test(gooddata-sdk): cover AAC round-trip for agg-aware datasets

YAML ↔ declarative round-trip tests for the three new aggregate-aware
shapes the WASM convertor handles end-to-end on 11.35.0a2:

- AUXILIARY datasets (no physical mapping; synthetic identity attrs).
- NORMAL pre-aggregation datasets with `aggregated_facts` — both the
  vanilla SUM-of-fact path and APPROXIMATE_COUNT-of-attribute (HLL
  synopses targeting the AUX identity attribute, requires
  `reference.type == "attribute"` per gdc-nas CQ-2147).
- NORMAL synthesized dim datasets backed by a `sql:` block.

These guard the SDK side of the AAC convertor pipeline; the heavy
lifting lives in `gooddata-code-convertors`.

risk: low

Author: jaceksan

packages/gooddata-sdk/tests/catalog/unit_tests/test_aac_agg_aware.py

@@ -0,0 +1,183 @@
+# (C) 2026 GoodData Corporation
+"""AAC YAML ↔ declarative round-trip tests for aggregate-aware dataset shapes.
+
+Exercises the SDK wrappers around the WASM convertor (gooddata-code-convertors)
+for the three new shapes that aggregate-aware LDMs introduce:
+
+- AUXILIARY datasets (no physical mapping, synthetic identity attributes).
+- NORMAL pre-aggregation datasets with `aggregated_facts` (SUM-of-fact and
+  APPROXIMATE_COUNT-of-attribute for HLL synopses).
+- NORMAL synthesized dim datasets backed by a `sql:` block.
+
+These tests guard the **SDK side** of the AAC convertor pipeline; the
+heavy lifting lives in the WASM module bumped to 11.33.0+.
+"""
+
+from __future__ import annotations
+
+import yaml
+from gooddata_sdk.catalog.workspace.aac import (
+    aac_dataset_to_declarative,
+    declarative_dataset_to_aac,
+)
+
+
+def _aac_from(declarative: dict) -> dict:
+    """Convert declarative dataset → AAC dict via the SDK wrapper."""
+    result = declarative_dataset_to_aac(declarative)
+    return yaml.safe_load(result["content"])
+
+
+def _entities_for(aac: dict) -> list[dict]:
+    """Build a one-element entities list (most round-trips need it)."""
+    return [{"id": aac["id"], "type": aac["type"], "path": f"{aac['id']}.yaml", "data": aac}]
+
+
+def test_auxiliary_dataset_round_trips_through_aac() -> None:
+    declarative_in = {
+        "id": "orders",
+        "title": "Orders",
+        "type": "AUXILIARY",
+        "attributes": [
+            {"id": "orders.unique_customer", "title": "Unique Customer", "labels": []},
+        ],
+        "facts": [],
+        "grain": [],
+        "references": [],
+        "aggregatedFacts": [],
+    }
+    aac = _aac_from(declarative_in)
+    # AUX is encoded with `dataset_type: auxiliary` on the AAC side.
+    assert aac["dataset_type"] == "auxiliary"
+    assert "orders.unique_customer" in aac["fields"]
+    assert aac["fields"]["orders.unique_customer"]["type"] == "attribute"
+    # AUX attributes must NOT carry source_column on the YAML side either.
+    assert "source_column" not in aac["fields"]["orders.unique_customer"]
+
+    declarative_out = aac_dataset_to_declarative(aac, _entities_for(aac))
+    assert declarative_out["type"] == "AUXILIARY"
+    attrs = declarative_out["attributes"]
+    assert any(a["id"] == "orders.unique_customer" for a in attrs)
+    # No physical-column mapping was injected on the way back.
+    assert all("sourceColumn" not in a for a in attrs)
+
+
+def test_pre_aggregation_sum_round_trips_through_aac() -> None:
+    """The vanilla pre-aggregation path: SUM-of-fact aggregated_facts."""
+    declarative_in = {
+        "id": "agg_orders_country_daily",
+        "title": "Orders by country (daily)",
+        "type": "NORMAL",
+        "dataSourceTableId": {
+            "dataSourceId": "demo-ds",
+            "id": "agg_orders_country_daily",
+            "type": "dataSource",
+            "path": ["agg_orders_country_daily"],
+        },
+        "precedence": 1,
+        "aggregatedFacts": [
+            {
+                "id": "agg_orders_country_daily.revenue",
+                "sourceColumn": "revenue",
+                "sourceFactReference": {
