最大时间的id oracle,，求ID相同的连续日期的最大值和最小值

SQL codecreate table tabname1 (id0 INT,id1 int ,month1 DATE);

insert into tabname1 values(1,1,to_date('2009-01-01','yyyy-mm-dd'));

insert into tabname1 values(1,1,to_date('2009-02-01','yyyy-mm-dd'));

insert into tabname1 values(1,1,to_date('2009-03-01','yyyy-mm-dd'));

insert into tabname1 values(1,2,to_date('2009-04-01','yyyy-mm-dd'));

insert into tabname1 values(1,2,to_date('2009-05-01','yyyy-mm-dd'));

insert into tabname1 values(1,2,to_date('2009-06-01','yyyy-mm-dd'));

insert into tabname1 values(1,1,to_date('2009-07-01','yyyy-mm-dd'));

insert into tabname1 values(1,1,to_date('2009-08-01','yyyy-mm-dd'));

insert into tabname1 values(2,1,to_date('2009-05-01','yyyy-mm-dd'));

insert into tabname1 values(2,1,to_date('2009-06-01','yyyy-mm-dd'));

insert into tabname1 values(2,3,to_date('2009-08-01','yyyy-mm-dd'));

insert into tabname1 values(2,3,to_date('2009-09-01','yyyy-mm-dd'));

insert into tabname1 values(2,4,to_date('2009-12-01','yyyy-mm-dd'));

insert into tabname1 values(2,4,to_date('2010-01-01','yyyy-mm-dd'));

select id0,id1,min(month1),max(month1)

from

(SELECT id0,id1,month1,COUNT(1) over(PARTITION BY id0,id1,dr) dr FROM

(SELECT id0,id1,

month1,

add_months(month1,- DENSE_RANK() OVER(PARTITION BY id0,id1 ORDER BY month1)) dr

FROM tabname1

)

)

group by id0,id1,dr

order by id0,id1

id0 id1 min(month1) max(month1)

--------------------

1 1 1 2009/7/1 2009/8/1

2 1 1 2009/1/1 2009/3/1

3 1 2 2009/4/1 2009/6/1

4 2 1 2009/5/1 2009/6/1

5 2 3 2009/8/1 2009/9/1

6 2 4 2009/12/1 2010/1/1