Consider a situation where you have 2 receive locations, each receiving it's own unique message type. There is an orchestration with a parallel correlation happening based on a shared unique value in each of these messages.
Once a correlation set occurs, the orchestration runs and the job of it is to merge data from the 2 messages and create 1 from them. My idea was to use a map which takes in 2 input messages: 1 of each type in the correlation. The destination schema just happens to be the same as one of the input schemas (so we're basically just adding data to one of them from the other)
I can create the map, choose the 2 input message schema and the destination schema. The mapper than opens up and on the source side looks like so:
Which is quite alright.
The problem comes when you start expanding the nodes, they only seem to go 1 level deep. For example, here is the source and destination side-by-side, the same schema, except one is Part 1 of a 2 part source and the other is the single destination part:
This is just one example, but compare EVN_5. On the left it doesn't have children, on the right it does. It's the same schema, but one is part of a multi-input source and the other is the destination.
Is there any way to fix this, or is it not possible? Doing a link by name/structure results in missing data because the source "thinks" it's not there.
Edit: I just wanted to add the detail that this problem of only showing one-level deep worth of elements in the mapper is happening for both of the input schemas.
