这段sql怎么写？ 要用排序，统计。

[码畜二代]

select distinct (t.mc) mc,
t.c_id c_id,
t.pm pm,
t.s_id s_id,
to_char(t.rq, 'yyyy') rq,
t.mhb mhb,
r.parent_id parent_id

from pfsc t, jgjk_cpfl r
where t.c_id in (100100, 100111, 100112)
and t.s_id = r.id
这是sql
差出来的数据

现在要统计 所有市场，一级市场，二级市场的相对应parent_id相同的个数 根据日期 和市场 MC，RQ 分组。

[一颗向往数据的心丶]

select mc,rq,parent_id,count(1) as p_num from pfsc group by mc,rq,parent_id;

[一颗向往数据的心丶]

引用 1 楼 z_shousi 的回复:

select mc,rq,parent_id,count(1) as p_num from pfsc group by mc,rq,parent_id;

 select t.mc,
       to_char(t.rq, 'yyyy') rq,
       r.parent_id,
       count(1) as p_num
  from pfsc t, jgjk_cpfl r
 where t.c_id in (100100, 100111, 100112)
   and t.s_id = r.id
 group by t.mc,to_char(t.rq, 'yyyy'),r.parent_id

[阿三]

用分析函数试试

[shenlele088]

引用 2 楼 z_shousi 的回复:

[quote=引用 1 楼 z_shousi 的回复:]

select mc,rq,parent_id,count(1) as p_num from pfsc group by mc,rq,parent_id;

 select t.mc,
       to_char(t.rq, 'yyyy') rq,
       r.parent_id,
       count(1) as p_num
  from pfsc t, jgjk_cpfl r
 where t.c_id in (100100, 100111, 100112)
   and t.s_id = r.id
 group by t.mc,to_char(t.rq, 'yyyy'),r.parent_id

[/quote] +1