Library

• Scans src/faebryk/library/*.py, detects whether the file contains a same-named class, and writes _F.py.
• Orders exports via a topological sort of F.<Name> references to avoid import-order cycles.

Dependants (Call Sites)

• User Code: atopile projects heavily import from faebryk.library._F (aliased as F).
• Compiler: The compiler maps ato built-ins to these classes.

How to Work With / Develop / Test

Core Concepts

• Traits vs Components: Use Traits for behavior (what it can do like can_bridge) and Components for physical things (what it is like Resistor).
• Export model: _F.py is a generated “barrel” module; importing it is intentionally convenient but can be heavyweight.

Development Workflow

1. New Component: Create a new file MyComponent.py in src/faebryk/library/. Inherit from Node (or a more specific base).
2. Naming Convention: Class names should match the file basename (usually).
3. Regenerate _F.py: run python tools/library/gen_F.py and commit the updated src/faebryk/library/_F.py.

Testing

• Library tests live under test/library/ (including test/library/nodes/).
• A good smoke test for new modules is:
  • ato dev test --llm test/library/test_instance_library_modules.py -q

Best Practices

• Atomic Parts: Mark leaf components (specifically verified part numbers) with the is_atomic_part trait.
• Parameters: Use F.Parameters to define physical properties like resistance, capacitance, etc.
• Documentation: Add docstrings to components explaining their ports and parameters.