Description

This refactors Producer to operate primarily against a Type rather than requiring a strict Artifact subclass (with a nested Type). The Type is derived from the parameter type annotations, by one of the following mechanisms:

• inferring from the type hint (via python_type_system) (int)
• using a Type instance embedded in an Annotated type hint (Annotated[int, Int64()])
• using a Artifact class with a type default set embedded in an Annotated type hint (Annotated[int, MyArtifact])

Relaxing the requirement for Artifact subclass improves simple cases of Producer and Graph creation significantly and allows direct Artifact instantiation. As requested in #260, this simple example now works as expected:

from arti import Graph, producer

@producer()
def increment(numbers: list[int], by: int) -> list[int]:
    return [n + by for n in numbers]

with Graph(name="Demo") as g:
    g.artifacts.step, g.artifacts.numbers = 5, [1, 2, 3]
    g.artifacts.incremented = increment(numbers=g.artifacts.numbers, by=g.artifacts.step)

if __name__ == "__main__":
    g.build()
    print(g.read(g.artifacts.numbers, annotation=list[int]))
    print(g.read(g.artifacts.incremented, annotation=list[int]))

, yielding:

[1, 2, 3]
[6, 7, 8]

Resolves #260 and supersedes #259 (pickling now works fine w/o the dynamic classes).

Type of change

• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
  • Artifact.for_literal and Artifact.from_type have been removed
    • these were likely only used internally (via Artifact.cast).
  • Most of the pseudo-private Producer._*_ class vars have been replaced

How Has This Been Tested?

Tests have been updated and added to maintain coverage.

Checklist:

• My code follows the style guidelines of this project
• I have performed a self-review of my own code
• I have commented my code, particularly in hard-to-understand areas
• I have made corresponding changes to the documentation
• My changes generate no new warnings
• I have added tests that prove my fix is effective or that my feature works
• New and existing unit tests pass locally with my changes
• Any dependent changes have been merged and published in upstream modules