[EPIC] Short-Term Development Plan for EasyDiffraction

[AndrewSazonov]

This issue outlines the short-term strategy for evolving the EasyDiffraction library towards a cleaner, more stable, and maintainable design. It acts as a checklist of high-level development tasks and priorities. Sub-tasks should be tracked as individual issues linked here.

Goal

• Finalize and stabilize the new API
• Improve developer experience and code quality
• Prepare for integration with EasyScience shared components
• Lay the foundation for a long-term roadmap

Checklist

• Finalize the new API draft

  • Fix all related issues added to the project board

• Polish the new API design

  • Ensure clean, simple, and maintainer-friendly interfaces
  • No new features unless necessary for testing structure
  • Add:
    • PdfFit calculator as a cross-check for calculator integration
    • Single crystal diffraction as a test case for extensibility

• Freeze the API

  • Add type annotations
  • Add unit tests
  • Add complete and clear docstrings

• Set up CI quality enforcement

  • Add static analysis tools (ruff, mypy, etc.)
  • Block PRs that reduce code quality or test coverage

• Adopt EasyScience organization templates

  • Wait for finalized org-wide templates (currently in development)
  • Apply to include:
    • GitHub workflows
    • pyproject.toml
  • Continue using requirements.txt until switch is made

• Improve issue tracking

  • Create issues for missing features, bugs, and tasks

• Organize and prioritize work

  • Group issues into categories
  • Define milestones
  • Assign priority levels

• Prepare for EasyScience core class integration

  • Add strong unit test coverage
  • Add functional and performance tests
  • Ensure no performance regressions from refactor

• Create long-term roadmap

  • Based on outcomes of this short-term plan
  • Define direction for further development, growth, and community usage

Notes

• This issue will remain pinned and act as a living task board.
• Linked sub-issues will provide additional detail and progress tracking.

[rozyczko]

I think that the EasyCore/EasyCrystallography integration for

• use of Parameters and Descriptors
• use of symmetry code and cif parsers
• use of constraints

should be of the highest priority, since it will also modify the API (your Parameter is not 100% EasyScience Parameter, no CIF readers etc)

[AndrewSazonov]

I totally agree that integrating with EasyCore and EasyCrystallography (for things like Parameters, symmetry, etc.) is a high priority long term.

But I still think it makes sense to treat that as the second milestone, after we’ve wrapped up the current API stabilization work. At the moment, the API is still evolving a lot, and it’s much easier to tweak and improve things while everything is still internal to EasyDiffraction.

Once the API is stable, we’ll have a clear view of what parts should/can move into EasyScience as reusable components. That’ll make the migration smoother and more meaningful.

So, the migration is definitely on the roadmap, but I think it makes sense to get things stable first.

[AndrewSazonov]

Also, before starting the migration, I really want to have solid unit test coverage, and especially performance tests, in place. That way, we can be confident that nothing breaks or slows down during the switch.