Question

我正在尝试从plsql程序中调用teradata中定义的宏，通过在线搜索各种博客，我能够设计下面的进程来调用并从所有宏中获取值，这些宏只包含一个选择查询，但是如果我使用相同的技术从宏中获取可能包含多个查询的值，然后只获取第一个查询的结果。

宏（单个查询） -

replace macro a_macro1 (arg1 (char(9)), arg2 (char(8)), arg3 (char(8))) as 
(select ret1 (char(5)), ret2 (char(3)) 
 from table1 where arg1 = :arg1 and arg2 = :arg2 and arg3 = :arg3;);

宏（多个查询） -

replace macro a_macro2 (arg1 (char(9)), arg2 (char(5)), arg3 (char(8)), arg4 
 (char(8))) as (
 select ret1, ret2, ret3, ret4, ret5, ret6, ret7  
 from table1 where table1.arg1 = :arg1 and table1.arg2 = :arg2 
 and table1.arg3 = :arg3 and table1.arg4 = :arg4;

 select ret1, ret2 from table2 where ret2 = 'm' and 
   table2.arg1 = :arg1 
 and table2.arg3 = :arg3 and table2.arg4 = :arg4;

  select ret1, ret2 from table3 where ret2 in ('a','z') 
   and table3.arg1 = :arg1 
  and table3.arg2 = :arg2 and table3.arg3 = :arg3 and table3.arg4 = 
   :arg4;

  select ret1, ret2, ret3, ret4, ret5, ret6, ret7, ret8
   from table4 where arg1 = :arg1 and arg2 = :arg2 and arg3 = :arg3 
   and arg4 = :arg4;);

使用PLSQL方法 -

c := DBMS_HS_PASSTHROUGH.OPEN_CURSOR@TERA_DBLINK;
DBMS_HS_PASSTHROUGH.PARSE@TERA_DBLINK(c, 'EXEC a_macro1(?,?,?)');
DBMS_HS_PASSTHROUGH.BIND_VARIABLE@TERA_DBLINK(c,1,'val1');
DBMS_HS_PASSTHROUGH.BIND_VARIABLE@TERA_DBLINK(c,2,'val2');
DBMS_HS_PASSTHROUGH.BIND_VARIABLE@TERA_DBLINK(c,3,'val3');
num_rows := DBMS_HS_PASSTHROUGH.EXECUTE_NON_QUERY@TERA_DBLINK(c);

DBMS_OUTPUT.PUT_LINE(num_rows || ' Rows fetched');


LOOP
    num_rows := DBMS_HS_PASSTHROUGH.FETCH_ROW@TERA_DBLINK(c);
    EXIT WHEN num_rows = 0;
    i := 1;
    loop 
        BEGIN
        exit when i = 99;
        DBMS_HS_PASSTHROUGH.GET_VALUE@TERA_DBLINK(c, i, val1);
        DBMS_OUTPUT.PUT_LINE(val1);
        i := i+1;
        EXCEPTION 
        WHEN OTHERS THEN
        BEGIN
            i:=99;
            END;
        END;
    end loop;
END LOOP;  

DBMS_HS_PASSTHROUGH.CLOSE_CURSOR@TERA_DBLINK(c);

此方法适用于从第一个宏获取值。如何更改它以使其适用于第二个宏？

另外，有没有办法获取列名？

GET_VALUE

仅提取值，而不是列名。

编辑1 - 这是我调用第二个宏的代码 -

set serveroutput on;
DECLARE
num_rows INTEGER;
c INTEGER;
i INTEGER;
val1  VARCHAR2(100);
BEGIN
c := DBMS_HS_PASSTHROUGH.OPEN_CURSOR@TERA_DBLINK;
DBMS_OUTPUT.PUT_LINE('Cursor opened');
DBMS_HS_PASSTHROUGH.PARSE@TERA_DBLINK(c, 'EXEC a_macro2 (?,?,?,?)');
DBMS_OUTPUT.PUT_LINE('command set for parse');
DBMS_HS_PASSTHROUGH.BIND_VARIABLE@TERA_DBLINK(c,1,'123456789');
DBMS_OUTPUT.PUT_LINE('first variable bound');
DBMS_HS_PASSTHROUGH.BIND_VARIABLE@TERA_DBLINK(c,2,'00001');
DBMS_OUTPUT.PUT_LINE('second variable bound');
DBMS_HS_PASSTHROUGH.BIND_VARIABLE@TERA_DBLINK(c,3,'20161215');
DBMS_OUTPUT.PUT_LINE('third variable bound');
DBMS_HS_PASSTHROUGH.BIND_VARIABLE@TERA_DBLINK(c,4,'15462250');
DBMS_OUTPUT.PUT_LINE('fourth variable bound');
num_rows := DBMS_HS_PASSTHROUGH.EXECUTE_NON_QUERY@TERA_DBLINK(c);
DBMS_OUTPUT.PUT_LINE(num_rows || ' Rows fetched');

LOOP
    num_rows := DBMS_HS_PASSTHROUGH.FETCH_ROW@TERA_DBLINK(c);
    EXIT WHEN num_rows = 0;
    i := 1;
    loop 
        BEGIN
        exit when i = 99;
        DBMS_HS_PASSTHROUGH.GET_VALUE@TERA_DBLINK(c, i, val1);
        DBMS_OUTPUT.PUT_LINE(val1);
        i := i+1;
        EXCEPTION 
        WHEN OTHERS THEN
        BEGIN
            i:=99;
            END;
        END;
    end loop;
END LOOP;  

DBMS_HS_PASSTHROUGH.CLOSE_CURSOR@TERA_DBLINK(c);
END;
/

输出 -

 Cursor opened
 command set for parse
 first variable bound
 second variable bound
 third variable bound
 fourth variable bound
 1 Rows fetched
 ABCDEF02AB8
 ABCDEF02ABC
 1234
 T1F
 1F
 A
 Y

仅返回7个值，这些值来自第二个宏的第一个查询。

Answer 1

要使代码适用于下一个宏，只需在此步骤进行更改：

DBMS_HS_PASSTHROUGH.PARSE@TERA_DBLINK(c, 'EXEC a_macro2(?,?,?,?,?)');

不要忘记添加bind varaibles作为传递的参数数量。

从PL / SQL调用具有多个选择查询的teradata宏

1 个答案: