MySQL field type and corresponding length & bytes

Byte chapter

Byte history

• Americans first coded English letters , It's the earliest ascii code , Use the bottom of a byte pair 7 In English 128 Characters .
• Later the Europeans discovered 128 It's not enough , There are also phonetic symbols in French. How to distinguish them? The higher one should be included , In this way, Europe generally uses a full byte encoding , At most, it can express 256 position . European and American people just like to go straight , Fewer characters , The number of bits used for coding is small ;
• But even if the number of digits is small , Different countries and regions use different character codes , although 0–127 The symbols are the same , however 128–255 The explanation of this paragraph is completely out of order , Even if 2 The base numbers are exactly the same , The characters represented are completely different , such as 135 In French , Hebrew , Russian code is completely different symbols ;
• More troubling , Chinese culture is broad and profound. We find that we have 10 More than ten thousand Chinese characters , You are in Europe and America 256 It's not enough to plug the teeth . So we invented GB2312 These Chinese characters are encoded , Typical use 2 Bytes to represent most commonly used Chinese characters , At most, it can mean 65536 Chinese characters , In this way, it is not difficult to understand that some Chinese characters can be found in Xinhua dictionary , But if you don't deal with it on the computer, you won't be able to display it .
• Now each character set is used to encode , How can the world be unified ？ The Russians sent a letter email To the Chinese , The character set codes on both sides are different , Nima shows that they are all garbled . In order to unify , So we invented unicode, Include all the symbols in the world , Each symbol is given a unique code , Now? unicode It can hold 100 More than ten thousand symbols , The coding of each symbol is different , This can be unified , All languages can communicate , A web page can display the words of different countries at the same time .
• However ,unicode Although it unifies the binary coding of characters all over the world , But it doesn't stipulate how to store it , Pro - .x86 and amd The small end sequence and the large end sequence of the computer architecture are indistinguishable , Don't mention how computers recognize what is unicode still acsii 了 . If Unicode Uniform rules , Each symbol is represented by three or four bytes , Then every letter must be preceded by two or three bytes 0, The size of the text file will therefore be two or three times larger , It's a huge waste of storage . This leads to a consequence ： There is Unicode A variety of storage methods
• The rise of the Internet , Various characters should be displayed on the web page , Must be unified , Pro - .utf-8 Namely Unicode One of the most important implementations . And then there is utf-16、utf-32 etc. .UTF-8 Not a fixed word length encoding , It is a kind of variable length coding . It can be used 1~4 Bytes represent a symbol , The length of the bytes varies according to the symbol . This is a clever design , If the first bit of a byte is 0, Then this byte is just a character ; If the first one is 1, How many in a row 1, It means how many bytes the current character occupies .
• Be careful unicode Character encoding and utf-8 The storage code representation of is different , for example ” yan ” The word Unicode Code is 4E25,UTF-8 Encoding is E4B8A5, This 7 It explains ,UTF-8 Coding not only considers coding , Storage is also considered ,E4B8A5 On the basis of storing the identification code 4E25.
• UTF-8 Use one to four bytes to encode each character .128 individual ASCII character （Unicode Range from U+0000 to U+007F） Just one byte , Latin with diacritic symbols 、 Greek 、 Cyrillic alphabet 、 Armenian 、 Hebrew 、 arabic 、 Syrian and Maldivian （Unicode Range from U+0080 to U+07FF） Two bytes are required , Other basic multilingual planes （BMP） The characters in （CJK Belong to this category -Qieqie notes ） Use three bytes , other Unicode The characters of the auxiliary plane are encoded in four bytes .

English letter ：

code	byte
GB2312	1
GBK	1
GB18030	1
ISO-8859-1	1
UTF-8	1
UTF-16	4
UTF-16BE	2
UTF-16LE	2

Chinese characters

code	byte
GB2312	2
GBK	2
GB18030	2
ISO-8859-1	1
UTF-8	3
UTF-16	4
UTF-16BE	2
UTF-16LE	2

Mysql The field type corresponds to the length

Numeric type

Column type	The amount of storage needed
TINYINT	1 Bytes
SMALLINT	2 Bytes
MEDIUMNT	3 Bytes
INT	4 Bytes
INTEGER	4 Bytes
BIGINT	8 Bytes
FLOAT(X)	4 Bytes
FLOAT	4 Bytes
DOUBLE	8 Bytes
DOUBLE PRECISION	8 Bytes
REAL	8 Bytes
DECIMAL(M,D)	M Bytes (D+2, If M < D)
NUMERIC(M,D)	M Bytes (D+2, If M < D)

The date type

type	The amount of storage needed
DATE	3 Bytes
DATETIME	8 Bytes
TIMESTAMP	4 Bytes
TIME	3 Bytes
YEAR	1 byte

String type

type	The amount of storage needed
CHAR(M)	M byte ,1 <= M <= 255
VARCHAR(M)	L+1 byte , Here it is L <= M and 1 <= M <= 255
TINYBLOB,TINYTEXT	L+1 byte , Here it is L< 2 ^ 8
BLOB,TEXT	L+2 byte , Here it is L< 2 ^ 16
MEDIUMBLOB,MEDIUMTEXT	L+3 byte , Here it is L< 2 ^ 24
LONGBLOB,LONGTEXT	L+4 byte , Here it is L< 2 ^ 32
ENUM(‘value1’, ‘value2’, …)	1 or 2 Bytes , Depends on the number of enumeration values ( Maximum 65535)
SET(‘value1’, ‘value2’)	1,2,3,4 or 8 Bytes , Depends on the number of members of the collection ( most 64 Members )

 Reference resources ：
https://blog.csdn.net/yaomingyang/article/details/79374209
https://blog.csdn.net/lijinzhou2017/article/details/81062877