FIND_IN_SET() alternative?

Question

I have a query that currently looks like:

SELECT [column a], [column b], [column c], [column d]
FROM [table] 
WHERE FIND_IN_SET(2, column d)
ORDER BY [colu         


        

Answer 1

Alternative: properly normalize the schema.

FIND_IN_SET is not Sargable and the index cannot be used.

Answer 2

One possible optimization is to define [column d] as type SET. As documentation says:

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.

Here's a quick and simple example:

CREATE TABLE `tbl_name` (
  `id` int(11) NOT NULL AUTO_INCREMENT,
  `set_col` set('a','b','c','d') NOT NULL,
  PRIMARY KEY (`id`),
  KEY `set_col_idx` (`set_col`)
) ENGINE=InnoDB DEFAULT CHARSET=utf8;

INSERT INTO `tbl_name` (`set_col`) VALUES
('a'),
('b'),
('c'),
('d'),
('a,b'),
('a,c'),
('a,d'),
('b,c'),
('b,d'),
('c,d'),
('a,b,c'),
('a,b,d'),
('a,c,d'),
('b,c,d'),
('a,b,c,d');

A column set_col specified as SET('a','b','c','d') have members with following decimal and binary values:

╔════════════╦═══════════════╦══════════════╗
║ SET Member ║ Decimal Value ║ Binary Value ║
╠════════════╬═══════════════╬══════════════╣
║     'a'    ║       1       ║     0001     ║
╠════════════╬═══════════════╬══════════════╣
║     'b'    ║       2       ║     0010     ║
╠════════════╬═══════════════╬══════════════╣
║     'c'    ║       4       ║     0100     ║
╠════════════╬═══════════════╬══════════════╣
║     'd'    ║       8       ║     1000     ║
╚════════════╩═══════════════╩══════════════╝

So, if you need to retrive records with value a,c,d, it's first, third and fourth member, which is 1 + 4 + 8, which is 13.

If you run query:

EXPLAIN SELECT * FROM `tbl_name` WHERE `tbl_name`.`set_col` = 13;

You'll get:

╔════╦═════════════╦══════════╦══════╦═══════════════╦═════════════╦═════════╦═══════╦══════╦═════════════╗
║ id ║ select_type ║ table    ║ type ║ possible_keys ║ key         ║ key_len ║ ref   ║ rows ║ Extra       ║
╠════╬═════════════╬══════════╬══════╬═══════════════╬═════════════╬═════════╬═══════╬══════╬═════════════╣
║  1 ║ SIMPLE      ║ tbl_name ║ ref  ║ set_col_idx   ║ set_col_idx ║ 1       ║ const ║ 1    ║ Using index ║
╚════╩═════════════╩══════════╩══════╩═══════════════╩═════════════╩═════════╩═══════╩══════╩═════════════╝

You don't need to manualy get know decimal values of SET options – you can use SUM, e.g.:

SELECT *
FROM `tbl_name`
WHERE `set_col` = (SELECT SUM(`set_col`)
 FROM `tbl_name`
 WHERE `set_col` IN ('a', 'c', 'd')
);

╔════╦═════════════╦══════════╦═══════╦═══════════════╦═════════════╦═════════╦═══════╦══════╦══════════════════════════╗
║ id ║ select_type ║ table    ║ type  ║ possible_keys ║ key         ║ key_len ║ ref   ║ rows ║ Extra                    ║
╠════╬═════════════╬══════════╬═══════╬═══════════════╬═════════════╬═════════╬═══════╬══════╬══════════════════════════╣
║  1 ║ PRIMARY     ║ tbl_name ║ index ║ set_col_idx   ║ set_col_idx ║ 1       ║ NULL  ║ 15   ║ Using where; Using index ║
╠════╬═════════════╬══════════╬═══════╬═══════════════╬═════════════╬═════════╬═══════╬══════╬══════════════════════════╣
║  1 ║ SUBQUERY    ║ tbl_name ║ range ║ set_col_idx   ║ set_col_idx ║ 1       ║ NULL  ║ 3    ║ Using where; Using index ║
╚════╩═════════════╩══════════╩═══════╩═══════════════╩═════════════╩═════════╩═══════╩══════╩══════════════════════════╝

Answer 3

1- fulltext index is not a good idea in this case because: length of the string you searching for is small (1) in this case, and this won't be found (It is configurable though, but not a good idea)

2- If this query is frequent, I suggest changing the tables' structure as follows:

• table1 (col_a PK, col_b, col_c)
• table2 (col_a FK, value_of_sub_d) where value_of_sub_d is one of (2, 3, ...)

In this one-to-many relationship, you can either do a join, or get the PK from table2 where condition is met, and select that ID's row from table1 Example:

Select * from table1 t1 inner join table2 t2 on t1.col_a=t2.col_a WHERE t2.value_of_sub_d=2