Generating Sequence Number....

Hi,
I got stuck with a trivial issue. I have a table named T1 having column as Col1 (INT). I have another table named T2 having columns Col1(Int), COl2(Int). Following are the reacords in my table T1.

Col1
--
1
1
1
1
2
2
3
3
3
3
3
4
5
6
6
7

And I want to migrate the data from T1 to T2 where the Col1 data of T1 will get migrated to Col1 of T1 and after the migration is done the T2 should have the data like this.......

Col1 Col2
-- --
1 1
1 2
1 3
1 4
2 1
2 2
3 1
3 2
3 3
3 4
3 5
4 1
5 1
6 1
6 2
7 1

Thanks In Advance,
Rahul JhaI thought this would be a fun one to try - so here's a very poor attempt using an icky cursor... I don't think I have to explain to a DBA such as yourself why you shouldn't implement this method ;)

DECLARE @.col1 int
DECLARE @.col2 int
DECLARE @.id int
DECLARE @.i int
DECLARE @.x int
DECLARE @.myTable table (
id int identity(1,1)
, col1 int
, col2 int
)

SET NOCOUNT ON

INSERT INTO @.myTable(col1) VALUES(1)
INSERT INTO @.myTable(col1) VALUES(1)
INSERT INTO @.myTable(col1) VALUES(1)
INSERT INTO @.myTable(col1) VALUES(1)
INSERT INTO @.myTable(col1) VALUES(2)
INSERT INTO @.myTable(col1) VALUES(2)
INSERT INTO @.myTable(col1) VALUES(3)
INSERT INTO @.myTable(col1) VALUES(3)
INSERT INTO @.myTable(col1) VALUES(3)
INSERT INTO @.myTable(col1) VALUES(3)
INSERT INTO @.myTable(col1) VALUES(3)
INSERT INTO @.myTable(col1) VALUES(4)
INSERT INTO @.myTable(col1) VALUES(5)
INSERT INTO @.myTable(col1) VALUES(6)
INSERT INTO @.myTable(col1) VALUES(6)
INSERT INTO @.myTable(col1) VALUES(7)

SET @.i = 1
SET @.x = 1

DECLARE theIckyCursor cursor
FOR
SELECT id
, col1
FROM @.myTable
ORDER
BY col1 ASC

OPEN theIckyCursor

FETCH NEXT FROM theIckyCursor INTO @.id, @.col1

WHILE @.@.Fetch_Status <> -1 BEGIN
IF @.x <> @.col1 BEGIN
SET @.x = @.x + 1
SET @.i = 1
END

UPDATE @.myTable
SET col2 = @.i
WHERE id = @.id

SET @.i = @.i + 1
FETCH NEXT FROM theIckyCursor INTO @.id, @.col1
END

CLOSE theIckyCursor
DEALLOCATE theIckyCursor

SELECT col1
, col2
FROM @.myTable

SET NOCOUNT OFF

You know... If you provide us with *real* data, rather than just T1, T2 etc then a better answer could be provided.

(Basically, what you're trying to do sounds very wrong!)|||thnkx georgy. Even I was trying on the same line :-). But what I am looking forward is some process (without cursor) or just the query that can do the needfull. there is a huge data migration to be done. hence suggest me some thing that doesn't eat much of the system resource.|||a table with only one column where there are duplicate values cannot possibly have a primary key, so you cannot expect a good solution

export your data to excel, use a macro to apply the sequence numbers

otherwise, tell us your real situation, not this fantasy of only one column|||Ok, let's go back to the real world scenario.
Can you provide proper column headings etc and describe WHY you want to add a sequence number?
If you explain your justification for this design; perhaps we can suggest a better solution :)
EDIT: Sniped!|||Goergy I won't able able to give you the real world schema as it's highly confidential. Scenario is exactly teh same as i said. the only difference is that there are more columns than I said in both the tables.

Why I need this, is because of i am migrating the data from and de-normalised database to a normalised one.

Thanks,
Rahul Jha|||can you use an ORDER BY to get the rows into the necessary sequence for numbering them?|||select col1,row_number()over(partition by col1 order by col1)col2 into t2 from t1|||row_number?

Can you make it more clear?

Thanks,
Rahul Jha|||row_number() (http://www.databasejournal.com/features/mssql/article.php/3572301)|||pdreyer's solution actually works. I have learned something new today :)|||IVON, can you paste the code for the same. I am not getting the meaning of row_number(). How to use this function.

I'll be obliged if any one let me know on the query that pdreyer has written.

Thanks,
Rahul Jha|||I googled for ROW_NUMBER() and it seems that this is in SS 2005. This is not the part of SS 2000. And my DB is in SS 2000.

