计算SQL Server中的每个不同值

Question

I have a table tbl_demo like this:

              aliasname is_correct
                    10  INCORRECT
                    10  INCORRECT
                    10  INCORRECT
                    10  CORRECT
                    10  CORRECT
                    10  NOT SOLVED
                    9   INCORRECT
                    9   INCORRECT
                    9   CORRECT

I created a query:

select 
    aliasname, 
    count(*) as quest_count,  
    is_correct 
from 
    tbl_demo 
group by 
    aliasname, is_correct order by aliasname

and I get output like this

              aliasname quest_count  is_correct
              ---------------------------------
              10          2            CORRECT
              10          3            INCORRECT
              10          1            NOT SOLVED
               9          1            CORRECT
               9          2            INCORRECT

but I want 0 for in 9 and NOT solved

Like one more row like this

                9         0            NOT SOLVED

Can this be achieved easily, and how?

Answer 1

You can do this by generating all possible combinations using cross join , and the using left join to bring in the original data. The rest is just aggregation:

select a.aliasname, count(d.aliasname) as quest_count, ic.is_correct
from (select distinct aliasname from tbl_demo
     ) a cross join
     (select distinct is_correct from tbl_demo
     ) ic left outer join
     tbl_demo d
     on d.aliasname = a.aliasname and d.is_correct = ic.is_correct
group by a.aliasname, ic.is_correct;