feat: add null values to pdm and sql scans

okrcmaGD · okrcmaGD · commit 95b2cea31b84 · 2026-06-15T13:09:33.000+02:00
JIRA: CQ-2472
risk: low
diff --git a/gooddata-api-client/docs/ActionsApi.md b/gooddata-api-client/docs/ActionsApi.md
@@ -4214,6 +4214,7 @@ with gooddata_api_client.ApiClient() as api_client:
                             data_type="INT",
                             description="Customer unique identifier",
                             name="customer_id",
+                            null_value="null_value_example",
                         ),
                     ],
                     statement="select * from abc",
@@ -4242,6 +4243,7 @@ with gooddata_api_client.ApiClient() as api_client:
                             is_nullable=True,
                             is_primary_key=True,
                             name="customer_id",
+                            null_value="null_value_example",
                             referenced_table_column="customer_id",
                             referenced_table_id="customers",
                         ),
diff --git a/gooddata-api-client/docs/DeclarativeColumn.md b/gooddata-api-client/docs/DeclarativeColumn.md
@@ -10,6 +10,7 @@ Name | Type | Description | Notes
 **description** | **str** | Column description/comment from database | [optional] 
 **is_nullable** | **bool** | Column is nullable | [optional] 
 **is_primary_key** | **bool** | Is column part of primary key? | [optional] 
+**null_value** | **str** | Value used as sentinel for nullable columns | [optional] 
 **referenced_table_column** | **str** | Referenced table (Foreign key) | [optional] 
 **referenced_table_id** | **str** | Referenced table (Foreign key) | [optional] 
 **any string name** | **bool, date, datetime, dict, float, int, list, str, none_type** | any string name can be used but the value must be the correct type | [optional]
diff --git a/gooddata-api-client/docs/GenerateLogicalDataModelApi.md b/gooddata-api-client/docs/GenerateLogicalDataModelApi.md
@@ -55,6 +55,7 @@ with gooddata_api_client.ApiClient() as api_client:
                             data_type="INT",
                             description="Customer unique identifier",
                             name="customer_id",
+                            null_value="null_value_example",
                         ),
                     ],
                     statement="select * from abc",
@@ -83,6 +84,7 @@ with gooddata_api_client.ApiClient() as api_client:
                             is_nullable=True,
                             is_primary_key=True,
                             name="customer_id",
+                            null_value="null_value_example",
                             referenced_table_column="customer_id",
                             referenced_table_id="customers",
                         ),
diff --git a/gooddata-api-client/docs/SqlColumn.md b/gooddata-api-client/docs/SqlColumn.md
@@ -8,6 +8,7 @@ Name | Type | Description | Notes
 **data_type** | **str** | Column type | 
 **name** | **str** | Column name | 
 **description** | **str** | Column description/comment from database | [optional] 
+**null_value** | **str** | Value used as sentinel for null values in the column | [optional] 
 **any string name** | **bool, date, datetime, dict, float, int, list, str, none_type** | any string name can be used but the value must be the correct type | [optional]
 
