One relationship of complementary

Single-table queries 
a syntax sequentially 
select distinct query field 1, 2 ,. query field . . from table 
    before the packet filter condition where 
    the filter conditions after having packet 
    group by group by 
    order by sort field 
    number of limit display; 

Second, the order of execution 

    select distinct query field 1, 2 ,. query field . . from table name 
    is first performed from determines to perform is the file 
    and then executed before where determining a packet filter condition 
    and then perform group by group 
    and then performing a select screened 
    before performing order by row number sequence 
    and then execute limit restriction results show Article number 

three, according to the priority level of SQL statements to write 
a, first determine which tables db39.emp from 
B, whether a filter condition like name WHERE '% I%' 
. . . 
z, discharge function select 

four, where the filter 
where the words may be used: 
1. The comparison operators:> <!> = <= <> is not equal to a = #! = Not <> 
   SELECT ID, name from db39.emp WHERE ID> = ID. 3 and <=. 6

BETWEEN 80 and 100 2. 
   SELECT * WHERE ID BETWEEN from db39.emp. 3 and. 6; #> =. 3 and <=. 6 

3. in (80,90,100) is 80 or 90 or 100 
   SELECT * from the salary in EMP WHERE ( 20000,18000,17000); SELECT * # EMP WHERE from the salary or the salary = 20000 or 18000 = 17000 = the salary; 

4. like 'Egon%',% or pattern may be _,% represents any number of characters, a character represented _ 
   select name, salary from db39.emp where name like '% i%' # claim: employee name in the employee name query comprising i letters their salary 
   select name, salary from db39.emp where name like '____'; # requirements: query employee name is a four-character name of its employee payroll 
   select name, salary from db39.emp where char_length (name) = 4; # consistent results over a 

5 logical operators: can be used directly in a plurality of logic conditions Not or operator and 
   SELECT * WHERE from db39.emp Not BETWEEN. 3 and ID. 6; 
   SELECT * from EMP WHERE Not in the salary (20000,18000,17000);

   Requirements: query job description empty employee name and job name 
   select name, post from db39.emp where post_comment is NULL; # for NULL must be is, can not be used = 
   the SELECT name, POST from db39.emp the WHERE IS post_comment not NULL; 
   #NULL refers not take any storage, the string is hollow mysql occupy storage space, i.e. not empty (NULL) 

five, group by group 
    , if not set to only_full_group_by mode, the sub-group was complete default withdrawn * is the first personal data within the group. But an element within minutes after the finish to take a separate group group does not make sense, therefore, of the packet, the general pattern will be processed as follows 
    # sql_mode set to only_full_group_by, means that the future whenever a packet, packet basis can only take into 
    mysql> Global the sql_mode = sET "STRICT_TRANS_TABLES, only_full_group_by"; 

    # polymerizable function group function (typically in conjunction with a packet) 
    SELECT POST, max (the salary) from group EMP by POST; # could not take the name element in the group, age .., can only take group name (packet based) or polymerization function 

    SELECT POST, min (the salary) from group EMP by POST; 

    SELECT POST, AVG (the salary) from group EMP by POST;

    POST SELECT, SUM (the salary) from Group EMP by POST; 

    SELECT POST, COUNT (ID) Group from EMP by POST; 

    #group_concat (packet after use): extraction with the desired information; string concatenation operator 
    select post, group_concat (name) from EMP Group by POST; 
    SELECT POST, GROUP_CONCAT (name, "_ SB") from EMP Group by POST; 
    SELECT POST, GROUP_CONCAT (name, ":", the salary) from EMP Group by POST; 
    SELECT POST, GROUP_CONCAT ( the salary) from Group EMP by POST; 

    # supplemented concat (when not using packets): string concatenation operator 
    select concat ( "nAME:", name) as the name, concat ( "SAL:", salary) as EMP pay from; 

    # added as syntax: taken as a field or a table alias 
    select name as the name, salary as salary from emp; # as may be omitted 
    mysql> select emp.id, emp.name from emp as t1; # error
    mysql> select t1.id, t1.name from emp as t1; # with MySQL> the SELECT the above mentioned id, name from emp AS T1; 

    # query four operations 
    the SELECT name, salary * 12 AS annual_salary from emp; 

    # group exercises 
    select post, group_concat (name) from emp group by post ; # query job name and the names of all staff positions included in 

    select post, count (id) from emp group by post; # query job name and the number of employees in each job included 

    select sex, count (id) from emp group by sex ; # query within the company male employees and female employees the number of 

    select post, avg (salary) from emp group by post; # query job name, and the average salary of each position 

    select sex, avg (salary ) from emp group by sex; # inquiry average salary of male employees and male employees, the average salary of female employees and female employees 

    select post, avg (salary) from emp where age> = 30 group by post; # statistical departments age employees aged 30 years and average wages 

six, having filter (must use the group name (group by) or aggregate functions) 
    having the syntax and where exactly the same, but after having carried out the packet filter further
    That is where can not use aggregate functions, and having that can aggregate function, which is the two of them the biggest difference 

    department # statistics departments older workers the average wage over the age of 30, and retains the average wage of more than 10,000 of 
    select post, avg (salary ) from emp the WHERE Age> = 30 Group by POST the HAVING AVG (salary)> 10000; 

