SQL查询以跨多行显示“总列”

Question

假设我有下表

claim_id | person_id | age_category | amount
--------------------------------------------
       1 |         1 |        adult |  10.00
       1 |         2 |        adult |  10.00
       1 |         3 |     juvenile |   8.00
       1 |         4 |        child |   5.00
       2 |         5 |        adult |  15.00
       3 |         6 |        adult |  12.00
       3 |         7 |        child |   6.00
     ...
     100 |       250 |        child |   7.00

因此有几个人可能属于同一个主张。 我已经设法实现的是这样一个结果表：

category | total people |     amount
------------------------------------
adult    |          150 | 300'000.00
juvenile |           20 |  40'000.00
child    |           80 | 160'000.00

使用以下查询：

select 
    age_category as "category"
    count(*) as "total people",
    sum(amount) as "amount"
from
    my_table
group by
    age_category

有没有办法计算索赔数量并将其显示在同一个结果表中？例如。类似的东西：

category | total claims | total people |     amount
---------------------------------------|-----------
adult    |          100 |          150 | 300'000.00
juvenile |              |           20 |  40'000.00
child    |              |           80 | 160'000.00

感谢任何提示！

P.S。：我正在使用DB2

Answer 1

试试这个：

select 
    age_category as "category",
    COUNT(distinct claim_id ) as "total_claims", --  <-- add this line 
    count(*) as "total people",
    sum(amount) as "amount"

from
    my_table
group by
    age_category

编辑：

根据您的评论，请使用此查询

select 
    age_category as "category",
    case when age_category='adult' then COUNT(distinct claim_id ) end 
                                                         as "total_claims" , 
    count(*) as "total people",
    sum(amount) as "amount"

from
    my_table
group by
    age_category  

SQL Fiddle Demo

Answer 2

select 
    age_category as "category",
    sum(case when amount>0 then 1 else 0 end ) as "total claims",
    count(*) as "total people",
    sum(amount) as "amount"
from
    my_table
group by
    age_category

Answer 3

尝试使用count（DISTINCT）：

select 
    age_category as "category"
    count(*) as "total people",
    sum(amount) as "amount",
    count(distinct claim_id) as "total_claims"
from
    my_table
group by
    age_category

或者也尝试使用它：

select t.*,
(select count(*) from

(select distinct claim_id from my_table 
   where my_table.age_category=t.category) as d
 ) as "total_claims"
from
(
select 
    age_category as "category",
    count(*) as "total people",
    sum(amount) as "amount"
from
    my_table
group by
    age_category
  ) as t