Ephys

[pauladkisson]

This PR adds an OpenEphysRecordingInterface and PhySortingInterface tested on a single example session.

[h-mayorquin]

Everything here looks good to me.

As a comment that might be useful:

Lately I have started working using a converter pipe instead of the NWBConverter object. I think that some of the things that you do here like manipulating the properties of the recording object are easier to write and understand with that workflow. Plus, if something is wrong you know earlier. Example:

https://github.com/catalystneuro/dicarlo-lab-to-nwb/blob/39656bcd239a405362596c9693eb03cc3ccf8f04/src/dicarlo_lab_to_nwb/conversion/convert_session.py#L66-L74

[pauladkisson]

Lately I have started working using a converter pipe instead of the NWBConverter object. I think that some of the things that you do here like manipulating the properties of the recording object are easier to write and understand with that workflow. Plus, if something is wrong you know earlier.

Interesting, I didn't even know this kind of alternative workflow was possible. Before, if I was making too many changes to an interface I would usually just define a child Interface, but using a ConverterPipe could be a good alternative for some cases.

I think in this case, I'll just stick with the NWBConverter, at least for now, but thank you for the heads up.

As a side note, if we want to normalize using ConverterPipes instead of NWBConverters (in some cases ofc), we should really have some documentation on them. I couldn't find anything about them on the docs

[h-mayorquin]

I think in this case, I'll just stick with the NWBConverter, at least for now, but thank you for the heads up.

I think this is wise, probably too much friction to change in the middle of an ongoing project.