ENUM and SET Constraints

 ENUM and SET columns provide an efficient way to define columns that can contain only a given set of values. See Section 11.3.5, “The ENUM Type”, and Section 11.3.6, “The SET Type”.

Unless strict mode is disabled (not recommended, but see Section 5.1.11, “Server SQL Modes”), the definition of a ENUM or SET column acts as a constraint on values entered into the column. An error occurs for values that do not satisfy these conditions:

• An ENUM value must be one of those listed in the column definition, or the internal numeric equivalent thereof. The value cannot be the error value (that is, 0 or the empty string). For a column defined as ENUM('a','b','c'), values such as '', 'd', or 'ax' are invalid and are rejected.

• A SET value must be the empty string or a value consisting only of the values listed in the column definition separated by commas. For a column defined as SET('a','b','c'), values such as 'd' or 'a,b,c,d' are invalid and are rejected.

Errors for invalid values can be suppressed in strict mode if you use INSERT IGNORE or UPDATE IGNORE. In this case, a warning is generated rather than an error. For ENUM, the value is inserted as the error member (0). For SET, the value is inserted as given except that any invalid substrings are deleted. For example, 'a,x,b,y' results in a value of 'a,b'.

The ENUM Type

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

See Section 11.3.1, “String Data Type Syntax” for ENUM type syntax and length limits.

The ENUM type has these advantages:

• Compact data storage in situations where a column has a limited set of possible values. The strings you specify as input values are automatically encoded as numbers. See Section 11.7, “Data Type Storage Requirements” for storage requirements for the ENUM type.

• Readable queries and output. The numbers are translated back to the corresponding strings in query results.

and these potential issues to consider:

• If you make enumeration values that look like numbers, it is easy to mix up the literal values with their internal index numbers, as explained in Enumeration Limitations.

• Using ENUM columns in ORDER BY clauses requires extra care, as explained in Enumeration Sorting.

• Creating and Using ENUM Columns

• Index Values for Enumeration Literals

• Handling of Enumeration Literals

• Empty or NULL Enumeration Values

• Enumeration Sorting

• Enumeration Limitations
  
  

InnoDB and MyISAM Index Statistics Collection

 Storage engines collect statistics about tables for use by the optimizer. Table statistics are based on value groups, where a value group is a set of rows with the same key prefix value. For optimizer purposes, an important statistic is the average value group size.

MySQL uses the average value group size in the following ways:

• To estimate how many rows must be read for each ref access

• To estimate how many rows a partial join produces, that is, the number of rows produced by an operation of the form

  (...) JOIN tbl_name ON tbl_name.key = expr

As the average value group size for an index increases, the index is less useful for those two purposes because the average number of rows per lookup increases: For the index to be good for optimization purposes, it is best that each index value target a small number of rows in the table. When a given index value yields a large number of rows, the index is less useful and MySQL is less likely to use it.

The average value group size is related to table cardinality, which is the number of value groups. The SHOW INDEX statement displays a cardinality value based on N/S, where N is the number of rows in the table and S is the average value group size. That ratio yields an approximate number of value groups in the table.

For a join based on the <=> comparison operator, NULL is not treated differently from any other value: NULL <=> NULL, just as N <=> N for any other N.

However, for a join based on the = operator, NULL is different from non-NULL values: expr1 = expr2 is not true when expr1 or expr2 (or both) are NULL. This affects ref accesses for comparisons of the form tbl_name.key = expr: MySQL does not access the table if the current value of expr is NULL, because the comparison cannot be true.

For = comparisons, it does not matter how many NULL values are in the table. For optimization purposes, the relevant value is the average size of the non-NULL value groups. However, MySQL does not currently enable that average size to be collected or used.

For InnoDB and MyISAM tables, you have some control over collection of table statistics by means of the innodb_stats_method and myisam_stats_method system variables, respectively. These variables have three possible values, which differ as follows:

• When the variable is set to nulls_equal, all NULL values are treated as identical (that is, they all form a single value group).

  If the NULL value group size is much higher than the average non-NULL value group size, this method skews the average value group size upward. This makes index appear to the optimizer to be less useful than it really is for joins that look for non-NULL values. Consequently, the nulls_equal method may cause the optimizer not to use the index for ref accesses when it should.

