SQL Joins where join-condition column contains a NULL
From: Jeff Zucker <jeff_at_vpservices.com>
Date: Sat, 06 Oct 2001 17:46:39 -0700
Message-ID: <3BBFA5EF.B9C519BC_at_vpservices.com>
I am working on joins for the perl module SQL::Statement and I have some questions regarding NULLs in SQL92 (as a standard, I don't really care about particular implementations).
Date: Sat, 06 Oct 2001 17:46:39 -0700
Message-ID: <3BBFA5EF.B9C519BC_at_vpservices.com>
I am working on joins for the perl module SQL::Statement and I have some questions regarding NULLs in SQL92 (as a standard, I don't really care about particular implementations).
Given two tables called "lower" and "upper", each composed of a single row:
lower : number | lowcase
-------+--------
NULL | x
upper : number | upcase
-------+--------
NULL | Y
And Given the statement template:
SELECT lowcase, upcase
FROM lower <join_type> JOIN upper USING (number)
Is this results table correct (omiting the USING clause with UNION)?
join_type results
INNER 0 rows LEFT 1 row: x,NULL RIGHT 1 row: NULL,Y FULL 2 rows: 1)x,NULL 2)NULL,Y UNION 2 rows: 1)x,NULL 2)NULL,Y
And would the results be the same if NATURAL or ON lower.number = upper.number were used instead of the USING clause?
-- JeffReceived on Sun Oct 07 2001 - 02:46:39 CEST