Reputation: 65183
I have Data Flow Task A
which merges a bunch of tables together.
I want to use the output from Data Flow Task A
in Data Flow Task B
to continue merging tables.
How do I do this?
The reason the tasks in Data Flow Task A
are not in Data Flow Task B
, is because Data Flow Task A
is also going to provide data for Data Flow Task C
Upvotes: 1
Views: 1967
Reputation: 61269
You don't. At least, not directly.
At our disposal we have 3 output classifications: table, file and a local in-memory recordset object.
Since you want to use the rows as a source, we can already throw out the recordset because while we have a "Recordset Destination", lamentably we have no "Recordset Source"
That leaves us with tables versus files. This is dealer's choice. I favor tables because that's usually easier if I have to debug something but writing to a file works just fine. If you do write to a file, the Raw file option will provide better performance over a flat file as it will be in a binary format which eliminates the processing overhead of parsing and interpreting the data against locale rules.
I guess a final option could be to add a MultiCast into your data flow A and shoehorn the logic for B and C into it. The challenge with this approach is that it quickly becomes painful to edit the data flow due to dependencies between componentns
Upvotes: 2