• When the variable is set to nulls_unequal, NULL values are not considered the same. Instead, each NULL value forms a separate value group of size 1.

  If you have many NULL values, this method skews the average value group size downward. If the average non-NULL value group size is large, counting NULL values each as a group of size 1 causes the optimizer to overestimate the value of the index for joins that look for non-NULL values. Consequently, the nulls_unequal method may cause the optimizer to use this index for ref lookups when other methods may be better.

• When the variable is set to nulls_ignored, NULL values are ignored.

If you tend to use many joins that use <=> rather than =, NULL values are not special in comparisons and one NULL is equal to another. In this case, nulls_equal is the appropriate statistics method.

Multiple-Column Indexes

 MySQL can create composite indexes (that is, indexes on multiple columns). An index may consist of up to 16 columns. For certain data types, you can index a prefix of the column (see Section 8.3.5, “Column Indexes”).

MySQL can use multiple-column indexes for queries that test all the columns in the index, or queries that test just the first column, the first two columns, the first three columns, and so on. If you specify the columns in the right order in the index definition, a single composite index can speed up several kinds of queries on the same table.

A multiple-column index can be considered a sorted array, the rows of which contain values that are created by concatenating the values of the indexed columns.

Note

As an alternative to a composite index, you can introduce a column that is “hashed” based on information from other columns. If this column is short, reasonably unique, and indexed, it might be faster than a “wide” index on many columns. In MySQL, it is very easy to use this extra column:

SELECT * FROM tbl_name
  WHERE hash_col=MD5(CONCAT(val1,val2))
  AND col1=val1 AND col2=val2;

Suppose that a table has the following specification:

CREATE TABLE test (
    id         INT NOT NULL,
    last_name  CHAR(30) NOT NULL,
    first_name CHAR(30) NOT NULL,
    PRIMARY KEY (id),
    INDEX name (last_name,first_name)
);

The name index is an index over the last_name and first_name columns. The index can be used for lookups in queries that specify values in a known range for combinations of last_name and first_name values. It can also be used for queries that specify just a last_name value because that column is a leftmost prefix of the index (as described later in this section). Therefore, the name index is used for lookups in the following queries:

SELECT * FROM test WHERE last_name='Jones';

SELECT * FROM test
  WHERE last_name='Jones' AND first_name='John';

SELECT * FROM test
  WHERE last_name='Jones'
  AND (first_name='John' OR first_name='Jon');

SELECT * FROM test
  WHERE last_name='Jones'
  AND first_name >='M' AND first_name < 'N';

However, the name index is not used for lookups in the following queries:

SELECT * FROM test WHERE first_name='John';

SELECT * FROM test
  WHERE last_name='Jones' OR first_name='John';

Suppose that you issue the following SELECT statement:

SELECT * FROM tbl_name
  WHERE col1=val1 AND col2=val2;

If a multiple-column index exists on col1 and col2, the appropriate rows can be fetched directly. If separate single-column indexes exist on col1 and col2, the optimizer attempts to use the Index Merge optimization (see Section 8.2.1.3, “Index Merge Optimization”), or attempts to find the most restrictive index by deciding which index excludes more rows and using that index to fetch the rows.

If the table has a multiple-column index, any leftmost prefix of the index can be used by the optimizer to look up rows. For example, if you have a three-column index on (col1, col2, col3), you have indexed search capabilities on (col1), (col1, col2), and (col1, col2, col3).

MySQL cannot use the index to perform lookups if the columns do not form a leftmost prefix of the index. Suppose that you have the SELECT statements shown here:

SELECT * FROM tbl_name WHERE col1=val1;
SELECT * FROM tbl_name WHERE col1=val1 AND col2=val2;

SELECT * FROM tbl_name WHERE col2=val2;
SELECT * FROM tbl_name WHERE col2=val2 AND col3=val3;

If an index exists on (col1, col2, col3), only the first two queries use the index. The third and fourth queries do involve indexed columns, but do not use an index to perform lookups because (col2) and (col2, col3) are not leftmost prefixes of (col1, col2, col3).