oracle sql汇总帮助

Question

I've written a query to collect some data to display in an auto-updating box & whisker graph in excel. I'd like to use rollup to create summary rows for each type of train_line (PF and MJ) to include in the excel chart.

Can I do that using Rollup?

I've tried to get my head around Rollup, but I'm not getting far. I've tried just wrapping it around random things in my group-by but it hasn't done what I've wanted.

Here's what the first few columns of results look like.

DUMP_YEAR   DUMP_WEEK    LINE   MINE PRODUCT    CODE
2009             30        MJ      MJ   C        MJ-C
2009             30        PF      BR   F        BR-F
2009             30        PF      BR   L        BR-L
2009             30        PF      HD   F        HD-F
2009             30        PF      HD   L        HD-L
2009             30        PF      MA   F        MA-F
2009             30        PF      MA   L        MA-L
2009             30        PF      NM   F        NM-F
2009             30        PF      NM   L        NM-L
2009             30        PF      PA   F        PA-F
2009             30        PF      PA   L        PA-L
2009             30        PF      TP   F        TP-F
2009             30        PF      TP   L        TP-L
2009             30        PF      WA   F        WA-F
2009             30        PF      WA   L        WA-L
2009             30        PF      YA   F        YA-F

And here's my SQL query.

 select t.dump_year,
       t.dump_week,
       (case when t.product = 'L' or t.product = 'F' then 'PF'
             when t.product = 'C' then 'MJ'
             else null
        end) as train_line,    
       t.mine_id,
       t.product,
       t.mine_id||'-'||t.product as code,
       count(distinct t.tpps_train_id) as trains,
       count(1) as wagons,
       count(CASE WHEN w.tonnes >= 1121 THEN w.tonnes END) as overload,
       round(count(CASE WHEN w.tonnes >= 1121 THEN w.tonnes END)/count(1)*100,1) as pct_ol,
       min(t.dump_date) as first_train,
       max(t.dump_date) as last_train,     
       119 as u_limit,
       100 as target,    

       round(avg(w.tonnes),2) as average,
       round(stddev(w.tonnes),2) as deviation,
       round(min(w.tonnes),2) as minimum,
       round(max(w.tonnes),2) as maximum,
      round(percentile_disc(0.99) within group (order by (w.tonnes) desc),2) as pct_1st,
      round((percentile_disc(0.75) within group (order by (w.tonnes) desc)),2)-round((percentile_disc(0.99) within group (order by (w.tonnes) desc)),2) as whisker1,
      round(percentile_disc(0.75) within group (order by (w.tonnes) desc),2) as pct_25th,
      round((percentile_disc(0.50) within group (order by (w.tonnes) desc)),2)-round((percentile_disc(0.75) within group (order by (w.tonnes) desc)),2) as box50,
      round((percentile_disc(0.25) within group (order by (w.tonnes) desc)),2)-round(percentile_disc(0.50) within group (order by (w.tonnes) desc),2) as box75,
      round((percentile_disc(0.01) within group (order by (w.tonnes) desc)),2)-round((percentile_disc(0.25) within group (order by (w.tonnes) desc)),2) as whisker99,
      round(percentile_disc(0.50) within group (order by (w.tonnes) desc),2) as pct_50th,
      round(percentile_disc(0.25) within group (order by (w.tonnes) desc),2) as pct_75th,
      round(percentile_disc(0.01) within group (order by (w.tonnes) desc),2) as pct_99th

   from 

    (
        select trn.mine_code as mine_id,
               substr(trn.train_control_id,2,1) as port,
               trn.train_tpps_id as tpps_train_id,      
               con.weight_total-con.empty_weight_total as tonnes     
        from  widsys.train trn
                  INNER JOIN widsys.consist con
                      USING (train_record_id)

        where trn.direction = 'N'
              and (con.weight_total-con.empty_weight_total) > 10
              and trn.num_cars > 10 
       ) w,

        (
         select td.datetime_act_comp_dump as dump_date,
                to_char(td.datetime_act_comp_dump-7/24, 'IYYY') as dump_year,
                to_char(td.datetime_act_comp_dump-7/24, 'IW') as dump_week,
                td.mine_code as mine_id,
                td.train_id as tpps_train_id,
                pt.product_type_code as product
         from tpps.train_details td
              inner join tpps.ore_products op
              using (ore_product_key)
              inner join tpps.product_types pt
              using (product_type_key)
         where to_char(td.datetime_act_comp_dump-7/24, 'IYYY') = 2009
               and to_char(td.datetime_act_comp_dump-7/24, 'IW') = 30
         order by td.datetime_act_comp_dump asc
    ) t 
   where w.mine_id = t.mine_id
      and w.tpps_train_id = t.tpps_train_id

 --having t.product is not null or t.mine_id is null 
   group by 
         t.dump_year,
         t.dump_week, 
       (case when t.product = 'L' or t.product = 'F' then 'PF'when t.product = 'C' then 'MJ'else null end),       
         t.mine_id,
         t.product


order by train_line asc

Answer 1

You would use ROLLUP to generate hierarchical subtotals for your query, ie:

SQL> WITH DATA AS (
  2     SELECT 'i' || MOD(ROWNUM, 1) dim1,
  3            'j' || MOD(ROWNUM, 2) dim2,
  4            'k' || MOD(ROWNUM, 3) dim3,
  5            ROWNUM qty
  6       FROM dual
  7     CONNECT BY LEVEL <= 100
  8  )
  9  SELECT dim1, dim2, dim3, SUM(qty) tot
 10    FROM DATA
 11   GROUP BY dim1, rollup(dim2,dim3)
 12   ORDER BY 1, 2, 3;

DIM1  DIM2  DIM3         TOT
----- ----- ----- ----------
i0    j0    k0           816
i0    j0    k1           884
i0    j0    k2           850
i0    j0                2550 (*)
i0    j1    k0           867
i0    j1    k1           833
i0    j1    k2           800
i0    j1                2500 (*)
i0                      5050 (*)

The ROLLUP clause generated the lines marked (*)

If you only want to get a set of subtotals and not all the hierarchical levels, you can use the GROUPING SETS clause, ie:

SQL> WITH DATA AS (
  2     SELECT 'i' || MOD(ROWNUM, 1) dim1,
  3            'j' || MOD(ROWNUM, 2) dim2,
  4            'k' || MOD(ROWNUM, 3) dim3,
  5            ROWNUM qty
  6       FROM dual
  7     CONNECT BY LEVEL <= 100
  8  )
  9  SELECT dim1, dim2, dim3, SUM(qty) tot
 10    FROM DATA
 11   GROUP BY GROUPING SETS (
 12     (dim1, dim2, dim3), -- detail
 13     (dim1) -- total
 14   )
 15   ORDER BY 1, 2, 3;

DIM1  DIM2  DIM3         TOT
----- ----- ----- ----------
i0    j0    k0           816
i0    j0    k1           884
i0    j0    k2           850
i0    j1    k0           867
i0    j1    k1           833
i0    j1    k2           800
i0                      5050

Answer 2

有“多维数据集”，“汇总”和“分组集”： http : //download.oracle.com/docs/cd/B28359_01/server.111/b28313/aggreg.htm#i1007462