Add regression coverage for explicit empty record selectors

Author: qorud02

tests/test_records_key.py

@@ -0,0 +1,86 @@
+"""Explicit JSON selectors distinguish an empty key from an absent option."""
+import json
+import os
+from pathlib import Path
+import subprocess
+import sys
+import tempfile
+import unittest
+
+
+ROOT = Path(__file__).resolve().parents[1]
+
+
+class ExplicitRecordsKeyTests(unittest.TestCase):
+    def setUp(self):
+        temporary = tempfile.TemporaryDirectory()
+        self.addCleanup(temporary.cleanup)
+        self.root = Path(temporary.name)
+        self.contract = self.root / "contract.json"
+        self.contract.write_text('{"columns":{"id":{"type":"string"}}}', encoding="utf-8")
+
+    def run_cli(self, name, content, options=()):
+        source = self.root / name
+        source.write_bytes(content.encode("utf-8"))
+        environment = dict(os.environ, PYTHONPATH=str(ROOT / "src"),
+                           PYTHONDONTWRITEBYTECODE="1", PYTHONIOENCODING="utf-8")
+        return subprocess.run(
+            [sys.executable, "-m", "public_data_sentinel.cli", str(source),
+             "--contract", str(self.contract), *options],
+            cwd=self.root, env=environment, capture_output=True, text=True,
+            encoding="utf-8", timeout=30,
+        )
+
+    def test_explicit_empty_key_selects_valid_and_empty_record_arrays(self):
+        for records, expected in (([{"id": "001"}], 0), ([], 1)):
+            with self.subTest(records=records):
+                result = self.run_cli(
+                    "envelope.json", json.dumps({"": records, "other": [{"id": 123}]}),
+                    ("--records-key", ""),
+                )
+                self.assertEqual(result.returncode, expected, result.stderr)
+                self.assertEqual(result.stderr, "")
+                report = json.loads(result.stdout)
+                self.assertEqual(report["records_checked"], len(records))
+                self.assertEqual(report["valid"], expected == 0)
+                if not records:
+                    self.assertEqual(report["issues"][0]["code"], "empty_data")
+
+    def test_explicit_empty_key_rejects_missing_or_nonarray_selection(self):
+        for content in ('[{"id":"001"}]', '{"items":[{"id":"001"}]}', '{"":{}}'):
+            with self.subTest(content=content):
+                result = self.run_cli("invalid-selection.json", content, ("--records-key", ""))
+                self.assertEqual(result.returncode, 2, result.stderr)
+                self.assertEqual(result.stdout, "")
+                self.assertTrue(result.stderr.startswith("data-sentinel: "))
+                self.assertNotIn("Traceback", result.stderr)
+
+    def test_delimited_inputs_reject_empty_and_named_selectors(self):
+        for suffix in ("csv", "tsv"):
+            for key in ("", "items"):
+                with self.subTest(suffix=suffix, key=key):
+                    result = self.run_cli("data." + suffix, "id\n001\n", ("--records-key", key))
+                    self.assertEqual(result.returncode, 2, result.stderr)
+                    self.assertEqual(result.stdout, "")
+                    self.assertIn("--records-key applies only to JSON", result.stderr)
+
+    def test_absent_and_named_selectors_keep_existing_behavior(self):
+        cases = (
+            ("array.json", '[{"id":"001"}]', ()),
+            ("envelope.json", '{"items":[{"id":"001"}]}', ("--records-key", "items")),
+            ("data.csv", "id\n001\n", ()),
+            ("data.tsv", "id\n001\n", ()),
+        )
+        for name, content, options in cases:
+            with self.subTest(name=name, options=options):
+                result = self.run_cli(name, content, options)
+                self.assertEqual(result.returncode, 0, result.stderr)
+                self.assertEqual(result.stderr, "")
+                report = json.loads(result.stdout)
+                self.assertTrue(report["valid"])
+                self.assertEqual(report["records_checked"], 1)
+                self.assertEqual(report["error_count"], 0)
+
+
+if __name__ == "__main__":
+    unittest.main()