From 8e8a7c3808f1098172bb4212fb393c81ba5a33d0 Mon Sep 17 00:00:00 2001
From: Deepjyoti-Sarmah <deepjyotisarmah37@gmail.com>
Date: Thu, 22 Oct 2026 04:12:32 +0530
Subject: [PATCH] test: one node per SQL table across files

A table no file defines, read from N files, becomes N per-file stub
nodes, and a schema-qualified read never binds to its definition in
another file. A SQL read of `Orders` also binds to a Python class
Orders. Guards cover same-file binding and different schemas.

(cherry picked from commit 1dfc710497a753cc663a2cefc3ece724414be784)
---
 tests/test_sql_cross_file_tables.py | 95 +++++++++++++++++++++++++++++
 1 file changed, 95 insertions(+)
 create mode 100644 tests/test_sql_cross_file_tables.py

diff --git a/tests/test_sql_cross_file_tables.py b/tests/test_sql_cross_file_tables.py
new file mode 100644
index 000000000..9f1394e83
--- /dev/null
+++ b/tests/test_sql_cross_file_tables.py
@@ -0,0 +1,95 @@
+"""A table referenced from several files is one node, bound to its definition when one exists."""
+from __future__ import annotations
+
+from pathlib import Path
+
+import pytest
+
+pytest.importorskip("tree_sitter_sql")
+
+from graphify.extract import extract  # noqa: E402
+
+
+def _project(tmp_path: Path, files: dict[str, str]) -> dict:
+    paths = []
+    for rel, text in files.items():
+        p = tmp_path / rel
+        p.parent.mkdir(parents=True, exist_ok=True)
+        p.write_text(text)
+        paths.append(p)
+    return extract(paths, root=tmp_path)
+
+
+def _norm(label: str) -> str:
+    return label.replace("[", "").replace("]", "").lower()
+
+
+def _ids(result: dict, name: str) -> set[str]:
+    return {n["id"] for n in result["nodes"] if _norm(n["label"]) == name}
+
+
+def _read_targets(result: dict, view: str) -> set[str]:
+    src = _ids(result, view)
+    return {e["target"] for e in result["edges"] if e["source"] in src and e["relation"] == "reads_from"}
+
+
+VIEWS = {
+    "v/a.sql": "CREATE VIEW dbo.VA AS SELECT id FROM {t};\n",
+    "v/b.sql": "CREATE VIEW dbo.VB AS SELECT id FROM {t};\n",
+}
+
+
+# bug
+@pytest.mark.parametrize("table", ["dbo.Orders", "Orders"])
+def test_undefined_table_from_two_files_is_one_node(tmp_path, table):
+    r = _project(tmp_path, {k: v.format(t=table) for k, v in VIEWS.items()})
+    assert len(_ids(r, table.lower())) == 1
+    assert _read_targets(r, "dbo.va") == _read_targets(r, "dbo.vb")
+
+
+# bug
+@pytest.mark.parametrize("table", ["dbo.Orders", "Orders"])
+def test_reads_bind_to_table_defined_in_another_file(tmp_path, table):
+    files = {"t/tables.sql": f"CREATE TABLE {table} (id INT);\n"}
+    files.update({k: v.format(t=table) for k, v in VIEWS.items()})
+    r = _project(tmp_path, files)
+    defined = _ids(r, table.lower())
+    assert len(defined) == 1
+    assert _read_targets(r, "dbo.va") == defined
+    assert _read_targets(r, "dbo.vb") == defined
+
+
+# bug
+def test_bracket_and_plain_names_across_files_are_one_table(tmp_path):
+    r = _project(tmp_path, {
+        "t/tables.sql": "CREATE TABLE [dbo].[Orders] (id INT);\n",
+        "v/a.sql": "CREATE VIEW dbo.VA AS SELECT id FROM dbo.Orders;\n",
+    })
+    assert _read_targets(r, "dbo.va") == _ids(r, "dbo.orders")
+    assert len(_ids(r, "dbo.orders")) == 1
+
+
+# guard
+def test_same_file_definition_still_binds(tmp_path):
+    r = _project(tmp_path, {"a.sql": "CREATE TABLE dbo.Orders (id INT);\nCREATE VIEW dbo.VA AS SELECT id FROM dbo.Orders;\n"})
+    assert _read_targets(r, "dbo.va") == _ids(r, "dbo.orders")
+
+
+# guard
+def test_different_schemas_stay_different_tables(tmp_path):
+    r = _project(tmp_path, {
+        "v/a.sql": "CREATE VIEW dbo.VA AS SELECT id FROM dbo.Orders;\n",
+        "v/b.sql": "CREATE VIEW dbo.VB AS SELECT id FROM sales.Orders;\n",
+    })
+    assert _read_targets(r, "dbo.va").isdisjoint(_read_targets(r, "dbo.vb"))
+
+
+# guard
+def test_sql_read_does_not_bind_to_python_class(tmp_path):
+    r = _project(tmp_path, {
+        "app/models.py": "class Orders:\n    pass\n",
+        "v/a.sql": "CREATE VIEW dbo.VA AS SELECT id FROM Orders;\n",
+    })
+    py = {n["id"] for n in r["nodes"] if n.get("source_file", "").endswith(".py") and n["label"] == "Orders"}
+    assert py
+    assert _read_targets(r, "dbo.va").isdisjoint(py)