 [[Back to Model list]](../README.md#documentation-for-models) [[Back to API list]](../README.md#documentation-for-api-endpoints) [[Back to README]](../README.md)
diff --git a/gooddata-api-client/gooddata_api_client/model/declarative_column.py b/gooddata-api-client/gooddata_api_client/model/declarative_column.py
@@ -109,6 +109,7 @@ def openapi_types():
             'description': (str,),  # noqa: E501
             'is_nullable': (bool,),  # noqa: E501
             'is_primary_key': (bool,),  # noqa: E501
+            'null_value': (str,),  # noqa: E501
             'referenced_table_column': (str,),  # noqa: E501
             'referenced_table_id': (str,),  # noqa: E501
         }
@@ -124,6 +125,7 @@ def discriminator():
         'description': 'description',  # noqa: E501
         'is_nullable': 'isNullable',  # noqa: E501
         'is_primary_key': 'isPrimaryKey',  # noqa: E501
+        'null_value': 'nullValue',  # noqa: E501
         'referenced_table_column': 'referencedTableColumn',  # noqa: E501
         'referenced_table_id': 'referencedTableId',  # noqa: E501
     }
@@ -176,6 +178,7 @@ def _from_openapi_data(cls, data_type, name, *args, **kwargs):  # noqa: E501
             description (str): Column description/comment from database. [optional]  # noqa: E501
             is_nullable (bool): Column is nullable. [optional]  # noqa: E501
             is_primary_key (bool): Is column part of primary key?. [optional]  # noqa: E501
+            null_value (str): Value used as sentinel for nullable columns. [optional]  # noqa: E501
             referenced_table_column (str): Referenced table (Foreign key). [optional]  # noqa: E501
             referenced_table_id (str): Referenced table (Foreign key). [optional]  # noqa: E501
         """
@@ -272,6 +275,7 @@ def __init__(self, data_type, name, *args, **kwargs):  # noqa: E501
             description (str): Column description/comment from database. [optional]  # noqa: E501
             is_nullable (bool): Column is nullable. [optional]  # noqa: E501
             is_primary_key (bool): Is column part of primary key?. [optional]  # noqa: E501
+            null_value (str): Value used as sentinel for nullable columns. [optional]  # noqa: E501
             referenced_table_column (str): Referenced table (Foreign key). [optional]  # noqa: E501
             referenced_table_id (str): Referenced table (Foreign key). [optional]  # noqa: E501
         """
diff --git a/gooddata-api-client/gooddata_api_client/model/sql_column.py b/gooddata-api-client/gooddata_api_client/model/sql_column.py
@@ -95,6 +95,7 @@ def openapi_types():
             'data_type': (str,),  # noqa: E501
             'name': (str,),  # noqa: E501
             'description': (str,),  # noqa: E501
+            'null_value': (str,),  # noqa: E501
         }
 
     @cached_property
@@ -106,6 +107,7 @@ def discriminator():
         'data_type': 'dataType',  # noqa: E501
         'name': 'name',  # noqa: E501
         'description': 'description',  # noqa: E501
+        'null_value': 'nullValue',  # noqa: E501
     }
 
     read_only_vars = {
@@ -154,6 +156,7 @@ def _from_openapi_data(cls, data_type, name, *args, **kwargs):  # noqa: E501
                                 through its discriminator because we passed in
                                 _visited_composed_classes = (Animal,)
             description (str): Column description/comment from database. [optional]  # noqa: E501
+            null_value (str): Value used as sentinel for null values in the column. [optional]  # noqa: E501
         """
 
         _check_type = kwargs.pop('_check_type', True)
@@ -246,6 +249,7 @@ def __init__(self, data_type, name, *args, **kwargs):  # noqa: E501
                                 through its discriminator because we passed in
                                 _visited_composed_classes = (Animal,)
             description (str): Column description/comment from database. [optional]  # noqa: E501
+            null_value (str): Value used as sentinel for null values in the column. [optional]  # noqa: E501
         """
 
         _check_type = kwargs.pop('_check_type', True)
diff --git a/packages/gooddata-sdk/src/gooddata_sdk/catalog/data_source/action_model/sql_column.py b/packages/gooddata-sdk/src/gooddata_sdk/catalog/data_source/action_model/sql_column.py
@@ -11,6 +11,7 @@
 class SqlColumn(Base):
     data_type: str
     name: str
+    null_value: str | None = None
 
