cancel
Showing results for 
Search instead for 
Did you mean: 

Source Table

Former Member
0 Kudos

Hello everyone,

I have 10 source tables but the operation i am going to do with that is common. I dont want to have 10 different transforms to perform the same action.

Is there any way i can pass the table name over run time ? so that all my work will be done in one go.

Any leads ?

Thanks,

AJ.

Accepted Solutions (0)

Answers (5)

Answers (5)

Former Member
0 Kudos

Hi Arun,

One  way to do this is to use an SQL transform as the source and specify the table name as a variable in the SQL string.  For each iteration of the dataflow - parse the table name to be used in the SQL string.

Hope this helps,

Ian.

Former Member
0 Kudos

Arun,

As already suggested, either do a union all in the database or use the Merge transform in Data Services so that you are using a single query transform.

If you have to populate each of these ten tables in 10 separate target tables, you can use a

the table name as an identifier (Include an extra column with your table name?) in your where clause in a query transform after the merge transform.You may still need 10 query transforms but it will save you the hassle of writing complex transforms which you will be doing in your query transform.

10 Sources > Merge > Single Query transform with complex mappings > 10 query xfms with table name in where clause ( If you have separate targets)

Cheers,

Chaitanya

Former Member
0 Kudos

Hi Arun,

if the source tables all have the same columns, you can merge them using the MERGE Transformation and than connect the MERGE Transformation to a Query (where you can manipulate the data)

Best regards

Goran

former_member187605
Active Contributor
0 Kudos

Create a union_all view in your database and use that view as source in a single data flow.

severin_thelen
Contributor
0 Kudos

Hello Joseph,

but every source table has a different target table? Then you have to create the query 10 times.

If all tables have the same target table you could merge the tables (merge transform) and after that you just need one query.

Regards

Severin