Hi All, I am trying to join two output tMap flows and haven't succeeded yet. I wanted to know if this is possible or may be I should redesign my job flow. This is my use case: In the job flow below, the tNormalize(tNormalize_2) normalizes a field and then perform a lookup on the unique normalized values. I have some other computation in the tMap (tMap_4) in case i dont find a match in my lookup. Once this lookup is completed I need to join back to the tNormalized output (tNormalize_2) so that I can output the lookup value got earlier and the remaining fields at the tNormalize. For this purpose I put a tMap (tMap_5) basically to create two similar flows - one which does the lookup and the other that could be used to join after the lookup (ie after tMap_4). This is where I am stuck as I am not able to join two tMap flows. I tried using both tMap and tJoin to join them but it doesn't work. Here's an example of what I am trying to achieve: tNormalize_2 output: 1 fam 2 ffn 7 bbz 7 fam This is replicated into two flows by tMap_5 - One goes to tUniqRow_2 tUniqRow_2 acts on the second field to generate: fam ffn bbz tMap_4 does a lookup from the db and generates a default if the lookup fails such as: fam 1 ffn 2 (default value) bbz 3 This has to join to the other output flow of tMap_5 so that I can finally output: How can this be done? 1 1 2 2 7 3 7 1 I really appreciate any help here. Sid
You cannot split flows and then rejoin them later; search the forum for "cycle" for many many posts on this. You have to instead store the data temporarily to a table, file or hash to re-read in a subsequent subjob. Having said that, why do you need to tUnique in the first place? Why can't you just do the lookup in tMap5?