Chapter 4 of getting started with MySQL: data types stored in data tables

MySQL Support multiple data types , It can be roughly divided into three categories , They are numeric types 、 Date and time type 、 character string （ character ) type .

1.1 value type

MySQL Support all standards SQL Numerical data type . These types include strict numerical data types （INTEGER,
SMALLINT、TINYINT、MEDIUMINT and BIGINT), Approximate numerical data type （FLOAT、REAL and DOUBLE), And fixed-point number type （DECIMAL).

Be careful : keyword INT yes INTEGER A synonym for , keyword DEC yes DECIMAL
A synonym for .

MySQL Provide multiple integer types , Different data types provide different value ranges , The larger the range of values that can be stored , The more storage space it needs , Therefore, we should choose the appropriate data type according to the actual needs .

The following figure shows the storage requirements and value range of each integer type .

Type the name	explain	Storage demand byte	The value range of signed number	The value range of the unsigned number
TINYINT	Very small integers	1	128～127	0～255
SMALLINT	Small integers	2	32768～32767	0～65535
MEDIUMINT	An integer of medium size	3	-8388608~8388607	0～16777215
INT (INTEGER）	An integer of normal size	4	-2147483648～2147483647	0～4294967295
BIGINT	Large integer	8	9223372036854775808~9223372036854775807	0～18446744073709551615

MSQL It supports the selection of the display width of integer values specified in parentheses behind keywords of this type （ for example INT(4)).INT(M) Medium M Indicates the maximum display width , The maximum effective display width is 4, It should be noted that , The display width is independent of the range of values contained in the storage size or type .

1.2 Floating point type

In real life, many situations need to store numerical values with decimal parts , This requires a floating point number type , Such as FLOAT and DOUBLE. among ,FLOAT It is a single precision floating-point number type ;DOUBLE Double precision floating-point number type . Floating point number type can be used (M,D) To express , among M It's called precision , The total number of digits ;D Is called a scale , Represents the number of decimal places . The following table shows the storage requirements and value range of each floating-point number type .

Type the name	Storage requirements / byte	Signed value range	The range of unsigned values
FLOAT	4	-3.402823466E+38～-1.175494351E-38	0 and 1.175494351E-38~3.402823466E+38
DOUBLE	8	-1.7976931348623157E+308～ -2.2250738585072014E-308	0 and 2.2250738585072014E-308～ 1.7976931348623157E+308

Be careful :M and D stay FLOAT and DOUBLE Is optional ,FLOAT and DOUBLE The type will be saved to the maximum precision supported by the hardware .

1.3 Fixed point number type

MySQL in , Except that floating-point numbers are used to represent decimals , You can also use fixed-point numbers to represent decimals , There is only one type of fixed-point number :DECIMAL. Fixed point number type can also be used (M,D) To express , among M It's called precision , The total number of digits ;D Is called a scale , Represents the number of decimal places .DECIMAL Default D The value is 0,M The value is 10. The following table shows the storage requirements and value range of fixed-point number types .
DECIMAL Different types of dry FLOAT and DECIMAL, DECIMAL It is actually stored as a string .DECIMAL The valid value range of is determined by M and D Value determination of . If you change M And fixed D, Then its value range will follow M Become bigger and change

1.4 Date and time type

MySQL in , The date and time type that represent the time value are DATETIME、DATA、TIMESTAMP,TIME and YEAR. for example , Just record the year information , It can be used only YEAR type . Each type has a legal value range .

Type the name	Date format	Date range	Storage requirements / byte
YEAR	YYYY	1901～2155	1
TIME	HH:MM:SS	-838:59:59-838:59:59	3
DATE	YYYY-MM-DD	1000-01-019999-12-31	3
DATETIME	YYYY-MM-DD HH:MM:SS	1000-01-01 00:00:009999-12-3123:59:59	8
TIMESTAMP	YYYY-MM-DD HH:MM:SS	1970-01-01 00:00:001～2038-01-19 03:14:07	4

1.5 String type

String type is used to store string data ,MySQL Support two types of string data : Text strings and binary strings . Text strings can be used for case sensitive or case insensitive string comparison , You can also search for pattern matching .MySQL The string type in refers to CHAR、VARCHAR、TINYTEXT、TEXT、MEDIUMTEXT、LONGTEXT,ENUM and SET. The following table lists them MySQL String data type in .

