Add test for v3 pipeline runner

Author: xiangtinghe616-blip

tests/test_v3_pipeline_runner.py

@@ -0,0 +1,39 @@
+from pathlib import Path
+import sys
+
+import pandas as pd
+
+
+REPO_ROOT = Path(__file__).resolve().parents[1]
+sys.path.insert(0, str(REPO_ROOT))
+
+from versions.v3.src.reconciliation.run_v3_pipeline import run_v3_pipeline  # noqa: E402
+
+
+def test_run_v3_pipeline_creates_expected_outputs():
+    result = run_v3_pipeline()
+
+    output_dir = REPO_ROOT / "versions" / "v3" / "output"
+
+    expected_outputs = [
+        "validation_issues.csv",
+        "canonical_bank_transactions.csv",
+        "canonical_internal_transactions.csv",
+        "pipeline_run_summary.csv",
+    ]
+
+    for filename in expected_outputs:
+        assert (output_dir / filename).exists(), f"Missing pipeline output: {filename}"
+
+    assert result["run_id"].startswith("v3_local_run_")
+    assert result["canonical_bank_count"] == 595
+    assert result["canonical_ledger_count"] == 600
+    assert result["validation_issue_count"] >= 1
+
+    summary = pd.read_csv(output_dir / "pipeline_run_summary.csv")
+
+    assert set(summary["stage"]) == {
+        "schema_validation",
+        "bank_standardization",
+        "ledger_standardization",
+    }