    # stressed: having to be used later group by (do not recognize the default packet) 
    the SELECT * from emp the HAVING AVG (salary)> 10000; # error 

seven , distinct deduplication (performed after having, and post, name, etc. belonging to the same execution level) 
   SELECT DISTINCT POST, AVG (the salary) from EMP WHERE Age> = 30 Group by POST HAVING AVG (the salary)> 10000; 

eight, order by sorting (default ascending) 
   SELECT * from EMP Order by the salary ASC; # default ascending row 
   select * from emp order by salary desc ; # descending row 
   select * from emp order by age desc ; # descending row 
   select * from emp order by age desc , salary asc; # age descending according to the first row, and then according to pay in ascending

   # Statistics of the department over the age of 10 year-old employee average wage, and keep the average wage of more than 1000 of the department, then on average wages sort 
   select post, avg (salary) from emp where age> 10 group by post having avg (salary) > 1000 order by avg (salary) ; 

nine, limit restrictions show the number of; paging 
   the SELECT * from emp limit 3; 
   the SELECT * from emp the Order by salary desc limit 1; # display the highest paid person's information 
   select * from emp limit 0, 5; # tab, starting from 0, take 5 (1-5) 
   SELECT * from limit EMP 5, 5; # tab 5 from the beginning, taking 5 (6-10) 
   tab equation: 
       If each of the information article referred to as a count number 
       then the n-th information also is the start of the (n-1) * count information pieces in total, taking the future count article 
       is 
           select * from emp limit (n- 1) * count, n ; 

X. regular expression 
   select * from emp where name regexp ' ^ jin * (n | g) $.'; # adjusting regular; regular expressions common 
   
multi-table join query 
a Cartesian product 
   from emp, dep, dep2, ...
   
Second, the connector: the two tables of correspondence between a virtual table records connected 
   SELECT * from the Join EMP Inner DEP = ON emp.dep_id dep.id; 

# Applications: 
    SELECT * from EMP, DEP DEP = WHERE emp.dep_id .id and dep.name = "technology"; # is not recommended; do not even live to do with where the table 
    select * from emp inner join dep on emp.dep_id = dep.id where dep.name = " technology"; # logic and on a consistent 

three left connection: connection basis, including on the left there is no corresponding record retention relationship 
   select * from emp left join dep on emp.dep_id = dep.id; 

Fourth, the right connection: including a connection basis, do not correspond to retain the right record 
   select * from emp right join dep on emp.dep_id = dep.id; 

five full connection: connection basis, including on the reservation left and right do not correspond to the record 
   select * from emp left join ON emp.dep_id = dep.id DEP 
   of Union # to re- 
   select * from emp right join dep on emp.dep_id = dep.id; 

six multi-table joins can be a single table constantly connected to the virtual table 
   # Find all sectors the highest salary
   T1 SELECT. * T1 from EMP AS 
   Inner the Join 
   (SELECT POST, max (the salary) from EMP Group AS MS by POST) # AS T2 commission virtual table T2 
   ON = t2.post t1.post 
   WHERE t1.salary = t2.ms 
   ; 

   SELECT * from EMP AS T1 T1. 
   Inner the Join 
   (SELECT POST, max (the salary) from EMP Group AS MS by POST) AS T2 
   ON t1.salary = t2.ms 
   ; 
subquery (a problem solving a problem) 
   to a query statement parentheses, as an additional query terms to use, called a subquery 

   select name from emp where dep_id = ( select id from dep where name = " technology"); # subquery 
   select emp.name from emp inner join dep on emp.dep_id = dep.id where dep.name = " technology"; # list 

   # query an average age of over 25-year-old department name of
   select name from dep where id in ( select dep_id from emp group by dep_id having avg (age)> 25); # subquery 
   select dep.name from emp inner join dep on emp.dep_id = dep.id group by dep.name having avg (age)> 25; # list 

   # 2 people view the lack of a department name (subquery get is someone's department the above mentioned id) 
   the SELECT * from emp the WHERE EXISTS (DEP the WHERE from the SELECT the above mentioned id the above mentioned id> 3); #exists usage, when () return True, the outer query will query; when the return value is False, the outer query is not a query (empty the SET) 

   # query each department newest recruits employees who 
   select t1.id, t1. name, t1.post, t1.hire_date, t2.post, t2.max_date from emp as t1 inner join (select post, max (hire_date) as max_date from emp group by post) as t2 on t1.post = t2.post where t1.hire_date = t2.max_date;