Add rerun_evaluator/tests/test_core.py

rerun_evaluator/tests/test_core.py

@@ -0,0 +1,55 @@
+import pytest
+from rerun_evaluator import core
+
+@pytest.fixture
+def sample_runs():
+    return [
+        core.RunData(run_id="1", label_triggered=True, flappy=False, pinned=False, unknown_rate=0.1),
+        core.RunData(run_id="2", label_triggered=False, flappy=False, pinned=True, unknown_rate=0.0),
+        core.RunData(run_id="3", label_triggered=True, flappy=True, pinned=False, unknown_rate=0.3),
+        core.RunData(run_id="4", label_triggered=False, flappy=False, pinned=False, unknown_rate=0.0),
+    ]
+
+@pytest.fixture
+def empty_runs():
+    return []
+
+def test_evaluate_rerun_needs_basic(sample_runs):
+    result = core.evaluate_rerun_needs(sample_runs)
+    assert isinstance(result, dict)
+    # Expected keys
+    for key in ("total_runs", "label_trigger_rate", "flappy_rate", "avg_unknown_rate", "rerun_recommended"):
+        assert key in result
+    assert result["total_runs"] == len(sample_runs)
+    assert 0 <= result["label_trigger_rate"] <= 1
+    assert 0 <= result["flappy_rate"] <= 1
+    assert 0 <= result["avg_unknown_rate"] <= 1
+    assert isinstance(result["rerun_recommended"], bool)
+
+def test_evaluate_rerun_needs_empty(empty_runs):
+    result = core.evaluate_rerun_needs(empty_runs)
+    assert isinstance(result, dict)
+    assert result["total_runs"] == 0
+    assert result["label_trigger_rate"] == 0
+    assert result["flappy_rate"] == 0
+    assert result["avg_unknown_rate"] == 0
+    assert result["rerun_recommended"] is False
+
+def test_rundata_initialization():
+    rd = core.RunData(run_id="X99", label_triggered=True, flappy=False, pinned=True, unknown_rate=0.15)
+    assert rd.run_id == "X99"
+    assert isinstance(rd.label_triggered, bool)
+    assert isinstance(rd.flappy, bool)
+    assert isinstance(rd.pinned, bool)
+    assert isinstance(rd.unknown_rate, float)
+
+@pytest.mark.parametrize("unknown_rate,expected_recommendation", [
+    (0.0, False),
+    (0.5, True),
+    (1.0, True),
+])
+def test_evaluate_rerun_needs_unknown_rate_threshold(unknown_rate, expected_recommendation):
+    runs = [core.RunData(run_id=str(i), label_triggered=False, flappy=False, pinned=False, unknown_rate=unknown_rate) for i in range(5)]
+    result = core.evaluate_rerun_needs(runs)
+    assert isinstance(result, dict)
+    assert result["rerun_recommended"] == expected_recommendation