Felix Musil

I am not quite sure how to label this issue but it certainly is not a bug, merely something that might hinder portability. I would keep this flag by default...

> (BTW for json-formatted reference configurations it's even easier to add a comment field. Just add a new entry labelled "comment" and all our readers should ignore it.) I agree...

I would prefer to avoid changing the fundamental of the structure that is provided. We do it for isolated structures because there it's strictly equivalent but if the user has...

> emitting a warning if the cell is non zero but PBC is false There is no reason to emit a warning in this case since defining a unit cell...

> in principle, there's no point in defining a cell if your system is non-periodic. The only reason we need a cell is an artifact of how we implement the...

Should you have a look at this small PR ? Or should it be deleted ? @ceriottm @max-veit

> I would compute and set the "fictitious cell" non destructively Would a preliminary copy of the input object be sufficient to address your comment ?

I agree that such doc is needed. My concern is the duplication of the information so I would either have a full doc on the subject and reference it in...

I am not sure to follow what you mean, could you elaborate a bit more what you have in mind ?

It is a bug in the sense that I expected the feature to be working as of now but it actually does not. We are not really using it for...