I thought feret_diameter_max might make more sense than max_feret_diameter, because that way all potential Feret diameters would be grouped, in alphabetic property lists (e.g. documentation).

@maxfrei750 It looks like some tests are failing. I’m on my phone so I can’t investigate right now. The changes look ok to me.

@scikit-image/core I’m agnostic on max_feret vs feret_max. max- sounds better but I agree with @maxfrei750 that the grouping with -max/-min is a nice idea.

On the other hand, we have max-intensity, min-intensity, and mean-intensity...

Any thoughts?

@jni I already saw the errors, however I don't understand what they are caused by. There are loads of key errors like the following one in the "collection" section:
ERROR feature/tests/test_util.py - KeyError: 'feret_diameter_max'

Concerning max_feret vs feret_max: I agree that a consistency with min-intensity, etc. would be nice. However, I still like the idea of similar properties being grouped. Especially, since the Feret diameters would be mixed with the intensities, if both were to begin with min, mean, max. Therefore, IMHO it would be good to rename the intensities as well. Of course, breaking the API is quite drastic, so maybe for now, we could go with max_feret and change it in 1.0, along with the intensities.

The docstring in regionprops needs to be updated as well. At the moment it still states

    **feret_diameter** : float
        Maximum Feret's diameter computed as the longest distance between

which doesn't match the update name. There is some code in _parse_docs that parses the regionprops docstring and might be the source of the errors you are seeing.

@grlee77 Thanks for the hint. Changing the docstring has fixed the problems.

I'd like to finalize this PR. Any final thoughts concerning the naming?

I like the way it is now---it is more easily discoverable.

Looks like we have a consensus!!! Thank you @maxfrei750!