Question

一周前我问过以下问题：

Compare Rows in oracle table and update matching ones

我收到了非常有用的答案，但我再次陷入困境，因为我需要修改收到的答案，并添加另一个过滤器。

除了具有不同的BuySell字段之外，还有一个不能相等的Description字段。与只能采用两个值的BuySell字段不同，描述可以是任何值。

以下是我一直在使用的@ MT0的答案，我希望修改：

MERGE INTO your_table dst 
USING (   SELECT ROW_NUMBER() OVER (
             PARTITION BY tDate, Product, Price, Quantity, BuySell
             ORDER BY ID
           ) AS idx,
         COUNT( CASE BuySell WHEN 'Buy' THEN 1 END ) OVER (
             PARTITION BY tDate, Product, Price, Quantity
           ) AS num_buy,
         COUNT( CASE BuySell WHEN 'Sell' THEN 1 END ) OVER (
             PARTITION BY tDate, Product, Price, Quantity
           ) AS num_sell   FROM   your_table

) src ON ( src.ROWID = dst.ROWID AND src.idx <= LEAST( src.num_buy, src.num_sell ) ) 
WHEN MATCHED THEN   UPDATE SET Status = 'Matched';

编辑：以下是一个例子

我有一张如下表：

**ID      tDate       Product   Description  Price    Quantity    BuySell    Status**
  1     10-May-17       pppp       p1        $12        20         Buy       Null
  2     12-May-17       tttt       t1        $10        20         Sell      Null
  3     12-May-17       tttt       t2        $10        20         Buy       Null
  4     18-May-17       pppp       p2        $14        20         Sell      Null
  5     18-May-17       pppp       p3        $14        20         Buy       Null
  6     19-May-17       xxxx       x4        $11        10         Sell      Null
  7     19-May-17       xxxx       x4        $11        10         Buy       Null

我需要更新名为STATUS的字段，并将其设置为“匹配”，只要找到一个具有相同tDate，产品，价格和数量且不等于BuySell AND Description的对。

以下是期望的结果：

**ID      tDate       Product   Description  Price    Quantity    BuySell    Status**
  1     10-May-17       pppp      p1         $12        20         Buy       Null
  2     12-May-17       tttt      t1         $10        20         Sell      Matched
  3     12-May-17       tttt      t2         $10        20         Buy       Matched
  4     18-May-17       pppp      p2         $14        20         Sell      Matched
  5     18-May-17       pppp      p3         $14        20         Buy       Matched
  6     19-May-17       xxxx      x4         $11        10         Sell      Null
  7     19-May-17       xxxx      x4         $11        10         Buy       Null

请注意，6和7不匹配，因为它们具有相同的描述。

Answer 1

update your_table
   set status='Matched'
 where id in(
    with IDS(id1,id2,grp,num,rnum) as(
     select a.id,b.id,
            rank() over(order by a.tdate,a.price,a.product,a.quantity),
            case when a.id=lag(a.id) over(partition by a.tdate,a.price,a.product,a.quantity order by a.id)
            then 0
            else dense_rank() over(partition by a.tdate,a.price,a.product,a.quantity order by a.id) end,
            row_number() over(partition by a.tdate,a.price,a.product,a.quantity order by a.id)
       from your_table a, your_table b
      where a.tdate=b.tdate and a.price=b.price and a.product=b.product 
        and a.quantity=b.quantity
        and a.buysell='Sell' and b.buysell='Buy' and a.description!=b.description
    ),
    Q(id1,id2,grp,num,used) as(
      select id1,id2,grp,num,','||id2||','
        from IDS where rnum=1
      union all
      select I.id1,R.column_value,Q.grp,Q.num+1,Q.used||R.column_value||','
        from Q, IDS I,
             table(cast(multiset(
                select min(id2) id from IDS N
                 where N.id1=I.id1
                   and Q.used not like '%,'||N.id2||',%'
             ) as sys.ODCINumberList)) R
       where I.grp=Q.grp and I.num=Q.num+1
    )
 select decode(X,1,id1,id2)
   from Q,
        (select 1 X from DUAL union all select 2 from DUAL)
  where id2 is not null
)

后续问题：比较Oracle表中的行并更新匹配行

1 个答案: