按照單個字符查詢匹配

begin 
declare
cursor myemp_cur
is select * from table_a a where a.type1=‘user‘;
type myemp_tab is table of table_a%rowtype;
myemp_rd myemp_tab;
begin
open myemp_cur;
loop --循環1
fetch myemp_cur bulk collect into myemp_rd limit 20;
for i in 1..myemp_rd.count loop
dbms_output.put_line(‘**********************************************************‘);
dbms_output.put_line(
 
‘目標對象：‘||myemp_rd(i).MKT_TRADE_TYPE_DESC);
dbms_output.put_line(‘結果對象：‘);
--循環2
declare i_num number;
begin
  select length(myemp_rd(i).MKT_TRADE_TYPE_DESC) into i_num from dual;
  for i1 in 1..i_num loop

dbms_output.put_line(i1||‘--‘);

--循環3
    declare
    cursor myemp_cur1
    is select * from table_a a where
 
 a.DK_SYSTEM_OF_UPD！=‘FCMS‘ and 
    a.mkt_trade_type_desc like ‘%‘||substr(myemp_rd(i).MKT_TRADE_TYPE_DESC,i1,1)||‘%‘;
    type myemp_tab1 is table of table_a%rowtype;
    myemp_rd1 myemp_tab1;
    begin
    open myemp_cur1;
    loop
    fetch myemp_cur1 bulk collect into myemp_rd1 limit 20;
    for
 
 i3 in 1..myemp_rd1.count loop
    dbms_output.put_line(i3||‘:‘||myemp_rd1(i3).MKT_TRADE_TYPE_DESC);
    end loop;
    exit when myemp_cur1%notfound;
    end loop;
    end;
--
end loop;
end;
--
end loop;
exit when myemp_cur%notfound;
end loop;
end;
end;

結果:
不太盡人意。換成兩個字符匹配後：

看起來明了多了。

oracle利用循環批量檢索對應的數據