Review of MySQL knowledge points

MYSQL Study

install ：sudo apt install mysql-server

https://blog.csdn.net/qq_38505969/article/details/109957055

mysql -h Host name -p port -u user name -p password （ Be careful p No space after ）

or mysql -u user name -p password

or mysql -u user name -p

-u： user name ,-p： password

sign out ：exit、quit、ctrl+d

see mysql process ：ps -aux | grep mysql

see mysql Service status ：sudo service mysql status

stop it mysql service ：sudo service mysql stop

start-up mysql service ：sudo service mysql start

restart mysql service ：sudo service mysql restart

mysql Profile path ：/etc/mysql/mysql.conf.d/mysqld.cnf

View currently used databases ：select database();

Delete database ：drop database Library name ;

View database creation information ：show create database Library name ;

Modify database character set ：alter database Library name charset=utf8;

Simply create a data table ：create table user(id int,name char(7),age tinyint);

Create standard data tables ：create table person(id int(5) unsigned zerofill primary key auto_increment,name char(5) not null,height float(5,2) default 0.0,gender enum(‘ male ’,‘ Woman ’) default ‘ male ’);

View table structure ：desc Table name ;

See the information for creating the table ：show creat table Table name ;

Add fields to table ：alter table Table name add weight float(5,2) not null;

Change the type for the table ：alter table Table name modify weight int(5);

Modify the field names and constraints of the table ：alter table Table name change weight tizhong float(5,3) default 99.99;

When the field type changes , If there is data in the original field , Then the number string and the number can be converted

You can modify multiple fields and constraints at the same time , Separated by commas .

Delete table fields alter table Table name drop tizhong;

Delete table ：drop table Table name ;

Inquire about ：

​ select * from person;

Insert ：

​ insert into Table name values(1,‘tom’,default,male),(2,‘Jack’,178,male); If you don't want to provide data , Also use default placeholder

​ insert into Table name (name) values(‘Lucy’),(‘Tom’) ;

​ insert into Table name (id,name,gender) values(5,‘Bulke’,1);

Update data ：

​ update Table name set Field name = value ;

​ update Table name set Field name = value where Conditions ;

Delete data ：

​ delete from Table name ;

​ delete from Table name where Conditions ;

Alias ：

​ select id as Serial number ,name as full name ,gender as Gender from person;(as It can be omitted )

Data De duplication ：

​ select distinct name from person;

Fuzzy query ：like,%(0 Or more characters ),_( A character ),

​ select * from person where name like ‘ Grandchildren %’;

​ select * from person where name like ‘ Grandchildren _’;

​ select * from person where name like ‘% Grandchildren %’;

Range queries ：between…and…( In a continuous range ),in( A discontinuous range )

​ select * from person where height between 170 and 180;

​ select * from person where id in (1,7,9);

Empty judgment ：is null,is not null

​ select * from person where height is null;

​ Be careful ：where height is not null; and where not height is null; The results are the same but the execution logic is different , The latter is to take out the empty data first and then invert the selection .

Sort ：order by

​ select * from Table name order by Name asc： Ascending order , Don't write asc Default ascending order

​ select * from Table name order by Name desc： Descending order

​ select * from Table name order by Column 1 desc, Column 2 asc： When the same value in column 1 cannot be sorted, use column 2 to sort

​ where Conditions To write in order by In front of

Paging query ：limit,start—— Index start position ,count—— Number of pieces

​ select * from Table name limit start,count from start+1 Start reading count Data ,start Do not write default is 0,count Show as many as you have when you don't have enough

​ select * from student limit (n-1)*m,m For the first n Page data

​ select * from person where gender=‘ Woman ’ order by gender desc limit 5; Query the five oldest people of female sex

​ where Conditions ——order by——limit

Aggregate functions ： Used for statistics and calculation of data

​ count(col)： Find the total row number of the specified column ( Do not count rows with null values )、max(col)、min(col)、sum(col)、avg(col)： Average the specified column

​ round(), Specify the number of decimal places to keep , Such as round(avg(height),2)

​ select count(height),max(height),min(height),sum(height),avg(height) from person;

​ It's generally used count(*), Null values can also be counted

ifnull(height,0)： If height It's empty , Then use 0 Instead of , Can be used with aggregate functions , Such as count(ifnull(height,0));

Group query ：group by

​ select gender from person group by gender; You can query through any group ——select A from Table name group by A, This method can only see the grouped categories , You can't see the members of each group

​ group by And group_concat() Use it together to view the members of each group ：select gender ,group_concat(name) from person group by gender;

​ having Filter grouping data （having It can only be used for group by）：select gender,group_concat(name) from person where age>50 group by gender having gender=‘ Woman ’; Displays age greater than 50 Name of the female member of

​ Used with aggregate functions ：select gender,count(age),max(age),min(age),sum(age),avg(age) from person where age is not null group by gender;

​ with rollup：select gender,count(age),max(age),min(age),sum(age),avg(age) from person where age is not null group by gender with rollup; Add a new line after the last record , Show select Statistics and calculation results of aggregate functions during query

Internal connection 、 Left connection 、 The right connection 、 Self connection query where Must be written after the join statement

Internal connection ：inner join on according to on The following conditions take the values of two tables “ intersection ”,on Omission

​ select * from student inner join class; In this case, the number of items queried is student Number of entries in class The product of the number of pieces
​ select * from student inner join class on student.class_id=class.id; on Following condition

​ select s.id,s.name,c.class_name class from student as s inner join class c on s.class_id=c.id; as It can be omitted

The left outer join ：left join on Mainly in the left table , Query the data in the right table according to the criteria , If the data in the right table does not exist, use null fill ,on Do not omit

​ select Field from The left table left join Right table on The left table . Field 1= Right table . Field 2;

Right connection ：right join on Mainly in the right table , Query the data in the left table according to the criteria , If the data in the left table does not exist, use null fill ,on Do not omit

​ select Field from The left table right join Right table on The left table . Field 1= Right table . Field 2;

​ Swap the left table with the right table ,right and left Replace , The result of the left connection query is the same as that of the right connection query

Self join ： The left table and the right table are the same table , Internal connection 、 Left connection 、 Right connection can realize self connection

​ select p.name Provincial name c.name City name from areas p inner join areas c on p.id=c.pid; pid It means the superior place name of the changed place name id

Subquery ： The subquery is a complete select sentence , Embed in main query

​ select * from student where age > (select avg(age) from student); Query students older than the average

​ select name from class where id in (select distinct class_id from student where class_id is not null); Query which classes students belong to

​ select * from student where (age,height) = (select max(age),max(height) from student); Find the oldest and tallest student

Foreign key constraints ：foreign key(col) reference

​ When the value of the foreign key field is updated and inserted , The fields in the table will be referenced for data validation , If the data does not conform to the law, it will fail , Ensure the validity of data

​ A table can have multiple foreign keys

​ Write directly when creating a table ：create table Test(id int(5) not null primary key auto_increment, Field 1 char(5),foreign key( Field 1) references surface 2( Lot 2)); Field 1 And field 2 The type should be consistent , Referenced fields , That is, the fields 2 It must be a watch 2 Primary key of

​ Be careful ： After defining foreign keys , If you want to delete the referenced table or the data in the table , You need to delete the table that defines the foreign key or the data in the table , Just like if you want to dissolve a company , First dismiss the employee . Or delete the foreign key constraint first .

​ Delete foreign key constraint ：

​ 1. Get foreign key name （ Automatic system generation ）：show create table Table name ; View foreign key name

​ 2. Delete foreign key constraint ：alter table Table name drop foreign key Foreign key name ;

​ Add a foreign key constraint ： If the field 1 There are fields in that cannot be associated with 2 Matching data , Adding a foreign key will fail .

​ alter table Table name add constraint fk_class_id foreign key( Field 1) references surface 2( Field 2); fk_class_id Is the foreign key name ,constraint fk_class_id It can be omitted , If omitted, the system automatically generates the foreign key name , The foreign key name is usually fk_+ Field name

Insert the query results as data into another table ：

​ insert into … select…; No need to write values()

Use the join to update the data in the table ：

​ update surface 1 inner/left/right join surface 2 on…set…;

When creating a table, add data by querying ：create table…select…

​ create table Table name (id int primary key auto_increment,name char(10) not null) select name from Table name ; perhaps select Field name as name from Table name .id You can omit it , By default, the matching starts from the second field , Be careful ：select The field name that follows must be the same as the field name that was added when the table was created .

