Question

我正在尝试导入使用Bteq脚本导出的数据，但无法加载到另一个表中。

.EXPORT report file=E:\t\tgt\report_excel_092702.csv;
.set width 200;
.SET TITLEDASHES OFF;
.SET SEPARATOR ','; 
select  empno,coalesce(ename,'no_name') as ename,coalesce(job,'no_job') as job ,coalesce(hiredate, date) as hiredate,COALESCE(mgr,0) as mgr,COALESCE(sal, 0) as sal,COALESCE(comm, 0) as comm,deptno from emps;
--.EXPORT reset
.LOGOFF
.EXIT

数据库teradata1

report_excel_092702.csv（已导出.csv）

 EMPNO  ename       job     hiredate     mgr     sal    comm DEPTNO
7698    BLAKE   MANAGER     81/11/17    7839    2850       0    30
7839    KING    PRESIDENT   81/11/17       0    50000      0    10
7654    MARTIN  SALESMAN    81/09/28    7698    1250    1400    30

Bteq脚本......

.import vartext ',' file='E:\t\tgt\report_excel_092702.csv',skip=1;

 DATABASE tetra;
 delete from emp_csv;
.QUIET ON 
.REPEAT * 
 USING 
        empno    (varchar(100))
        ,ename   (varchar(100))
        ,hiredate (varchar(100))

 INSERT INTO emp_csv(empno,ename,hiredate)
        VALUES (
        :empno,
        :ename,
        Cast((:hiredate (Date,Format 'dd.mm.yyyy')) As Date) 
        --cast(hiredate 'yy/mm/dd') as :hiredate
        --CAST( CAST( :hiredate AS DATE FORMAT 'mm/dd/yyyy') AS DATE FORMAT 'mm/dd/yyyy')
        );
.QUIT
.LOGOFF

Answer 1

我想知道为什么不进行简单的CASing，因为在BTET或ANSI模式下yy / mm / dd隐式转换为DATE？它适用于我，使用与您相同的日期格式。

。LOGON dbc, dbc; .import vartext '|' FILE = /root/ETL/new_revision/report.csv, skip=1; .REPEAT * USING ( EmpNo VARCHAR(2000) ,Name VARCHAR(2000) ,hire_date VARCHAR(2000) ) INSERT INTO retail.employee_import VALUES ( :EmpNo, :Name, CAST(:hire_date AS DATE), ); .QUIT .LOGOFF

如何使用Bteq从.CSV导入日期到Teradata表？

1 个答案: