Add audit_analysis/tests/test_core.py

audit_analysis/tests/test_core.py

@@ -0,0 +1,77 @@
+import pytest
+import pandas as pd
+from pathlib import Path
+from audit_analysis import core
+
+
+def _create_sample_csv(tmp_path: Path) -> str:
+    data = [
+        {"id": 1, "status": "unknown", "reason": "artefact_missing"},
+        {"id": 2, "status": "unknown", "reason": "io_failure"},
+        {"id": 3, "status": "passed", "reason": ""},
+        {"id": 4, "status": "unknown", "reason": "contract_error"},
+        {"id": 5, "status": "unknown", "reason": "artefact_missing"},
+    ]
+    df = pd.DataFrame(data)
+    file_path = tmp_path / "audit.csv"
+    df.to_csv(file_path, index=False)
+    return str(file_path)
+
+
+def test_analyze_audit_nominal(tmp_path):
+    csv_path = _create_sample_csv(tmp_path)
+    result = core.analyze_audit(csv_path)
+    assert isinstance(result, dict), "Result should be a dictionary"
+
+    # expect three classes: artefact_missing, io_failure, contract_error
+    classes = {entry['class'] for entry in result.get('entries', [])}
+    expected_classes = {"artefact_missing", "io_failure", "contract_error"}
+    assert expected_classes.issubset(classes)
+
+    # percentages add up to approx 100
+    total_percentage = sum(entry['percentage'] for entry in result['entries'])
+    assert pytest.approx(total_percentage, rel=0.01) == 100.0
+
+    # each entry has required keys
+    for entry in result['entries']:
+        assert set(entry.keys()) == {"class", "percentage", "action"}
+        assert isinstance(entry['class'], str)
+        assert isinstance(entry['percentage'], (float, int))
+        assert isinstance(entry['action'], str)
+
+
+def test_analyze_audit_invalid_path():
+    with pytest.raises((FileNotFoundError, ValueError)):
+        core.analyze_audit("/nonexistent/file.csv")
+
+
+def test_analyze_audit_empty_file(tmp_path):
+    file_path = tmp_path / "empty.csv"
+    file_path.write_text("")
+    with pytest.raises((ValueError, pd.errors.EmptyDataError)):
+        core.analyze_audit(str(file_path))
+
+
+def test_analyze_audit_all_unknown_same_class(tmp_path):
+    df = pd.DataFrame(
+        [
+            {"id": i, "status": "unknown", "reason": "io_failure"} for i in range(5)
+        ]
+    )
+    csv_path = tmp_path / "audit_uniform.csv"
+    df.to_csv(csv_path, index=False)
+    result = core.analyze_audit(str(csv_path))
+    entries = result.get('entries', [])
+    assert len(entries) == 1
+    assert entries[0]['class'] == 'io_failure'
+    assert pytest.approx(entries[0]['percentage'], rel=0.01) == 100.0
+    assert isinstance(entries[0]['action'], str)
+
+
+def test_analyze_audit_action_field_content(tmp_path):
+    csv_path = _create_sample_csv(tmp_path)
+    result = core.analyze_audit(csv_path)
+    actions = {entry['action'] for entry in result.get('entries', [])}
+    # expect actions to be from a finite set e.g. WARN or FAIL
+    for act in actions:
+        assert act in {"WARN", "FAIL", "RETRY"} or isinstance(act, str)