Merge Join Output Bug?

I've run into something that looks like a bug to me but I wanted to run it by the board:

Merge join 2 sorted tables.

Table1: ColumnA : Sort Order 1, ColumnB Sort Order 2

Table2 : ColumnA: Sort Order 1, ColumnB Sort Order 2, ColumnC not sorted

Merge Join the two tables on ColumnA and ColumnB...

Choose the following as output columns

A + B + C = works

C = works

A + C = works

B + C = NOT work.. error message: The column with the SortKeyPosition value of 0 is not valid. It should be 2.

Basically if you choose one or more of the sorted columns in the output at least one of them has to be the column with Sort position 1 or you'll get that error.

Is this a bug or intentional? If you do not have sort column 1 in the output that output could no longer be considered sorted... so perhaps the error is related to that (instead of error I'd expect some warning about the sorting). Interesting that it lets you choose C only becuase that also makes the output unsorted.

I see your point Chris.

I think it is intential -

. The reason why B+C not work is because column B has a non-zero sortKeyPosition which indicates the output (to which B belongs) should be sorted (in other words, the output's "isSorted" property is true), but the output can not find a column with SortKeyPosition 1

. As for why C column only works is because the output is then not sorted.

If you think the error message is not very helpful, please log a customer issue through our connect website http://connect.microsoft.com/SQLServer and your request will be addressed soon as appropriate.

Thanks

wenyang

Merge Join Output Bug?

I've run into something that looks like a bug to me but I wanted to run it by the board:

Merge join 2 sorted tables.

Table1: ColumnA : Sort Order 1, ColumnB Sort Order 2

Table2 : ColumnA: Sort Order 1, ColumnB Sort Order 2, ColumnC not sorted

Merge Join the two tables on ColumnA and ColumnB...

Choose the following as output columns

A + B + C = works

C = works

A + C = works

B + C = NOT work.. error message: The column with the SortKeyPosition value of 0 is not valid. It should be 2.

Basically if you choose one or more of the sorted columns in the output at least one of them has to be the column with Sort position 1 or you'll get that error.

Is this a bug or intentional? If you do not have sort column 1 in the output that output could no longer be considered sorted... so perhaps the error is related to that (instead of error I'd expect some warning about the sorting). Interesting that it lets you choose C only becuase that also makes the output unsorted.

I see your point Chris.

I think it is intential -

. The reason why B+C not work is because column B has a non-zero sortKeyPosition which indicates the output (to which B belongs) should be sorted (in other words, the output's "isSorted" property is true), but the output can not find a column with SortKeyPosition 1

. As for why C column only works is because the output is then not sorted.

If you think the error message is not very helpful, please log a customer issue through our connect website http://connect.microsoft.com/SQLServer and your request will be addressed soon as appropriate.

Thanks

wenyang

Merge Join - HELP

Hi,

I have a SQL Statatment:

SELECT * FROM TABLE1 AS A
JOIN TABLE2 AS B
ON A.X= B.X
AND A.Y= B.Y

When i execute this code in sql server returns 549 lines. I created a package with two oledb sources one for each table, sorted the tables with fields X and Y after placed a Merge Join with the fields:

A.Y join B.Y order 1

A.X join B.X order 2

both fields with the Join Key checked

But my package return 411 lines.

What's happened? :(

When a i have the code:

SELECT A.X, A.Y, B.X, B.Y
FROM TABLE1 AS A
JOIN TABLE2 AS B
ON A.X= B.X

When i did the join only one field SSIS worked fine, sql server returns 622 and SSIS returns 622 lines.

Please help-me...

Thanks,

Andr

Are your join keys all integers or does one of them have characters. SSIS is case senstive while SQL Server by default isn't. You may want to check this.|||

Also check that the inputs of the Merge Join are sorted...

Rafael Salas

|||

Thanks Jay,

My columns are characters type and my oledb source 1 columns is lowercase and oledb source 2 columns is uppercase.

I created a Derived Column replaced my column to uppercase. :)

|||Character Map Transform might be more appropriate to change case|||

Rafael Salas wrote:

Also check that the inputs of the Merge Join are sorted...

Rafael Salas

Very good point. And note that setting IsSorted=1 on an output does NOT sort the data.

-Jamie

Merge Join - HELP

Hi,

I have a SQL Statatment:

SELECT * FROM TABLE1 AS A
JOIN TABLE2 AS B
ON A.X= B.X
AND A.Y= B.Y

When i execute this code in sql server returns 549 lines. I created a package with two oledb sources one for each table, sorted the tables with fields X and Y after placed a Merge Join with the fields:

A.Y join B.Y order 1

A.X join B.X order 2

both fields with the Join Key checked

But my package return 411 lines.

What's happened? :(

When a i have the code:

SELECT A.X, A.Y, B.X, B.Y
FROM TABLE1 AS A
JOIN TABLE2 AS B
ON A.X= B.X

When i did the join only one field SSIS worked fine, sql server returns 622 and SSIS returns 622 lines.

Please help-me...

Thanks,

Andr

Are your join keys all integers or does one of them have characters. SSIS is case senstive while SQL Server by default isn't. You may want to check this.|||

Also check that the inputs of the Merge Join are sorted...

Rafael Salas

|||

Thanks Jay,

My columns are characters type and my oledb source 1 columns is lowercase and oledb source 2 columns is uppercase.

I created a Derived Column replaced my column to uppercase. :)

|||Character Map Transform might be more appropriate to change case|||

Rafael Salas wrote:

Also check that the inputs of the Merge Join are sorted...

Rafael Salas

Very good point. And note that setting IsSorted=1 on an output does NOT sort the data.

-Jamie

Merge always deletes inserts on subscriber

Hey.
Table1 replicates to table3
Table2, just another table with the same schema as Table1 and 3.
In my setup, I created a copy table, copied 66 rows from table1 on publisher
to table2 on subscriber and deleted all 66 rows. The changes were replicated
to the subscriber to Table2(66 deletes made). Now i insert all 66 rows from
Table2 to table3 on the subscriber side. The change should have gone to
publisher. Instead, the inserts get deleted on the subscriber... Why would
SQL do this?
The inserts done later should be taken as new changes on subscriber and
should be replicated to publisher, isn't it?
Forgot to add, it took it as a conflict between pub and sub and for me, by
default, the pub wins... The problem with this is though, it shouldn't be a
conflict. I've ADDED these rows right now and ONLY to the subscriber...
Any clues.