Split one message and create N Files on target side based on FieldName

[Former Member]

Hi Experts

How to split one message into N messages

I have used BPM and I have put

One Receiver Step

One Transformation Step and

One Send Step.

In Receiver step I have used Correlation as Field2

In Tansformation Step I have done One to One Mapping

But In receiver Side only one File is creating.

Message Structure is

<Main_MT>.... 1...1

........<test>01</test>.... 1...1

...........<Sub_MT>...... 0...Unbound

................<Field1> </Field1>

................<Field2>123</Field>

..........</Sub_MT>

..........<Sub_MT>...... 0...Unbound

................<Field1> </Field1>

................<Field2>234</Field>

</Sub_MT>

</Main_MT>

How to resolve this problem

Thanks & Regards

Sowmya

[Former Member]

You don't need BPM.

In ID, create one receiver determination and multiple interface determinations, choose a interface mapping to map the source message to the corresponding message (interface). if target file name is based on a field contained in the message, then you might define UDF to set the target file name and use it in the receiver CC.

BTW, Pooja, I think multiple mapping (search in SDN using keywork multiple mapping) won't work, as this will still save splitted message into one single target file.

Jayson

[Former Member]

Hey Jayson,

I dint search the blogs using the keyword " mutiple mapping". The given two blogs I had used for creating my first 1:n mapping, so their names are by heart to me.

I believe both the blogs provided by me help in creating the mutiple seperate messages on the target end.

The Claus's blog covers more detailed mapping logic for message split and the Jin's blog covers the ID pieces needed to be configured when we want to create mutiple messages on the target end.

Pardon me if I have misconstrued your point.

Thanks,

Pooja

[Former Member]

Hi Pooja,

sorry, my mistake, you are right, the multiple mapping would work better. And it's easy to do.

Jayson

[Former Member]

Also check out the blog : /people/claus.wallacher/blog/2006/06/29/message-splitting-using-the-graphical-mapping-tool

for more detailed message split mapping.

Thanks,

Pooja

[prateek]

Many flaws.

1. You said you have to split single file. This would mean that only one file is to be received by BPM. In this case, correlation is not required.

2. In case you need to split the file, why don't you split it in transform step? Where are you actually splitting the file.

3. I guess you misunderstood the use of correlation. It is Not used to split the message based on fieldname. Rather it is used to merge various incoming files based on some common fieldnames in the file.

Regards,

Prateek

[Former Member]

Hi Sowmya,

First of all you dont have to use BPM for 1:n mapping.

Check out this blog for multimapping without BPM L /people/jin.shin/blog/2006/02/07/multi-mapping-without-bpm--yes-it146s-possible

Thanks,

Pooja