feat: add daemon sql batch analysis

python/ktx-daemon/tests/test_app.py

@@ -280,6 +280,37 @@ def test_sql_parse_table_identifier_endpoint() -> None:
     assert body["results"]["template"]["reason"] == "looker_template_unresolved"
 
 
+def test_sql_analyze_batch_endpoint_returns_per_item_results() -> None:
+    client = TestClient(create_app())
+
+    response = client.post(
+        "/sql/analyze-batch",
+        json={
+            "dialect": "postgres",
+            "max_workers": 1,
+            "items": [
+                {
+                    "id": "orders",
+                    "sql": "select status from public.orders where created_at is not null",
+                },
+                {"id": "broken", "sql": "select * from where"},
+            ],
+        },
+    )
+
+    assert response.status_code == 200
+    body = response.json()
+    assert body["results"]["orders"]["tables_touched"] == ["public.orders"]
+    assert body["results"]["orders"]["columns_by_clause"] == {
+        "select": ["status"],
+        "where": ["created_at"],
+    }
+    assert body["results"]["orders"]["error"] is None
+    assert body["results"]["broken"]["tables_touched"] == []
+    assert body["results"]["broken"]["columns_by_clause"] == {}
+    assert body["results"]["broken"]["error"] is not None
+
+
 def test_semantic_query_endpoint_returns_sql() -> None:
     client = TestClient(create_app())
 

python/ktx-daemon/tests/test_sql_analysis.py

@@ -0,0 +1,53 @@
+from __future__ import annotations
+
+from ktx_daemon.sql_analysis import (
+    AnalyzeSqlBatchItem,
+    AnalyzeSqlBatchRequest,
+    analyze_sql_batch_response,
+)
+
+
+def test_analyze_sql_batch_extracts_tables_and_clause_columns() -> None:
+    response = analyze_sql_batch_response(
+        AnalyzeSqlBatchRequest(
+            dialect="postgres",
+            items=[
+                AnalyzeSqlBatchItem(
+                    id="orders_by_customer",
+                    sql=(
+                        "select o.status, count(*) "
+                        "from public.orders o "
+                        "join public.customers c on o.customer_id = c.id "
+                        "where o.created_at >= current_date - interval '30 day' "
+                        "group by o.status"
+                    ),
+                )
+            ],
+            max_workers=1,
+        )
+    )
+
+    result = response.results["orders_by_customer"]
+    assert result.error is None
+    assert result.tables_touched == ["public.orders", "public.customers"]
+    assert result.columns_by_clause == {
+        "select": ["status"],
+        "where": ["created_at"],
+        "join": ["customer_id", "id"],
+        "groupBy": ["status"],
+    }
+
+
+def test_analyze_sql_batch_returns_per_item_parse_errors() -> None:
+    response = analyze_sql_batch_response(
+        AnalyzeSqlBatchRequest(
+            dialect="postgres",
+            items=[AnalyzeSqlBatchItem(id="broken", sql="select * from where")],
+            max_workers=1,
+        )
+    )
+
+    result = response.results["broken"]
+    assert result.tables_touched == []
+    assert result.columns_by_clause == {}
+    assert result.error is not None