如何在sql中没有数据的地方显示零

Question

我正在运行一个查询，我正在做一些结果的联合。现在查询的工作原理，默认情况下有6行。现在联合会给那些行提供6个结果。如果所有联合都有数据那么没问题，我得到6行，但问题是如果任何一个联盟没有数据显示然后我没有得到6行，而是我得到4行或有时2。但我想要哪个查询不会有数据它将返回0那里。我发布我的查询

select count(distinct CANDYS.candidateId) from
(Select C.candidateID,J.joborderID from db_candidatenote C, db_jobordernote   J 
where C.noteid=J.noteid and C.action like 'Engaged' 
and C.dateadded between '2014-04-01' and '2014-04-30' ) N,
(Select distinct Note.candidateId, JNote.jobOrderID 
from db_candidatenote Note,db_jobordernote JNote, corporateuser User where 
Note.noteid=JNote.noteid
and Note.commentingPersonID = User.userid 
and User.firstName='Avishek'
and Note.action='Sourcing') CANDYS
where N.candidateID= CANDYS.candidateId
and N.joborderid=CANDYS.jobOrderID

union

 select count(distinct CANDYS.candidateId) from
(Select C.candidateID,J.joborderID from db_candidatenote C, db_jobordernote J 
 where C.noteid=J.noteid and C.action like 'Qualification' 
and C.dateadded between '2014-04-01' and '2014-04-30' ) N,
(Select distinct Note.candidateId, JNote.jobOrderID 
from db_candidatenote Note,db_jobordernote JNote, corporateuser User where 
Note.noteid=JNote.noteid
and Note.commentingPersonID = User.userid 
and User.firstName='Avishek'
and Note.action='Sourcing') CANDYS
where N.candidateID= CANDYS.candidateId
and N.joborderid=CANDYS.jobOrderID

union

select count(distinct CANDYS.candidateId) from
(Select C.candidateID,J.joborderID from db_candidatenote C, db_jobordernote J 
where C.noteid=J.noteid and C.action like 'Internal Submission' 
and C.dateadded between '2014-04-01' and '2014-04-30' ) N,
(Select distinct Note.candidateId, JNote.jobOrderID 
from db_candidatenote Note,db_jobordernote JNote, corporateuser User where 
Note.noteid=JNote.noteid
and Note.commentingPersonID = User.userid 
and User.firstName='Avishek'
and Note.action='Sourcing') CANDYS
where N.candidateID= CANDYS.candidateId
and N.joborderid=CANDYS.jobOrderID

union

select count(distinct CANDYS.candidateId) from sendout S,
(Select distinct Note.candidateId, JNote.jobOrderID 
from db_candidatenote Note,db_jobordernote JNote, corporateuser User where 
Note.noteid=JNote.noteid
and Note.commentingPersonID = User.userid 
and User.firstName='Avishek'
and Note.action='Sourcing' ) CANDYS
where S.candidateID= CANDYS.candidateId
and S.joborderid=CANDYS.jobOrderID
and S.dateadded between '2014-04-01' and '2014-04-30'

union

  select count(distinct CANDYS.candidateId) from
 (Select C.candidateID,J.joborderID from db_candidatenote C, db_jobordernote J 
  where C.noteid=J.noteid and C.action like '%Interview%' 
  and C.dateadded between '2014-04-01' and '2014-05-30' ) N,
 (Select distinct Note.candidateId, JNote.jobOrderID 
 from db_candidatenote Note,db_jobordernote JNote, corporateuser User where 
 Note.noteid=JNote.noteid
 and Note.commentingPersonID = User.userid 
 and User.firstName='Avishek'
 and Note.action='Sourcing') CANDYS
 where N.candidateID= CANDYS.candidateId
 and N.joborderid=CANDYS.jobOrderID

 union

select count(distinct CANDYS.candidateId) from
(Select C.candidateID,J.joborderID from db_candidatenote C, db_jobordernote J 
where C.noteid=J.noteid and C.action like 'Support' 
and C.dateadded between '2014-04-01' and '2014-04-30' ) N,
(Select distinct Note.candidateId, JNote.jobOrderID 
from db_candidatenote Note,db_jobordernote JNote, corporateuser User where 
Note.noteid=JNote.noteid
and Note.commentingPersonID = User.userid 
and User.firstName='Avishek'
and Note.action='Sourcing') CANDYS
where N.candidateID= CANDYS.candidateId
and N.joborderid=CANDYS.jobOrderID

union

select count(distinct CANDYS.candidateId )from placement P,(Select distinct      Note.candidateId, JNote.jobOrderID 
from db_candidatenote Note,db_jobordernote JNote, corporateuser User where 
Note.noteid=JNote.noteid
and Note.commentingPersonID = User.userid 
and User.firstName='Avishek'
and Note.action='Sourcing') CANDYS
where P.candidateID= CANDYS.candidateId
and P.joborderid=CANDYS.jobOrderID
and P.dateadded between '2014-04-01' and '2014-04-30'

这是我的查询。因此，此查询将给出每个部分的结果。现在，如果任何查询没有数据，则它不应该为空，而是在结果处显示0。如果这样做。有人请帮帮忙。

就像它显示2行的图像一样。也不确定它显示的是哪个行数据。

Answer 1

为简化问题，让我们假设这个架构：

create table t ( i int, a char(1) );
insert into t values ( 1, 'a');

在这种情况下，当where子句中没有包含该行时，我们需要一个带有zero的虚拟行作为数据。我们可以通过vars实现这一目标：

SET @f=0;
select i,a, @f:=1
from t
where i<>1                     --< row not included
union
select 0,'0', @f as f          --< fictitious row,only if f=0
from ( select 1 ) tt
having f=0                     --< f is 0

结果：

i     a
0     '0'

当包含一行时：

SET @f=0;
select i,a, @f:=1
from t
where i=1                     --< row is included
union
select 0,'0', @f as f
from ( select 1 ) tt
having f=0                     --< f is 1

结果：

i     a
1     'a'

通过此剪辑，您可以推断代码中所需的更改。用于简化问题以简化帮助。 了解How to create a Minimal, Complete, and Verifiable example