有 字段A 和B
比如数据如下
A B
1 2
1 3
1 4
2 2
2 3
统计出的sql结果:
A count
1 3
2 2
select a,count(b) from t group by a;
--或count(b) over (partition by a)
如统计一级会计科目数量的sql:
select fo,count(dispname)from( select substr(subjcode,1,4) fo,bd_accsubj.dispname, bd_accsubj.pk_accsubj, bd_accsubj.pk_glorgbook from bd_accsubj where pk_glorgbook in(select pk_glorgbook from bd_glorgbook where glorgbookcode= '010502-0001' ) order by dispname) group by fo order by fo
方法二 partition 这个时候后面不需要group by 但是需要distinct fo 否则见下图
select id,count(*) over(partition by pro_id) from sal;
以pro_id分组,统计分组后每个pro_id的记录总数及对应的id;
类似还有count(*) over(order by ……)、sum(amount) over(partition by ……)等,略有区别,挺有意思,有兴趣可以小研究下
select distinct fo,count(dispname) over( partition by fo) num1 from ( select substr(subjcode,1,4) fo,bd_accsubj.dispname, bd_accsubj.pk_accsubj, bd_accsubj.pk_glorgbook from bd_accsubj where pk_glorgbook in(select pk_glorgbook from bd_glorgbook where glorgbookcode= '010502-0001' ) order by dispname) order by fo
如果group by fo ,这里不能group by,总结:partition是group by的一种形式,group 不需要重复
不加distinct fo
更新2012-11-9 19:12:增加一列显示一级科目名称 需要外面嵌套两个select
select fo, (select subjname from bd_accsubj where subjcode = fo and pk_glorgbook = '0001E1100000000000MX') aa,num1 from (select fo, count(dispname)num1 from (select substr(subjcode, 1, 4) fo, bd_accsubj.dispname, bd_accsubj.pk_accsubj, bd_accsubj.pk_glorgbook from bd_accsubj where pk_glorgbook in (select pk_glorgbook from bd_glorgbook where glorgbookcode = '010101-0001') order by dispname) group by fo) order by fo
统计结算中心制单由哪些部分组成
select pk_system, count (PK_SYSTEM) from gl_voucher where pk_corp= '1162' and dr= '0' GROUP BY pk_system |
2012-12-21 10:39:58 折腾到现在终于发现这个29462就是我结算凭证的数量
上面的gl是期初产生的,GL是手工录入,EC是单据管理付款来的(当地行支付,应收应付报账中心)
CV是协同凭证,FTS是结算中心的资金结算
结算中心从开始到现在制单有29474=29462(FTS)+6(EC)+(8-2)(GL,其中有下面两个不知道为什么废了)