Usage of case when then else end statement

1. SELECT a.managecom,    
2.        a.subtype,    
3.        count(*) loadsucc,    
4.        sum(case when a.state in  ('4', '5', '6', '7', '8', '9') then 1 else 0 end) recogsucc,    
5.        sum(case when a.state in  ('3', '12', '13') then 1 else 0 end) recogfail,    
6.        sum(case when a.state in  ('1', '2') then 1 else 0 end) waitrecog    
7.   FROM ocr_docdetail a, ocr_loaddetail c    
8.  WHERE 1 = 1    
9.    and a.managecom like '86%'    
10.    and a.managecom = c.managecom    
11.    and a.bussno = c.bussno    
12.    and a.subtype = c.subtype    
13.    and c.loadstate = 0    
14.    and c.scandate >= date '2012-07-29'    
15.    and c.scandate <= date '2013-01-07'    
16.  group by a.managecom, a.subtype    
17.  order by a.managecom, a.subtype;  

case There are two formats . Simple case Functions and case The search function .

-- Simple case function

            case sex

            when '1' then ' male '

            when '2' then ' Woman '

            else ' other ' end

--case The search function

            case when sex = '1' then ' male '

            when sex = '2' then ' Woman '

            else ' other ' end

These two ways , Can achieve the same function . Simple case The writing of function is relatively simple , But and case Search function , There will be some functional limitations , For example, write a decision form . 
There is another issue that needs attention ,case Function returns only the first value that matches the condition , The rest case Parts will be automatically ignored .

-- for instance , The following paragraph sql, You'll never get “ The second category ” This result

            case when col_1 in ( 'a', 'b') then' The first category '

            when col_1 in ('a')       then ' The second category '

            else ' other ' end

So let's see , Use case Functions can do something . 

One , The known data is grouped in another way , analysis . 

Here's the data :( To see better , I didn't use the country code , It's the name of the country primary key)

Country （country）	Population （population）
China	600
The United States	100
Canada	100
The British	200
The French	300
Japan	250
Germany	200
Mexico	50
India	250

According to the population data of this country , Count the population of Asia and North America . We should get the following result .

continent	Population
Asia	1100
In North America	250
other	700

I want to solve this problem , What would you do ？ Generate a continent with code Of view, It's a solution , But it's hard to change the way of statistics dynamically . 
If used case function ,sql The code is as follows :

select sum(population),

            case country

            when ' China '     then' Asia '

            when ' India '     then' Asia '

            when ' Japan '     then' Asia '

            when ' The United States '     then' In North America '

            when ' Canada '  then' In North America '

            when ' Mexico '  then' In North America '

            else ' other ' end

            from   table_a

            group by case country

            when ' China '     then' Asia '

            when ' India '     then' Asia '

            when ' Japan '     then' Asia '

            when ' The United States '     then' In North America '

            when ' Canada '  then' In North America '

            when ' Mexico '  then' In North America '

            else ' other ' end;

notes ：

In the example above , Actually select There are two fields ：sum(population), case country;

country There are many values in the field , Yes “ China , The United States , Japan , Canada ...” wait ,select After case when then else end In fact, it is equivalent to country There are three types of values for ：" Asia , In North America , other ";

There is no from hinder case when sentence , The result set is like this ,sum What you get is the population population Total of , So we need to population grouping , There's the back group by：

hinder case when sentence , In fact, the fields queried earlier case country Grouping , Divide into three groups ：" Asia , In North America , other ", Then you can get the result

alike , We can also use this method to determine the salary grade , And count the number of people at each level .sql The code is as follows ;

select

            case when salary <= 500 then '1'

            when salary > 500 and salary <= 600  then '2'

            when salary > 600 and salary <= 800  then '3'

            when salary > 800 and salary <= 1000 then '4'

            else null end salary_class,

            count(*)

            from   table_a

            group by

            case when salary <= 500 then '1'

            when salary > 500 and salary <= 600  then '2'

            when salary > 600 and salary <= 800  then '3'

            when salary > 800 and salary <= 1000 then '4'

            else null end;

Two , Use one sql Statement to complete the grouping of different conditions . 

Here's the data

Country （country）	Gender （sex）	Population （population）
China	1	340
China	2	260
The United States	1	45
The United States	2	55
Canada	1	51
Canada	2	49
The British	1	40
The British	2	60

Grouped by country and gender , The results are as follows

Country	male	Woman
China	340	260
The United States	45	55
Canada	51	49
The British	40	60

In general , use union You can also query with a single statement . But that increases consumption ( Two select part ), and sql The statements will be longer . 
Here's one that uses case Function to complete this function

select country,

            sum( case when sex = '1' then

            population else 0 end),  -- Male population 

            sum( case when sex = '2' then

            population else 0 end)   -- Female population 

            from table_a

            group by country;

So we use select, Complete the output form of the two-dimensional table , It fully shows that case The power of functions . 

3、 ... and , stay check Use in case function . 

stay check Use in case Functions are very good solutions in many cases . Maybe a lot of people don't have to check, Well, I suggest you try the following example as well sql Use in check.
Let's take an example  
company a, This company has a policy , The salary of female staff must be higher than 1000 block . If used check and case To show , As shown below

constraint check_salary check

            ( case when sex = '2'

            then case when salary > 1000

            then 1 else 0 end

            else 1 end = 1 )

If only used check, As shown below

constraint check_salary check

            ( sex = '2' and salary > 1000 )

The conditions of female staff are met , Male staff can't input .