Thanks,
Rahul Jha|||please answer the question in post #7|||This one ought to leave a mark!SELECT 1 AS ick
INTO #junque
UNION ALL SELECT 1
UNION ALL SELECT 1
UNION ALL SELECT 1
UNION ALL SELECT 2
UNION ALL SELECT 2
UNION ALL SELECT 3
UNION ALL SELECT 3
UNION ALL SELECT 3
UNION ALL SELECT 3
UNION ALL SELECT 3
UNION ALL SELECT 4
UNION ALL SELECT 5
UNION ALL SELECT 6
UNION ALL SELECT 6
UNION ALL SELECT 7

SELECT ick, Identity(INT, 1, 1) AS bar, 0 AS bat
INTO #foo
FROM #junque

UPDATE #foo
SET bat = (SELECT Count(*)
FROM #foo AS baz
WHERE baz.ick = #foo.ick
AND baz.bar <= #foo.bar)

SELECT ick, bat
FROM #foo

DROP TABLE #foo
DROP TABLE #junque-PatP|||ORDER BY on COL1 can be done......... Let me know if i hv cleared ur doubt or not............|||have you seen pats post on page 1|||This one ought to leave a mark!ouch!

[applies mark remover]|||I googled for ROW_NUMBER() and it seems that this is in SS 2005. This is not the part of SS 2000.
True, true.
And my DB is in SS 2000.
We didn't know that.|||ORDER BY on COL1 can be doneexcellent

here is your solution:

select ( select count(*) + 1
from daTable
where Col1 < T.Col1 ) as rownum
, Col1
from daTable as T
order
by Col1

rownum Col1
1 1
1 1
1 1
1 1
5 2
5 2
7 3
7 3
7 3
7 3
7 3
12 4
13 5
14 6
14 6
16 7
here is pdreyer's ROW_NUMBER() solution for comparison --

select row_number() over
( partition by Col1
order by Col1 ) as rownum
, Col1
from daTable
order
by Col1

rownum Col1
1 1
2 1
3 1
4 1
1 2
2 2
1 3
2 3
3 3
4 3
5 3
1 4
1 5
1 6
2 6
1 7
if you remove the PARTITION BY clause, then the result is

rownum Col1
1 1
2 1
3 1
4 1
5 2
6 2
7 3
8 3
9 3
10 3
11 3
12 4
13 5
14 6
15 6
16 7|||You guys rock............

Thanks for the solution.......... Thanks patty, r937 and pdreyer......... gr8 help that was.

Thanks,
Rahul Jha|||SELECT ick, Identity(INT, 1, 1) AS bar, 0 AS bat
INTO #foo
FROM #junqueI had no idea you could insert identity values that way.|||Thanks patty

*laughs out loud*|||:) :) Don Laugh Georgy............... :) :)

I really learnt few things thru this thread.........

Thanks guys once again...... You all are awesome......... Nothing to fear Rahul till the time you guys are there......... :) :) :)

Thanks,
Rahul Jha|||I had no idea you could insert identity values that way.Identity function - I think it is new to 2005. The devs here love it because they love select into too.|||that's a negative poots. the identity function is in SQL 2k. I have used it before.|||My mistake. I came across it about the time we migrated so probably got my wires crossed there.|||well, be careful for what you wish for

http://weblogs.sqlteam.com/brettk/archive/2004/06/09/1530.aspx|||-- Prepare sample data
CREATE TABLE #Sample
(
Col1 INT,
Col2 INT
)

INSERT #Sample
(
Col1
)
SELECT 1 UNION ALL
SELECT 1 UNION ALL
SELECT 1 UNION ALL
SELECT 1 UNION ALL
SELECT 2 UNION ALL
SELECT 2 UNION ALL
SELECT 3 UNION ALL
SELECT 3 UNION ALL
SELECT 3 UNION ALL
SELECT 3 UNION ALL
SELECT 3 UNION ALL
SELECT 4 UNION ALL
SELECT 5 UNION ALL
SELECT 6 UNION ALL
SELECT 6 UNION ALL
SELECT 7

-- Check precondition
SELECT * FROM #Sample

-- Setup staging mechanism
CREATE CLUSTERED INDEX IX_Sample ON #Sample (Col1)

DECLARE @.Index INT,
@.Col INT

SELECT TOP 1 @.Index = 0,
@.Col = Col1
FROM #Sample
ORDER BY Col1

UPDATE #Sample
SET @.Index = Col2 = CASE WHEN Col1 = @.Col THEN @.Index + 1 ELSE 1 END,
@.Col = Col1

DROP INDEX #Sample.IX_Sample

-- Check postcondition
SELECT * FROM #Sample

-- Clean up
DROP TABLE #Sample