Expansion of project

[clbarnes]

Would you consider a PR to expand this into more of a "regular" python package? Potential avenues would be

• [ ] requirements.txt (or conda environment.yml) for consistent development environment
  • e.g. linters which catch certain classes of bug before runtime, and improve readability
• [ ] setup.py for consistent deployment environment (could be linked to requirements.txt using extreqs)
  • Means that installation is a single command, rather than manually installing packages listed in README.md and then doing manual PYTHON_PATH manipulations to get the package in scope
• [ ] Split FIBSEM_SIFT_gs.py into modules
  • easier to navigate and discover functionality, as well as extend with e.g. N5/zarr support, rather than a single 10,000 line file
  • internal helper functions don't pollute the export namespace so easier for users
  • Import paths (e.g. for dependent scripts) can be kept identical using __init__.__all__
• [ ] automated docs using pdoc3 or sphinx
  • means that the full context of the docstring (rather than the digest currently in the readme) can be read and linked to, without any need to manually update the readme

[gleb-shtengel]

All good points and I would be interested. But, there are few issues with this:

• Unfortunately I am not a good python programmer (as I am sure you have already figured out), so this would be slow. But I am happy to learn.
• This is not my highest priority project at the moment. The code does work and does what it was intended to do. I am not sure how much need there is in it outside Janelia and perhaps one or two other labs.

[clbarnes]

Most of that is pretty small work I'd be happy to do myself and raise a PR. I am in one of the other handful of labs using these microscopes and am keen to see how much of the pipeline could benefit us as well - another benefit of modularising it would be that we could mix and match a bit more rather than treat it as a monolithic black box.