Question

我正在尝试将列标题和相应字段插入另一个表中。

Table1 :

col1    col2   col3   col4
1       2       3      4

输出应如下所示：

COL_A   COL_B   COL_C     COL_D    COL_E     COL_F      COL_G    COL_H 
col1    1       col2       2       col3       3         col4     4

表1将动态创建，因此表1的结构可能会有所不同，例如。有时，可能是

Case1:   

col1    col2   col3   col4
1       2       3      4

对于Case1，输出应如下所示：

COL_A   COL_B   COL_C     COL_D    COL_E     COL_F      COL_G    COL_H 
col1    1       col2       2       col3       3         col4     4

或者它可能是情况2：

  col1    col2   
    1       2

对于Case2，输出应如下所示：

COL_A   COL_B   COL_C     COL_D    
col1    1       col2       2

我在Stack Overflow的另一个问题中得到了类似要求的静态查询的帮助，但我试图将静态查询转换为动态查询。静态查询：

静态表结构：

CREATE TABLE Table1
    (col1 int, col2 int, col3 int, col4 int)
;

INSERT ALL 
    INTO Table1 (col1, col2, col3, col4)
         VALUES (1, 2, 3, 4)
SELECT * FROM dual
;

静态查询：

SELECT * FROM
 (
    SELECT *
    FROM Table1
    UNPIVOT(val FOR col IN (
                COL1
                ,COL2
                ,COL3
                ,COL4
                ))
    )
PIVOT(  MAX(COl) as C, MAX(VAL) as V FOR COL IN (
             'COL1' as VAL1
            ,'COL2' as VAL2
            ,'COL3' as VAL3
            ,'COL4' as VAL4
            ));

输出：

VAL1_C   VAL1_V    VAL2_C    VAL2_V  VAL3_C   VAL3_V   VAL4_C   VAL4_V
COL1      1        COL2        2       COL3       3      COL4      4

当我试图使静态查询动态化时，我尝试了这个：

SELECT * FROM
 (
    SELECT *
    FROM Table1
    UNPIVOT(val FOR col IN (
                select regexp_substr((select * from (SELECT LISTAGG(COLUMN_NAME, ',') WITHIN GROUP ( order by COLUMN_ID) aa
FROM (select * from all_tab_columns where table_name = 'TABLE1'))),'[^,]+', 1, level) "yyy" from dual
   connect by regexp_substr((select * from (SELECT LISTAGG(COLUMN_NAME, ',') WITHIN GROUP ( order by COLUMN_ID) aa
FROM (select * from all_tab_columns where table_name = 'TABLE1'))), '[^,]+', 1, level) is not null
                ))
    )
PIVOT(  MAX(COl) as C, MAX(VAL) as V FOR COL IN (
             select regexp_substr((select '''' || AA ||'''' from (SELECT LISTAGG(COLUMN_NAME, ''',''') WITHIN GROUP ( order by COLUMN_ID) aa
FROM (select * from all_tab_columns where table_name = 'TABLE1'))),'[^,]+', 1, level) "yyy"  from dual
   connect by regexp_substr((select '''' || AA ||'''' from (SELECT LISTAGG(COLUMN_NAME, ''',''') WITHIN GROUP ( order by COLUMN_ID) aa
FROM (select * from all_tab_columns where table_name = 'TABLE1'))), '[^,]+', 1, level) is not null
            ));

但这引发了错误：

ORA-00904: : invalid identifier
00904. 00000 -  "%s: invalid identifier"
*Cause:    
*Action:
Error at Line: 6 Column: 5

我尝试使用动态sql .Below是我使用的动态sql：

declare
  sql_stmt     varchar2(4000);
  pivot_clause1 varchar2(4000);
  pivot_clause2 varchar2(4000);


begin
  SELECT LISTAGG(COLUMN_NAME, ',') WITHIN GROUP ( order by COLUMN_ID) aa
into   pivot_clause1
  FROM (select * from all_tab_columns where table_name = 'TEST11') ;

 select * into pivot_clause2 from 
 (  
 select  AA ||'''' aa

 from (SELECT ''''||LISTAGG(COLUMN_NAME, ''',''') WITHIN GROUP ( order by COLUMN_ID) aa 
FROM (select * from all_tab_columns where table_name = 'TEST11')) 

   );

  dbms_output.put_line(pivot_clause1);
   dbms_output.put_line(pivot_clause2);


sql_stmt := 'create table test13 as ('|| 'SELECT * FROM
 (
    SELECT *
    FROM TEST11
    UNPIVOT(val FOR col IN ( '
                || pivot_clause1 || '
                ))
    )
   PIVOT(  MAX(COl) as C, MAX(VAL) as V FOR COL IN (' || 
           pivot_clause2   || '
            )) 

    )';
   execute immediate sql_stmt  ;


end;
/

对于具有以下描述的表格罚款：

Name Null Type       
---- ---- ---------- 
COL1      NUMBER(38) 
COL2      NUMBER(38) 
COL3      NUMBER(38) 
COL4      NUMBER(38)

但不适用于具有说明的表格

COL1                          DATE          
COL1                          VARCHAR2(100) 
COL1                         VARCHAR2(100) 
COL1                         NUMBER        
COL1                         NUMBER

我收到以下错误：

Error report -
ORA-01790: expression must have same datatype as corresponding expression
ORA-06512: at line 38
01790. 00000 -  "expression must have same datatype as corresponding expression"
*Cause:    
*Action:

我刚学会 unpivot不能与具有不同数据类型的列一起使用，有什么办法吗？有人可以帮忙!!

如何在Oracle

0 个答案: