需求:
将以下文件中每行数据，按列拆分为多行。如将文件a.txt插入表中，
文件a.txt内容：
a 1 2 3
b 11 12 13
c 21 22 23

插入后的效果：
a col1 1
b col1 11
c col1 21
a col2 2
b col2 12
c col2 22
a col3 3
b col3 13
c col3 23

--目标表
SQL> desc tmp_mh
 Name        Null?    Type
 ----------------------------------------- -------- ----------------------------
 X          VARCHAR2(10)
 KEY          VARCHAR2(10)
 VALUE          NUMBER(38)

--控制文件:tmp.ctl
--只有使用position,才会从行首读取数据
LOAD DATA
INTO TABLE tmp_mh
( x POSITION(1:1),
  key constant "col1",
  value TERMINATED BY WHITESPACE,
  x3 filler TERMINATED BY WHITESPACE,
  x4 filler TERMINATED BY WHITESPACE
)
INTO TABLE tmp_mh
( x POSITION(1:1),
  key constant "col2",
  x2 filler TERMINATED BY WHITESPACE,
  value TERMINATED BY WHITESPACE,
  x4 filler TERMINATED BY WHITESPACE
)
INTO TABLE tmp_mh
( x POSITION(1:1),
  key constant "col3",
  x2 filler TERMINATED BY WHITESPACE,
  x3 filler TERMINATED BY WHITESPACE,
  value TERMINATED BY WHITESPACE
)

sqlldr userid/userid control=tmp.ctl data=a.txt log=a.log