You mean to manipulate the lattice files in python?

I am not completely sure of the scope of this repository. I think the main purpose of this repo should be to specify/define the latticeJSON format and it should be independent of Python.

Even though the latticeJSON tool to validate/convert the lattice files is written in Python, I think this should only be an implementation detail (and no concern to the user).

The functionality to load lattice files into Python, manipulate them and save them is already implemented in apace.

The way is see it there are two options:

1. We move the code of apace which can load/save/manipulate latticeJSON files to a seperate repo (e.g. named pylatticejson). Then apace and your tools would be user of this separate library.

2. I have to design apace in such a way that it can be used as library by your and other tools.

A concern I have about the first approach is, that is it maybe a bit over-engineered and could be confusing to a user which has to import different libraries.

I would prefer the second approach, as I think it not so easy to complete separate the lattice-logic and from the rest of the code of apace. For example the signal system or the print_tree function. Would they belong to the pylatticejson library or to my apace code?

But if you have a strong opinion that there should be a separate python library, we can do this. But we have to discuss what belongs in there.

Ok, sorry read issue #3 first. I'm not sure, the idea of a pylatticejson repo does not sound to bad to me. If we want to keep all the packages useful for other users, pylatticejson might not be a bad idea, maybe not every user needs the full apace functionalities? I think some general tools would go well in such a repo (read, insert, upsert, convert,...). Also how would you drop the Python dependence for the validation?