比如：表pat_visit中，要算出7月份每天在院人数的累加数据，即要算出7月1号的数据：
select count(*)
from pat_visit p
where p.admission_date_time <= to_date('2014-07-01', 'yyyy-MM-dd')
and p.discharge_date_time >= to_date('2014-07-01', 'yyyy-MM-dd')；
再算出7月2号的数据，直到31号结束，并把1号到31号的数据累加起来；
sql语句怎么写啊？
------解决方案--------------------
select sum(
least(to_date('2014-07-31', 'yyyy-MM-dd'),trunc(discharge_date_time))
- greatest(to_date('2014-07-01', 'yyyy-MM-dd'),trunc(admission_date_time))
+1
)
from pat_visit p
where trunc(p.admission_date_time) <= to_date('2014-07-31', 'yyyy-MM-dd')
and trunc(p.discharge_date_time) >= to_date('2014-07-01', 'yyyy-MM-dd')；
------解决方案--------------------
select count(*), to_char(p.admission_date_time, 'yyyy-mm')
from pat_visit p
group by to_char(p.admission_date_time, 'yyyy-mm')
--不可以酱紫？
------解决方案--------------------
题目意思，我觉得是 每天的量加一个汇总。 记得好像 roll up over 可以这样了
------解决方案--------------------
是Oracle数据库吧，试试下面的：
select to_date('2014-07-01', 'yyyy-MM-dd') as rq ,count(*)
from pat_visit p
group by to_date('2014-07-01', 'yyyy-MM-dd')
with rollup
------解决方案--------------------
计算每天多少人
select B.RQ,COUNT(1)
from pat_visit A,(select to_date('20140701','yyyymmdd')+rownum-1 RQ from dual connect by rownum<=31)B
WHERE TRUNC(A.admission_date_time)<=B.RQ AND TRUNC(A.discharge_date_time)>=B.RQ
GROUP BY B.RQ;
计算月度总和
select sum(COUNT(1))
from pat_visit A,(select to_date('20140701','yyyymmdd')+rownum-1 RQ from dual connect by rownum<=31)B
WHERE TRUNC(A.admission_date_time)<=B.RQ AND TRUNC(A.discharge_date_time)>=B.RQ
GROUP BY B.RQ;
------解决方案--------------------
用rollup的写法
select B.RQ,COUNT(1)
from pat_visit A,(select to_date('20140701','yyyymmdd')+rownum-1 RQ from dual connect by rownum<=31)B
WHERE TRUNC(A.admission_date_time)<=B.RQ AND TRUNC(A.discharge_date_time)>=B.RQ
GROUP BY ROLLUP(B.RQ);