[mretegan]

The PR updates the script to write a NeXus file following the current definition proposed here nexusformat/definitions#1640.

Since things are still changing in the definitions, keeping all the converted file updated is tricky, so for the moment only I focused on the 3 Fe XDI files.

@newville Could you have a closer look and let me know what you think?

[newville]

This looks mostly okay to me.

Would`root["definition"] = "NXxas" allowed? If not, why not? That is, it's OK to have subclasses, but are they required? That does sort of put an extra burden on defining the data, and on using it.

also: What happens here is mono crystall and/or d_spacing cannot be determined from the "mono" attributes? A reasonable default might be helpful?

[mretegan]

Having root["definition"] = "NXxas is allowed, of course, but in this case we know it is transmission data; it makes sense to use the NXxas_trans subclass.

I am not sure that we should have defaults for things that are not in the parent file. As you said many times, this could be used for reprocessing, and if the default is not appropriate for that specific case, it might cause more harm than good.

[newville]

@mretegan thanks -- I agree with all of that... just want to make sure that "NXxas" is not invalid...