These seem to be conflicting statements. Which document type are you expecting TN to see? I assume EDIMetaEnvelope.
For that doc type, how are the attribute extractions for sender, receiver, doc ID, etc. defined? For the sender and receiver, which external ID are they intended to match (for XML, they can only match one type, unless you’re doing a custom transformation)?
How are the external IDs defined for each sender and receiver? When you map the sender/receiver IDs from the EDI string to the XML fields, are you accounting for the qualifiers?
It seems like there may be a mixing of EDI-oriented configuration and XML configuration, where EDI identifiers are being used in the doc and defined in the profiles, but the doc recognition step assumes a consistent external ID type.
What sort of issues? It seems like you’re going through a lot of work to make EDI function like XML.
“Better” is always a subjective thing and any qualitative evaluation needs to consider all the info available–how’s that for a disclaimer!
. Given the limited info I have about your situation it’s risky to make a recommendation but I’ll offer one anyway–you should try to treat EDI as EDI, not XML. There are a lot of things that WmEDI and WmEDIforTN give you out of the box and if you avoid using them, then you’ll either forego those benefits or have to reimplement them yourself.
Perhaps this thread can turn toward addressing the issues you faced when submitting the EDI string directly to TN?
#webMethods#B2B-Integration#Integration-Server-and-ESB