added tests

Author: codinglabsong

.github/workflows/smoke-tests.yaml

@@ -31,5 +31,5 @@ jobs:
           pre-commit install
           pre-commit run --all-files
 
-      # - name: Run tests
-      #   run: pytest
+      - name: Run tests
+        run: pytest

tests/conftest.py

@@ -0,0 +1,4 @@
+import sys
+from pathlib import Path
+
+sys.path.insert(0, str(Path(__file__).resolve().parents[1] / "src"))

tests/test_indexing.py

@@ -0,0 +1,24 @@
+from any_chatbot.indexing import _tbl, build_duckdb_and_summary_cards
+from pathlib import Path
+
+
+def test_tbl_cleaning():
+    assert _tbl("my table!") == "my_table"
+    assert _tbl("123name") == "t_123name"
+    assert _tbl("##$") == "t_"
+
+
+def test_build_duckdb_and_summary_cards_csv(tmp_path: Path):
+    csv_path = tmp_path / "data.csv"
+    csv_path.write_text("a,b\n1,2\n3,4")
+    db_path = tmp_path / "db.duckdb"
+
+    cards = build_duckdb_and_summary_cards(tmp_path, db_path)
+    assert len(cards) == 1
+
+    card = cards[0]
+    assert card.metadata["source_type"] == "table_summary"
+    assert card.metadata["db_path"] == str(db_path)
+    assert card.metadata["table"] == "data"
+    assert "TABLE CARD" in card.page_content
+    assert "a:BIGINT" in card.page_content

tests/test_tools.py

@@ -0,0 +1,28 @@
+from any_chatbot.tools import initialize_retrieve_tool, is_safe_sql
+from langchain.schema import Document
+
+
+class DummyStore:
+    def __init__(self):
+        self.calls = []
+
+    def similarity_search(self, query: str, k: int = 5, filter=None):
+        self.calls.append((query, k, filter))
+        return [Document(page_content="foo", metadata=filter)]
+
+
+def test_initialize_retrieve_tool_invokes_vectorstore():
+    store = DummyStore()
+    retrieve = initialize_retrieve_tool(store)
+    text, docs = retrieve.func("hello", "text_chunk")
+
+    assert store.calls == [("hello", 5, {"source_type": "text_chunk"})]
+    assert docs[0].metadata["source_type"] == "text_chunk"
+    assert "foo" in text
+
+
+def test_is_safe_sql():
+    assert is_safe_sql("SELECT * FROM tbl")
+    assert is_safe_sql("SELECT updated_at FROM tbl")
+    assert not is_safe_sql("DROP TABLE tbl")
+    assert not is_safe_sql("UPDATE tbl SET a=1")