Type the name	explain	Storage requirements
CHAR M)	Fixed length non binary string	M byte ,1<=M<=255
VARCHAR(M)	Variable length non binary strings	L+1 byte , Here it is L<=M and 1<=M<=255
TINYTEXT	Very small non binary strings	L+1 byte , Here it is L<28
TEXT	Small non binary strings	L+2 byte , Here it is L<216
MEDIUMTEXT	Medium sized non binary strings	L+3 byte , Here it is L<224
LONGTEXT	Large non binary strings	L+4 byte , Here it is L<232
ENUM	Enumeration type , There can only be one enumeration string value	1 or 2 byte , Depends on the number of enumeration values （ Maximum 65535)
SET	A collection , String objects can have zero or more SET member	1,2,3,4, or 8 byte , Depends on the number of members of the collection ( most 64 Members ）

1.6 Data type selection

MySQL Provides a large number of data types , To optimize storage , Improve database performance , In any case, the most precise type should be used , That is, in all types that can represent the value of the column , This type uses the least storage

1. Integers and floating point numbers
   If you don't need a fraction , Use integers to hold data ; If necessary, the decimal part , Floating point number type . For floating point data columns , The stored values will round the decimal places defined in the column . for example , If the range of values of the column is 1~99999, If you use integers , be MEDIUMINT UNSIGNED It's the best type ; If you need to store decimals , Then use FLOAT type .
   Floating point types include FLOAT and DOUBLE type .DOUBLE The type accuracy is better than FLOAT Type high , therefore , If high storage accuracy is required , Choose DOUBLE type .

2. Floating point and fixed point numbers
   Floating point numbers FLOAT and DOUBLE Relative to the fixed number DECIMAL The advantage is : In the case of a certain length , Floating point numbers can represent a larger range of data . But because floating point numbers are prone to errors , Therefore, when the accuracy requirement is relatively high , It is recommended to use DECIMAL To store .DECIMAL stay MySQL Is stored as a string , Used to define currency and other data requiring high accuracy . In data migration ,FLOAT(M,D) Right or wrong SQL Definition , There may be problems with database migration , It's best not to use . in addition , It is also easy to have problems when subtracting and comparing two floating-point numbers , So in the calculation , Be careful . If you do a numerical comparison , Best use DECIMAL type .

3. Date and time type
   MySQL There are many data types for different kinds of dates and times , such as YEAR and TIME, If you only need to record the year , Then use YEAR The type is enough ; If you only record the time , Just use TIME type . If you need to record the date and time at the same time , You can use TIMESTAMP perhaps DATETIME type . because TIMESTAMP The value range of column is less than DATETIME Value range of , Therefore, it is best to use the date with a large storage range DATETIME.TIMESTAMP There is also a DATETIME Attributes that you don't have . By default , When a record is inserted but not specified TIMESTAMP When this column value ,MySQL Will be able to TIMESTAMP The column is set to the current time . Therefore, when it is necessary to insert the current time while inserting the record , Use TIMESTAMP It's convenient , in addition TIMESTAMP In space than DATETIME More effective .

4. CHAR And VARCHAR Between the characteristics and choices CHAR and VARCHAR The difference is as follows :
   (1）CHAR It's a fixed length character ,VARCHAR It's a variable length character ;
   (2）CHAR The trailing space of the inserted data will be deleted automatically ,VARCHAR Does not remove trailing spaces ;
   (3）CHAR It's a fixed length , So it's faster than VARCHAR It's faster , But its disadvantage is a waste of storage space . So it's not a big memory , But those with speed requirements can be used CHAR type ;
   conversely , have access to VARCHAR Type to implement