Hello:

I have 3 table with this description:

Table1
--------
id: int
value1: int

Table2
--------
id: int
value1: int
title as varchar(80)

Table3
--------
id: int
norden: int
text : varchar


I have to transform data from Table1 to Table2 but get some value from
Table3 too, I explain what I want specific.
in Table3 I have for every id about 24 text, and id of each text is norden,
then I have to get from Table1 id and value1, to import in id and value1 to
Table2, but then get the first text in Table3 for current id in progress
that have more than 40 character, how can I do that.

Thanks for advantage,
Best regards,
Owen.

If they all are in the same DB on the same server then you should do
this in the SQL Statement you use as the source for your datapump.

Allan

"Owen" <anibal (AT) prensa-latina (DOT) cu> wrote

Please can you explain me yout idea step by step, because I'm a begginer in
DTS please.

Best regards,
Owen.

"Allan Mitchell" <allan (AT) no-spam (DOT) sqldts.com> wrote

You want something very simialr to

This does not really have a lot to do with DTS per se as this could be
done without it. You want something very similar to

(PS you do not tell is if these tables are in the same DB or even on the
same server)

INSERT TABLE2(Column List)
SELECT (Columns)
FROM TABLE2 T2 JOIN TABLE3 T3
ON T2.ID = T3.ID


Allan


"owen" <anibal (AT) prensa-latina (DOT) cu> wrote

I appreciate your answer to my questions, but let me ask other about the
same problem, I make query like you teach me but I get some row duplicate,
how I can remove this duplicate row

remember that in table 3 for one id have a lot of norden, I make this
query:

select *
from table1 inner join table3 on (table1.id = table3.id)
where len(table3.text) > 50
order by norden //to get the first row that get this goal, because I want to
get the first order by norden

then I want to elimate the duplicate id that return this query, can you help
again please.

Best regards,
Owen.


"Allan Mitchell" <allan (AT) no-spam (DOT) sqldts.com> wrote

So by joining onto Table3 you get many rows back for the same id because
table3 has many entries for that id?

You only want to return the one?

Which one?

There must be a reason you join onto Table 3 so what exactly do you want
from there? The latest entry in date? The first alphabetically


Let me give you an example

CREATE TABLE T1(T1PK int primary Key, col2 int)
CREATE TABLE T2(T2PK int primary Key, col2 int, T1PK int references
T1(T1PK))

INSERT T1(T1PK,Col2) VALUES(1,1)
INSERT T2(T2PK,Col2,T1PK) VALUES(1,1,1)
INSERT T2(T2PK,Col2,T1PK) VALUES(2,2,1)
INSERT T1(T1PK,Col2) VALUES(2,1)
INSERT T2(T2PK,Col2,T1PK) VALUES(3,3,2)
INSERT T2(T2PK,Col2,T1PK) VALUES(4,4,2)
INSERT T2(T2PK,Col2,T1PK) VALUES(5,5,2)

--This will return all rows
SELECT
T1.col2,T2.col2
FROM
T1 JOIN T2
ON
T1.T1PK = T2.T1PK

--This will return only one row for each ID based on the MIN(col2) in T2
SELECT
T1.col2,T2.col2
FROM
T1 JOIN T2
ON
T1.T1PK = T2.T1PK
AND
T2.Col2 = (SELECT MIN(col2) FROM T2 WHERE T1PK = T1.T1PK)


Allan


"owen" <anibal (AT) prensa-latina (DOT) cu> wrote