Chapter聽11.聽Data Types - MySQL 5.0参考手册英文版

A summary of the numeric data types follows. For additional information, see Section聽11.2, 鈥淣umeric Types鈥�. Storage requirements are given in Section聽11.5, 鈥淒ata Type Storage Requirements鈥�.

M indicates the maximum display width for integer types. The maximum legal display width is 255. Display width is unrelated to the range of values a type can contain, as described in Section聽11.2, 鈥淣umeric Types鈥�. For floating-point and fixed-point types, M is the total number of digits that can be stored.

If you specify ZEROFILL for a numeric column, MySQL automatically adds the UNSIGNED attribute to the column.

Numeric data types that allow the UNSIGNED attribute also allow SIGNED. However, these data types are signed by default, so the SIGNED attribute has no effect.

SERIAL is an alias for BIGINT UNSIGNED NOT NULL AUTO_INCREMENT UNIQUE.

SERIAL DEFAULT VALUE in the definition of an integer column is an alias for NOT NULL AUTO_INCREMENT UNIQUE.

Warning: When you use subtraction between integer values where one is of type UNSIGNED, the result is unsigned unless the NO_UNSIGNED_SUBTRACTION SQL mode is enabled. See Section聽12.9, 鈥淐ast Functions and Operators鈥�.

A summary of the temporal data types follows. For additional information, see Section聽11.3, 鈥淒ate and Time Types鈥�. Storage requirements are given in Section聽11.5, 鈥淒ata Type Storage Requirements鈥�. Functions that operate on temporal values are described at Section聽12.6, 鈥淒ate and Time Functions鈥�.

For the DATETIME and DATE range descriptions, 鈥�supported鈥� means that although earlier values might work, there is no guarantee.

The SUM() and AVG() aggregate functions do not work with temporal values. (They convert the values to numbers, which loses the part after the first non-numeric character.) To work around this problem, you can convert to numeric units, perform the aggregate operation, and convert back to a temporal value. Examples:

SELECT SEC_TO_TIME(SUM(TIME_TO_SEC(time_col))) FROM tbl_name;
SELECT FROM_DAYS(SUM(TO_DAYS(date_col))) FROM tbl_name;

A summary of the string data types follows. For additional information, see Section聽11.4, 鈥淪tring Types鈥�. Storage requirements are given in Section聽11.5, 鈥淒ata Type Storage Requirements鈥�.

In some cases, MySQL may change a string column to a type different from that given in a CREATE TABLE or ALTER TABLE statement. See Section聽13.1.8.1, 鈥淪ilent Column Specification Changes鈥�.

In MySQL 4.1 and up, string data types include some features that you may not have encountered in working with versions of MySQL prior to 4.1:

Chapter聽10, Character Set Support, provides additional information about use of character sets in MySQL.

The DEFAULT value clause in a data type specification indicates a default value for a column. With one exception, the default value must be a constant; it cannot be a function or an expression. This means, for example, that you cannot set the default for a date column to be the value of a function such as NOW() or CURRENT_DATE. The exception is that you can specify CURRENT_TIMESTAMP as the default for a TIMESTAMP column. See Section聽11.3.1.1, 鈥�TIMESTAMP Properties as of MySQL 4.1鈥�.

BLOB and TEXT columns cannot be assigned a default value.

If a column definition includes no explicit DEFAULT value, MySQL determines the default value as follows:

If the column can take NULL as a value, the column is defined with an explicit DEFAULT NULL clause.

If the column cannot take NULL as the value, MySQL defines the column with no explicit DEFAULT clause. For data entry, if an INSERT or REPLACE statement includes no value for the column, MySQL handles the column according to the SQL mode in effect at the time:

Suppose that a table t is defined as follows:

CREATE TABLE t (i INT NOT NULL);

In this case, i has no explicit default, so in strict mode each of the following statements produce an error and no row is inserted. When not using strict mode, only the third statement produces an error; the implicit default is inserted for the first two statements, but the third fails because DEFAULT(i) cannot produce a value:

INSERT INTO t VALUES();
INSERT INTO t VALUES(DEFAULT);
INSERT INTO t VALUES(DEFAULT(i));

See Section聽5.2.6, 鈥淪QL Modes鈥�.

