[ease-dev] Jupyter Kernel Result Adapters

[Martin Kloesch <martin.kloesch@xxxxxxxxx>]

Hi,

I got the mentioned adapters working today (thanks Christian for the 
activation tip).
To have a quick test I implemented a dummy adapter that casts Strings to 
their IJupyterPublishable representation. This is working nicely and 
shows that the general flow for adapters is working as well.

Since the fallback mechanism for casting result objects uses Strings 
anyway, I do not want to actually check in this "unused" code.
As I still want to show how to implement these adapters, I am not quite 
sure how to proceed. I can either create a reference implementation for 
something you would like to have, still check in the dummy 
implementation or write a tutorial somewhere. In the end it would be 
cool if module developers would provide adapters for their custom result 
classes that they can be integrated nicely in our Jupyter framework. If 
we can come up with an easy entry point for devs I guess that more of 
you might be inclined to use this... In the end it is very easy to write 
the adapters, but there are a couple of pitfalls that might surprise 
some of you.

What do you suggest or prefer?

Best regards,
Martin