DSXchange

I have one Binary file that contains multiple layout coming from mainframe. First few colums in all layout are same, rest are not. We can differentiate between different layouts from 'TXN_Type' colum.

'TXN_Type' is TXN## (FIELD 3) in following example.
All 20s,40s,50s have same Layout respectively.

sample file

FIELD 1, FIELD 2, TXN20, FIELD 4, FIELD 5
FIELD 1, FIELD 2, TXN21, FIELD 4, FIELD 5
FIELD 1, FIELD 2, TXN22, FIELD 4, FIELD 5
FIELD 1, FIELD 2, TXN40, FIELD 4
FIELD 1, FIELD 2, TXN42, FIELD 4
FIELD 1, FIELD 2, TXN50, FIELD 4, FIELD 5, FIELD 6
FIELD 1, FIELD 2, TXN52, FIELD 4, FIELD 5, FIELD 6
FIELD 1, FIELD 2, TXN51, FIELD 4, FIELD 5, FIELD 6

Options I have (as far as my knowledge goes):

1) I can split the file on Mainframe into seperate file based on Transaction Type 20s and 40s... and copy one by one to the DS SE Server. VERDICT: Not desriable since volume is under 25k

2) I prefer to copy the whole file to DS SE and then use DS split Job to split the file based on TXN_Type. Since 'ID Field' 'Value' under 'Selection Criteria' tab of CFF stage only pick one layout not all 20s and so on. VERDICT: best way to move forward.

In EE we can read EBCIDIC file in Sequential file stage, read only one colum e.g. TXN_TYPE, and leave the rest as is and split the file but please let me know how can I do this in SE.

I have tried searching but EE solutions came up...feel free to show me if you found one... thanks in advance...

I haven't used the CFF in Server but having a quick peek at the stage, I would say you need to create an output for each of your types, you then define the file layout for each of these types in their respective output tabs

Kryt0n wrote:I haven't used the CFF in Server but having a quick peek at the stage, I would say you need to create an output for each of your types, you then define the file layout for each of these types in their respective output tabs

So your suggestion is to create an output for each of types from Mainframe? unless someone knows how to split it on DS that seems to be the only option to me...

No, I meant the CFF file can do the split based on the field you specify but you have to have an output link for each of these types

CFF are best bet for multi record format . or you can use sequential file stage as varchar and after that filter or transformer to split into different files ( all records of same format into one file )

nagarjuna wrote:CFF are best bet for multi record format . or you can use sequential file stage as varchar and after that filter or transformer to split into different files ( all records of same format into one file )

Thanks to all for reply. What i did not know that CFF can directly split files into multiple files and can have diferent layouts for each output, and offcourse I have the option to filter specific value to these output links in Selection Criteria... Thanks again...