MySQL 數據類型（float）的注意事項

知識點：
      float：浮點數，單精度，佔4字節。

測試

root@localhost : test 05:49:32>create table fl(id int,fl float);
Query OK, 0 rows affected (0.05 sec)

root@localhost : test 05:49:40>insert into fl values(1,123456);
Query OK, 1 row affected (0.00 sec)

root@localhost : test 05:49:51>insert into fl values(2,123.456);
Query OK, 1 row affected (0.00 sec)

root@localhost : test 05:49:57>insert into fl values(3,1234.567);#7位了
Query OK, 1 row affected (0.00 sec)

root@localhost : test 05:50:13>insert into fl values(4,1234567);#7位了
Query OK, 1 row affected (0.00 sec)

root@localhost : test 05:50:26>select * from fl;
+------+-------------+
| id   | fl          |
+------+-------------+
|    1 |      123456 |
|    2 |     123.456 |
|    3 |     1234.57 |
|    4 | 1.23457e+06 |
+------+-------------+
4 rows in set (0.00 sec)

從上面看出：默認的float類型都只能存6個數字（包括小數點前後的位數），整數超過6位就被科學計數表示（id=4），小數位超出則需要四捨五入。那指定float(m)的精度呢？

root@localhost : test 05:59:08>create table fl(id int,fl float(20));
Query OK, 0 rows affected (0.04 sec)

root@localhost : test 05:59:28>show create table fl\G;
*************************** 1. row ***************************
Table: fl
Create Table: CREATE TABLE `fl` (
  `id` int(11) DEFAULT NULL,
  `fl` float DEFAULT NULL
) ENGINE=MyISAM DEFAULT CHARSET=utf8
1 row in set (0.00 sec)

root@localhost : test 05:59:35>insert into fl values(1,123456);
Query OK, 1 row affected (0.00 sec)

root@localhost : test 05:59:44>insert into fl values(2,123.456);
Query OK, 1 row affected (0.00 sec)

root@localhost : test 05:59:46>insert into fl values(3,1234.567);#7位了
Query OK, 1 row affected (0.00 sec)

root@localhost : test 05:59:50>insert into fl values(4,1234567);#7位了
Query OK, 1 row affected (0.00 sec)

root@localhost : test 05:59:53>select * from fl;
+------+-------------+
| id   | fl          |
+------+-------------+
|    1 |      123456 |
|    2 |     123.456 |
|    3 |     1234.57 |
|    4 | 1.23457e+06 |
+------+-------------+
4 rows in set (0.01 sec)

從上面看出：和默認的float一樣。所以：默認float和float(m)一樣 ，m<=24； 都是默認float類型，都只能存6個數字（包括小數點前後的位數），整數超過6位就被科學計數表示（id=4）；小數位超出則需要四捨五入。那指定他的標度float(m,d) 呢？

mysql> show create table fl\G;
*************************** 1. row ***************************
Table: fl
Create Table: CREATE TABLE `fl` (
  `id` int(11) DEFAULT NULL,
  `fl` float(7,4) DEFAULT NULL
) ENGINE=MyISAM DEFAULT CHARSET=utf8
1 row in set (0.02 sec)

ERROR: 
No query specified

mysql> insert into fl values(1,123456);#整數是3位，超出了
Query OK, 1 row affected, 1 warning (0.00 sec)

mysql> insert into fl values(3,123.4567);
Query OK, 1 row affected (0.00 sec)

mysql> insert into fl values(2,123.45678);#小數位超出，四捨五入
Query OK, 1 row affected (0.00 sec)

mysql> insert into fl values(4,12.456789);#小數位超出，四捨五入
Query OK, 1 row affected (0.00 sec)

mysql> select * from fl;
+------+----------+
| id   | fl       |
+------+----------+
|    1 | 999.9999 |
|    3 | 123.4567 |
|    2 | 123.4568 |
|    4 |  12.4568 |
+------+----------+
4 rows in set (0.00 sec)

從上面看出：float(m,d)：小數點後位數爲d，即整數位數爲（m-d），整數位超出則整數爲(m-d)個999，小數點後位數爲(d)個9999，不用科學計算了；若小數位超出，則需要四捨五入。要是float(m,d)遇到m和d一樣呢？

mysql> CREATE TABLE `fl` (
->   `id` int(11) DEFAULT NULL,
->   `fl` float(4,4) DEFAULT NULL
-> ) ENGINE=MyISAM DEFAULT CHARSET=utf8;
Query OK, 0 rows affected (0.04 sec)

mysql> insert into fl values(1,123456);
Query OK, 1 row affected, 1 warning (0.00 sec)

mysql> insert into fl values(2,123.45678);
Query OK, 1 row affected, 1 warning (0.00 sec)

mysql> insert into fl values(3,123.4567);
Query OK, 1 row affected, 1 warning (0.00 sec)

mysql> insert into fl values(4,12.456789);
Query OK, 1 row affected, 1 warning (0.00 sec)

mysql> select * from fl;
+------+--------+
| id   | fl     |
+------+--------+
|    1 | 0.9999 |
|    2 | 0.9999 |
|    3 | 0.9999 |
|    4 | 0.9999 |
+------+--------+
4 rows in set (0.00 sec)

從上面看出：float(m,d)：若m和d一樣，則先把小數位數先滿足了，再分配整數位數。所以只有小數點位數，而整數位是0。

之前提過的m<=24，要是大於24呢？

mysql> CREATE TABLE `fl` (
->   `id` int(11) DEFAULT NULL,
->   `fl` float(25) DEFAULT NULL
-> ) ENGINE=MyISAM DEFAULT CHARSET=utf8;
Query OK, 0 rows affected (0.01 sec)

mysql> show create table fl\G;
*************************** 1. row ***************************
Table: fl
Create Table: CREATE TABLE `fl` (
  `id` int(11) DEFAULT NULL,
  `fl` double DEFAULT NULL
) ENGINE=MyISAM DEFAULT CHARSET=utf8
1 row in set (0.00 sec)

從上面看出：在沒有標度的float中，從float轉換成了double。而有標度的 float(m,d)當中，m>24呢？

mysql> CREATE TABLE `fl` (
->   `id` int(11) DEFAULT NULL,
->   `fl` float(100,3) DEFAULT NULL
-> ) ENGINE=MyISAM DEFAULT CHARSET=utf8;
Query OK, 0 rows affected (0.01 sec)

mysql> show create table fl\G;
*************************** 1. row ***************************
Table: fl
Create Table: CREATE TABLE `fl` (
  `id` int(11) DEFAULT NULL,
  `fl` float(100,3) DEFAULT NULL
) ENGINE=MyISAM DEFAULT CHARSET=utf8

從上面看出：從float還是float，沒有受到影響。不過設置這麼大，在整數會變的不準；小數的位數還是m-d，位數不足用0填充（zerofill）。如下測試：

mysql> show create table fl\G;
*************************** 1. row ***************************
Table: fl
Create Table: CREATE TABLE `fl` (
  `id` int(11) DEFAULT NULL,
  `fl` float(100,3) DEFAULT NULL
) ENGINE=MyISAM DEFAULT CHARSET=utf8
1 row in set (0.00 sec)

mysql> insert into fl values(5,777777777777777777777777777.5555);
Query OK, 1 row affected (0.00 sec)

mysql> select * from fl;
+------+---------------------------------+
| id   | fl                              |
+------+---------------------------------+
|    5 | 777777744225350500000000000.000 |  #不準了
+------+---------------------------------+
1 row in set (0.00 sec)
mysql> drop table fl;
Query OK, 0 rows affected (0.00 sec)

mysql> CREATE TABLE `fl` (
->   `id` int(11) DEFAULT NULL,
->   `fl` float(25,23) DEFAULT NULL
-> ) ENGINE=MyISAM DEFAULT CHARSET=utf8;
Query OK, 0 rows affected (0.02 sec)

mysql> show create table fl\G;
*************************** 1. row ***************************
Table: fl
Create Table: CREATE TABLE `fl` (
  `id` int(11) DEFAULT NULL,
  `fl` float(25,23) DEFAULT NULL
) ENGINE=MyISAM DEFAULT CHARSET=utf8
1 row in set (0.00 sec)


mysql> insert into fl values(5,99.555555555555555);
Query OK, 1 row affected (0.00 sec)

mysql> select * from fl;
+------+----------------------------+
| id   | fl                         |
+------+----------------------------+
|    5 | 99.55555725097656000000000 | #小數位被0填充
+------+----------------------------+
1 row in set (0.00 sec)

注意：當float(m,d) 設置了標度之後，小數位一定會存在。當整數位沒有被超過，需要用0填充；整數位被超過了，需要用9填充，小數位數不能大於30否則報錯：

mysql> CREATE TABLE `fl` (
->   `id` int(11) DEFAULT NULL,
->   `fl` float(5,3) DEFAULT NULL
-> ) ENGINE=MyISAM DEFAULT CHARSET=utf8
-> ;
Query OK, 0 rows affected (0.03 sec)

mysql> insert into fl values(1,123);
Query OK, 1 row affected, 1 warning (0.00 sec)

mysql> select * from fl;
+------+--------+
| id   | fl     |
+------+--------+
|    1 | 99.999 |
+------+--------+
1 row in set (0.00 sec)

mysql> insert into fl values(2,13);
Query OK, 1 row affected (0.00 sec)

mysql> insert into fl values(3,3);
Query OK, 1 row affected (0.00 sec)

mysql> select * from fl;
+------+--------+
| id   | fl     |
+------+--------+
|    1 | 99.999 |    #9填充
|    2 | 13.000 |    #0填充  
|    3 |  3.000 |
+------+--------+
3 rows in set (0.00 sec)

mysql> CREATE TABLE `fl` (
->   `id` int(11) DEFAULT NULL,
->   `fl` float(102,100) DEFAULT NULL
-> ) ENGINE=MyISAM DEFAULT CHARSET=utf8
-> ) ENGINE=MyISAM DEFAULT CHARSET=utf8;
ERROR 1425 (42000): Too big scale 100 specified for column 'fl'. Maximum is 30.   #標度不能大於30，等於30可以建立
mysql> CREATE TABLE `fl` (
->   `id` int(11) DEFAULT NULL,
->   `fl` float(52,50) DEFAULT NULL
-> ) ENGINE=MyISAM DEFAULT CHARSET=utf8;
ERROR 1425 (42000): Too big scale 50 specified for column 'fl'. Maximum is 30.

總結：float存在這麼多潛在因素，用的時候需要注意，能避免就避免用。