create/read variables to the ctrlX Data Layer with Python

[yannic]

EDIT by CodeShepherd: Post was moved to own topic from this thread.

Hey CodeShepherd,

thanks for your solution! But how does it work with the python wheel in the SDK you´ve said?

Greetings 

Yannic

[CodeShepherd · Accepted solution]

You will have to build your own app and install it on the ctrlX CORE. This app then can be provider and client of data. Please have a look in the documentation of the SDK for ctrlX AUTOMATION. It will be available on GitHub soon.

[aFouraker]

@CodeShepherd, is it possible to load a pacake into a python instance without building it into a snap? If a root user installs a package is it availble to all python instances on the core? If a snap is created that has the required packages and installs them for the snap to run, are those packages now available for other scripts?

I am imagining an app that only serves to load packages for my python instances I run from the core independent of an app. 

[CodeShepherd]

The best way would always be to pack everything together in one app. If you have more than one application that use the same packages also this is the way to go. You can add several in one single app.

You cannot extend the standard python interpreter that is part of the ctrlX CORE - Python Runtime App. Also there is no mechanism to install a package globally on the ctrlX CORE.

There are two cases where you create dependencies that would dissent to the idea of the snap technology. (Snaps are app packages for desktop, cloud and IoT that are easy to install, secure, cross-platform and dependency-free.):

• Like mentioned in this thread, you could unpack a wheel and load the folders to the ctrlX CORE where an instance can access it.
• When you pack your own app you can open an interface (e.g. via plug and slot mechanism) where you can then pack several packages into several apps to provide the functionality (worst architecture possible).