加入不起作用的条件

Posted

tags:

篇首语:本文由小常识网(cha138.com)小编为大家整理,主要介绍了加入不起作用的条件相关的知识,希望对你有一定的参考价值。

我有条件在哪里条件我试图根据某些条件加入不同字段的表。但它不允许。是否有另一种方法来获得结果。你们任何人都可以帮我解决这个问题。

             select
                      s1.Seq1,
                      s1.Seq2,
                      s1.Transf_Date,
                      s1.Place_Name,
                      s1.Place_Code,
                      '0' as Completion_Flag,
                      s1.Flag   
                      from
                      (
                           select 
                                case when T.STOCK_TRANSF_SEQ > 0 then  T.STOCK_TRANSF_SEQ else T.PO_SEQ end as Seq1,
                                case when T.STOCK_TRANSF_SEQ > 0 then  T.STOCK_TRANSF_SEQ2 else T.PO_SEQ2 end as Seq2,  
                                case when T.STOCK_TRANSF_SEQ > 0 then  '0' else '1' end as Flag, 
                                T.STOCK_TRANSF_DATE as Transf_Date,
                                min(T.WHIN_PLACE_CD) as Place_Code,
                                min(C.CUSTOMER_NAME1) as Place_Name
                                from
                                T_TRANSFER_ORDER t  
                                join M_CUSTOMER c ON t.WHIN_PLACE_CD = c.CUSTOMER_CODE 
                                    and C.PLACE_F = '1' 
                                    and C.DELETE_FLAG = 0
                                    and C.OWN_WH = '1'
                                    and C.WAREHOUSE_CODE = '10'
                                join M_CUSTOMER C2 on T.WHOUT_PLACE_CD = C2.CUSTOMER_CODE
                                where (T.STOCK_TRANSF_SEQ > 0 or T.PO_SEQ > 0)
                                and T.WHOUT_PLACE_CD = '002001000'
                                group by T.STOCK_TRANSF_SEQ, T.STOCK_TRANSF_SEQ2, t.PO_SEQ,T.PO_SEQ2, t.STOCK_TRANSF_DATE
                                having min(t.RG_FIN_F) = 0
                            ) s1
                            join 
                            (
                            select res.Seq1,
                                res.Seq2,
                                case when res.CNT= res.CNT2 then 1 else 0 end  AS Completion_Flag
                            from (
                                    select
                                        case when T.STOCK_TRANSF_SEQ > 0 then  T.STOCK_TRANSF_SEQ else T.PO_SEQ end as Seq1,
                                        case when T.STOCK_TRANSF_SEQ > 0 then  T.STOCK_TRANSF_SEQ2 else T.PO_SEQ2 end as Seq2, 
                                        count(*)  CNT,
                                        (select count(DISTINCT ITEM_CD) from T_TRANSFER_RESULT  
                                         where case when T.STOCK_TRANSF_SEQ > 0 then (STOCK_TRANSF_SEQ = T.STOCK_TRANSF_SEQ 
                                                AND STOCK_TRANSF_SEQ2 = T.STOCK_TRANSF_SEQ2) ELSE( PO_SEQ = T.PO_SEQ AND PO_SEQ2 = T.PO_SEQ2) end
                                         and  OUTPUT_FLAG = '1' AND DELETE_FLAG = 0) as CNT2
                                    from
                                        T_TRANSFER_ORDER t
                                     group BY t.STOCK_TRANSF_SEQ, t.STOCK_TRANSF_SEQ2, t.PO_SEQ, t.PO_SEQ2
                                ) res
                            )
                            s2 on s1.Seq1 = s2.Seq1 and s1.Seq2 = s2.Seq2 and s2.Completion_Flag = 0
                            order BY s1.Transf_Date
答案

我改变了它的工作原理结构

select
                          s1.Seq1,
                          s1.Seq2,
                          s1.Transf_Date,
                          s1.Place_Name,
                          s1.Place_Code,
                          '0' as Completion_Flag,
                          s1.Flag   
                          from
                          (
                               select 
                                    case when T.STOCK_TRANSF_SEQ > 0 then  T.STOCK_TRANSF_SEQ else T.PO_SEQ end as Seq1,
                                    case when T.STOCK_TRANSF_SEQ > 0 then  T.STOCK_TRANSF_SEQ2 else T.PO_SEQ2 end as Seq2,  
                                    case when T.STOCK_TRANSF_SEQ > 0 then  '0' else '1' end as Flag, 
                                    T.STOCK_TRANSF_DATE as Transf_Date,
                                    min(T.WHIN_PLACE_CD) as Place_Code,
                                    min(C.CUSTOMER_NAME1) as Place_Name
                                    from
                                    T_TRANSFER_ORDER t  
                                    join M_CUSTOMER c ON t.WHIN_PLACE_CD = c.CUSTOMER_CODE 
                                        and C.PLACE_F = '1' 
                                        and C.DELETE_FLAG = 0
                                        and C.OWN_WH = '1'
                                        and C.WAREHOUSE_CODE = '10'
                                    join M_CUSTOMER C2 on T.WHOUT_PLACE_CD = C2.CUSTOMER_CODE
                                    where (T.STOCK_TRANSF_SEQ > 0 or T.PO_SEQ > 0)
                                    and T.WHOUT_PLACE_CD = '002001000'
                                    group by T.STOCK_TRANSF_SEQ, T.STOCK_TRANSF_SEQ2, t.PO_SEQ,T.PO_SEQ2, t.STOCK_TRANSF_DATE
                                    having min(t.RG_FIN_F) = 0
                                ) s1
                                join 
                                (
                                select res.Seq1,
                                    res.Seq2,
                                    case when res.CNT= res.CNT2 then 1 else 0 end  AS Completion_Flag
                                from (
                                        select
                                            case when T.STOCK_TRANSF_SEQ > 0 then  T.STOCK_TRANSF_SEQ else T.PO_SEQ end as Seq1,
                                            case when T.STOCK_TRANSF_SEQ > 0 then  T.STOCK_TRANSF_SEQ2 else T.PO_SEQ2 end as Seq2, 
                                            count(*)  CNT,
                                            (select count(DISTINCT ITEM_CD) from T_TRANSFER_RESULT  
                                             where ((T.STOCK_TRANSF_SEQ > 0 AND STOCK_TRANSF_SEQ = T.STOCK_TRANSF_SEQ 
                                                    AND STOCK_TRANSF_SEQ2 = T.STOCK_TRANSF_SEQ2 )
                                                    OR (T.STOCK_TRANSF_SEQ = 0 AND  PO_SEQ = T.PO_SEQ AND PO_SEQ2 = T.PO_SEQ2))
                                             and  OUTPUT_FLAG = '1' AND DELETE_FLAG = 0) as CNT2
                                        from
                                            T_TRANSFER_ORDER t
                                         group BY t.STOCK_TRANSF_SEQ, t.STOCK_TRANSF_SEQ2, t.PO_SEQ, t.PO_SEQ2
                                    ) res
                                )
                                s2 on s1.Seq1 = s2.Seq1 and s1.Seq2 = s2.Seq2 and s2.Completion_Flag = 0
                                order BY s1.Transf_Date
另一答案

问题似乎是在WHERE谓词中使用CASE,如下所示。

select count(DISTINCT ITEM_CD) 
  from T_TRANSFER_RESULT  
 where case when T.STOCK_TRANSF_SEQ > 0 
            then (STOCK_TRANSF_SEQ = T.STOCK_TRANSF_SEQ 
                 AND STOCK_TRANSF_SEQ2 = T.STOCK_TRANSF_SEQ2) 
            ELSE (PO_SEQ = T.PO_SEQ AND PO_SEQ2 = T.PO_SEQ2) end
   and OUTPUT_FLAG = '1' AND DELETE_FLAG = 0) as CNT2;

select * from mytab;

输出:

A    B
1    Balaji
2    Vanchi

方式不正确:

select *
     from mytab
    where case when a = 1 then b = 'Balaji' else b = 'Vanchi' end and a > 0;

输出:

ORA-00905: missing keyword

正确的方式:

select * from mytab where b = (case when a = 1 then 'Balaji' else 'Vanchi' end) and a > 0;

输出:

A    B
1    Balaji
2    Vanchi

以上是关于加入不起作用的条件的主要内容,如果未能解决你的问题,请参考以下文章

替换或删除后台堆栈上现有片段的代码不起作用

片段 getActivity 不起作用

Javascript代码片段在drupal中不起作用

从父片段到选项卡片段的接口侦听器不起作用

添加到片段时按钮不起作用

Visual Studio 自定义代码片段在方法定义的参数列表中不起作用