For a given table, you can use the SHOW CREATE TABLE statement to see which columns have an explicit DEFAULT clause.

Implicit defaults are defined as follows:

The DATETIME, DATE, and TIMESTAMP types are related. This section describes their characteristics, how they are similar, and how they differ.

The DATETIME type is used when you need values that contain both date and time information. MySQL retrieves and displays DATETIME values in 'YYYY-MM-DD HH:MM:SS' format. The supported range is '1000-01-01 00:00:00' to '9999-12-31 23:59:59'.

The DATE type is used when you need only a date value, without a time part. MySQL retrieves and displays DATE values in 'YYYY-MM-DD' format. The supported range is '1000-01-01' to '9999-12-31'.

For the DATETIME and DATE range descriptions, 鈥�supported鈥� means that although earlier values might work, there is no guarantee.

The TIMESTAMP data type has varying properties, depending on the MySQL version and the SQL mode the server is running in. These properties are described later in this section.

You can specify DATETIME, DATE, and TIMESTAMP values using any of a common set of formats:

A microseconds part is allowable in temporal values in some contexts, such as in literal values, and in the arguments to or return values from some temporal functions. Microseconds are specified as a trailing .uuuuuu part in the value. Example:

mysql> SELECT MICROSECOND('2010-12-10 14:12:09.019473');
+-------------------------------------------+
| MICROSECOND('2010-12-10 14:12:09.019473') |
+-------------------------------------------+
|                                     19473 | 
+-------------------------------------------+

However, microseconds cannot be stored into a column of any temporal data type. Any microseconds part is discarded.

Conversion of DATETIME values to numeric form (for example, by adding +0) results in a double value with a microseconds part of .000000:

mysql> SELECT NOW(), NOW()+0;
+---------------------+-----------------------+
| NOW()               | NOW()+0               |
+---------------------+-----------------------+
| 2007-04-23 14:21:52 | 20070423142152.000000 | 
+---------------------+-----------------------+

Illegal DATETIME, DATE, or TIMESTAMP values are converted to the 鈥�zero鈥� value of the appropriate type ('0000-00-00 00:00:00' or '0000-00-00').

For values specified as strings that include date part delimiters, it is not necessary to specify two digits for month or day values that are less than 10. '1979-6-9' is the same as '1979-06-09'. Similarly, for values specified as strings that include time part delimiters, it is not necessary to specify two digits for hour, minute, or second values that are less than 10. '1979-10-30 1:2:3' is the same as '1979-10-30 01:02:03'.

Values specified as numbers should be 6, 8, 12, or 14 digits long. If a number is 8 or 14 digits long, it is assumed to be in YYYYMMDD or YYYYMMDDHHMMSS format and that the year is given by the first 4 digits. If the number is 6 or 12 digits long, it is assumed to be in YYMMDD or YYMMDDHHMMSS format and that the year is given by the first 2 digits. Numbers that are not one of these lengths are interpreted as though padded with leading zeros to the closest length.

Values specified as non-delimited strings are interpreted using their length as given. If the string is 8 or 14 characters long, the year is assumed to be given by the first 4 characters. Otherwise, the year is assumed to be given by the first 2 characters. The string is interpreted from left to right to find year, month, day, hour, minute, and second values, for as many parts as are present in the string. This means you should not use strings that have fewer than 6 characters. For example, if you specify '9903', thinking that represents March, 1999, MySQL inserts a 鈥�zero鈥� date value into your table. This occurs because the year and month values are 99 and 03, but the day part is completely missing, so the value is not a legal date. However, you can explicitly specify a value of zero to represent missing month or day parts. For example, you can use '990300' to insert the value '1999-03-00'.

You can to some extent assign values of one date type to an object of a different date type. However, there may be some alteration of the value or loss of information:

Be aware of certain pitfalls when specifying date values:

mysql> SET GLOBAL sql_mode=MAXDB;

mysql> SET SESSION sql_mode=MAXDB;

CREATE TABLE t
(
  ts1 TIMESTAMP NULL DEFAULT NULL,
  ts2 TIMESTAMP NULL DEFAULT 0,
  ts3 TIMESTAMP NULL DEFAULT CURRENT_TIMESTAMP
);

CREATE TABLE t (ts TIMESTAMP NULL DEFAULT CURRENT_TIMESTAMP);

CREATE TABLE t1 (ts TIMESTAMP NULL DEFAULT NULL);
CREATE TABLE t2 (ts TIMESTAMP NULL DEFAULT '0000-00-00 00:00:00');

INSERT INTO t1 VALUES (NOW());
INSERT INTO t2 VALUES (CURRENT_TIMESTAMP);

MySQL retrieves and displays TIME values in 'HH:MM:SS' format (or 'HHH:MM:SS' format for large hours values). TIME values may range from '-838:59:59' to '838:59:59'. The hours part may be so large because the TIME type can be used not only to represent a time of day (which must be less than 24 hours), but also elapsed time or a time interval between two events (which may be much greater than 24 hours, or even negative).

You can specify TIME values in a variety of formats:

A trailing .uuuuuu microseconds part of TIME values is allowed under the same conditions as for other temporal values, as described in Section聽11.3.1, 鈥淭he DATETIME, DATE, and TIMESTAMP Types鈥�. This includes the property that any microseconds part is discarded from values stored into TIME columns.

For TIME values specified as strings that include a time part delimiter, it is not necessary to specify two digits for hours, minutes, or seconds values that are less than 10. '8:3:2' is the same as '08:03:02'.

Be careful about assigning abbreviated values to a TIME column. Without colons, MySQL interprets values using the assumption that the two rightmost digits represent seconds. (MySQL interprets TIME values as elapsed time rather than as time of day.) For example, you might think of '1112' and 1112 as meaning '11:12:00' (12 minutes after 11 o'clock), but MySQL interprets them as '00:11:12' (11 minutes, 12 seconds). Similarly, '12' and 12 are interpreted as '00:00:12'. TIME values with colons, by contrast, are always treated as time of the day. That is, '11:12' mean '11:12:00', not '00:11:12'.

By default, values that lie outside the TIME range but are otherwise legal are clipped to the closest endpoint of the range. For example, '-850:00:00' and '850:00:00' are converted to '-838:59:59' and '838:59:59'. Illegal TIME values are converted to '00:00:00'. Note that because '00:00:00' is itself a legal TIME value, there is no way to tell, from a value of '00:00:00' stored in a table, whether the original value was specified as '00:00:00' or whether it was illegal.

For more restrictive treatment of invalid TIME values, enable strict SQL mode to cause errors to occur. See Section聽5.2.6, 鈥淪QL Modes鈥�.

SERIAL DEFAULT VALUE in the definition of an integer column is an alias for NOT NULL AUTO_INCREMENT UNIQUE.

The YEAR type is a one-byte type used for representing years.

MySQL retrieves and displays YEAR values in YYYY format. The range is 1901 to 2155.

You can specify YEAR values in a variety of formats:

Illegal YEAR values are converted to 0000.

MySQL Server itself has no problems with Year 2000 (Y2K) compliance:

Although MySQL Server itself is Y2K-safe, you may run into problems if you use it with applications that are not Y2K-safe. For example, many old applications store or manipulate years using two-digit values (which are ambiguous) rather than four-digit values. This problem may be compounded by applications that use values such as 00 or 99 as 鈥�missing鈥� value indicators. Unfortunately, these problems may be difficult to fix because different applications may be written by different programmers, each of whom may use a different set of conventions and date-handling functions.

Thus, even though MySQL Server has no Y2K problems, it is the application's responsibility to provide unambiguous input. Any value containing a two-digit year is ambiguous, because the century is unknown. Such values must be interpreted into four-digit form because MySQL stores years internally using four digits.

For DATETIME, DATE, TIMESTAMP, and YEAR types, MySQL interprets dates with ambiguous year values using the following rules:

Remember that these rules are only heuristics that provide reasonable guesses as to what your data values mean. If the rules used by MySQL do not produce the correct values, you should provide unambiguous input containing four-digit year values.

ORDER BY properly sorts YEAR values that have two-digit years.

Some functions like MIN() and MAX() convert a YEAR to a number. This means that a value with a two-digit year does not work properly with these functions. The fix in this case is to convert the TIMESTAMP or YEAR to four-digit year format.

The CHAR and VARCHAR types are similar, but differ in the way they are stored and retrieved. They also differ in maximum length and in whether trailing spaces are retained. No lettercase conversion takes place during storage or retrieval.

The CHAR and VARCHAR types are declared with a length that indicates the maximum number of characters you want to store. For example, CHAR(30) can hold up to 30 characters.

The length of a CHAR column is fixed to the length that you declare when you create the table. The length can be any value from 0 to 255. When CHAR values are stored, they are right-padded with spaces to the specified length. When CHAR values are retrieved, trailing spaces are removed.

Values in VARCHAR columns are variable-length strings. The length can be specified as a value from 0 to 65,535. (The maximum effective length of a VARCHAR is determined by the maximum row size and the character set used. The maximum column length is subject to a row size of 65,532 bytes.)

In contrast to CHAR, VARCHAR values are stored using only as many characters as are needed, plus one byte to record the length (two bytes for columns that are declared with a length longer than 255).

VARCHAR values are not padded when they are stored. Trailing spaces are retained when values are stored and retrieved, in conformance with standard SQL.

If you assign a value to a CHAR or VARCHAR column that exceeds the column's maximum length, the value is truncated to fit. If the truncated characters are not spaces, a warning is generated. For truncation of non-space characters, you can cause an error to occur (rather than a warning) and suppress insertion of the value by using strict SQL mode. See Section聽5.2.6, 鈥淪QL Modes鈥�.

The following table illustrates the differences between CHAR and VARCHAR by showing the result of storing various string values into CHAR(4) and VARCHAR(4) columns:

Note that the values shown as stored in the last row of the table apply only when not using strict mode; if MySQL is running in strict mode, values that exceed the column length are not stored, and an error results.

If a given value is stored into the CHAR(4) and VARCHAR(4) columns, the values retrieved from the columns are not always the same because trailing spaces are removed from CHAR columns upon retrieval. The following example illustrates this difference:

mysql> CREATE TABLE vc (v VARCHAR(4), c CHAR(4));
Query OK, 0 rows affected (0.01 sec)

mysql> INSERT INTO vc VALUES ('ab  ', 'ab  ');
Query OK, 1 row affected (0.00 sec)

mysql> SELECT CONCAT('(', v, ')'), CONCAT('(', c, ')') FROM vc;
+---------------------+---------------------+
| CONCAT('(', v, ')') | CONCAT('(', c, ')') |
+---------------------+---------------------+
| (ab  )              | (ab)                |
+---------------------+---------------------+
1 row in set (0.06 sec)

Values in CHAR and VARCHAR columns are sorted and compared according to the character set collation assigned to the column.

Note that all MySQL collations are of type PADSPACE. This means that all CHAR and VARCHAR values in MySQL are compared without regard to any trailing spaces. For example:

mysql> CREATE TABLE names (myname CHAR(10), yourname VARCHAR(10));
Query OK, 0 rows affected (0.09 sec)

mysql> INSERT INTO names VALUES ('Monty ', 'Monty ');
Query OK, 1 row affected (0.00 sec)

mysql> SELECT myname = 'Monty  ', yourname = 'Monty  ' FROM names;
+--------------------+----------------------+
| myname = 'Monty  ' | yourname = 'Monty  ' |
+--------------------+----------------------+
|                  1 |                    1 |
+--------------------+----------------------+
1 row in set (0.00 sec)

Note that this is true for all MySQL versions, and that it is not affected by the server SQL mode.

For those cases where trailing pad characters are stripped or comparisons ignore them, if a column has an index that requires unique values, inserting into the column values that differ only in number of trailing pad characters will result in a duplicate-key error. For example, if a table contains 'a', an attempt to store 'a聽' causes a duplicate-key error.

The BINARY and VARBINARY types are similar to CHAR and VARCHAR, except that they contain binary strings rather than non-binary strings. That is, they contain byte strings rather than character strings. This means that they have no character set, and sorting and comparison are based on the numeric values of the bytes in the values.

The allowable maximum length is the same for BINARY and VARBINARY as it is for CHAR and VARCHAR, except that the length for BINARY and VARBINARY is a length in bytes rather than in characters.

The BINARY and VARBINARY data types are distinct from the CHAR BINARY and VARCHAR BINARY data types. For the latter types, the BINARY attribute does not cause the column to be treated as a binary string column. Instead, it causes the binary collation for the column character set to be used, and the column itself contains non-binary character strings rather than binary byte strings. For example, CHAR(5) BINARY is treated as CHAR(5) CHARACTER SET latin1 COLLATE latin1_bin, assuming that the default character set is latin1. This differs from BINARY(5), which stores 5-bytes binary strings that have no character set or collation.

When BINARY values are stored, they are right-padded with the pad value to the specified length. The pad value is 0x00 (the zero byte). Values are right-padded with 0x00 on insert, and no trailing bytes are removed on select. All bytes are significant in comparisons, including ORDER BY and DISTINCT operations. 0x00 bytes and spaces are different in comparisons, with 0x00 < space.

Example: For a BINARY(3) column, 'a聽' becomes 'a聽\0' when inserted. 'a\0' becomes 'a\0\0' when inserted. Both inserted values remain unchanged when selected.

For VARBINARY, there is no padding on insert and no bytes are stripped on select. All bytes are significant in comparisons, including ORDER BY and DISTINCT operations. 0x00 bytes and spaces are different in comparisons, with 0x00 < space.

For those cases where trailing pad bytes are stripped or comparisons ignore them, if a column has an index that requires unique values, inserting into the column values that differ only in number of trailing pad bytes will result in a duplicate-key error. For example, if a table contains 'a', an attempt to store 'a\0' causes a duplicate-key error.

You should consider the preceding padding and stripping characteristics carefully if you plan to use the BINARY data type for storing binary data and you require that the value retrieved be exactly the same as the value stored. The following example illustrates how 0x00-padding of BINARY values affects column value comparisons:

mysql> CREATE TABLE t (c BINARY(3));
Query OK, 0 rows affected (0.01 sec)

mysql> INSERT INTO t SET c = 'a';
Query OK, 1 row affected (0.01 sec)

mysql> SELECT HEX(c), c = 'a', c = 'a\0\0' from t;
+--------+---------+-------------+
| HEX(c) | c = 'a' | c = 'a\0\0' |
+--------+---------+-------------+
| 610000 |       0 |           1 |
+--------+---------+-------------+
1 row in set (0.09 sec)

If the value retrieved must be the same as the value specified for storage with no padding, it might be preferable to use VARBINARY or one of the BLOB data types instead.

A BLOB is a binary large object that can hold a variable amount of data. The four BLOB types are TINYBLOB, BLOB, MEDIUMBLOB, and LONGBLOB. These differ only in the maximum length of the values they can hold. The four TEXT types are TINYTEXT, TEXT, MEDIUMTEXT, and LONGTEXT. These correspond to the four BLOB types and have the same maximum lengths and storage requirements. See Section聽11.5, 鈥淒ata Type Storage Requirements鈥�. No lettercase conversion for TEXT or BLOB columns takes place during storage or retrieval.

BLOB columns are treated as binary strings (byte strings). TEXT columns are treated as non-binary strings (character strings). BLOB columns have no character set, and sorting and comparison are based on the numeric values of the bytes in column values. TEXT columns have a character set, and values are sorted and compared based on the collation of the character set.

If a TEXT column is indexed, index entry comparisons are space-padded at the end. This means that, if the index requires unique values, duplicate-key errors will occur for values that differ only in the number of trailing spaces. For example, if a table contains 'a', an attempt to store 'a聽' causes a duplicate-key error. This is not true for BLOB columns.

When not running in strict mode, if you assign a value to a BLOB or TEXT column that exceeds the data type's maximum length, the value is truncated to fit. If the truncated characters are not spaces, a warning is generated. You can cause an error to occur and the value to be rejected rather than to be truncated with a warning by using strict SQL mode. See Section聽5.2.6, 鈥淪QL Modes鈥�.

