Add artifact1/tests/test_core.py

artifact1/tests/test_core.py

@@ -0,0 +1,58 @@
+import pytest
+import pandas as pd
+from pathlib import Path
+import json
+
+import src.artifact1.core as core
+
+@pytest.fixture
+def sample_data():
+    return {
+        'runs': [
+            {'max_only_alerts': 3, 'outlier_frequency': 0.05, 'expires_at_dist_hours': 2.5, 'retry_total_overhead': 120.0},
+            {'max_only_alerts': 4, 'outlier_frequency': 0.07, 'expires_at_dist_hours': 1.8, 'retry_total_overhead': 98.0},
+            {'max_only_alerts': 2, 'outlier_frequency': 0.03, 'expires_at_dist_hours': 2.9, 'retry_total_overhead': 130.0}
+        ]
+    }
+
+
+def test_analyze_metrics_returns_dict(sample_data):
+    result = core.analyze_metrics(sample_data)
+    assert isinstance(result, dict), "The result should be a dictionary"
+    expected_keys = {'max_only_alerts', 'outlier_frequency', 'expires_at_dist_hours', 'retry_total_overhead'}
+    assert expected_keys.issubset(result.keys()), "All expected metrics should be present"
+
+
+def test_analyze_metrics_values_reasonable(sample_data):
+    result = core.analyze_metrics(sample_data)
+    assert result['max_only_alerts']['mean'] == pytest.approx(3.0, rel=1e-2)
+    assert result['outlier_frequency']['mean'] == pytest.approx(0.05, rel=1e-2)
+    assert result['expires_at_dist_hours']['mean'] == pytest.approx(2.4, rel=1e-2)
+    assert result['retry_total_overhead']['mean'] == pytest.approx(116.0, rel=1e-2)
+
+
+def test_analyze_metrics_handles_invalid_input():
+    with pytest.raises((KeyError, TypeError, ValueError)):
+        core.analyze_metrics({'invalid': 123})
+
+
+def test_analyze_metrics_empty_data():
+    with pytest.raises(ValueError):
+        core.analyze_metrics({'runs': []})
+
+
+def test_analyze_metrics_stat_consistency(sample_data):
+    result = core.analyze_metrics(sample_data)
+    for metric, values in result.items():
+        assert 'mean' in values and 'std' in values, f"Metric {metric} should have mean and std"
+        assert isinstance(values['mean'], (int, float))
+        assert not pd.isna(values['mean']), f"Mean should not be NaN for {metric}"
+
+
+def test_analyze_metrics_outlier_logic(sample_data):
+    result = core.analyze_metrics(sample_data)
+    assert 'anomalies' in result, "Result should contain an 'anomalies' section"
+    anomalies = result['anomalies']
+    assert isinstance(anomalies, dict), "Anomalies should be represented as dict"
+    for key, value in anomalies.items():
+        assert isinstance(value, bool), f"Anomaly flag for {key} should be boolean"