在SQL查询中同时使用count和union

Question

I have two SQL queries

select count ( distinct w.processno ) AS "Number of Processes", y.userid from upmfolder f, upmperson p, w2process w, w2processcheck y where f.personid = p.personid and f.folderid = w.keynumb and w.Processno = y.processid and w.keyobject ='UPMFolder' group by 2

and another similar one is :

select count ( distinct w.processno ) AS "Number of Processes", y.userid from upmfolder f, upmperson p, w2process w, w2processcheck y where f.personid = p.personid and f.payrollmemberid = w.keynumb and w.keyobject = 'UPMPayrollmember' and w.Processno = y.processid group by 2

How would I combine the two into just one query? So the count and group would still function correctly. I tried with union all , however this didn't work. What do I need to do?

Answer 1

You can do this with conditional aggregation. For a count(distinct) you need to supply a NULL value:

select y.userid,
       count(distinct case when w.keyobject ='UPMFolder' then w.processno end) as NumFolder, 
       count(distinct case when w.keyobject ='UPMPayrollMember' then w.processno end) as NumPayrollMember
from upmfolder f, upmperson p, w2process w, w2processcheck y
where f.personid = p.personid and
      f.folderid = w.keynumb and
      w.Processno = y.processid
group by y.userid

Note: You should also learn modern join syntax. I haven't changed the query, but the join conditions should really be in an on clause rather than in the where clause.