Fixes #1419: ensure correct run status on failure for MLflow, patch pre-execution NoneType crash

[dedyoc]

Fixes #1419

This PR resolves an issue where the MLflow plugin incorrectly reported failed Hamilton DAG as successful ("FINISHED"). Additionally, it patches a bug that caused the plugin to crash when no config or inputs were provided.

Changes

• Updated run_after_graph_execution in h_mlflow.py to explicitly pass the correct run status ("FINISHED" or "FAILED") to mlflow.end_run() as mlflow.end_run() defaulted to success.
• Added a fallback to the dictionary iteration in run_before_graph_execution to prevent an AttributeError when self.config or inputs evaluate to None.

How I tested this

1. Created a DAG using with MLFlowTracker adapter.
2. Intentionally raised a ValueError inside a node to simulate a crash.

import pandas as pd

def signups() -> pd.Series:
    return pd.Series([1, 10, 50, 100])

def spend() -> pd.Series:
    return pd.Series([10, 10, 20, 40])

def acquisition_cost(signups: pd.Series, spend: pd.Series) -> pd.Series:
    raise ValueError("intentional crash!!")
    return spend / signups

3. Verified via mlflow ui that the run correctly registered as FAILED.

5. Verified that running the DAG without external inputs no longer crashes the pre-execution hook.

Notes

• I detected the pre-execution crash while reproducing the issue with dummy DAGs that has no inputs.

Checklist

• PR has an informative and human-readable title (this will be pulled into the release notes)
• Changes are limited to a single goal (no scope creep)
• Code passed the pre-commit check & code is left cleaner/nicer than when first encountered.
• Any change in functionality is tested
• New functions are documented (with a description, list of inputs, and expected output)
• Placeholder code is flagged / future TODOs are captured in comments
• Project documentation has been updated if adding/changing functionality.

[skrawcz]

Great thanks. Is there a unit and/or integration test we could write for this?

[dedyoc]

I checked the tests/plugins directory and noticed there isn't an existing test folder for h_mlflow yet.
I'll go ahead and create one for it.

[skrawcz]

Test failed:
_______________ ERROR collecting tests/plugins/test_h_mlflow.py ________________
ImportError while importing test module '/home/runner/work/hamilton/hamilton/tests/plugins/test_h_mlflow.py'.
Hint: make sure your test modules/packages have valid Python names.
Traceback:
/opt/hostedtoolcache/Python/3.14.3/x64/lib/python3.14/importlib/init.py:88: in import_module
return _bootstrap._gcd_import(name[level:], package, level)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
tests/plugins/test_h_mlflow.py:21: in
from hamilton.plugins.h_mlflow import MLFlowTracker
hamilton/plugins/h_mlflow.py:23: in
import mlflow
E ModuleNotFoundError: No module named 'mlflow'

[dedyoc]

I have added pytest.importorskip("mlflow") since mlflow isn't installed for Python 3.14

[dedyoc]

Hi @skrawcz ,

I've verified that the tests are now passing successfully on my fork's Action

Could you please take another look when you have a moment?