Business ：

​ A transaction is a series of user-defined executions SQL Statement operation （ increase 、 Delete 、 Change ）, All operations in a transaction are either fully performed , Or not at all （ If you exit unexpectedly halfway through the execution, a rollback will occur ）, It is an indivisible work execution unit .

​ Execute the modification instruction after the transaction is started , The changed data will be saved to mysql In the server cache file , Instead of maintaining it in the physical table , Only execute commit Then the data in the local cache file will be submitted to the original physical table , Complete the data update .

​ Be careful ： If not shown, start a transaction , Every one of them sql Statements are treated as a transaction , And because of mysql The default mode is auto submit (autocommit), Therefore, the transaction submission operation will be performed automatically .

​ set autocommit=0; Indicates that the auto commit transaction mode is canceled , Manual required commit Complete transaction commit （ When a transaction is started , It's automatically set autocommit The value is 0）.

​ The execution of the display commit or rollback Indicates the end of the transaction

​ Four characteristics ：

​ Atomicity (Atomicity)： Emphasize that multiple operations in a transaction are a whole

​ Uniformity (Consistency)： Emphasize keeping consistent state in the database , for example ATM How much do you withdraw when you withdraw money , How much will the balance in the bank card be reduced

​ Isolation, (Isolation)： Emphasize that the operations between multiple transactions in the database are not visible to each other

​ persistence (Durability)： Emphasize that the database can store data permanently , Once submitted, it is irrevocable

​ Only InnoDB The engine can use transactions , The common table storage engine is InnoDB and MyISAM,MyISAM Unsupported transaction , The advantage is fast access .

​ Modify the storage engine of the table ：alter table Table name engine=‘MyISAM’;

Operation of transaction ：

​ sql sentence ：

​ Open transaction begin; or start transaction;

​ Commit transaction commit;

​ Roll back the transaction rollback;

​ After committing or rolling back a transaction , The current transaction will end

pymysql：

​ Connect to database objects .cursor() Create cursor object , Automatically and implicitly start a transaction

​ Commit transaction ： Connect to database objects .commit()

​ Roll back the transaction ： Connect to database objects .rollback()

Indexes ：

​ stay mysql Also known as ” key “, It's a special document , Keep all the location information recorded in the data sheet , It's like a book catalog , Can speed up the query speed of the database .

​ show index from Table name ; Look at the index （ The primary key column will automatically create an index ）

​ alter table test_index add index idx_title(title); Create index （ It's very time consuming ）, The name of a non unique index is generally set to idx_+ Field name , If you do not write, the default is the field name , After adding the index , Use where You only need to compare the data once .

​ Add... Before the query statement desc By looking at rows Property value to view the number of comparisons ,Extra Property value view query method

​ alter table test_index drop index idx_title; Delete index

​ set profiling=1; Start time monitoring

​ show profiles; View the recorded data of time monitoring

Joint index （ Composite index ）： An index covers two or more fields in a table , It is generally used when multiple fields are queried together

​ alter table Table name add index [ Index name ] ( Field 1, Field 2,…);

​ The principle of left ： If ( Field 1, Field 2, Field 3) Create a federated index , Then at the same time, it will give ( Field 1)、( Field 1, Field 2) Create index

​ The advantage is that the overhead of disk space is reduced , The disadvantage is that it takes a long time

The principle of using index ：

​ 1. Avoid creating too many indexes for frequently updated tables , Because updating the index is very time-consuming , You should create an index on frequently queried fields

​ 2. It is better not to create an index for a table with a small amount of data

​ 3. Do not create an index when there are many same values in the same field , for example “ Gender ” Field

​ 4. The more indexes, the better , A comprehensive trade-off should be made between the query speed and the time and space occupied by indexing

​

​

​

​