Delete duplicates in each column, not duplicate records

[watertraveller]

Hi all. My ultimate goal is to return two columns, where no single value appears anywhere twice. This means that not only do I want to check that nothing from column A appears in column B and vice-versa, but I also don't want the same value appearing twice in A and twice in B.

So far I have:

--Diff the columns
INSERT INTO @Table
SELECT One, Two
FROM @Column1 a
FULL OUTER JOIN @Column2 b
ON ((One NOT IN (Two)) OR (Two NOT IN (One)))
 

But then my trouble is removing the redundant values in each column. I tried splitting the table into single-column tables so that they would no longer be linked together, and then deleting duplicate records out of that single-column table. But it didn't work.

-- Attempt at removing duplicate values
DELETE FROM @Column1
DELETE FROM @Column2
INSERT INTO @Column1 SELECT One FROM @Table
INSERT INTO @Column2 SELECT Two FROM @Table
INSERT INTO @Column3 SELECT DISTINCT One FROM @Column1 
         ORDER BY One
INSERT INTO @Column4 SELECT DISTINCT Two FROM @Column2 
         ORDER BY Two
 

But when I display these two columns together there are duplicates, and when I display them individually they aren't even in order. What's wrong here?

Thanks,
-C

[ck9663]

Could you post some of your sample data? Let's see if we can find a pattern among the values.

-- CK