In most respects, you can regard a BLOB column as a VARBINARY column that can be as large as you like. Similarly, you can regard a TEXT column as a VARCHAR column. BLOB and TEXT differ from VARBINARY and VARCHAR in the following ways:

LONG and LONG VARCHAR map to the MEDIUMTEXT data type. This is a compatibility feature. If you use the BINARY attribute with a TEXT data type, the column is assigned the binary collation of the column character set.

MySQL Connector/ODBC defines BLOB values as LONGVARBINARY and TEXT values as LONGVARCHAR.

Because BLOB and TEXT values can be extremely long, you might encounter some constraints in using them:

Each BLOB or TEXT value is represented internally by a separately allocated object. This is in contrast to all other data types, for which storage is allocated once per column when the table is opened.

In some cases, it may be desirable to store binary data such as media files in BLOB or TEXT columns. You may find MySQL's string handling functions useful for working with such data. See Section聽12.4, 鈥淪tring Functions鈥�. For security and other reasons, it is usually preferable to do so using application code rather than allowing application users the FILE privilege. You can discuss specifics for various languages and platforms in the MySQL Forums (http://forums.mysql.com/).

An ENUM is a string object with a value chosen from a list of allowed values that are enumerated explicitly in the column specification at table creation time.

An enumeration value must be a quoted string literal; it may not be an expression, even one that evaluates to a string value. This means that you also may not employ a user variable as an enumeration value.

The value may also be the empty string ('') or NULL under certain circumstances:

Each enumeration value has an index:

For example, a column specified as ENUM('one', 'two', 'three') can have any of the values shown here. The index of each value is also shown:

An enumeration can have a maximum of 65,535 elements.

Trailing spaces are automatically deleted from ENUM member values in the table definition when a table is created.

When retrieved, values stored into an ENUM column are displayed using the lettercase that was used in the column definition. Note that ENUM columns can be assigned a character set and collation. For binary or case-sensitive collations, lettercase is taken into account when assigning values to the column.

If you retrieve an ENUM value in a numeric context, the column value's index is returned. For example, you can retrieve numeric values from an ENUM column like this:

mysql> SELECT enum_col+0 FROM tbl_name;

If you store a number into an ENUM column, the number is treated as the index into the possible values, and the value stored is the enumeration member with that index. (However, this does not work with LOAD DATA, which treats all input as strings.) If the numeric value is quoted, it is still interpreted as an index if there is no matching string in the list of enumeration values. For these reasons, it is not advisable to define an ENUM column with enumeration values that look like numbers, because this can easily become confusing. For example, the following column has enumeration members with string values of '0', '1', and '2', but numeric index values of 1, 2, and 3:

numbers ENUM('0','1','2')

If you store 2, it is interpreted as an index value, and becomes '1' (the value with index 2). If you store '2', it matches an enumeration value, so it is stored as '2'. If you store '3', it does not match any enumeration value, so it is treated as an index and becomes '2' (the value with index 3).

mysql> INSERT INTO t (numbers) VALUES(2),('2'),('3');
mysql> SELECT * FROM t;
+---------+
| numbers |
+---------+
| 1       | 
| 2       | 
| 2       | 
+---------+

ENUM values are sorted according to the order in which the enumeration members were listed in the column specification. (In other words, ENUM values are sorted according to their index numbers.) For example, 'a' sorts before 'b' for ENUM('a', 'b'), but 'b' sorts before 'a' for ENUM('b', 'a'). The empty string sorts before non-empty strings, and NULL values sort before all other enumeration values. To prevent unexpected results, specify the ENUM list in alphabetical order. You can also use GROUP BY CAST(col AS CHAR) or GROUP BY CONCAT(col) to make sure that the column is sorted lexically rather than by index number.

If you want to determine all possible values for an ENUM column, use SHOW COLUMNS FROM tbl_name LIKE enum_col and parse the ENUM definition in the Type column of the output.

A SET is a string object that can have zero or more values, each of which must be chosen from a list of allowed values specified when the table is created. SET column values that consist of multiple set members are specified with members separated by commas (鈥�,鈥�). A consequence of this is that SET member values should not themselves contain commas.

For example, a column specified as SET('one', 'two') NOT NULL can have any of these values:

''
'one'
'two'
'one,two'

A SET can have a maximum of 64 different members.

Trailing spaces are automatically deleted from SET member values in the table definition when a table is created.

When retrieved, values stored in a SET column are displayed using the lettercase that was used in the column definition. Note that SET columns can be assigned a character set and collation. For binary or case-sensitive collations, lettercase is taken into account when assigning values to the column.

MySQL stores SET values numerically, with the low-order bit of the stored value corresponding to the first set member. If you retrieve a SET value in a numeric context, the value retrieved has bits set corresponding to the set members that make up the column value. For example, you can retrieve numeric values from a SET column like this:

mysql> SELECT set_col+0 FROM tbl_name;

If a number is stored into a SET column, the bits that are set in the binary representation of the number determine the set members in the column value. For a column specified as SET('a','b','c','d'), the members have the following decimal and binary values:

If you assign a value of 9 to this column, that is 1001 in binary, so the first and fourth SET value members 'a' and 'd' are selected and the resulting value is 'a,d'.

For a value containing more than one SET element, it does not matter what order the elements are listed in when you insert the value. It also does not matter how many times a given element is listed in the value. When the value is retrieved later, each element in the value appears once, with elements listed according to the order in which they were specified at table creation time. For example, suppose that a column is specified as SET('a','b','c','d'):

mysql> CREATE TABLE myset (col SET('a', 'b', 'c', 'd'));

If you insert the values 'a,d', 'd,a', 'a,d,d', 'a,d,a', and 'd,a,d':

mysql> INSERT INTO myset (col) VALUES 
-> ('a,d'), ('d,a'), ('a,d,a'), ('a,d,d'), ('d,a,d');
Query OK, 5 rows affected (0.01 sec)
Records: 5  Duplicates: 0  Warnings: 0

Then all of these values appear as 'a,d' when retrieved:

mysql> SELECT col FROM myset;
+------+
| col  |
+------+
| a,d  |
| a,d  |
| a,d  |
| a,d  |
| a,d  |
+------+
5 rows in set (0.04 sec)

If you set a SET column to an unsupported value, the value is ignored and a warning is issued:

mysql> INSERT INTO myset (col) VALUES ('a,d,d,s');
Query OK, 1 row affected, 1 warning (0.03 sec)

mysql> SHOW WARNINGS;
+---------+------+------------------------------------------+
| Level   | Code | Message                                  |
+---------+------+------------------------------------------+
| Warning | 1265 | Data truncated for column 'col' at row 1 |
+---------+------+------------------------------------------+
1 row in set (0.04 sec)

mysql> SELECT col FROM myset;
+------+
| col  |
+------+
| a,d  |
| a,d  |
| a,d  |
| a,d  |
| a,d  |
| a,d  |
+------+
6 rows in set (0.01 sec)

If strict SQL mode is enabled, attempts to insert invalid SET values result in an error.

SET values are sorted numerically. NULL values sort before non-NULL SET values.

Normally, you search for SET values using the FIND_IN_SET() function or the LIKE operator:

mysql> SELECT * FROM tbl_name WHERE FIND_IN_SET('value',set_col)>0;
mysql> SELECT * FROM tbl_name WHERE set_col LIKE '%value%';

The first statement finds rows where set_col contains the value set member. The second is similar, but not the same: It finds rows where set_col contains value anywhere, even as a substring of another set member.

The following statements also are legal:

mysql> SELECT * FROM tbl_name WHERE set_col & 1;
mysql> SELECT * FROM tbl_name WHERE set_col = 'val1,val2';

The first of these statements looks for values containing the first set member. The second looks for an exact match. Be careful with comparisons of the second type. Comparing set values to 'val1,val2' returns different results than comparing values to 'val2,val1'. You should specify the values in the same order they are listed in the column definition.

If you want to determine all possible values for a SET column, use SHOW COLUMNS FROM tbl_name LIKE set_col and parse the SET definition in the Type column of the output.