I created a Jupiter notebook that processes neuron morphologies (from the neuroinformatics platform), and saves them to a zipfile. I copied some code from the ‘Export software catalog’ notebook example to make the zip-file available as a data-url. That works, but I am afraid that for multi-MB results this will fail. Is there another way to give the user download access to tempfiles created in python? I could use the document service, but am not sure if it is appropriate to use that as a caching system.
The notebook is this one:
https://collab.humanbrainproject.eu/#/collab/514/nav/4965