Originally posted by: SystemAdmin

Hi,
There are multiple input arguments being passed to a function map. Depending on the input data being processed, some of the function map arguments may not contain data, when the function map argument does not contain data the map does not execute any of the rules for this map. The workaround would be to call different maps based on whether certain data is present or not, but this doesn’t seems to be a scalable solution especially when dealing with complex data such as EDI formats.
Any suggestions.

Thanks

Mehedi
#DataExchange
#IBM-Websphere-Transformation-Extender
#IBMSterlingTransformationExtender

Originally posted by: repanzer

Anytime you have different data to pass to the output, you have to have a new functional map. What I do is use the EITHER command;

In the event input2 is not present;

=EITHER(f_map1(input1, input2), fmap2(input1)
But both maps needs to be created.

The only possible solution I can think of for your dilemma, is to use EITHER on the field level, if that's possible. I have not tested this nor had a need to.

Like:

=f_Map(input1, EITHER(input2,"placeholder"))

If it allows EITHER in the fmap statement, the challenge will be, that input2 and "placeholder"...HAS TO BE OF THE SAME TYPE. In this case, text. As far as I know, you can not do on the fly adjustment to the types in functional maps.

I know there's a function REFORMAT, but I've never used it (nor understand it).
#IBMSterlingTransformationExtender
#IBM-Websphere-Transformation-Extender
#DataExchange

Originally posted by: TarunB

Rep, I've used the =f_Map(input1, EITHER(input2,"placeholder")) design before and indeed it works just like you've suggested.

Reformat is useful when you have an input and output tree which is largely of same format but with a different TT filename or group name. It avoids the need for field-level mapping, allowing instead a REFORMAT call at the group level.
#DataExchange
#IBMSterlingTransformationExtender
#IBM-Websphere-Transformation-Extender

Originally posted by: repanzer

"Reformat is useful when you have an input and output tree which is largely of same format but with a different TT filename or group name. It avoids the need for field-level mapping, allowing instead a REFORMAT call at the group level."

If I had only known....THANK YOU...AGAIN!!
#IBMSterlingTransformationExtender
#DataExchange
#IBM-Websphere-Transformation-Extender

Originally posted by: Daniel_MK

try

f_map(Input1alwaysexists,either(input2mayexist," ")
#IBMSterlingTransformationExtender
#DataExchange
#IBM-Websphere-Transformation-Extender