feat(tests): scaffold + integration smoke tests (config, db, models, pipeline)

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

tests/scaffold/test_config.py

@@ -0,0 +1,35 @@
+"""Smoke tests for codex.config.Settings."""
+
+from __future__ import annotations
+
+from codex.config import Settings, get_settings
+
+
+def test_settings_defaults() -> None:
+    """Settings() is constructable without env vars; expected defaults are set."""
+    s = Settings()
+    assert s.database_url == "postgresql://researcher:change_me@localhost:5432/papers"
+    assert s.embedding_model == "BAAI/bge-m3"
+    assert s.embedding_dim == 1024
+
+
+def test_settings_env_override(monkeypatch: object) -> None:
+    """DATABASE_URL env var overrides the default database_url."""
+    import pytest
+
+    mp: pytest.MonkeyPatch = monkeypatch  # type: ignore[assignment]
+    mp.setenv("DATABASE_URL", "postgresql://x:y@h:5432/db")
+    s = Settings()
+    assert s.database_url == "postgresql://x:y@h:5432/db"
+
+
+def test_get_settings_is_singleton() -> None:
+    """get_settings() returns the same object on repeated calls; new object after cache_clear."""
+    get_settings.cache_clear()
+    first = get_settings()
+    second = get_settings()
+    assert first is second
+
+    get_settings.cache_clear()
+    third = get_settings()
+    assert third is not first

tests/scaffold/test_db.py

@@ -0,0 +1,35 @@
+"""Smoke tests for codex.db — get_conn and apply_schema."""
+
+from __future__ import annotations
+
+from unittest.mock import MagicMock, patch
+
+
+def test_get_conn_context_manager() -> None:
+    """get_conn() yields a context manager that calls psycopg.connect and cleans up."""
+    from codex.db import get_conn
+
+    mock_conn = MagicMock()
+    mock_connect = MagicMock()
+    # psycopg.connect is itself used as a context manager inside get_conn
+    mock_connect.return_value.__enter__ = MagicMock(return_value=mock_conn)
+    mock_connect.return_value.__exit__ = MagicMock(return_value=False)
+
+    with patch("codex.db.psycopg.connect", mock_connect), get_conn() as conn:
+        assert conn is mock_conn
+
+    mock_connect.assert_called_once()
+
+
+def test_apply_schema_executes_sql() -> None:
+    """apply_schema(conn) calls conn.execute and conn.commit."""
+    from codex.db import apply_schema
+
+    mock_conn = MagicMock()
+
+    # Provide a fake schema.sql so Path.read_text succeeds
+    with patch("codex.db.Path.read_text", return_value="CREATE TABLE IF NOT EXISTS test ();"):
+        apply_schema(mock_conn)
+
+    mock_conn.execute.assert_called_once()
+    mock_conn.commit.assert_called_once()

tests/scaffold/test_models.py

@@ -0,0 +1,34 @@
+"""Smoke tests for codex.models dataclasses."""
+
+from __future__ import annotations
+
+from codex.models import Chunk, Citation, CodeLink, Paper
+
+
+def test_paper_defaults() -> None:
+    """Paper(id=..., title=...) has correct optional-field defaults."""
+    p = Paper(id="2301.00001", title="T")
+    assert p.openalex_id is None
+    assert p.authors == []
+    assert p.year is None
+
+
+def test_chunk_required_fields() -> None:
+    """Chunk(paper_id=..., ord=..., content=...) has id=None and embedding=None."""
+    c = Chunk(paper_id="x", ord=0, content="c")
+    assert c.id is None
+    assert c.embedding is None
+
+
+def test_citation_fields() -> None:
+    """Citation(citing_id=..., cited_id=...) has context=None."""
+    cit = Citation(citing_id="a", cited_id="b")
+    assert cit.context is None
+
+
+def test_codelink_fields() -> None:
+    """CodeLink(symbol=...) has paper_id=None, role=None, and id=None."""
+    cl = CodeLink(symbol="foo")
+    assert cl.paper_id is None
+    assert cl.role is None
+    assert cl.id is None