由于列中的双引号超过 2 个，sqlldr 失败

Question

问题是当我在 NAME 列中放置两个以上的双引号时（检查下面的代码）然后 sqlldr 失败。 在 data.txt 文件中，前 3 条记录失败，因为 NAME 列（记录中的第二个位置）有超过 2 个双引号。 请协助。我认为我需要针对 NAME 列将一些函数放入控制文件中。

您可以运行以下代码并检查，前 3 条记录将失败，最后一条记录将被插入，我想插入所有带或不带双引号的记录：

--table creation
    create table employee
    (      
      name varchar2(4000)      
    );


--control file
load data
 infile '/tmp/swetabh/data.txt' 
 into table employee
 truncate
 fields terminated by "," OPTIONALLY ENCLOSED by '"'  TRAILING NULLCOLS 
 (
 name      
 )


--data.txt (SEE ALL RECORDS ARE HAVING MORE THAN 2 DOUBLE QUOTE AND WANT TO INSERT DATA TO THE TABLE WITH OR WITHOUT DOUBLE QUOTE)
""start with double quote       "
"end with double quote          ""
"double quote " in the middle   "

-bash-4.1$ sqlldr userid=xxx/xxx@xxxx control=/tmp/swetabh/control.txt log=/tmp/swetabh/control.log bad=/tmp/swetabh/bad.txt readsize=2000000000 bindsize =2000000000

Answer 1

您对流程的提取部分有任何控制权吗？如果是这样，您可以指定数据中的双引号加倍吗 （引）？或者更好的是，指定使用管道作为分隔符，并且不要使用双引号。

Answer 2

如何应用 REPLACE 函数将 替换 双引号 (CHR(34)) 与 nothing (NULL)？另外，删除OPTIONALLY ENCLOSED 子句。

控制文件：

load data 
infile *
replace
into table employee
fields terminated by "," TRAILING NULLCOLS 
(
name "replace(:name, chr(34), null)"    
)

begindata
""start with double quote       "
"end with double quote          ""
"double quote " in the middle   "

测试：

SQL> $sqlldr scott/tiger control=test04.ctl log=test04.log

SQL*Loader: Release 11.2.0.2.0 - Production on Uto O×u 19 19:14:49 2019

Copyright (c) 1982, 2009, Oracle and/or its affiliates.  All rights reserved.

Commit point reached - logical record count 2
Commit point reached - logical record count 3

SQL> select * From employee;

NAME
--------------------------------------------------
start with double quote
end with double quote
double quote  in the middle

SQL>