Skip to main content

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
[basyx-dev] Issue with Python SDK and Java Server interoperability

Dear Basyx-Dev Team,

I'm a student researcher at RWTH Aachen University working on an Industry4.0 demonstrator using your backend.
Our team have been struggling to make some things work as hoped, and I was hoping for some Input from your side.

Our overarching goal is to use the Basyx Java Server as our main backend while programming with the Python SDK for our own features and data transformations.

We are rather early in our implementations, but have run into some issues.

  1. The Python SDK's AASToJsonEncoder and AASFromJsonDecoder use very different "standards" for AAS files compared to the Java Server
    This leads to being unable to view AAS in the Viewer when exported from the Python backend, and vice versa them being unreadable to the python decoder when uploaded through the UI
    Why is that? Is there a way to instate a semblance of interoperability?
    (Some fiddling actually let me import Files as uploaded from the UI by modifying the SNAKE_CASE to CamelCase at a few key points and adding a small handful of fields that the Server doesn't seem to save. But in reverse, some critical data seems to get lost, and overall this feels incredibly hacky.)

  2. No apparent way to detect AAS that were added to the MongoBackend
    I assume this is 'intended', but would there be a way for the AAS-ENV to detect new AAS added to it's repo from the backend?
    Would, assuming Issue 1 is worked out, a PUT/POST to the ENV's REST interface solve this issue? (As in, POST a placeholder AAS with the ID to the REST interface, then filling out the data and performing further modifications directly on the MongoDB data)


Thank you very much for your time and consideration.

Kind Regards,
Frederik Oelert
ITA Institute for Textile Applications, RWTH Aachen University


Back to the top