     @staticmethod
     def client_class() -> type[ApiSqlColumn]:
diff --git a/packages/gooddata-sdk/tests/catalog/fixtures/data_sources/scan_scan_sql_without_preview.yaml b/packages/gooddata-sdk/tests/catalog/fixtures/data_sources/scan_scan_sql_without_preview.yaml
@@ -22,6 +22,7 @@ interactions:
           columns:
             - dataType: INT
               name: campaign_id
+              nullValue: '0'
       headers:
         Content-Type:
           - application/json
diff --git a/packages/gooddata-sdk/tests/catalog/fixtures/data_sources/scan_sql.yaml b/packages/gooddata-sdk/tests/catalog/fixtures/data_sources/scan_sql.yaml
@@ -22,10 +22,13 @@ interactions:
           columns:
             - dataType: INT
               name: product_id
+              nullValue: '0'
             - dataType: STRING
               name: product_name
+              nullValue: _
             - dataType: STRING
               name: category
+              nullValue: _
           dataPreview:
             - - '150'
               - Skirt
diff --git a/packages/gooddata-sdk/tests/catalog/fixtures/data_sources/scan_sql_with_nulls_in_preview.yaml b/packages/gooddata-sdk/tests/catalog/fixtures/data_sources/scan_sql_with_nulls_in_preview.yaml
@@ -23,6 +23,7 @@ interactions:
           columns:
             - dataType: INT
               name: campaign_id
+              nullValue: '0'
           dataPreview:
             - - null
             - - null
diff --git a/packages/gooddata-sdk/tests/catalog/test_catalog_data_source.py b/packages/gooddata-sdk/tests/catalog/test_catalog_data_source.py
@@ -728,9 +728,9 @@ def test_scan_sql(test_config: dict):
 
     assert len(response.columns) == 3
     assert response.columns == [
-        SqlColumn(name="category", data_type="STRING"),
-        SqlColumn(name="product_id", data_type="INT"),
-        SqlColumn(name="product_name", data_type="STRING"),
+        SqlColumn(name="category", data_type="STRING", null_value="_"),
+        SqlColumn(name="product_id", data_type="INT", null_value="0"),
+        SqlColumn(name="product_name", data_type="STRING", null_value="_"),
     ]
     assert len(response.data_preview) == 10
 
@@ -746,7 +746,7 @@ def test_scan_sql_with_nulls_in_preview(test_config: dict):
 
     assert len(response.columns) == 1
     assert response.columns == [
-        SqlColumn(name="campaign_id", data_type="INT"),
+        SqlColumn(name="campaign_id", data_type="INT", null_value="0"),
     ]
     assert len(response.data_preview) == 5
     assert [None] in response.data_preview
@@ -763,7 +763,7 @@ def test_scan_sql_without_preview(test_config: dict):
 
     assert len(response.columns) == 1
     assert response.columns == [
-        SqlColumn(name="campaign_id", data_type="INT"),
+        SqlColumn(name="campaign_id", data_type="INT", null_value="0"),
     ]
     assert response.data_preview is None
 
diff --git a/schemas/gooddata-api-client.json b/schemas/gooddata-api-client.json
@@ -5836,6 +5836,10 @@
             "maxLength": 255,
             "type": "string"
           },
+          "nullValue": {
+            "description": "Value used as sentinel for nullable columns",
+            "type": "string"
+          },
           "referencedTableColumn": {
             "description": "Referenced table (Foreign key)",
             "example": "customer_id",
@@ -34017,6 +34021,10 @@
             "description": "Column name",
             "example": "customer_id",
             "type": "string"
+          },
+          "nullValue": {
+            "description": "Value used as sentinel for null values in the column",
+            "type": "string"
           }
         },
         "required": [
diff --git a/schemas/gooddata-scan-client.json b/schemas/gooddata-scan-client.json
@@ -239,6 +239,10 @@
             "maxLength": 255,
             "type": "string"
           },
+          "nullValue": {
+            "description": "Value used as sentinel for nullable columns",
+            "type": "string"
+          },
           "referencedTableColumn": {
             "description": "Referenced table (Foreign key)",
             "example": "customer_id",
@@ -544,6 +548,10 @@
             "description": "Column name",
             "example": "customer_id",
             "type": "string"
+          },
+          "nullValue": {
+            "description": "Value used as sentinel for null values in the column",
+            "type": "string"
           }
         },
         "required": [