+                    "operation": "SUM",
+                    "reference": {"id": "orders.revenue", "type": "fact"},
+                },
+            },
+        ],
+        "attributes": [],
+        "facts": [],
+        "grain": [],
+        "references": [],
+    }
+    aac = _aac_from(declarative_in)
+    assert aac["dataset_type"] == "standard"
+    assert aac["precedence"] == 1
+    field = aac["fields"]["agg_orders_country_daily.revenue"]
+    assert field["type"] == "aggregated_fact"
+    assert field["aggregated_as"] == "SUM"
+    assert field["assigned_to"] == "orders.revenue"
+
+    declarative_out = aac_dataset_to_declarative(aac, _entities_for(aac), data_source_id="demo-ds")
+    assert declarative_out["type"] == "NORMAL"
+    assert declarative_out["precedence"] == 1
+    out_facts = declarative_out["aggregatedFacts"]
+    assert len(out_facts) == 1
+    assert out_facts[0]["sourceFactReference"]["operation"] == "SUM"
+    assert out_facts[0]["sourceFactReference"]["reference"]["type"] == "fact"
+
+
+def test_synthesized_dim_with_sql_round_trips_through_aac() -> None:
+    declarative_in = {
+        "id": "dim_country",
+        "title": "Country",
+        "type": "NORMAL",
+        "sql": {
+            "statement": "SELECT DISTINCT country FROM agg_orders_country_daily",
+            "dataSourceId": "demo-ds",
+        },
+        "attributes": [
+            {
+                "id": "dim_country.country",
+                "title": "Country",
+                "sourceColumn": "country",
+                "sourceColumnDataType": "STRING",
+                "labels": [],
+            },
+        ],
+        "facts": [],
+        "grain": [{"id": "dim_country.country", "type": "attribute"}],
+        "references": [],
+        "aggregatedFacts": [],
+    }
+    aac = _aac_from(declarative_in)
+    assert aac["sql"].startswith("SELECT DISTINCT")
+    assert aac["data_source"] == "demo-ds"
+
+    declarative_out = aac_dataset_to_declarative(aac, _entities_for(aac), data_source_id="demo-ds")
+    assert declarative_out["sql"]["statement"] == declarative_in["sql"]["statement"]
+    assert declarative_out["sql"]["dataSourceId"] == "demo-ds"
+    out_attrs = declarative_out["attributes"]
+    assert out_attrs[0]["sourceColumn"] == "country"
+
+
+def test_pre_aggregation_approximate_count_attribute_target_round_trips() -> None:
+    """HLL APPROXIMATE_COUNT references an attribute, not a fact.
+
+    The platform requires `aggregatedFacts[].sourceFactReference.reference.type
+    == "attribute"` for HLL synopses (gdc-nas CQ-2147).
+    """
+    aac = {
+        "type": "dataset",
+        "id": "agg_orders_country_daily",
+        "title": "Agg",
+        "table_path": "agg_orders_country_daily",
+        "data_source": "demo-ds",
+        "dataset_type": "standard",
+        "precedence": 1,
+        "fields": {
+            "agg_orders_country_daily.unique_customers_hll": {
+                "type": "aggregated_fact",
+                "source_column": "unique_customers_hll",
+                "data_type": "HLL",
+                "aggregated_as": "APPROXIMATE_COUNT",
+                "assigned_to": "attribute/orders.unique_customer",
+            },
+        },
+    }
+    aux = {
+        "type": "dataset",
+        "id": "orders",
+        "title": "Orders",
+        "dataset_type": "auxiliary",
+        "fields": {"unique_customer": {"type": "attribute", "title": "Unique Customer"}},
+    }
+    entities = [
+        {"id": "orders", "type": "dataset", "path": "orders.yaml", "data": aux},
+        {"id": aac["id"], "type": aac["type"], "path": f"{aac['id']}.yaml", "data": aac},
+    ]
+    declarative = aac_dataset_to_declarative(aac, entities, data_source_id="demo-ds")
+    af = declarative["aggregatedFacts"][0]
+    assert af["sourceFactReference"]["operation"] == "APPROXIMATE_COUNT"
+    assert af["sourceColumnDataType"] == "HLL"
+    assert af["sourceFactReference"]["reference"]["type"] == "attribute"