sql同类型数据没合并问题

如下图的查出来的结果是有问题的。 

select distinct(n.dict) name,nvl(zl.total,0) value from( select m.*,dict.* from sa.SJJK_TAG m
left join sa.ST_DICT dict on dict.dictid=m.TAGID where dict.kind ='SJJK_TAG' and dict.parentid ='01') n
left join sa.SJJK_SJZL zl on zl.sjzlno =n.sjzlno

select name,sum(value) value from ( 
select dict.dictid,dict.dict name,value from 
(select tagid,nvl(sum(total),0) value from sa.sjjk_sjzl a,sa.sjjk_tag b 
where a.sjzlno =b.sjzlno and b.tagid like '01%' group by tagid) m
left outer join sa.st_dict dict on dict.dictid=m.tagid and dict.kind ='SJJK_TAG'
union all
(
select dictid,dict name ,0 value from sa.st_dict dict where dict.kind ='SJJK_TAG' and dict.parentid ='01'
)
)

group by dictid,name
order by dictid

上面的sql为正解。把重复的数据先分组去重，再和别的表连接。

以防止完全没数据的情况，用

union all
(
select dictid,dict name ,0 value from sa.st_dict dict where dict.kind ='SJJK_TAG' and dict.parentid ='01'
)
给没值的字段赋0