我在Netezza工作,或者您知道的是Analytics(分析)的纯数据-nzsql,但是我认为这是一个ANSI SQL问题。这个问题是如此基本,我什至不知道如何搜索。
CREATE TEMPORARY TABLE DEMO1 AS SELECT 'SMORK' AS SMORK, 'PLONK' AS PLONK, 'SPROING' AS SPROING; SELECT SMORK AS PLONK, PLONK, SPROING AS CLUNK, CLUNK FROM DEMO1;
这将返回“ SMORK,PLONK,SPROING,SPROING”,也就是说,重用CLUNK别名可以很好地进行查询,但是源表中的列会覆盖PLONK别名。现在,如果我真的想要源表中的列,则可以编写SELECT SMORK AS PLONK,DEMO1.PLONK等,但是我不知道如何指定我更喜欢先前在同一目录中定义的别名。 SELECT子句。
有人知道吗?
在Netezza中,选择列时, Netezza将首先搜索表列,然后搜索别名。
示例:假设我们有以下语句:
CREATE TEMPORARY TABLE EMPLOYEES AS SELECT 1001 AS EMPLOYEE_ID ,'Alex' AS FIRST_NAME ,'Smith' AS LAST_NAME ,'Alex J. Smith' AS FULL_NAME; SELECT EMPLOYEE_ID ,FIRST_NAME ,LAST_NAME ,LAST_NAME||', '||FIRST_NAME AS FULL_NAME ,'My full name is :'||FULL_NAME AS DESCRIPTION FROM EMPLOYEES;
它将返回
**EMPLOYEE_ID FIRST_NAME LAST_NAME FULL_NAME说明** 1001 Alex Smith Smith,Alex我的全名是:Alex J. Smith
注意,在 DESCRIPTION中 ,FULL_NAME值是从表列而不是别名中选择的。
如果要使DESCRIPTION列使用别名“ FULL_NAME”中的值,则可以分两个步骤进行操作:
步骤1. 创建一个包含所有想要的列的子查询。对于要重用的所有别名,您需要将其命名为FROM子句中任何表列中都不存在的名称。
步骤2. 仅从子查询中选择所需的列。
CREATE TEMPORARY TABLE EMPLOYEES AS SELECT 1001 AS EMPLOYEE_ID, 'Alex' AS FIRST_NAME, 'Smith' AS LAST_NAME, 'Alex J. Smith' AS FULL_NAME; WITH EMPLOYESS_TMP AS ( SELECT EMPLOYEE_ID ,FIRST_NAME ,LAST_NAME ,LAST_NAME||', '||FIRST_NAME AS FULL_NAME2 ,FULL_NAME2 AS FULL_NAME ,'My full name is :'||FULL_NAME2 AS DESCRIPTION FROM EMPLOYEES) SELECT EMPLOYEE_ID ,FIRST_NAME ,LAST_NAME ,FULL_NAME ,DESCRIPTION FROM EMPLOYESS_TMP;
这将返回您想要的:
**EMPLOYEE_ID FIRST_NAME LAST_NAME FULL_NAME说明** 1001 Alex Smith Smith,Alex我的全名是:Smith,Alex