf16le COLLATE utf16le_general_ci
+NOT NULL DEFAULT 'passive'
+);
+SHOW CREATE TABLE t1;
+Table Create Table
+t1 CREATE TABLE `t1` (
+ `status` enum('active','passive') CHARACTER SET utf16le NOT NULL DEFAULT 'passive'
+) ENGINE=MyISAM DEFAULT CHARSET=latin1
+ALTER TABLE t1 ADD a int NOT NULL AFTER status;
+SHOW CREATE TABLE t1;
+Table Create Table
+t1 CREATE TABLE `t1` (
+ `status` enum('active','passive') CHARACTER SET utf16le NOT NULL DEFAULT 'passive',
+ `a` int(11) NOT NULL
+) ENGINE=MyISAM DEFAULT CHARSET=latin1
+DROP TABLE t1;
+#
+# Conversion FROM an UTF16LE string to a decimal column
+#
+CREATE TABLE t1 (a VARCHAR(64) CHARACTER SET utf16le, b DECIMAL(10,3));
+INSERT INTO t1 VALUES ("1.1", 0), ("2.1", 0);
+UPDATE t1 set b=a;
+SELECT *, HEX(a) FROM t1;
+a b HEX(a)
+1.1 1.100 31002E003100
+2.1 2.100 32002E003100
+DROP TABLE t1;
+#
+# Bug#9442 Set parameter make query fail if column CHARACTER SET is UCS2
+#
+CREATE TABLE t1 (utext VARCHAR(20) CHARACTER SET utf16le);
+INSERT INTO t1 VALUES ("lily");
+INSERT INTO t1 VALUES ("river");
+PREPARE stmt FROM 'SELECT utext FROM t1 where utext like ?';
+SET @param1='%%';
+EXECUTE stmt USING @param1;
+utext
+lily
+river
+EXECUTE stmt USING @param1;
+utext
+lily
+river
+SELECT utext FROM t1 where utext like '%%';
+utext
+lily
+river
+DROP TABLE t1;
+DEALLOCATE PREPARE stmt;
+#
+# Bug#22638 SOUNDEX broken for international characters
+#
+SET NAMES utf8, collation_connection=utf16le_general_ci;
+SELECT SOUNDEX(''),SOUNDEX('he'),SOUNDEX('hello all folks'),SOUNDEX('#3556 in bugdb');
+SOUNDEX('') SOUNDEX('he') SOUNDEX('hello all folks') SOUNDEX('#3556 in bugdb')
+ H000 H4142 I51231
+SELECT HEX(SOUNDEX('')),HEX(SOUNDEX('he')),HEX(SOUNDEX('hello all folks')),HEX(SOUNDEX('#3556 in bugdb'));
+HEX(SOUNDEX('')) HEX(SOUNDEX('he')) HEX(SOUNDEX('hello all folks')) HEX(SOUNDEX('#3556 in bugdb'))
+ 4800300030003000 48003400310034003200 490035003100320033003100
+SELECT 'mood' sounds like 'mud';
+'mood' sounds like 'mud'
+1
+# Cyrillic A, BE, VE
+SELECT HEX(SOUNDEX(_utf16le 0x041004110412));
+HEX(SOUNDEX(_utf16le 0x041004110412))
+0410300030003000
+# Make sure that "U+00BF INVERTED QUESTION MARK" is not considered as letter
+SELECT HEX(SOUNDEX(_utf16le 0x00BF00C0));
+HEX(SOUNDEX(_utf16le 0x00BF00C0))
+00BF300030003000
+#
+# Bug#14290: character_maximum_length for text fields
+#
+CREATE TABLE t1(a BLOB, b TEXT CHARSET utf16le);
+SELECT data_type, character_octet_length, character_maximum_length
+FROM information_schema.columns where table_name='t1';
+data_type character_octet_length character_maximum_length
+blob 65535 65535
+text 65535 32767
+DROP TABLE t1;
+SET NAMES utf8, collation_connection=utf16le_general_ci;
+#
+# Testing cs->coll->instr()
+#
+SELECT POSITION('bb' IN 'abba');
+POSITION('bb' IN 'abba')
+2
+#
+# Testing cs->coll->hash_sort()
+#
+SET NAMES utf8, collation_connection=utf16le_bin;
+SELECT @@collation_connection;
+@@collation_connection
+utf16le_bin
+CREATE TABLE t1 ENGINE=HEAP AS SELECT REPEAT (' ', 10) AS a LIMIT 0;
+ALTER TABLE t1 ADD KEY (a);
+CREATE TABLE t2 (a VARCHAR(10));
+INSERT INTO t2 VALUES ('0'),('1'),('2'),('3'),('4'),('5'),('6'),('7');
+INSERT INTO t2 VALUES ('8'),('9'),('A'),('B'),('C'),('D'),('E'),('F');
+INSERT INTO t1 SELECT CONCAT('a',t21.a,t22.a) FROM t2 t21, t2 t22 ORDER BY 1;
+DROP TABLE t2;
+INSERT INTO t1 VALUES ('a ');
+SELECT a, HEX(a) FROM t1 WHERE a='a';
+a HEX(a)
+a 61002000
+DROP TABLE t1;
+SET NAMES utf8, collation_connection=utf16le_general_ci;
+SELECT @@collation_connection;
+@@collation_connection
+utf16le_general_ci
+CREATE TABLE t1 ENGINE=HEAP AS SELECT REPEAT (' ', 10) AS a LIMIT 0;
+ALTER TABLE t1 ADD KEY (a);
+CREATE TABLE t2 (a VARCHAR(10));
+INSERT INTO t2 VALUES ('0'),('1'),('2'),('3'),('4'),('5'),('6'),('7');
+INSERT INTO t2 VALUES ('8'),('9'),('A'),('B'),('C'),('D'),('E'),('F');
+INSERT INTO t1 SELECT CONCAT('a',t21.a,t22.a) FROM t2 t21, t2 t22 ORDER BY 1;
+DROP TABLE t2;
+INSERT INTO t1 VALUES ('a ');
+SELECT a, HEX(a) FROM t1 WHERE a='a';
+a HEX(a)
+a 61002000
+DROP TABLE t1;
+#
+# Testing cs->cset->numchars()
+#
+SELECT CHAR_LENGTH('abcd'), OCTET_LENGTH('abcd');
+CHAR_LENGTH('abcd') OCTET_LENGTH('abcd')
+4 8
+SELECT CHAR_LENGTH(_utf16le 0x00D800DC), OCTET_LENGTH(_utf16le 0x00D800DC);
+CHAR_LENGTH(_utf16le 0x00D800DC) OCTET_LENGTH(_utf16le 0x00D800DC)
+1 4
+SELECT CHAR_LENGTH(_utf16le 0x7DD8FFDF), OCTET_LENGTH(_utf16le 0x7FD8DDDF);
+CHAR_LENGTH(_utf16le 0x7DD8FFDF) OCTET_LENGTH(_utf16le 0x7FD8DDDF)
+1 4
+#
+# Testing cs->cset->charpos()
+#
+SELECT LEFT('abcd',2);
+LEFT('abcd',2)
+ab
+SELECT HEX(LEFT(_utf16le 0x00D800DC7FD8FFDF, 1));
+HEX(LEFT(_utf16le 0x00D800DC7FD8FFDF, 1))
+00D800DC
+SELECT HEX(RIGHT(_utf16le 0x00D800DC7FD8FFDF, 1));
+HEX(RIGHT(_utf16le 0x00D800DC7FD8FFDF, 1))
+7FD8FFDF
+#
+# Testing cs->cset->well_formed_length()
+#
+CREATE TABLE t1 (a VARCHAR(10) CHARACTER SET utf16le);
+# Bad sequences
+INSERT INTO t1 VALUES (_utf16le 0x00D8);
+ERROR HY000: Invalid utf16le character string: '00D8'
+INSERT INTO t1 VALUES (_utf16le 0x00DC);
+ERROR HY000: Invalid utf16le character string: '00DC'
+INSERT INTO t1 VALUES (_utf16le 0x00D800D8);
+ERROR HY000: Invalid utf16le character string: '00D800'
+INSERT INTO t1 VALUES (_utf16le 0x00D800E8);
+ERROR HY000: Invalid utf16le character string: '00D800'
+INSERT INTO t1 VALUES (_utf16le 0x00D80008);
+ERROR HY000: Invalid utf16le character string: '00D800'
+# Good sequences
+INSERT INTO t1 VALUES (_utf16le 0x00D800DC);
+INSERT INTO t1 VALUES (_utf16le 0x00D8FFDC);
+INSERT INTO t1 VALUES (_utf16le 0xFFDB00DC);
+INSERT INTO t1 VALUES (_utf16le 0xFFDBFFDC);
+SELECT HEX(a) FROM t1;
+HEX(a)
+00D800DC
+00D8FFDC
+FFDB00DC
+FFDBFFDC
+DROP TABLE t1;
+#
+# Bug#32393 Character sets: illegal characters in utf16le columns
+#
+# Tests that cs->cset->wc_mb() doesn't accept surrogate parts
+#
+# via ALTER
+#
+CREATE TABLE t1 (s1 VARCHAR(50) CHARACTER SET ucs2);
+INSERT INTO t1 VALUES (0xDF84);
+ALTER TABLE t1 MODIFY column s1 VARCHAR(50) CHARACTER SET utf16le;
+Warnings:
+Warning 1366 Incorrect string value: '\xDF\x84' for column 's1' at row 1
+SELECT HEX(s1) FROM t1;
+HEX(s1)
+3F00
+DROP TABLE t1;
+#
+# via UPDATE
+#
+CREATE TABLE t1 (s1 VARCHAR(5) CHARACTER SET ucs2, s2 VARCHAR(5) CHARACTER SET utf16le);
+INSERT INTO t1 (s1) VALUES (0xdf84);
+UPDATE t1 set s2 = s1;
+Warnings:
+Warning 1366 Incorrect string value: '\xDF\x84' for column 's2' at row 1
+SELECT HEX(s2) FROM t1;
+HEX(s2)
+3F00
+DROP TABLE t1;
+#
+# Testing cs->cset->lengthsp()
+#
+CREATE TABLE t1 (a CHAR(10)) CHARACTER SET utf16le;
+INSERT INTO t1 VALUES ('a ');
+SELECT HEX(a) FROM t1;
+HEX(a)
+6100
+DROP TABLE t1;
+#
+# Testing cs->cset->caseup() and cs->cset->casedn()
+#
+SELECT UPPER('abcd'), LOWER('ABCD');
+UPPER('abcd') LOWER('ABCD')
+ABCD abcd
+#
+# Checking str_to_datetime()
+#
+select @@collation_connection;
+@@collation_connection
+utf16le_general_ci
+CREATE TABLE t1 (a date);
+INSERT INTO t1 VALUES ('2007-09-16');
+SELECT * FROM t1;
+a
+2007-09-16
+DROP TABLE t1;
+#
+# Testing cs->cset->ll10tostr
+#
+CREATE TABLE t1 (a VARCHAR(10) CHARACTER SET utf16le);
+INSERT INTO t1 VALUES (123456);
+SELECT a, HEX(a) FROM t1;
+a HEX(a)
+123456 310032003300340035003600
+DROP TABLE t1;
+#
+# Testing cs->cset->fill
+# SOUNDEX fills strings with DIGIT ZERO up to four characters
+#
+SELECT SOUNDEX('a'), HEX(SOUNDEX('a'));
+SOUNDEX('a') HEX(SOUNDEX('a'))
+A000 4100300030003000
+#
+# Testing cs->cset->strntoul
+#
+CREATE TABLE t1 (a enum ('a','b','c')) CHARACTER SET utf16le;
+INSERT INTO t1 VALUES ('1');
+SELECT * FROM t1;
+a
+a
+DROP TABLE t1;
+#
+# Testing cs->cset->strntoll and cs->cset->strntoull
+#
+SET NAMES latin1;
+SELECT HEX(CONV(CONVERT('123' USING utf16le), -10, 16));
+HEX(CONV(CONVERT('123' USING utf16le), -10, 16))
+3742
+SELECT HEX(CONV(CONVERT('123' USING utf16le), 10, 16));
+HEX(CONV(CONVERT('123' USING utf16le), 10, 16))
+3742
+SET NAMES utf8, collation_connection=utf16le_general_ci;
+#
+# Testing cs->cset->strntod
+#
+SET NAMES utf8, collation_connection=utf16le_general_ci;
+SELECT 1.1 + '1.2';
+1.1 + '1.2'
+2.3
+SELECT 1.1 + '1.2xxx';
+1.1 + '1.2xxx'
+2.3
+Warnings:
+Warning 1292 Truncated incorrect DOUBLE value: ''
+#
+# Testing cs->cset->strtoll10
+#
+SELECT LEFT('aaa','1');
+LEFT('aaa','1')
+a
+CREATE TABLE t1 AS SELECT REPEAT('abcd', 128) AS a;
+SELECT LEFT(a, '2') FROM t1;
+LEFT(a, '2')
+ab
+SELECT LEFT(a, ' \t \t 2') FROM t1;
+LEFT(a, ' \t \t 2')
+ab
+SELECT LEFT(a, ' \t \t +2') FROM t1;
+LEFT(a, ' \t \t +2')
+ab
+SELECT SUBSTR(a, '-2') FROM t1;
+SUBSTR(a, '-2')
+cd
+SELECT SUBSTR(a, ' \t \t -2') FROM t1;
+SUBSTR(a, ' \t \t -2')
+cd
+SELECT LEFT(a, '00002') FROM t1;
+LEFT(a, '00002')
+ab
+SELECT LEFT(a, ' \t \t 00002') FROM t1;
+LEFT(a, ' \t \t 00002')
+ab
+SELECT LEFT(a, ' \t \t +00002') FROM t1;
+LEFT(a, ' \t \t +00002')
+ab
+SELECT SUBSTR(a, '-00002') FROM t1;
+SUBSTR(a, '-00002')
+cd
+SELECT SUBSTR(a, ' \t \t -00002') FROM t1;
+SUBSTR(a, ' \t \t -00002')
+cd
+DROP TABLE t1;
+CREATE TABLE t1 AS SELECT REPEAT('abcd', 128) AS a LIMIT 0;
+INSERT INTO t1 VALUES ('255'), ('65535'),('16777215'),('4294967295'),
+('1099511627775'),('281474976710655'),('72057594037927935'),
+('1844674407370955161'),('18446744073709551614'), ('18446744073709551615');
+SELECT a, CAST(a AS SIGNED), CAST(a AS UNSIGNED) FROM t1;
+a CAST(a AS SIGNED) CAST(a AS UNSIGNED)
+255 255 255
+65535 65535 65535
+16777215 16777215 16777215
+4294967295 4294967295 4294967295
+1099511627775 1099511627775 1099511627775
+281474976710655 281474976710655 281474976710655
+72057594037927935 72057594037927935 72057594037927935
+1844674407370955161 1844674407370955161 1844674407370955161
+18446744073709551614 -2 18446744073709551614
+18446744073709551615 -1 18446744073709551615
+Warnings:
+Warning 1105 Cast to signed converted positive out-of-range integer to it's negative complement
+Warning 1105 Cast to signed converted positive out-of-range integer to it's negative complement
+UPDATE t1 SET a=CONCAT('-', a);
+SELECT a, CAST(a AS SIGNED) FROM t1;
+a CAST(a AS SIGNED)
+-255 -255
+-65535 -65535
+-16777215 -16777215
+-4294967295 -4294967295
+-1099511627775 -1099511627775
+-281474976710655 -281474976710655
+-72057594037927935 -72057594037927935
+-1844674407370955161 -1844674407370955161
+-18446744073709551614 -9223372036854775808
+-18446744073709551615 -9223372036854775808
+Warnings:
+Warning 1292 Truncated incorrect INTEGER value: '-18446744073709551614'
+Warning 1292 Truncated incorrect INTEGER value: '-18446744073709551615'
+DROP TABLE t1;
+#
+# Testing cs->cset->strntoull10rnd
+#
+CREATE TABLE t1 (a int);
+INSERT INTO t1 VALUES ('-1234.1e2');
+INSERT INTO t1 VALUES ('-1234.1e2xxxx');
+Warnings:
+Warning 1265 Data truncated for column 'a' at row 1
+INSERT INTO t1 VALUES ('-1234.1e2 ');
+INSERT INTO t1 VALUES ('123');
+INSERT INTO t1 VALUES ('-124');
+INSERT INTO t1 VALUES ('+125');
+INSERT INTO t1 VALUES (' \t \t 123');
+INSERT INTO t1 VALUES (' \t \t -124');
+INSERT INTO t1 VALUES (' \t \t +125');
+INSERT INTO t1 VALUES (' \t \t 000123');
+INSERT INTO t1 VALUES (' \t \t -000124');
+INSERT INTO t1 VALUES (' \t \t +000125');
+SELECT * FROM t1;
+a
+-123410
+-123410
+-123410
+123
+-124
+125
+123
+-124
+125
+123
+-124
+125
+DROP TABLE t1;
+CREATE TABLE t1 (a BIGINT UNSIGNED);
+INSERT INTO t1 VALUES ('255'), ('65535'),('16777215'),('4294967295'),
+('1099511627775'),('281474976710655'),('72057594037927935'),
+('1844674407370955161'),('18446744073709551614'), ('18446744073709551615');
+SELECT * FROM t1;
+a
+255
+65535
+16777215
+4294967295
+1099511627775
+281474976710655
+72057594037927935
+1844674407370955161
+18446744073709551614
+18446744073709551615
+DROP TABLE t1;
+#
+# Testing cs->cset->scan
+#
+CREATE TABLE t1 (a int);
+INSERT INTO t1 VALUES ('1 ');
+INSERT INTO t1 VALUES ('1 x');
+Warnings:
+Warning 1265 Data truncated for column 'a' at row 1
+SELECT * FROM t1;
+a
+1
+1
+DROP TABLE t1;
+#
+# Testing auto-conversion to TEXT
+#
+CREATE TABLE t1 (a VARCHAR(17000) CHARACTER SET utf16le);
+Warnings:
+Note 1246 Converting column 'a' from VARCHAR to TEXT
+SHOW CREATE TABLE t1;
+Table Create Table
+t1 CREATE TABLE `t1` (
+ `a` mediumtext CHARACTER SET utf16le
+) ENGINE=MyISAM DEFAULT CHARSET=latin1
+DROP TABLE t1;
+#
+# Testing that maximim possible key length is 1000 bytes
+#
+CREATE TABLE t1 (a VARCHAR(250) CHARACTER SET utf16le PRIMARY KEY);
+SHOW CREATE TABLE t1;
+Table Create Table
+t1 CREATE TABLE `t1` (
+ `a` varchar(250) CHARACTER SET utf16le NOT NULL,
+ PRIMARY KEY (`a`)
+) ENGINE=MyISAM DEFAULT CHARSET=latin1
+DROP TABLE t1;
+CREATE TABLE t1 (a VARCHAR(334) CHARACTER SET utf16le PRIMARY KEY);
+ERROR 42000: Specified key was too long; max key length is 1000 bytes
+#
+# Conversion to utf8
+#
+CREATE TABLE t1 (a CHAR(1) CHARACTER SET utf16le);
+INSERT INTO t1 VALUES (0x00D800DC),(0x00D8FFDC),(0x7FDB00DC),(0x7FDBFFDC);
+INSERT INTO t1 VALUES (0xC000), (0xFF00),(0x00E0), (0xFFFF);
+SELECT HEX(a), HEX(@a:=CONVERT(a USING utf8mb4)), HEX(CONVERT(@a USING utf16le)) FROM t1;
+HEX(a) HEX(@a:=CONVERT(a USING utf8mb4)) HEX(CONVERT(@a USING utf16le))
+00D800DC F0908080 00D800DC
+00D8FFDC F09083BF 00D8FFDC
+7FDB00DC F3AFB080 7FDB00DC
+7FDBFFDC F3AFB3BF 7FDBFFDC
+C000 C380 C000
+FF00 C3BF FF00
+00E0 EE8080 00E0
+FFFF EFBFBF FFFF
+DROP TABLE t1;
+#
+# Test basic regex functionality
+#
+SET NAMES utf8, collation_connection=utf16le_general_ci;
+drop table if exists t1;
+create table t1 as
+select repeat(' ', 64) as s1, repeat(' ',64) as s2
+union
+select null, null;
+show create table t1;
+Table Create Table
+t1 CREATE TABLE `t1` (
+ `s1` varchar(64) CHARACTER SET utf16le DEFAULT NULL,
+ `s2` varchar(64) CHARACTER SET utf16le DEFAULT NULL
+) ENGINE=MyISAM DEFAULT CHARSET=latin1
+delete from t1;
+insert into t1 values('aaa','aaa');
+insert into t1 values('aaa|qqq','qqq');
+insert into t1 values('gheis','^[^a-dXYZ]+$');
+insert into t1 values('aab','^aa?b');
+insert into t1 values('Baaan','^Ba*n');
+insert into t1 values('aaa','qqq|aaa');
+insert into t1 values('qqq','qqq|aaa');
+insert into t1 values('bbb','qqq|aaa');
+insert into t1 values('bbb','qqq');
+insert into t1 values('aaa','aba');
+insert into t1 values(null,'abc');
+insert into t1 values('def',null);
+insert into t1 values(null,null);
+insert into t1 values('ghi','ghi[');
+select HIGH_PRIORITY s1 regexp s2 from t1;
+s1 regexp s2
+1
+1
+1
+1
+1
+1
+1
+0
+0
+0
+NULL
+NULL
+NULL
+NULL
+drop table t1;
+#
+# Test how CHARACTER SET works with date/time
+#
+CREATE TABLE t1 AS SELECT repeat('a',20) AS s1 LIMIT 0;
+SET timestamp=1216359724;
+INSERT INTO t1 VALUES (current_date);
+INSERT INTO t1 VALUES (current_time);
+INSERT INTO t1 VALUES (current_timestamp);
+SELECT s1, hex(s1) FROM t1;
+s1 hex(s1)
+2008-07-18 32003000300038002D00300037002D0031003800
+08:42:04 300038003A00340032003A0030003400
+2008-07-18 08:42:04 32003000300038002D00300037002D00310038002000300038003A00340032003A0030003400
+DROP TABLE t1;
+SET timestamp=0;
+#
+# Bug#33073 Character sets: ordering fails with utf32
+#
+SET NAMES utf8, collation_connection=utf16le_general_ci;
+CREATE TABLE t1 AS SELECT REPEAT('a',2) as s1 LIMIT 0;
+SHOW CREATE TABLE t1;
+Table Create Table
+t1 CREATE TABLE `t1` (
+ `s1` varchar(2) CHARACTER SET utf16le NOT NULL DEFAULT ''
+) ENGINE=MyISAM DEFAULT CHARSET=latin1
+INSERT INTO t1 VALUES ('ab'),('AE'),('ab'),('AE');
+SELECT * FROM t1 ORDER BY s1;
+s1
+ab
+ab
+AE
+AE
+SET max_sort_length=4;
+SELECT * FROM t1 ORDER BY s1;
+s1
+ab
+ab
+AE
+AE
+DROP TABLE t1;
+SET max_sort_length=DEFAULT;
+#
+# Bug#52520 Difference in tinytext utf column metadata
+#
+CREATE TABLE t1 (
+s1 TINYTEXT CHARACTER SET utf16le,
+s2 TEXT CHARACTER SET utf16le,
+s3 MEDIUMTEXT CHARACTER SET utf16le,
+s4 LONGTEXT CHARACTER SET utf16le
+);
+SET NAMES utf8, @@character_set_results=NULL;
+SELECT *, HEX(s1) FROM t1;
+Catalog Database Table Table_alias Column Column_alias Type Length Max length Is_null Flags Decimals Charsetnr
+def test t1 t1 s1 s1 252 255 0 Y 16 0 56
+def test t1 t1 s2 s2 252 65535 0 Y 16 0 56
+def test t1 t1 s3 s3 252 16777215 0 Y 16 0 56
+def test t1 t1 s4 s4 252 4294967295 0 Y 16 0 56
+def HEX(s1) 253 6120 0 Y 0 0 33
+s1 s2 s3 s4 HEX(s1)
+SET NAMES latin1;
+SELECT *, HEX(s1) FROM t1;
+Catalog Database Table Table_alias Column Column_alias Type Length Max length Is_null Flags Decimals Charsetnr
+def test t1 t1 s1 s1 252 127 0 Y 16 0 8
+def test t1 t1 s2 s2 252 32767 0 Y 16 0 8
+def test t1 t1 s3 s3 252 8388607 0 Y 16 0 8
+def test t1 t1 s4 s4 252 2147483647 0 Y 16 0 8
+def HEX(s1) 253 2040 0 Y 0 0 8
+s1 s2 s3 s4 HEX(s1)
+SET NAMES utf8;
+SELECT *, HEX(s1) FROM t1;
+Catalog Database Table Table_alias Column Column_alias Type Length Max length Is_null Flags Decimals Charsetnr
+def test t1 t1 s1 s1 252 381 0 Y 16 0 33
+def test t1 t1 s2 s2 252 98301 0 Y 16 0 33
+def test t1 t1 s3 s3 252 25165821 0 Y 16 0 33
+def test t1 t1 s4 s4 252 4294967295 0 Y 16 0 33
+def HEX(s1) 253 6120 0 Y 0 0 33
+s1 s2 s3 s4 HEX(s1)
+CREATE TABLE t2 AS SELECT CONCAT(s1) FROM t1;
+SHOW CREATE TABLE t2;
+Table Create Table
+t2 CREATE TABLE `t2` (
+ `CONCAT(s1)` varchar(255) CHARACTER SET utf16le DEFAULT NULL
+) ENGINE=MyISAM DEFAULT CHARSET=latin1
+DROP TABLE t1, t2;
+#
+# WL#3664 WEIGHT_STRING
+#
+SET NAMES utf8, collation_connection=utf16le_general_ci;
+select @@collation_connection;
+@@collation_connection
+utf16le_general_ci
+select hex(weight_string('a'));
+hex(weight_string('a'))
+0041
+select hex(weight_string('A'));
+hex(weight_string('A'))
+0041
+select hex(weight_string('abc'));
+hex(weight_string('abc'))
+004100420043
+select hex(weight_string('abc' as char(2)));
+hex(weight_string('abc' as char(2)))
+00410042
+select hex(weight_string('abc' as char(3)));
+hex(weight_string('abc' as char(3)))
+004100420043
+select hex(weight_string('abc' as char(5)));
+hex(weight_string('abc' as char(5)))
+00410042004300200020
+select hex(weight_string('abc', 1, 2, 0xC0));
+hex(weight_string('abc', 1, 2, 0xC0))
+00
+select hex(weight_string('abc', 2, 2, 0xC0));
+hex(weight_string('abc', 2, 2, 0xC0))
+0041
+select hex(weight_string('abc', 3, 2, 0xC0));
+hex(weight_string('abc', 3, 2, 0xC0))
+004100
+select hex(weight_string('abc', 4, 2, 0xC0));
+hex(weight_string('abc', 4, 2, 0xC0))
+00410042
+select hex(weight_string('abc', 5, 2, 0xC0));
+hex(weight_string('abc', 5, 2, 0xC0))
+0041004200
+select hex(weight_string('abc',25, 2, 0xC0));
+hex(weight_string('abc',25, 2, 0xC0))
+00410042002000200020002000200020002000200020002000
+select hex(weight_string('abc', 1, 3, 0xC0));
+hex(weight_string('abc', 1, 3, 0xC0))
+00
+select hex(weight_string('abc', 2, 3, 0xC0));
+hex(weight_string('abc', 2, 3, 0xC0))
+0041
+select hex(weight_string('abc', 3, 3, 0xC0));
+hex(weight_string('abc', 3, 3, 0xC0))
+004100
+select hex(weight_string('abc', 4, 3, 0xC0));
+hex(weight_string('abc', 4, 3, 0xC0))
+00410042
+select hex(weight_string('abc', 5, 3, 0xC0));
+hex(weight_string('abc', 5, 3, 0xC0))
+0041004200
+select hex(weight_string('abc',25, 3, 0xC0));
+hex(weight_string('abc',25, 3, 0xC0))
+00410042004300200020002000200020002000200020002000
+select hex(weight_string('abc', 1, 4, 0xC0));
+hex(weight_string('abc', 1, 4, 0xC0))
+00
+select hex(weight_string('abc', 2, 4, 0xC0));
+hex(weight_string('abc', 2, 4, 0xC0))
+0041
+select hex(weight_string('abc', 3, 4, 0xC0));
+hex(weight_string('abc', 3, 4, 0xC0))
+004100
+select hex(weight_string('abc', 4, 4, 0xC0));
+hex(weight_string('abc', 4, 4, 0xC0))
+00410042
+select hex(weight_string('abc', 5, 4, 0xC0));
+hex(weight_string('abc', 5, 4, 0xC0))
+0041004200
+select hex(weight_string('abc',25, 4, 0xC0));
+hex(weight_string('abc',25, 4, 0xC0))
+00410042004300200020002000200020002000200020002000
+select @@collation_connection;
+@@collation_connection
+utf16le_general_ci
+select hex(weight_string(cast(_latin1 0x80 as char)));
+hex(weight_string(cast(_latin1 0x80 as char)))
+20AC
+select hex(weight_string(cast(_latin1 0x808080 as char)));
+hex(weight_string(cast(_latin1 0x808080 as char)))
+20AC20AC20AC
+select hex(weight_string(cast(_latin1 0x808080 as char) as char(2)));
+hex(weight_string(cast(_latin1 0x808080 as char) as char(2)))
+20AC20AC
+select hex(weight_string(cast(_latin1 0x808080 as char) as char(3)));
+hex(weight_string(cast(_latin1 0x808080 as char) as char(3)))
+20AC20AC20AC
+select hex(weight_string(cast(_latin1 0x808080 as char) as char(5)));
+hex(weight_string(cast(_latin1 0x808080 as char) as char(5)))
+20AC20AC20AC00200020
+select hex(weight_string(cast(_latin1 0x808080 as char), 1, 2, 0xC0));
+hex(weight_string(cast(_latin1 0x808080 as char), 1, 2, 0xC0))
+20
+select hex(weight_string(cast(_latin1 0x808080 as char), 2, 2, 0xC0));
+hex(weight_string(cast(_latin1 0x808080 as char), 2, 2, 0xC0))
+20AC
+select hex(weight_string(cast(_latin1 0x808080 as char), 3, 2, 0xC0));
+hex(weight_string(cast(_latin1 0x808080 as char), 3, 2, 0xC0))
+20AC20
+select hex(weight_string(cast(_latin1 0x808080 as char), 4, 2, 0xC0));
+hex(weight_string(cast(_latin1 0x808080 as char), 4, 2, 0xC0))
+20AC20AC
+select hex(weight_string(cast(_latin1 0x808080 as char), 5, 2, 0xC0));
+hex(weight_string(cast(_latin1 0x808080 as char), 5, 2, 0xC0))
+20AC20AC00
+select hex(weight_string(cast(_latin1 0x808080 as char),25, 2, 0xC0));
+hex(weight_string(cast(_latin1 0x808080 as char),25, 2, 0xC0))
+20AC20AC002000200020002000200020002000200020002000
+select hex(weight_string(cast(_latin1 0x808080 as char), 1, 3, 0xC0));
+hex(weight_string(cast(_latin1 0x808080 as char), 1, 3, 0xC0))
+20
+select hex(weight_string(cast(_latin1 0x808080 as char), 2, 3, 0xC0));
+hex(weight_string(cast(_latin1 0x808080 as char), 2, 3, 0xC0))
+20AC
+select hex(weight_string(cast(_latin1 0x808080 as char), 3, 3, 0xC0));
+hex(weight_string(cast(_latin1 0x808080 as char), 3, 3, 0xC0))
+20AC20
+select hex(weight_string(cast(_latin1 0x808080 as char), 4, 3, 0xC0));
+hex(weight_string(cast(_latin1 0x808080 as char), 4, 3, 0xC0))
+20AC20AC
+select hex(weight_string(cast(_latin1 0x808080 as char), 5, 3, 0xC0));
+hex(weight_string(cast(_latin1 0x808080 as char), 5, 3, 0xC0))
+20AC20AC20
+select hex(weight_string(cast(_latin1 0x808080 as char),25, 3, 0xC0));
+hex(weight_string(cast(_latin1 0x808080 as char),25, 3, 0xC0))
+20AC20AC20AC00200020002000200020002000200020002000
+select hex(weight_string(cast(_latin1 0x808080 as char), 1, 4, 0xC0));
+hex(weight_string(cast(_latin1 0x808080 as char), 1, 4, 0xC0))
+20
+select hex(weight_string(cast(_latin1 0x808080 as char), 2, 4, 0xC0));
+hex(weight_string(cast(_latin1 0x808080 as char), 2, 4, 0xC0))
+20AC
+select hex(weight_string(cast(_latin1 0x808080 as char), 3, 4, 0xC0));
+hex(weight_string(cast(_latin1 0x808080 as char), 3, 4, 0xC0))
+20AC20
+select hex(weight_string(cast(_latin1 0x808080 as char), 4, 4, 0xC0));
+hex(weight_string(cast(_latin1 0x808080 as char), 4, 4, 0xC0))
+20AC20AC
+select hex(weight_string(cast(_latin1 0x808080 as char), 5, 4, 0xC0));
+hex(weight_string(cast(_latin1 0x808080 as char), 5, 4, 0xC0))
+20AC20AC20
+select hex(weight_string(cast(_latin1 0x808080 as char),25, 4, 0xC0));
+hex(weight_string(cast(_latin1 0x808080 as char),25, 4, 0xC0))
+20AC20AC20AC00200020002000200020002000200020002000
+SELECT HEX(WEIGHT_STRING(_utf16le 0x00D800DC));
+HEX(WEIGHT_STRING(_utf16le 0x00D800DC))
+FFFD
+SELECT HEX(WEIGHT_STRING(_utf16le 0x00D801DC));
+HEX(WEIGHT_STRING(_utf16le 0x00D801DC))
+FFFD
+select @@collation_connection;
+@@collation_connection
+utf16le_general_ci
+select hex(weight_string('a' LEVEL 1));
+hex(weight_string('a' LEVEL 1))
+0041
+select hex(weight_string('A' LEVEL 1));
+hex(weight_string('A' LEVEL 1))
+0041
+select hex(weight_string('abc' LEVEL 1));
+hex(weight_string('abc' LEVEL 1))
+004100420043
+select hex(weight_string('abc' as char(2) LEVEL 1));
+hex(weight_string('abc' as char(2) LEVEL 1))
+00410042
+select hex(weight_string('abc' as char(3) LEVEL 1));
+hex(weight_string('abc' as char(3) LEVEL 1))
+004100420043
+select hex(weight_string('abc' as char(5) LEVEL 1));
+hex(weight_string('abc' as char(5) LEVEL 1))
+00410042004300200020
+select hex(weight_string('abc' as char(5) LEVEL 1 REVERSE));
+hex(weight_string('abc' as char(5) LEVEL 1 REVERSE))
+20002000430042004100
+select hex(weight_string('abc' as char(5) LEVEL 1 DESC));
+hex(weight_string('abc' as char(5) LEVEL 1 DESC))
+FFBEFFBDFFBCFFDFFFDF
+select hex(weight_string('abc' as char(5) LEVEL 1 DESC REVERSE));
+hex(weight_string('abc' as char(5) LEVEL 1 DESC REVERSE))
+DFFFDFFFBCFFBDFFBEFF
+SET NAMES utf8, collation_connection=utf16le_bin;
+select @@collation_connection;
+@@collation_connection
+utf16le_bin
+select hex(weight_string('a'));
+hex(weight_string('a'))
+000061
+select hex(weight_string('A'));
+hex(weight_string('A'))
+000041
+select hex(weight_string('abc'));
+hex(weight_string('abc'))
+000061000062000063
+select hex(weight_string('abc' as char(2)));
+hex(weight_string('abc' as char(2)))
+000061000062
+select hex(weight_string('abc' as char(3)));
+hex(weight_string('abc' as char(3)))
+000061000062000063
+select hex(weight_string('abc' as char(5)));
+hex(weight_string('abc' as char(5)))
+000061000062000063000020000020
+select hex(weight_string('abc', 1, 2, 0xC0));
+hex(weight_string('abc', 1, 2, 0xC0))
+00
+select hex(weight_string('abc', 2, 2, 0xC0));
+hex(weight_string('abc', 2, 2, 0xC0))
+0000
+select hex(weight_string('abc', 3, 2, 0xC0));
+hex(weight_string('abc', 3, 2, 0xC0))
+000061
+select hex(weight_string('abc', 4, 2, 0xC0));
+hex(weight_string('abc', 4, 2, 0xC0))
+00006100
+select hex(weight_string('abc', 5, 2, 0xC0));
+hex(weight_string('abc', 5, 2, 0xC0))
+0000610000
+select hex(weight_string('abc',25, 2, 0xC0));
+hex(weight_string('abc',25, 2, 0xC0))
+00006100006200002000002000002000002000002000002000
+select hex(weight_string('abc', 1, 3, 0xC0));
+hex(weight_string('abc', 1, 3, 0xC0))
+00
+select hex(weight_string('abc', 2, 3, 0xC0));
+hex(weight_string('abc', 2, 3, 0xC0))
+0000
+select hex(weight_string('abc', 3, 3, 0xC0));
+hex(weight_string('abc', 3, 3, 0xC0))
+000061
+select hex(weight_string('abc', 4, 3, 0xC0));
+hex(weight_string('abc', 4, 3, 0xC0))
+00006100
+select hex(weight_string('abc', 5, 3, 0xC0));
+hex(weight_string('abc', 5, 3, 0xC0))
+0000610000
+select hex(weight_string('abc',25, 3, 0xC0));
+hex(weight_string('abc',25, 3, 0xC0))
+00006100006200006300002000002000002000002000002000
+select hex(weight_string('abc', 1, 4, 0xC0));
+hex(weight_string('abc', 1, 4, 0xC0))
+00
+select hex(weight_string('abc', 2, 4, 0xC0));
+hex(weight_string('abc', 2, 4, 0xC0))
+0000
+select hex(weight_string('abc', 3, 4, 0xC0));
+hex(weight_string('abc', 3, 4, 0xC0))
+000061
+select hex(weight_string('abc', 4, 4, 0xC0));
+hex(weight_string('abc', 4, 4, 0xC0))
+00006100
+select hex(weight_string('abc', 5, 4, 0xC0));
+hex(weight_string('abc', 5, 4, 0xC0))
+0000610000
+select hex(weight_string('abc',25, 4, 0xC0));
+hex(weight_string('abc',25, 4, 0xC0))
+00006100006200006300002000002000002000002000002000
+select @@collation_connection;
+@@collation_connection
+utf16le_bin
+select hex(weight_string(cast(_latin1 0x80 as char)));
+hex(weight_string(cast(_latin1 0x80 as char)))
+0020AC
+select hex(weight_string(cast(_latin1 0x808080 as char)));
+hex(weight_string(cast(_latin1 0x808080 as char)))
+0020AC0020AC0020AC
+select hex(weight_string(cast(_latin1 0x808080 as char) as char(2)));
+hex(weight_string(cast(_latin1 0x808080 as char) as char(2)))
+0020AC0020AC
+select hex(weight_string(cast(_latin1 0x808080 as char) as char(3)));
+hex(weight_string(cast(_latin1 0x808080 as char) as char(3)))
+0020AC0020AC0020AC
+select hex(weight_string(cast(_latin1 0x808080 as char) as char(5)));
+hex(weight_string(cast(_latin1 0x808080 as char) as char(5)))
+0020AC0020AC0020AC000020000020
+select hex(weight_string(cast(_latin1 0x808080 as char), 1, 2, 0xC0));
+hex(weight_string(cast(_latin1 0x808080 as char), 1, 2, 0xC0))
+00
+select hex(weight_string(cast(_latin1 0x808080 as char), 2, 2, 0xC0));
+hex(weight_string(cast(_latin1 0x808080 as char), 2, 2, 0xC0))
+0020
+select hex(weight_string(cast(_latin1 0x808080 as char), 3, 2, 0xC0));
+hex(weight_string(cast(_latin1 0x808080 as char), 3, 2, 0xC0))
+0020AC
+select hex(weight_string(cast(_latin1 0x808080 as char), 4, 2, 0xC0));
+hex(weight_string(cast(_latin1 0x808080 as char), 4, 2, 0xC0))
+0020AC00
+select hex(weight_string(cast(_latin1 0x808080 as char), 5, 2, 0xC0));
+hex(weight_string(cast(_latin1 0x808080 as char), 5, 2, 0xC0))
+0020AC0020
+select hex(weight_string(cast(_latin1 0x808080 as char),25, 2, 0xC0));
+hex(weight_string(cast(_latin1 0x808080 as char),25, 2, 0xC0))
+0020AC0020AC00002000002000002000002000002000002000
+select hex(weight_string(cast(_latin1 0x808080 as char), 1, 3, 0xC0));
+hex(weight_string(cast(_latin1 0x808080 as char), 1, 3, 0xC0))
+00
+select hex(weight_string(cast(_latin1 0x808080 as char), 2, 3, 0xC0));
+hex(weight_string(cast(_latin1 0x808080 as char), 2, 3, 0xC0))
+0020
+select hex(weight_string(cast(_latin1 0x808080 as char), 3, 3, 0xC0));
+hex(weight_string(cast(_latin1 0x808080 as char), 3, 3, 0xC0))
+0020AC
+select hex(weight_string(cast(_latin1 0x808080 as char), 4, 3, 0xC0));
+hex(weight_string(cast(_latin1 0x808080 as char), 4, 3, 0xC0))
+0020AC00
+select hex(weight_string(cast(_latin1 0x808080 as char), 5, 3, 0xC0));
+hex(weight_string(cast(_latin1 0x808080 as char), 5, 3, 0xC0))
+0020AC0020
+select hex(weight_string(cast(_latin1 0x808080 as char),25, 3, 0xC0));
+hex(weight_string(cast(_latin1 0x808080 as char),25, 3, 0xC0))
+0020AC0020AC0020AC00002000002000002000002000002000
+select hex(weight_string(cast(_latin1 0x808080 as char), 1, 4, 0xC0));
+hex(weight_string(cast(_latin1 0x808080 as char), 1, 4, 0xC0))
+00
+select hex(weight_string(cast(_latin1 0x808080 as char), 2, 4, 0xC0));
+hex(weight_string(cast(_latin1 0x808080 as char), 2, 4, 0xC0))
+0020
+select hex(weight_string(cast(_latin1 0x808080 as char), 3, 4, 0xC0));
+hex(weight_string(cast(_latin1 0x808080 as char), 3, 4, 0xC0))
+0020AC
+select hex(weight_string(cast(_latin1 0x808080 as char), 4, 4, 0xC0));
+hex(weight_string(cast(_latin1 0x808080 as char), 4, 4, 0xC0))
+0020AC00
+select hex(weight_string(cast(_latin1 0x808080 as char), 5, 4, 0xC0));
+hex(weight_string(cast(_latin1 0x808080 as char), 5, 4, 0xC0))
+0020AC0020
+select hex(weight_string(cast(_latin1 0x808080 as char),25, 4, 0xC0));
+hex(weight_string(cast(_latin1 0x808080 as char),25, 4, 0xC0))
+0020AC0020AC0020AC00002000002000002000002000002000
+select @@collation_connection;
+@@collation_connection
+utf16le_bin
+select hex(weight_string('a' LEVEL 1));
+hex(weight_string('a' LEVEL 1))
+000061
+select hex(weight_string('A' LEVEL 1));
+hex(weight_string('A' LEVEL 1))
+000041
+select hex(weight_string('abc' LEVEL 1));
+hex(weight_string('abc' LEVEL 1))
+000061000062000063
+select hex(weight_string('abc' as char(2) LEVEL 1));
+hex(weight_string('abc' as char(2) LEVEL 1))
+000061000062
+select hex(weight_string('abc' as char(3) LEVEL 1));
+hex(weight_string('abc' as char(3) LEVEL 1))
+000061000062000063
+select hex(weight_string('abc' as char(5) LEVEL 1));
+hex(weight_string('abc' as char(5) LEVEL 1))
+000061000062000063000020000020
+select hex(weight_string('abc' as char(5) LEVEL 1 REVERSE));
+hex(weight_string('abc' as char(5) LEVEL 1 REVERSE))
+200000200000630000620000610000
+select hex(weight_string('abc' as char(5) LEVEL 1 DESC));
+hex(weight_string('abc' as char(5) LEVEL 1 DESC))
+FFFF9EFFFF9DFFFF9CFFFFDFFFFFDF
+select hex(weight_string('abc' as char(5) LEVEL 1 DESC REVERSE));
+hex(weight_string('abc' as char(5) LEVEL 1 DESC REVERSE))
+DFFFFFDFFFFF9CFFFF9DFFFF9EFFFF
+#
+# End of 5.6 tests
+#
=== added file 'mysql-test/t/ctype_utf16le.test'
--- a/mysql-test/t/ctype_utf16le.test 1970-01-01 00:00:00 +0000
+++ b/mysql-test/t/ctype_utf16le.test 2010-12-03 10:10:05 +0000
@@ -0,0 +1,729 @@
+-- source include/have_ucs2.inc
+-- source include/have_utf16.inc
+-- source include/have_utf32.inc
+-- source include/have_utf8mb4.inc
+
+
+--disable_warnings
+DROP TABLE IF EXISTS t1;
+--enable_warnings
+
+--echo #
+--echo # Start of 5.6 tests
+--echo #
+
+SET NAMES utf8, collation_connection=utf16le_general_ci;
+SELECT HEX('a'), HEX('a ');
+-- source include/endspace.inc
+
+
+--echo #
+--echo # Check that incomplete utf16le characters in HEX notation
+--echo # are left-padded with zeros
+--echo #
+SELECT HEX(_utf16le 0x44);
+SELECT HEX(_utf16le 0x3344);
+SELECT HEX(_utf16le 0x113344);
+
+--echo #
+--echo # Check that 0x20 is only trimmed when it is
+--echo # a part of real SPACE character, not just a part
+--echo # of a multibyte sequence.
+--echo # Note, CYRILLIC LETTER ER is used as an example, which
+--echo # is stored as 0x0420 in utf16le, thus contains 0x20 in the
+--echo # low byte. The second character is THREE-PER-M, U+2004,
+--echo # which contains 0x20 in the high byte.
+--echo #
+
+CREATE TABLE t1 (word VARCHAR(64), word2 CHAR(64)) CHARACTER SET utf16le;
+INSERT INTO t1 VALUES (_koi8r 0xF2, _koi8r 0xF2), (_ucs2 X'2004',_ucs2 X'2004');
+SELECT HEX(word) FROM t1 ORDER BY word;
+SELECT HEX(word2) FROM t1 ORDER BY word2;
+DELETE FROM t1;
+
+--echo #
+--echo # Check that real spaces are correctly trimmed.
+--echo #
+INSERT INTO t1 VALUES (_ucs2 X'042000200020', _ucs2 X'042000200020');
+INSERT INTO t1 VALUES (_ucs2 X'200400200020', _ucs2 X'200400200020');
+SELECT HEX(word) FROM t1 ORDER BY word;
+SELECT HEX(word2) FROM t1 ORDER BY word2;
+DROP TABLE t1;
+
+
+--echo #
+--echo # Check LPAD/RPAD
+--echo #
+CREATE TABLE t1 (a VARCHAR(10), pad INT, b VARCHAR(10)) CHARACTER SET utf16le;
+INSERT INTO t1 VALUES (_ucs2 X'0420', 10, _ucs2 X'0421');
+INSERT INTO t1 VALUES (_ucs2 X'0420', 10, _ucs2 X'04210422');
+INSERT INTO t1 VALUES (_ucs2 X'0420', 10, _ucs2 X'042104220423');
+INSERT INTO t1 VALUES (_ucs2 X'0420042104220423042404250426042704280429042A042B',10,_ucs2 X'042104220423');
+INSERT INTO t1 VALUES (_utf32 X'010000', 10, _ucs2 X'0421');
+INSERT INTO t1 VALUES (_ucs2 X'0421', 10, _utf32 X'010000');
+SELECT a, pad, b, LPAD(a, pad, b), HEX(LPAD(a, pad, b)) FROM t1;
+DROP TABLE t1;
+
+
+CREATE TABLE t1 SELECT
+LPAD(_utf16le X'2004',10,_utf16le X'2104') l,
+RPAD(_utf16le X'2004',10,_utf16le X'2104') r;
+SHOW CREATE TABLE t1;
+SELECT HEX(l), HEX(r) FROM t1;
+DROP TABLE t1;
+
+CREATE TABLE t1 (f1 CHAR(30));
+INSERT INTO t1 VALUES ("103000"), ("22720000"), ("3401200"), ("78000");
+SELECT LPAD(f1, 12, "-o-/") FROM t1;
+DROP TABLE t1;
+
+--echo #
+--echo # Testing LIKE
+--echo #
+
+SET NAMES utf8, collation_connection=utf16le_general_ci;
+--source include/ctype_like.inc
+
+CREATE TABLE t1 (a VARCHAR(10) CHARACTER SET utf16le);
+INSERT INTO t1 VALUES ('фыва'),('Фыва'),('фЫва'),('ES ('фывапролдж'),('Фывапролдж'),('фЫвапролдж'),('фыВапролдж');
+INSERT INTO t1 VALUES ('фывАпролдж'),('фываПрол+INSERT INTO t1 VALUES ('фывапроЛдж'),('фывЛДЖ');
+SELECT * FROM t1 WHERE a LIKE '%фЫва%' ORDER BY BINARY a;
+SELECT * FROM t1 WHERE a LIKE '%фЫв%' ORDER BY BINARY a;
+SELECT * FROM t1 WHERE a LIKE 'фЫфЫва%' COLLATE utf16le_bin ORDER BY BINARY a;
+DROP TABLE t1;
+
+CREATE TABLE t1 (word VARCHAR(64) NOT NULL, PRIMARY KEY (word))
+ENGINE=MyISAM CHARACTER SET utf16le;
+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'; # "c%"
+SELECT * FROM t1 WHERE word LIKE _ucs2 x'00630061005F'; # "ca_"
+DROP TABLE t1;
+
+
+--echo #
+--echo # Check that INSERT() works fine.
+--echo # This invokes charpos() function.
+--echo #
+CREATE TABLE t1 (
+ a VARCHAR(10) CHARACTER SET utf16le,
+ b VARCHAR(10) CHARACTER SET utf16le);
+INSERT INTO t1 VALUES ('abc', 'def');
+SELECT INSERT(a, 10, 2, b) FROM t1;
+SELECT INSERT(a, 1, 2, b) FROM t1;
+DROP TABLE t1;
+
+--echo #
+--echo # Bug#1264
+--echo #
+# 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 utf8, collation_connection=utf16le_general_ci;
+
+--echo #
+--echo # Two fields, index
+--echo #
+
+CREATE TABLE t1 (
+ word VARCHAR(64),
+ bar INT(11) DEFAULT 0,
+ PRIMARY KEY (word))
+ ENGINE=MyISAM
+ CHARSET utf16le
+ COLLATE utf16le_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;
+
+
+--echo #
+--echo # One field, index
+--echo #
+
+CREATE TABLE t1 (
+ word VARCHAR(64) ,
+ PRIMARY KEY (word))
+ ENGINE=MyISAM
+ CHARSET utf16le
+ COLLATE utf16le_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;
+
+
+--echo #
+--echo # Two fields, no index
+--echo #
+
+CREATE TABLE t1 (
+ word TEXT,
+ bar INT(11) AUTO_INCREMENT,
+ PRIMARY KEY (bar))
+ ENGINE=MyISAM
+ CHARSET utf16le
+ COLLATE utf16le_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;
+
+--echo #
+--echo # END OF Bug 1264 test
+--echo #
+
+
+--echo #
+--echo # Check alignment for from-binary-conversion with CAST and CONVERT
+--echo #
+SELECT HEX(CAST(0xAA as char CHARACTER SET utf16le));
+SELECT HEX(CONVERT(0xAA USING utf16le));
+
+--echo #
+--echo # Check alignment for string types
+--echo #
+CREATE TABLE t1 (a CHAR(10) CHARACTER SET utf16le);
+INSERT INTO t1 VALUES (0x1),(0x11),(0x111),(0x1111),(0x11111);
+SELECT HEX(a) FROM t1;
+DROP TABLE t1;
+
+CREATE TABLE t1 (a VARCHAR(10) CHARACTER SET utf16le);
+INSERT INTO t1 VALUES (0x1),(0x11),(0x111),(0x1111),(0x11111);
+SELECT HEX(a) FROM t1;
+DROP TABLE t1;
+
+CREATE TABLE t1 (a TEXT CHARACTER SET utf16le);
+INSERT INTO t1 VALUES (0x1),(0x11),(0x111),(0x1111),(0x11111);
+SELECT HEX(a) FROM t1;
+DROP TABLE t1;
+
+CREATE TABLE t1 (a MEDIUMTEXT CHARACTER SET utf16le);
+INSERT INTO t1 VALUES (0x1),(0x11),(0x111),(0x1111),(0x11111);
+SELECT HEX(a) FROM t1;
+DROP TABLE t1;
+
+CREATE TABLE t1 (a LONGTEXT CHARACTER SET utf16le);
+INSERT INTO t1 VALUES (0x1),(0x11),(0x111),(0x1111),(0x11111);
+SELECT HEX(a) FROM t1;
+DROP TABLE t1;
+
+
+--echo #
+--echo # Bug#5081 : UCS2 fields are filled with '0x2020'
+--echo # after extending field length
+--echo #
+
+CREATE TABLE t1(a CHAR(1)) DEFAULT CHARSET utf16le;
+INSERT INTO t1 VALUES ('a'),('b'),('c');
+ALTER TABLE t1 MODIFY a CHAR(5);
+SELECT a, HEX(a) FROM t1;
+DROP TABLE t1;
+
+--echo #
+--echo # Check prepare statement from an UTF16 string
+--echo #
+SET NAMES latin1;
+SET @ivar= 1234;
+SET @str1 = 'SELECT ?';
+SET @str2 = CONVERT(@str1 USING utf16le);
+PREPARE stmt1 FROM @str2;
+EXECUTE stmt1 USING @ivar;
+
+--echo #
+--echo # Check that utf16le works with ENUM and SET type
+--echo #
+SET NAMES utf8, collation_connection=utf16le_general_ci;
+CREATE TABLE t1 (a ENUM('x','y','z') CHARACTER SET utf16le);
+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 utf16le;
+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 utf16le);
+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;
+
+--echo #
+--echo # Bug#7302 UCS2 data in ENUM fields get truncated when new column is added
+--echo #
+CREATE TABLE t1(a ENUM('a','b','c')) DEFAULT CHARACTER SET utf16le;
+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 NAMES utf8, collation_connection='utf16le_general_ci';
+-- source include/ctype_filesort.inc
+-- source include/ctype_like_escape.inc
+SET NAMES utf8, collation_connection='utf16le_bin';
+-- source include/ctype_filesort.inc
+-- source include/ctype_filesort2.inc
+-- source include/ctype_like_escape.inc
+
+--echo #
+--echo # Bug#10344 Some string functions fail for UCS2
+--echo #
+CREATE TABLE t1 (a VARCHAR(10) CHARACTER SET utf16le, pos INT);
+INSERT INTO t1 VALUES (_ucs2 0x00e400e50068,1);
+INSERT INTO t1 VALUES (_ucs2 0x00e400e50068,2);
+INSERT INTO t1 VALUES (_ucs2 0x00e400e50068,3);
+INSERT INTO t1 VALUES (_ucs2 0x00e400e50068,-1);
+INSERT INTO t1 VALUES (_ucs2 0x00e400e50068,-2);
+INSERT INTO t1 VALUES (_ucs2 0x00e400e50068,-3);
+INSERT INTO t1 VALUES (_utf32 0x000000e4000000e500010000, 1);
+INSERT INTO t1 VALUES (_utf32 0x000000e4000000e500010000, 2);
+INSERT INTO t1 VALUES (_utf32 0x000000e4000000e500010000, 3);
+INSERT INTO t1 VALUES (_utf32 0x000000e4000000e500010000, -1);
+INSERT INTO t1 VALUES (_utf32 0x000000e4000000e500010000, -2);
+INSERT INTO t1 VALUES (_utf32 0x000000e4000000e500010000, -3);
+SELECT HEX(SUBSTR(a, pos)), SUBSTR(a, pos) FROM t1;
+DROP TABLE t1;
+
+SET NAMES utf8, collation_connection=utf16le_general_ci;
+
+--echo #
+--echo # Bug#9442 Set parameter make query fail if column CHARACTER SET is UCS2
+--echo #
+CREATE TABLE t1 (utext VARCHAR(20) CHARACTER SET utf16le);
+INSERT INTO t1 VALUES ("lily");
+INSERT INTO t1 VALUES ("river");
+PREPARE stmt FROM 'SELECT utext FROM t1 where utext like ?';
+SET @param1='%%';
+EXECUTE stmt USING @param1;
+EXECUTE stmt USING @param1;
+SELECT utext FROM t1 where utext like '%%';
+DROP TABLE t1;
+DEALLOCATE PREPARE stmt;
+
+--echo #
+--echo # Bug#22052 Trailing spaces are not removed FROM UNICODE fields in an index
+--echo #
+CREATE TABLE t1 (
+ a CHAR(10) CHARACTER SET utf16le NOT NULL,
+ INDEX a (a)
+) engine=myisam;
+INSERT INTO t1 VALUES (REPEAT(_ucs2 0x201f, 10));
+INSERT INTO t1 VALUES (REPEAT(_ucs2 0x2020, 10));
+INSERT INTO t1 VALUES (REPEAT(_ucs2 0x2021, 10));
+--echo # make sure "index read" is used
+explain SELECT HEX(a) FROM t1 ORDER BY a;
+SELECT HEX(a) FROM t1 ORDER BY a;
+ALTER TABLE t1 DROP INDEX a;
+SELECT HEX(a) FROM t1 ORDER BY a;
+DROP TABLE t1;
+
+--echo #
+--echo # Bug #20108: corrupted default enum value for a ucs2 field
+--echo #
+CREATE TABLE t1 (
+ status ENUM('active','passive') CHARACTER SET utf16le COLLATE utf16le_general_ci
+ NOT NULL DEFAULT 'passive'
+);
+SHOW CREATE TABLE t1;
+ALTER TABLE t1 ADD a int NOT NULL AFTER status;
+SHOW CREATE TABLE t1;
+DROP TABLE t1;
+
+
+--echo #
+--echo # Conversion FROM an UTF16LE string to a decimal column
+--echo #
+CREATE TABLE t1 (a VARCHAR(64) CHARACTER SET utf16le, b DECIMAL(10,3));
+INSERT INTO t1 VALUES ("1.1", 0), ("2.1", 0);
+UPDATE t1 set b=a;
+SELECT *, HEX(a) FROM t1;
+DROP TABLE t1;
+
+--echo #
+--echo # Bug#9442 Set parameter make query fail if column CHARACTER SET is UCS2
+--echo #
+CREATE TABLE t1 (utext VARCHAR(20) CHARACTER SET utf16le);
+INSERT INTO t1 VALUES ("lily");
+INSERT INTO t1 VALUES ("river");
+PREPARE stmt FROM 'SELECT utext FROM t1 where utext like ?';
+SET @param1='%%';
+EXECUTE stmt USING @param1;
+EXECUTE stmt USING @param1;
+SELECT utext FROM t1 where utext like '%%';
+DROP TABLE t1;
+DEALLOCATE PREPARE stmt;
+
+--echo #
+--echo # Bug#22638 SOUNDEX broken for international characters
+--echo #
+SET NAMES utf8, collation_connection=utf16le_general_ci;
+SELECT SOUNDEX(''),SOUNDEX('he'),SOUNDEX('hello all folks'),SOUNDEX('#3556 in bugdb');
+SELECT HEX(SOUNDEX('')),HEX(SOUNDEX('he')),HEX(SOUNDEX('hello all folks')),HEX(SOUNDEX('#3556 in bugdb'));
+SELECT 'mood' sounds like 'mud';
+--echo # Cyrillic A, BE, VE
+SELECT HEX(SOUNDEX(_utf16le 0x041004110412));
+--echo # Make sure that "U+00BF INVERTED QUESTION MARK" is not considered as letter
+SELECT HEX(SOUNDEX(_utf16le 0x00BF00C0));
+
+--echo #
+--echo # Bug#14290: character_maximum_length for text fields
+--echo #
+CREATE TABLE t1(a BLOB, b TEXT CHARSET utf16le);
+SELECT data_type, character_octet_length, character_maximum_length
+ FROM information_schema.columns where table_name='t1';
+DROP TABLE t1;
+
+
+SET NAMES utf8, collation_connection=utf16le_general_ci;
+
+--echo #
+--echo # Testing cs->coll->instr()
+--echo #
+SELECT POSITION('bb' IN 'abba');
+
+--echo #
+--echo # Testing cs->coll->hash_sort()
+--echo #
+SET NAMES utf8, collation_connection=utf16le_bin;
+--source include/ctype_heap.inc
+SET NAMES utf8, collation_connection=utf16le_general_ci;
+--source include/ctype_heap.inc
+
+--echo #
+--echo # Testing cs->cset->numchars()
+--echo #
+SELECT CHAR_LENGTH('abcd'), OCTET_LENGTH('abcd');
+SELECT CHAR_LENGTH(_utf16le 0x00D800DC), OCTET_LENGTH(_utf16le 0x00D800DC);
+SELECT CHAR_LENGTH(_utf16le 0x7DD8FFDF), OCTET_LENGTH(_utf16le 0x7FD8DDDF);
+
+--echo #
+--echo # Testing cs->cset->charpos()
+--echo #
+SELECT LEFT('abcd',2);
+SELECT HEX(LEFT(_utf16le 0x00D800DC7FD8FFDF, 1));
+SELECT HEX(RIGHT(_utf16le 0x00D800DC7FD8FFDF, 1));
+
+--echo #
+--echo # Testing cs->cset->well_formed_length()
+--echo #
+CREATE TABLE t1 (a VARCHAR(10) CHARACTER SET utf16le);
+--echo # Bad sequences
+--error ER_INVALID_CHARACTER_STRING
+INSERT INTO t1 VALUES (_utf16le 0x00D8);
+--error ER_INVALID_CHARACTER_STRING
+INSERT INTO t1 VALUES (_utf16le 0x00DC);
+--error ER_INVALID_CHARACTER_STRING
+INSERT INTO t1 VALUES (_utf16le 0x00D800D8);
+--error ER_INVALID_CHARACTER_STRING
+INSERT INTO t1 VALUES (_utf16le 0x00D800E8);
+--error ER_INVALID_CHARACTER_STRING
+INSERT INTO t1 VALUES (_utf16le 0x00D80008);
+--echo # Good sequences
+INSERT INTO t1 VALUES (_utf16le 0x00D800DC);
+INSERT INTO t1 VALUES (_utf16le 0x00D8FFDC);
+INSERT INTO t1 VALUES (_utf16le 0xFFDB00DC);
+INSERT INTO t1 VALUES (_utf16le 0xFFDBFFDC);
+SELECT HEX(a) FROM t1;
+DROP TABLE t1;
+
+--echo #
+--echo # Bug#32393 Character sets: illegal characters in utf16le columns
+--echo #
+--echo # Tests that cs->cset->wc_mb() doesn't accept surrogate parts
+--echo #
+--echo # via ALTER
+--echo #
+CREATE TABLE t1 (s1 VARCHAR(50) CHARACTER SET ucs2);
+INSERT INTO t1 VALUES (0xDF84);
+ALTER TABLE t1 MODIFY column s1 VARCHAR(50) CHARACTER SET utf16le;
+SELECT HEX(s1) FROM t1;
+DROP TABLE t1;
+
+--echo #
+--echo # via UPDATE
+--echo #
+CREATE TABLE t1 (s1 VARCHAR(5) CHARACTER SET ucs2, s2 VARCHAR(5) CHARACTER SET utf16le);
+INSERT INTO t1 (s1) VALUES (0xdf84);
+UPDATE t1 set s2 = s1;
+SELECT HEX(s2) FROM t1;
+DROP TABLE t1;
+
+
+--echo #
+--echo # Testing cs->cset->lengthsp()
+--echo #
+CREATE TABLE t1 (a CHAR(10)) CHARACTER SET utf16le;
+INSERT INTO t1 VALUES ('a ');
+SELECT HEX(a) FROM t1;
+DROP TABLE t1;
+
+
+--echo #
+--echo # Testing cs->cset->caseup() and cs->cset->casedn()
+--echo #
+SELECT UPPER('abcd'), LOWER('ABCD');
+
+--echo #
+--echo # Checking str_to_datetime()
+--echo #
+select @@collation_connection;
+CREATE TABLE t1 (a date);
+INSERT INTO t1 VALUES ('2007-09-16');
+SELECT * FROM t1;
+DROP TABLE t1;
+
+
+--echo #
+--echo # Testing cs->cset->ll10tostr
+--echo #
+CREATE TABLE t1 (a VARCHAR(10) CHARACTER SET utf16le);
+INSERT INTO t1 VALUES (123456);
+SELECT a, HEX(a) FROM t1;
+DROP TABLE t1;
+
+--echo #
+--echo # Testing cs->cset->fill
+--echo # SOUNDEX fills strings with DIGIT ZERO up to four characters
+--echo #
+SELECT SOUNDEX('a'), HEX(SOUNDEX('a'));
+
+
+--echo #
+--echo # Testing cs->cset->strntoul
+--echo #
+CREATE TABLE t1 (a enum ('a','b','c')) CHARACTER SET utf16le;
+INSERT INTO t1 VALUES ('1');
+SELECT * FROM t1;
+DROP TABLE t1;
+
+
+--echo #
+--echo # Testing cs->cset->strntoll and cs->cset->strntoull
+--echo #
+SET NAMES latin1;
+SELECT HEX(CONV(CONVERT('123' USING utf16le), -10, 16));
+SELECT HEX(CONV(CONVERT('123' USING utf16le), 10, 16));
+SET NAMES utf8, collation_connection=utf16le_general_ci;
+
+
+--echo #
+--echo # Testing cs->cset->strntod
+--echo #
+SET NAMES utf8, collation_connection=utf16le_general_ci;
+SELECT 1.1 + '1.2';
+SELECT 1.1 + '1.2xxx';
+
+
+--echo #
+--echo # Testing cs->cset->strtoll10
+--echo #
+SELECT LEFT('aaa','1');
+CREATE TABLE t1 AS SELECT REPEAT('abcd', 128) AS a;
+SELECT LEFT(a, '2') FROM t1;
+SELECT LEFT(a, ' \t \t 2') FROM t1;
+SELECT LEFT(a, ' \t \t +2') FROM t1;
+SELECT SUBSTR(a, '-2') FROM t1;
+SELECT SUBSTR(a, ' \t \t -2') FROM t1;
+SELECT LEFT(a, '00002') FROM t1;
+SELECT LEFT(a, ' \t \t 00002') FROM t1;
+SELECT LEFT(a, ' \t \t +00002') FROM t1;
+SELECT SUBSTR(a, '-00002') FROM t1;
+SELECT SUBSTR(a, ' \t \t -00002') FROM t1;
+DROP TABLE t1;
+CREATE TABLE t1 AS SELECT REPEAT('abcd', 128) AS a LIMIT 0;
+INSERT INTO t1 VALUES ('255'), ('65535'),('16777215'),('4294967295'),
+('1099511627775'),('281474976710655'),('72057594037927935'),
+('1844674407370955161'),('18446744073709551614'), ('18446744073709551615');
+SELECT a, CAST(a AS SIGNED), CAST(a AS UNSIGNED) FROM t1;
+UPDATE t1 SET a=CONCAT('-', a);
+SELECT a, CAST(a AS SIGNED) FROM t1;
+DROP TABLE t1;
+
+
+--echo #
+--echo # Testing cs->cset->strntoull10rnd
+--echo #
+CREATE TABLE t1 (a int);
+INSERT INTO t1 VALUES ('-1234.1e2');
+INSERT INTO t1 VALUES ('-1234.1e2xxxx');
+INSERT INTO t1 VALUES ('-1234.1e2 ');
+INSERT INTO t1 VALUES ('123');
+INSERT INTO t1 VALUES ('-124');
+INSERT INTO t1 VALUES ('+125');
+INSERT INTO t1 VALUES (' \t \t 123');
+INSERT INTO t1 VALUES (' \t \t -124');
+INSERT INTO t1 VALUES (' \t \t +125');
+INSERT INTO t1 VALUES (' \t \t 000123');
+INSERT INTO t1 VALUES (' \t \t -000124');
+INSERT INTO t1 VALUES (' \t \t +000125');
+SELECT * FROM t1;
+DROP TABLE t1;
+CREATE TABLE t1 (a BIGINT UNSIGNED);
+INSERT INTO t1 VALUES ('255'), ('65535'),('16777215'),('4294967295'),
+('1099511627775'),('281474976710655'),('72057594037927935'),
+('1844674407370955161'),('18446744073709551614'), ('18446744073709551615');
+SELECT * FROM t1;
+DROP TABLE t1;
+
+
+--echo #
+--echo # Testing cs->cset->scan
+--echo #
+CREATE TABLE t1 (a int);
+INSERT INTO t1 VALUES ('1 ');
+INSERT INTO t1 VALUES ('1 x');
+SELECT * FROM t1;
+DROP TABLE t1;
+
+
+--echo #
+--echo # Testing auto-conversion to TEXT
+--echo #
+CREATE TABLE t1 (a VARCHAR(17000) CHARACTER SET utf16le);
+SHOW CREATE TABLE t1;
+DROP TABLE t1;
+
+
+--echo #
+--echo # Testing that maximim possible key length is 1000 bytes
+--echo #
+CREATE TABLE t1 (a VARCHAR(250) CHARACTER SET utf16le PRIMARY KEY);
+SHOW CREATE TABLE t1;
+DROP TABLE t1;
+--error ER_TOO_LONG_KEY
+CREATE TABLE t1 (a VARCHAR(334) CHARACTER SET utf16le PRIMARY KEY);
+
+
+--echo #
+--echo # Conversion to utf8
+--echo #
+CREATE TABLE t1 (a CHAR(1) CHARACTER SET utf16le);
+INSERT INTO t1 VALUES (0x00D800DC),(0x00D8FFDC),(0x7FDB00DC),(0x7FDBFFDC);
+INSERT INTO t1 VALUES (0xC000), (0xFF00),(0x00E0), (0xFFFF);
+SELECT HEX(a), HEX(@a:=CONVERT(a USING utf8mb4)), HEX(CONVERT(@a USING utf16le)) FROM t1;
+DROP TABLE t1;
+
+
+--echo #
+--echo # Test basic regex functionality
+--echo #
+SET NAMES utf8, collation_connection=utf16le_general_ci;
+--source include/ctype_regex.inc
+
+
+--echo #
+--echo # Test how CHARACTER SET works with date/time
+--echo #
+--source include/ctype_datetime.inc
+
+
+--echo #
+--echo # Bug#33073 Character sets: ordering fails with utf32
+--echo #
+SET NAMES utf8, collation_connection=utf16le_general_ci;
+CREATE TABLE t1 AS SELECT REPEAT('a',2) as s1 LIMIT 0;
+SHOW CREATE TABLE t1;
+INSERT INTO t1 VALUES ('ab'),('AE'),('ab'),('AE');
+SELECT * FROM t1 ORDER BY s1;
+SET max_sort_length=4;
+SELECT * FROM t1 ORDER BY s1;
+DROP TABLE t1;
+SET max_sort_length=DEFAULT;
+
+
+--echo #
+--echo # Bug#52520 Difference in tinytext utf column metadata
+--echo #
+CREATE TABLE t1 (
+ s1 TINYTEXT CHARACTER SET utf16le,
+ s2 TEXT CHARACTER SET utf16le,
+ s3 MEDIUMTEXT CHARACTER SET utf16le,
+ s4 LONGTEXT CHARACTER SET utf16le
+);
+--enable_metadata
+SET NAMES utf8, @@character_set_results=NULL;
+SELECT *, HEX(s1) FROM t1;
+SET NAMES latin1;
+SELECT *, HEX(s1) FROM t1;
+SET NAMES utf8;
+SELECT *, HEX(s1) FROM t1;
+--disable_metadata
+CREATE TABLE t2 AS SELECT CONCAT(s1) FROM t1;
+SHOW CREATE TABLE t2;
+DROP TABLE t1, t2;
+
+
+--echo #
+--echo # WL#3664 WEIGHT_STRING
+--echo #
+
+SET NAMES utf8, collation_connection=utf16le_general_ci;
+--source include/weight_string.inc
+--source include/weight_string_euro.inc
+SELECT HEX(WEIGHT_STRING(_utf16le 0x00D800DC));
+SELECT HEX(WEIGHT_STRING(_utf16le 0x00D801DC));
+--source include/weight_string_l1.inc
+
+SET NAMES utf8, collation_connection=utf16le_bin;
+--source include/weight_string.inc
+--source include/weight_string_euro.inc
+--source include/weight_string_l1.inc
+
+--echo #
+--echo # End of 5.6 tests
+--echo #
=== modified file 'mysys/charset-def.c'
--- a/mysys/charset-def.c 2010-11-11 10:40:41 +0000
+++ b/mysys/charset-def.c 2010-12-03 10:10:05 +0000
@@ -326,6 +326,8 @@ my_bool init_compiled_charsets(myf flags
#ifdef HAVE_CHARSET_utf16
add_compiled_collation(&my_charset_utf16_general_ci);
add_compiled_collation(&my_charset_utf16_bin);
+ add_compiled_collation(&my_charset_utf16le_general_ci);
+ add_compiled_collation(&my_charset_utf16le_bin);
#ifdef HAVE_UCA_COLLATIONS
add_compiled_collation(&my_charset_utf16_unicode_ci);
add_compiled_collation(&my_charset_utf16_german2_uca_ci);
=== modified file 'sql/item_func.cc'
--- a/sql/item_func.cc 2010-11-26 15:45:26 +0000
+++ b/sql/item_func.cc 2010-12-03 10:10:05 +0000
@@ -971,13 +971,11 @@ longlong Item_func_signed::val_int_from_
value= cs->cset->strtoll10(cs, start, &end, error);
if (*error > 0 || end != start+ length)
{
- char err_buff[128];
- String err_tmp(err_buff,(uint32) sizeof(err_buff), system_charset_info);
- err_tmp.copy(start, length, system_charset_info);
+ ErrConvString err(res);
push_warning_printf(current_thd, MYSQL_ERROR::WARN_LEVEL_WARN,
ER_TRUNCATED_WRONG_VALUE,
ER(ER_TRUNCATED_WRONG_VALUE), "INTEGER",
- err_tmp.c_ptr());
+ err.ptr());
}
return value;
}
=== modified file 'sql/sql_string.cc'
--- a/sql/sql_string.cc 2010-09-13 09:58:11 +0000
+++ b/sql/sql_string.cc 2010-12-03 10:10:05 +0000
@@ -219,8 +219,8 @@ bool String::copy_aligned(const char *st
CHARSET_INFO *cs)
{
/* How many bytes are in incomplete character */
- offset= cs->mbmaxlen - offset; /* How many zeros we should prepend */
- DBUG_ASSERT(offset && offset != cs->mbmaxlen);
+ offset= cs->mbminlen - offset; /* How many zeros we should prepend */
+ DBUG_ASSERT(offset && offset != cs->mbminlen);
uint32 aligned_length= arg_length + offset;
if (alloc(aligned_length))
=== modified file 'sql/sys_vars.cc'
--- a/sql/sys_vars.cc 2010-12-02 15:05:07 +0000
+++ b/sql/sys_vars.cc 2010-12-03 10:10:05 +0000
@@ -463,7 +463,7 @@ static bool check_charset(sys_var *self,
if (var->value->result_type() == STRING_RESULT)
{
String str(buff, sizeof(buff), system_charset_info), *res;
- if (!(res=var->value->val_str(&str)))
+ if (!(res=var->value->val_str_ascii(&str)))
var->save_result.ptr= NULL;
else if (!(var->save_result.ptr= get_charset_by_csname(res->c_ptr(),
MY_CS_PRIMARY,
@@ -581,7 +581,7 @@ static bool check_collation_not_null(sys
if (var->value->result_type() == STRING_RESULT)
{
String str(buff, sizeof(buff), system_charset_info), *res;
- if (!(res= var->value->val_str(&str)))
+ if (!(res= var->value->val_str_ascii(&str)))
var->save_result.ptr= NULL;
else if (!(var->save_result.ptr= get_charset_by_name(res->c_ptr(), MYF(0))))
{
=== modified file 'strings/ctype-ucs2.c'
--- a/strings/ctype-ucs2.c 2010-12-03 09:24:07 +0000
+++ b/strings/ctype-ucs2.c 2010-12-03 10:10:05 +0000
@@ -709,7 +709,7 @@ cnv:
#ifdef HAVE_CHARSET_mb2
static longlong
-my_strtoll10_mb2(CHARSET_INFO *cs __attribute__((unused)),
+my_strtoll10_mb2(CHARSET_INFO *cs,
const char *nptr, char **endptr, int *error)
{
const char *s, *end, *start, *n_end, *true_end;
@@ -718,6 +718,8 @@ my_strtoll10_mb2(CHARSET_INFO *cs __attr
ulonglong li;
int negative;
ulong cutoff, cutoff2, cutoff3;
+ my_wc_t wc;
+ int res;
s= nptr;
/* If fixed length string */
@@ -725,10 +727,15 @@ my_strtoll10_mb2(CHARSET_INFO *cs __attr
{
/* Make sure string length is even */
end= s + ((*endptr - s) / 2) * 2;
- while (s < end && !s[0] && (s[1] == ' ' || s[1] == '\t'))
- s+= 2;
- if (s == end)
- goto no_conv;
+ for ( ; s < end; ) /* Skip leading spaces and tabs */
+ {
+ res= cs->cset->mb_wc(cs, &wc, (const uchar *) s, (const uchar *) end);
+ if (res < 0)
+ goto no_conv;
+ s+= res;
+ if (wc != ' ' && wc != '\t')
+ break;
+ }
}
else
{
@@ -738,13 +745,14 @@ my_strtoll10_mb2(CHARSET_INFO *cs __attr
/* Check for a sign. */
negative= 0;
- if (!s[0] && s[1] == '-')
+ if (wc == '-')
{
*error= -1; /* Mark as negative number */
negative= 1;
- s+= 2;
- if (s == end)
+ res= cs->cset->mb_wc(cs, &wc, (const uchar *) s, (const uchar *) end);
+ if (res < 0)
goto no_conv;
+ s+= res;
cutoff= MAX_NEGATIVE_NUMBER / LFACTOR2;
cutoff2= (MAX_NEGATIVE_NUMBER % LFACTOR2) / 100;
cutoff3= MAX_NEGATIVE_NUMBER % 100;
@@ -752,46 +760,55 @@ my_strtoll10_mb2(CHARSET_INFO *cs __attr
else
{
*error= 0;
- if (!s[0] && s[1] == '+')
+ if (wc == '+')
{
- s+= 2;
- if (s == end)
+ res= cs->cset->mb_wc(cs, &wc, (const uchar *) s, (const uchar *) end);
+ if (res < 0)
goto no_conv;
+ s+= res;
}
cutoff= ULONGLONG_MAX / LFACTOR2;
cutoff2= ULONGLONG_MAX % LFACTOR2 / 100;
cutoff3= ULONGLONG_MAX % 100;
}
+
/* Handle case where we have a lot of pre-zero */
- if (!s[0] && s[1] == '0')
+ if (wc == '0')
{
i= 0;
- do
+ for ( ; ; s+= res)
{
- s+= 2;
if (s == end)
goto end_i; /* Return 0 */
+ res= cs->cset->mb_wc(cs, &wc, (const uchar *) s, (const uchar *) end);
+ if (res < 0)
+ goto no_conv;
+ if (wc != '0')
+ break;
}
- while (!s[0] && s[1] == '0');
+ while (wc == '0');
n_end= s + 2 * INIT_CNT;
}
else
{
/* Read first digit to check that it's a valid number */
- if (s[0] || (c= (s[1]-'0')) > 9)
+ if ((c= (wc - '0')) > 9)
goto no_conv;
i= c;
- s+= 2;
n_end= s + 2 * (INIT_CNT-1);
}
/* Handle first 9 digits and store them in i */
if (n_end > end)
n_end= end;
- for (; s != n_end ; s+= 2)
+ for ( ; ; )
{
- if (s[0] || (c= (s[1]-'0')) > 9)
+ res= cs->cset->mb_wc(cs, &wc, (const uchar *) s, (const uchar *) n_end);
+ if (res < 0)
+ break;
+ s+= res;
+ if ((c= (wc - '0')) > 9)
goto end_i;
i= i*10+c;
}
@@ -806,10 +823,13 @@ my_strtoll10_mb2(CHARSET_INFO *cs __attr
n_end= end;
do
{
- if (s[0] || (c= (s[1]-'0')) > 9)
+ res= cs->cset->mb_wc(cs, &wc, (const uchar *) s, (const uchar *) end);
+ if (res < 0)
+ goto no_conv;
+ s+= res;
+ if ((c= (wc - '0')) > 9)
goto end_i_and_j;
j= j*10+c;
- s+= 2;
} while (s != n_end);
if (s == end)
{
@@ -817,20 +837,28 @@ my_strtoll10_mb2(CHARSET_INFO *cs __attr
goto end_i_and_j;
goto end3;
}
- if (s[0] || (c= (s[1]-'0')) > 9)
+ res= cs->cset->mb_wc(cs, &wc, (const uchar *) s, (const uchar *) end);
+ if (res < 0)
+ goto no_conv;
+ s+= res;
+ if ((c= (wc - '0')) > 9)
goto end3;
/* Handle the next 1 or 2 digits and store them in k */
k=c;
- s+= 2;
- if (s == end || s[0] || (c= (s[1]-'0')) > 9)
+ if (s == end)
+ goto end4;
+ res= cs->cset->mb_wc(cs, &wc, (const uchar *) s, (const uchar *) end);
+ if (res < 0)
+ goto no_conv;
+ s+= res;
+ if ((c= (wc - '0')) > 9)
goto end4;
k= k*10+c;
- s+= 2;
*endptr= (char*) s;
/* number string should have ended here */
- if (s != end && !s[0] && (c= (s[1]-'0')) <= 9)
+ if (s != end && (c= (wc - '0')) <= 9)
goto overflow;
/* Check that we didn't get an overflow with the last digit */
@@ -878,19 +906,23 @@ no_conv:
static size_t
-my_scan_mb2(CHARSET_INFO *cs __attribute__((unused)),
+my_scan_mb2(CHARSET_INFO *cs,
const char *str, const char *end, int sequence_type)
{
const char *str0= str;
- end--; /* for easier loop condition, because of two bytes per character */
-
+ my_wc_t wc;
+ int res;
+
switch (sequence_type)
{
case MY_SEQ_SPACES:
- for ( ; str < end; str+= 2)
+ for (res= cs->cset->mb_wc(cs, &wc,
+ (const uchar *) str, (const uchar *) end);
+ res > 0 && wc == ' ';
+ str+= res,
+ res= cs->cset->mb_wc(cs, &wc,
+ (const uchar *) str, (const uchar *) end))
{
- if (str[0] != '\0' || str[1] != ' ')
- break;
}
return (size_t) (str - str0);
default:
@@ -1041,11 +1073,26 @@ my_lengthsp_mb2(CHARSET_INFO *cs __attri
DB80..DBFF - Private surrogate high (128 pages)
DC00..DFFF - Surrogate low (1024 codes in a page)
*/
+#define MY_UTF16_SURROGATE_HIGH_FIRST 0xD800
+#define MY_UTF16_SURROGATE_HIGH_LAST 0xDBFF
+#define MY_UTF16_SURROGATE_LOW_FIRST 0xDC00
+#define MY_UTF16_SURROGATE_LOW_LAST 0xDFFF
#define MY_UTF16_HIGH_HEAD(x) ((((uchar) (x)) & 0xFC) == 0xD8)
#define MY_UTF16_LOW_HEAD(x) ((((uchar) (x)) & 0xFC) == 0xDC)
#define MY_UTF16_SURROGATE(x) (((x) & 0xF800) == 0xD800)
+#define MY_UTF16_WC2(a, b) ((a << 8) + b)
+
+/*
+ a= 110110?? (<< 18)
+ b= ???????? (<< 10)
+ c= 110111?? (<< 8)
+ d= ???????? (<< 0)
+*/
+#define MY_UTF16_WC4(a, b, c, d) (((a & 3) << 18) + (b << 10) + \
+ ((c & 3) << 8) + d + 0x10000)
+
static int
my_utf16_uni(CHARSET_INFO *cs __attribute__((unused)),
my_wc_t *pwc, const uchar *s, const uchar *e)
@@ -1067,23 +1114,14 @@ my_utf16_uni(CHARSET_INFO *cs __attribut
if (!MY_UTF16_LOW_HEAD(s[2])) /* Broken surrigate pair */
return MY_CS_ILSEQ;
- /*
- s[0]= 110110?? (<< 18)
- s[1]= ???????? (<< 10)
- s[2]= 110111?? (<< 8)
- s[3]= ???????? (<< 0)
- */
-
- *pwc= ((s[0] & 3) << 18) + (s[1] << 10) +
- ((s[2] & 3) << 8) + s[3] + 0x10000;
-
+ *pwc= MY_UTF16_WC4(s[0], s[1], s[2], s[3]);
return 4;
}
if (MY_UTF16_LOW_HEAD(*s)) /* Low surrogate part without high part */
return MY_CS_ILSEQ;
-
- *pwc= (s[0] << 8) + s[1];
+
+ *pwc= MY_UTF16_WC2(s[0], s[1]);
return 2;
}
@@ -1165,10 +1203,10 @@ my_caseup_utf16(CHARSET_INFO *cs, char *
DBUG_ASSERT(src == dst && srclen == dstlen);
while ((src < srcend) &&
- (res= my_utf16_uni(cs, &wc, (uchar *)src, (uchar*) srcend)) > 0)
+ (res= cs->cset->mb_wc(cs, &wc, (uchar *) src, (uchar *) srcend)) > 0)
{
my_toupper_utf16(uni_plane, &wc);
- if (res != my_uni_utf16(cs, wc, (uchar*) src, (uchar*) srcend))
+ if (res != cs->cset->wc_mb(cs, wc, (uchar *) src, (uchar *) srcend))
break;
src+= res;
}
@@ -1182,13 +1220,11 @@ my_hash_sort_utf16(CHARSET_INFO *cs, con
{
my_wc_t wc;
int res;
- const uchar *e= s+slen;
+ const uchar *e= s + cs->cset->lengthsp(cs, (const char *) s, slen);
MY_UNICASE_INFO *uni_plane= cs->caseinfo;
- while (e > s + 1 && e[-1] == ' ' && e[-2] == '\0')
- e-= 2;
-
- while ((s < e) && (res= my_utf16_uni(cs, &wc, (uchar *)s, (uchar*)e)) > 0)
+ while ((s < e) && (res= cs->cset->mb_wc(cs, &wc,
+ (uchar *) s, (uchar *) e)) > 0)
{
my_tosort_utf16(uni_plane, &wc);
n1[0]^= (((n1[0] & 63) + n2[0]) * (wc & 0xFF)) + (n1[0] << 8);
@@ -1212,10 +1248,10 @@ my_casedn_utf16(CHARSET_INFO *cs, char *
DBUG_ASSERT(src == dst && srclen == dstlen);
while ((src < srcend) &&
- (res= my_utf16_uni(cs, &wc, (uchar*) src, (uchar*) srcend)) > 0)
+ (res= cs->cset->mb_wc(cs, &wc, (uchar *) src, (uchar *) srcend)) > 0)
{
my_tolower_utf16(uni_plane, &wc);
- if (res != my_uni_utf16(cs, wc, (uchar*) src, (uchar*) srcend))
+ if (res != cs->cset->wc_mb(cs, wc, (uchar *) src, (uchar *) srcend))
break;
src+= res;
}
@@ -1237,8 +1273,8 @@ my_strnncoll_utf16(CHARSET_INFO *cs,
while (s < se && t < te)
{
- s_res= my_utf16_uni(cs, &s_wc, s, se);
- t_res= my_utf16_uni(cs, &t_wc, t, te);
+ s_res= cs->cset->mb_wc(cs, &s_wc, s, se);
+ t_res= cs->cset->mb_wc(cs, &t_wc, t, te);
if (s_res <= 0 || t_res <= 0)
{
@@ -1307,8 +1343,8 @@ my_strnncollsp_utf16(CHARSET_INFO *cs,
while (s < se && t < te)
{
- int s_res= my_utf16_uni(cs, &s_wc, s, se);
- int t_res= my_utf16_uni(cs, &t_wc, t, te);
+ int s_res= cs->cset->mb_wc(cs, &s_wc, s, se);
+ int t_res= cs->cset->mb_wc(cs, &t_wc, t, te);
if (s_res <= 0 || t_res <= 0)
{
@@ -1348,7 +1384,7 @@ my_strnncollsp_utf16(CHARSET_INFO *cs,
for ( ; s < se; s+= s_res)
{
- if ((s_res= my_utf16_uni(cs, &s_wc, s, se)) < 0)
+ if ((s_res= cs->cset->mb_wc(cs, &s_wc, s, se)) < 0)
{
DBUG_ASSERT(0);
return 0;
@@ -1362,22 +1398,11 @@ my_strnncollsp_utf16(CHARSET_INFO *cs,
static uint
-my_ismbchar_utf16(CHARSET_INFO *cs __attribute__((unused)),
- const char *b __attribute__((unused)),
- const char *e __attribute__((unused)))
+my_ismbchar_utf16(CHARSET_INFO *cs, const char *b, const char *e)
{
- if (b + 2 > e)
- return 0;
-
- if (MY_UTF16_HIGH_HEAD(*b))
- {
- return (b + 4 <= e) && MY_UTF16_LOW_HEAD(b[2]) ? 4 : 0;
- }
-
- if (MY_UTF16_LOW_HEAD(*b))
- return 0;
-
- return 2;
+ my_wc_t wc;
+ int res= cs->cset->mb_wc(cs, &wc, (const uchar *) b, (const uchar *) e);
+ return (uint) (res > 0 ? res : 0);
}
@@ -1385,6 +1410,7 @@ static uint
my_mbcharlen_utf16(CHARSET_INFO *cs __attribute__((unused)),
uint c __attribute__((unused)))
{
+ DBUG_ASSERT(0);
return MY_UTF16_HIGH_HEAD(c) ? 4 : 2;
}
@@ -1478,8 +1504,8 @@ my_strnncoll_utf16_bin(CHARSET_INFO *cs,
while ( s < se && t < te )
{
- s_res= my_utf16_uni(cs,&s_wc, s, se);
- t_res= my_utf16_uni(cs,&t_wc, t, te);
+ s_res= cs->cset->mb_wc(cs, &s_wc, s, se);
+ t_res= cs->cset->mb_wc(cs, &t_wc, t, te);
if (s_res <= 0 || t_res <= 0)
{
@@ -1517,8 +1543,8 @@ my_strnncollsp_utf16_bin(CHARSET_INFO *c
while (s < se && t < te)
{
- int s_res= my_utf16_uni(cs, &s_wc, s, se);
- int t_res= my_utf16_uni(cs, &t_wc, t, te);
+ int s_res= cs->cset->mb_wc(cs, &s_wc, s, se);
+ int t_res= cs->cset->mb_wc(cs, &t_wc, t, te);
if (s_res <= 0 || t_res <= 0)
{
@@ -1555,7 +1581,7 @@ my_strnncollsp_utf16_bin(CHARSET_INFO *c
for ( ; s < se; s+= s_res)
{
- if ((s_res= my_utf16_uni(cs, &s_wc, s, se)) < 0)
+ if ((s_res= cs->cset->mb_wc(cs, &s_wc, s, se)) < 0)
{
DBUG_ASSERT(0);
return 0;
@@ -1569,17 +1595,11 @@ my_strnncollsp_utf16_bin(CHARSET_INFO *c
static void
-my_hash_sort_utf16_bin(CHARSET_INFO *cs __attribute__((unused)),
- const uchar *key, size_t len,ulong *nr1, ulong *nr2)
+my_hash_sort_utf16_bin(CHARSET_INFO *cs,
+ const uchar *pos, size_t len, ulong *nr1, ulong *nr2)
{
- const uchar *pos = key;
-
- key+= len;
-
- while (key > pos + 1 && key[-1] == ' ' && key[-2] == '\0')
- key-= 2;
-
- for (; pos < (uchar*) key ; pos++)
+ const uchar *end= pos + cs->cset->lengthsp(cs, (const char *) pos, len);
+ for ( ; pos < end ; pos++)
{
nr1[0]^= (ulong) ((((uint) nr1[0] & 63) + nr2[0]) *
((uint)*pos)) + (nr1[0] << 8);
@@ -1719,6 +1739,175 @@ CHARSET_INFO my_charset_utf16_bin=
&my_collation_utf16_bin_handler
};
+
+static int
+my_utf16le_uni(CHARSET_INFO *cs __attribute__((unused)),
+ my_wc_t *pwc, const uchar *s, const uchar *e)
+{
+ my_wc_t lo;
+
+ if (s + 2 > e)
+ return MY_CS_TOOSMALL2;
+
+ if ((*pwc= uint2korr(s)) < MY_UTF16_SURROGATE_HIGH_FIRST ||
+ (*pwc > MY_UTF16_SURROGATE_LOW_LAST))
+ return 2; /* [0000-D7FF,E000-FFFF] */
+
+ if (*pwc >= MY_UTF16_SURROGATE_LOW_FIRST)
+ return MY_CS_ILSEQ; /* [DC00-DFFF] Low surrogate part without high part */
+
+ if (s + 4 > e)
+ return MY_CS_TOOSMALL4;
+
+ s+= 2;
+
+ if ((lo= uint2korr(s)) < MY_UTF16_SURROGATE_LOW_FIRST ||
+ lo > MY_UTF16_SURROGATE_LOW_LAST)
+ return MY_CS_ILSEQ; /* Expected low surrogate part, got something else */
+
+ *pwc= 0x10000 + (((*pwc & 0x3FF) << 10) | (lo & 0x3FF));
+ return 4;
+}
+
+
+static int
+my_uni_utf16le(CHARSET_INFO *cs __attribute__((unused)),
+ my_wc_t wc, uchar *s, uchar *e)
+{
+ if (wc < MY_UTF16_SURROGATE_HIGH_FIRST ||
+ (wc > MY_UTF16_SURROGATE_LOW_LAST &&
+ wc <= 0xFFFF))
+ {
+ if (s + 2 > e)
+ return MY_CS_TOOSMALL2;
+ int2store(s, wc);
+ return 2; /* [0000-D7FF,E000-FFFF] */
+ }
+
+ if (wc < 0xFFFF || wc > 0x10FFFF)
+ return MY_CS_ILUNI; /* [D800-DFFF,10FFFF+] */
+
+ if (s + 4 > e)
+ return MY_CS_TOOSMALL4;
+
+ wc-= 0x10000;
+ int2store(s, (0xD800 | ((wc >> 10) & 0x3FF))); s+= 2;
+ int2store(s, (0xDC00 | (wc & 0x3FF)));
+ return 4; /* [010000-10FFFF] */
+}
+
+
+static size_t
+my_lengthsp_utf16le(CHARSET_INFO *cs __attribute__((unused)),
+ const char *ptr, size_t length)
+{
+ const char *end= ptr + length;
+ while (end > ptr + 1 && uint2korr(end - 2) == 0x20)
+ end-= 2;
+ return (size_t) (end - ptr);
+}
+
+
+static MY_CHARSET_HANDLER my_charset_utf16le_handler=
+{
+ NULL, /* init */
+ my_ismbchar_utf16,
+ my_mbcharlen_utf16,
+ my_numchars_utf16,
+ my_charpos_utf16,
+ my_well_formed_len_utf16,
+ my_lengthsp_utf16le,
+ my_numcells_mb,
+ my_utf16le_uni, /* mb_wc */
+ my_uni_utf16le, /* wc_mb */
+ my_mb_ctype_mb,
+ my_caseup_str_mb2_or_mb4,
+ my_casedn_str_mb2_or_mb4,
+ my_caseup_utf16,
+ my_casedn_utf16,
+ my_snprintf_mb2,
+ my_l10tostr_mb2_or_mb4,
+ my_ll10tostr_mb2_or_mb4,
+ my_fill_mb2,
+ my_strntol_mb2_or_mb4,
+ my_strntoul_mb2_or_mb4,
+ my_strntoll_mb2_or_mb4,
+ my_strntoull_mb2_or_mb4,
+ my_strntod_mb2_or_mb4,
+ my_strtoll10_mb2,
+ my_strntoull10rnd_mb2_or_mb4,
+ my_scan_mb2
+};
+
+
+CHARSET_INFO my_charset_utf16le_general_ci=
+{
+ 56,0,0, /* number */
+ MY_CS_COMPILED|MY_CS_PRIMARY|MY_CS_STRNXFRM|MY_CS_UNICODE|MY_CS_NONASCII,
+ "utf16le", /* cs name */
+ "utf16le_general_ci",/* name */
+ "UTF-16LE Unicode", /* comment */
+ NULL, /* tailoring */
+ NULL, /* ctype */
+ NULL, /* to_lower */
+ NULL, /* to_upper */
+ NULL, /* sort_order */
+ NULL, /* uca */
+ NULL, /* tab_to_uni */
+ NULL, /* tab_from_uni */
+ &my_unicase_default, /* caseinfo */
+ NULL, /* state_map */
+ NULL, /* ident_map */
+ 1, /* strxfrm_multiply */
+ 1, /* caseup_multiply */
+ 1, /* casedn_multiply */
+ 2, /* mbminlen */
+ 4, /* mbmaxlen */
+ 0, /* min_sort_char */
+ 0xFFFF, /* max_sort_char */
+ ' ', /* pad char */
+ 0, /* escape_with_backslash_is_dangerous */
+ 1, /* levels_for_compare */
+ 1, /* levels_for_order */
+ &my_charset_utf16le_handler,
+ &my_collation_utf16_general_ci_handler
+};
+
+
+CHARSET_INFO my_charset_utf16le_bin=
+{
+ 62,0,0, /* number */
+ MY_CS_COMPILED|MY_CS_BINSORT|MY_CS_UNICODE|MY_CS_NONASCII,
+ "utf16le", /* cs name */
+ "utf16le_bin", /* name */
+ "UTF-16LE Unicode", /* comment */
+ NULL, /* tailoring */
+ NULL, /* ctype */
+ NULL, /* to_lower */
+ NULL, /* to_upper */
+ NULL, /* sort_order */
+ NULL, /* uca */
+ NULL, /* tab_to_uni */
+ NULL, /* tab_from_uni */
+ &my_unicase_default, /* caseinfo */
+ NULL, /* state_map */
+ NULL, /* ident_map */
+ 1, /* strxfrm_multiply */
+ 1, /* caseup_multiply */
+ 1, /* casedn_multiply */
+ 2, /* mbminlen */
+ 4, /* mbmaxlen */
+ 0, /* min_sort_char */
+ 0xFFFF, /* max_sort_char */
+ ' ', /* pad char */
+ 0, /* escape_with_backslash_is_dangerous */
+ 1, /* levels_for_compare */
+ 1, /* levels_for_order */
+ &my_charset_utf16le_handler,
+ &my_collation_utf16_bin_handler
+};
+
+
#endif /* HAVE_CHARSET_utf16 */
Attachment: [text/bzr-bundle] bzr/bar@mysql.com-20101203101005-h851i1eqa6u17brw.bundle
