removing duplicates from table without using temporary table - sql

Question

I've a table(TableA) with contents like this:
Col1
-----
A
B
B
B
C
C
D
i want to remove just the duplicate values without using temporary table in Microsoft SQL Server. can anyone help me?
the final table should look like this:
Col1
-----
A
B
C
D
thanks :)

Answer 1

WITH TableWithKey AS (
SELECT ROW_NUMBER() OVER (ORDER BY Col1) As id, Col1 As val
FROM TableA
)
DELETE FROM TableWithKey WHERE id NOT IN
(
SELECT MIN(id) FROM TableWithKey
GROUP BY val
)

Answer 2

Can you use the row_number() function (http://msdn.microsoft.com/en-us/library/ms186734.aspx) to partition by the columns you're looking for dupes on, and delete where row number isn't 1?

Answer 3

I completely agree that having a unique identifier will save you a lot of time.
But if you can't use one (or if this is purely hypothetical), here's an alternative: Determine the number of rows to delete (the count of each distinct value -1), then loop through and delete top X for each distinct value.
Note that I'm not responsible for the number of kittens that are killed every time you use dynamic SQL.
declare #name varchar(50)
declare #sql varchar(max)
declare #numberToDelete varchar(10)
declare List cursor for
select name, COUNT(name)-1 from #names group by name
OPEN List
FETCH NEXT FROM List
INTO #name,#numberToDelete
WHILE ##FETCH_STATUS = 0
BEGIN
IF #numberToDelete > 0
BEGIN
set #sql = 'delete top(' + #numberToDelete + ') from #names where name=''' + #name + ''''
print #sql
exec(#sql)
END
FETCH NEXT FROM List INTO #name,#numberToDelete
END
CLOSE List
DEALLOCATE List
Another alternative would to be create a view with a generated identity. In this way you could map the values to a unique identifer (allowing for conventional delete) without making a permanent addition to your table.

Answer 4

Select grouped data to temp table, then truncate original, after that move back it to original.
Second solution, I am not sure will it work but you can try open table directly from SQL Management Studio and use CTRL + DEL on selected rows to delete them. That is going to be extremely slowly because you need to delete every single row by hands.

Answer 5

You can remove duplicate rows using a cursor and DELETE .. WHERE CURRENT OF.
CREATE TABLE Client ([name] varchar(100))
INSERT Client VALUES('Bob')
INSERT Client VALUES('Alice')
INSERT Client VALUES('Bob')
GO
DECLARE #history TABLE (name varchar(100) not null)
DECLARE #cursor CURSOR, #name varchar(100)
SET #cursor = CURSOR FOR SELECT name FROM Client
OPEN #cursor
FETCH NEXT FROM #cursor INTO #name
WHILE ##FETCH_STATUS = 0
BEGIN
IF #name IN (SELECT name FROM #history)
DELETE Client WHERE CURRENT OF #cursor
ELSE
INSERT #history VALUES (#name)
FETCH NEXT FROM #cursor INTO #name
END