swarmauri_evaluator_externalimports 0.3.0.dev46

Evaluator that detects and penalizes non-built-in Python imports in source files

---

Swarmauri Evaluator External Imports

Evaluator that detects and penalizes non–standard-library imports in Python source files.

Purpose

ExternalImportsEvaluator gauges dependency hygiene by walking the Python sources referenced by a swarmauri_core.programs.IProgram and flagging modules that are not part of the Python standard library.

What it does

1. Recursively walks every .py file beneath program.path.
2. Parses each file's abstract syntax tree to collect import and from ... import ... statements.
3. Compares imported modules against the Python standard library (including built-ins and a curated set of commonly used packages).
4. Returns both a score and metadata describing any external dependencies that were found.

How scoring works

• The evaluator starts at a perfect score of 1.0.
• Each unique external module subtracts 0.1 from the score.
• The score never drops below 0.0 regardless of how many modules are flagged.

Metadata provided

evaluate returns the score alongside a dictionary with:

• total_imports: Count of all import statements encountered.
• external_imports_count: Number of non-standard imports detected.
• unique_external_modules: How many distinct external modules were used.
• external_modules: A list of the external module names.
• external_imports_details: Rich per-import entries containing the module, alias, line number, import type, and whether the module was considered standard.
• files_analyzed and python_files: Insights into the scan coverage.
• execution_time: Added by EvaluatorBase to show runtime of the analysis.

Installation

Pick the workflow that matches your toolchain:

# pip
pip install swarmauri_evaluator_externalimports

# Poetry
poetry add swarmauri_evaluator_externalimports

# uv (install the tool if you have not already)
curl -Ls https://astral.sh/uv/install.sh | sh
uv add swarmauri_evaluator_externalimports

Example

The snippet below creates a temporary project, evaluates it, and prints the results. You can drop it into a file such as example.py and run python example.py.

from pathlib import Path
import tempfile
from typing import Any

from swarmauri_core.programs.IProgram import IProgram
from swarmauri_evaluator_externalimports import ExternalImportsEvaluator


class DirectoryProgram(IProgram):
    """Minimal IProgram implementation backed by a filesystem path."""

    def __init__(self, path: str):
        self.path = path

    def diff(self, other: "IProgram"):
        raise NotImplementedError("Diffing is outside the scope of this example")

    def apply_diff(self, diff: Any) -> "IProgram":
        raise NotImplementedError("Diffing is outside the scope of this example")

    def validate(self) -> bool:
        return True

    def clone(self) -> "IProgram":
        return DirectoryProgram(self.path)


def evaluate_example() -> tuple[float, dict[str, Any]]:
    evaluator = ExternalImportsEvaluator()

    with tempfile.TemporaryDirectory() as workdir:
        project = Path(workdir)
        project.joinpath("app.py").write_text(
            """import os\nimport numpy\n\nprint(os.name, numpy.__version__)\n""",
            encoding="utf-8",
        )

        program = DirectoryProgram(workdir)
        score, details = evaluator.evaluate(program)

        print("Score:", score)
        print("External modules:", sorted(details["external_modules"]))
        return score, details


if __name__ == "__main__":
    evaluate_example()

Want to help?

If you want to contribute to swarmauri-sdk, read up on our guidelines for contributing.