Question

我的表格报告包含：

rid     reparti
R00001  Ortopedi
R00002  Kardiologji
R00003  Gjinekologji

表医生包含：

did      demri    dmbiemri  titullli      rruga        qyteti    kp     shteti  mob         fix       email                 rid     paga
D00001   Shaban   Merovci   Dr Familjar   Mic Sokoli   Vushtrri  42000  Kosove  37744842203 28570369  shmerovci@gmail.com   R00001  1000

表约会：

tid     pid     did     rid     data_koha  konfirmimi
T00001  P00001  D00010  R00002  2015-12-20 08:30:00 1
T00002  P00009  D00004  R00003  2015-12-25 10:30:00 1

表pospilization（没有预约，你可以住院）：

  hpid    pid     did     data_shtrimi  sid      terapia    diagnoza         tid
HP0001  P00009  D00004  2015-12-13    S00001   Barera     Hemorislapaskus  T00002
HP0002  P00006  D00002  2015-12-22    S00010   Kimoterapi Fragaria vesca   T00008
HP0003  P00001  D00002  2015-12-20    S00008   Barera     Depresion        T00001

表格账单包含：

 fid     pid     hpid    cmimi_patvsh  tvsh  cmimi_metvsh data_fatures
    F00001  P00002  HP0005  85.00         0.18  100.30       2016-02-13
    F00002  P00009  HP0001  700.00        0.18  826.00       2016-12-10
    F00003  P00001  HP0003  120.00        0.18  141.60       2015-12-29

我的目的是找到每个报告的医生数量（确实）作为NrDoktoreve工作，avg（paga）作为PagaMesatare的那个部分，住院数量为NrHospitalizimeve今年实现的，账单总和（值cmimi_patvsh，价值观cmimi_metvsh）今年实现了。所有这些都应该在一个数据集列表中。

当然，这些表包含了更多数据，我在这里放了一些例子。

Answer 1

查询不是太复杂，方法是识别连接列（表格放在一起），写入连接，按RID对表进行分组（因为你想要每个RID的所有结果），最后选择字段。我没有可用于测试的MySQL，但查询应如下所示：

select 
    rid,
    count(distinct did) NrDoktoreve, 
    avg(cmimi_metvsh) as PagaMesatare,
    count(distinct hpid) as NrHospitalizimeve,
    sum(cmimi_patvsh),
    sum(cmimi_metvsh)
        from reparts r 
        left join appointments a on r.rid = a.rid 
        left join hospitalization h on a.tod = h.tid 
        left join doctors d on r.rid = d.rid
        left join bills b on a.pid = b.pid
    group by rid
    order by rid;

如果您对编写SQL感到不自在，可以考虑像DbSchema（http://www.dbschema.com/）这样的图形工具（只有很多可用工具中的一个例子，但它适用于MySQL）。

Answer 2

这是我的问题的解决方案

select t1.rid,t1.reparti,t1.NrDoktoreve,t1.PagaMesatare,t2.NrHospitalizimeve,t3.VleraPaTvsh,t3.VleraMeTvsh from(
(select r.rid,r.reparti,count(d.did) as NrDoktoreve,avg(d.paga) as PagaMesatare from repartet r
left join doktoret d on r.rid=d.rid
group by r.rid order by r.rid ) as t1 left join 
(
select count(h.hpid) as NrHospitalizimeve ,r.rid
 from faturat f
inner join hospitalizimet h on f.hpid=h.hpid
inner join terminet t on h.tid=t.tid
inner join repartet r on t.rid=r.rid
where h.data_shtrimit between (curdate()-interval 1 year) and curdate() 
group by r.rid order by r.rid) as t2 on t1.rid=t2.rid left join
(select sum(f.cmimi_patvsh) as VleraPaTvsh,sum(f.cmimi_metvsh) as VleraMeTvsh,r.rid
from repartet r
left join terminet t on r.rid=t.rid
left join hospitalizimet h on t.tid=h.tid
left join faturat f on h.hpid=f.hpid
where data_fatures between (curdate()-interval 1 year) and curdate()
group by r.rid order by r.rid)as t3 on t2.rid=t3.rid)

如何写这样复杂的查询？

2 个答案: