How can I join two flows into one? (Unite does not work)
Hi,
I found several times that a flow must branch to follow different processing but after a few components they are ready to join again.
Currently I handle this situation generating several files and then processing. But, Is there any better way to joint to flows?
I attach a screenshot.
Thank you
I was able to make something like this work for a reconciliation program. Not sure its exactly what you're looking for....
Ignore the names and look at the flows. The tmap first outputs my main/rejected outputs, they each go to an output (I only kept one of them to keep file size small), I sorted/filtered them and then I output them to a temporary csv file.
Then I input those csv files and was able to join them however I wanted. Please note that the second image is a subjob of the first image.
Talend does not allow a "closed circuit". You can unite separte sources into one flow but cannot - start,split,unite- again. In the OP's picture I would use a second tFileOutputDelimited in append mode. Using temp files or use of child jobs and tBufferOutput then tUnite are the usual ways to accomplish these kinds of operations.
Warning:tUnite cannot exist in a data flow loop. For instance, if a data flow goes through several tMap components to generate two flows, they cannot be fed to tUnite.
This statement is from talend. I am stuck in the similar position where i am creating two flow from a tmap and then need to take the union of these 2 flows ,, i tried to use tunite but on research i got to know the above mentioned statement. Please help is there any other possible way or any other component to achieve this ????
Your response would be very much appreciated i anyone can help. Talend Team???
you're on the right road
but you have to create your 2 ouptput as a single ouptut !
Create your first output as usual , create second outpout as a 'join table' .
You will have a unique output at the end.
It's very useful, when you have to do different things on data depending on filter, values & so on , but result should be in the same flow.
your 2 differents output need to have same schema of course