HJ-20 remove requirement in fides and fideslang to not have datacategories on subfields

CHANGELOG.md

@@ -18,8 +18,6 @@ The types of changes are:
 ## [Unreleased](https://github.com/ethyca/fidesplus/compare/2.49.0...main)
 
 
-
-
 ## [2.49.0](https://github.com/ethyca/fidesplus/compare/2.48.2...2.49.0)
 
 ### Added
@@ -31,6 +29,7 @@ The types of changes are:
 - Added a security setting that must be set to true to enable the access request download feature [#5451](https://github.com/ethyca/fides/pull/5451)
 - Preventing erasures for the Zendesk integration if there are any open tickets [#5429](https://github.com/ethyca/fides/pull/5429)
 - Updated look/feel of all badges in the Data map report [#5464](https://github.com/ethyca/fides/pull/5464)
+- Allow adding data categories to nested fields [#5434](https://github.com/ethyca/fides/pull/5434)
 
 ### Fixed
  - Fix rendering of subfield names in D&D tables [#5439](https://github.com/ethyca/fides/pull/5439)

requirements.txt

@@ -17,7 +17,6 @@ types-defusedxml==0.7.0.20240218
 expandvars==0.9.0
 fastapi[all]==0.111.0
 fastapi-pagination[sqlalchemy]==0.12.25
-fideslang==3.0.7
 fideslog==1.2.10
 firebase-admin==5.3.0
 GitPython==3.1.41
@@ -72,3 +71,4 @@ twilio==7.15.0
 typing-extensions==4.12.2
 validators==0.20.0
 versioneer==0.19
+fideslang==3.0.8

src/fides/api/graph/config.py

@@ -338,22 +338,6 @@ class ObjectField(Field):
 
     fields: Dict[str, Field]
 
-    @field_validator("data_categories")
-    @classmethod
-    def validate_data_categories(
-        cls, value: Optional[List[FidesKey]]
-    ) -> Optional[List[FidesKey]]:
-        """To prevent mismatches between data categories on an ObjectField and a nested ScalarField, only
-        allow data categories to be defined on the individual fields.
-
-        This shouldn't be hit unless an ObjectField is declared directly.
-        """
-        if value:
-            raise ValueError(
-                "ObjectFields cannot be given data_categories; annotate the sub-fields instead."
-            )
-        return value
-
     def cast(self, value: Dict[str, Any]) -> Optional[Dict[str, Any]]:
         """Cast the input value into the form represented by data_type."""
 

tests/conftest.py

@@ -413,6 +413,26 @@ def resources_dict():
                             path="another.another.path",
                             data_categories=["user.contact.email"],
                         ),
+                        models.DatasetField(
+                            name="address",
+                            description="example top level field for nesting",
+                            path="table.address",
+                            data_categories=["user.contact.address"],
+                            fields=[
+                                models.DatasetField(
+                                    name="city",
+                                    description="example city field",
+                                    path="table.address.city",
+                                    data_categories=["user.contact.address.city"],
+                                ),
+                                models.DatasetField(
+                                    name="state",
+                                    description="example state field",
+                                    path="table.address.state",
+                                    data_categories=["user.contact.address.state"],
+                                ),
+                            ],
+                        ),
                     ],
                 )
             ],

tests/ops/graph/test_config.py

@@ -659,21 +659,21 @@ def test_generate_object_field_with_data_categories(self):
             data_categories=["user.contact.address.street"],
         )
 
-        with pytest.raises(pydantic.ValidationError):
-            generate_field(
-                name="obj",
-                data_categories=["A.B.C"],
-                identity="identity",
-                data_type_name="object",
-                references=[],
-                is_pk=False,
-                length=0,
-                is_array=False,
-                sub_fields=[apt_no_sub_field],
-                return_all_elements=None,
-                read_only=False,
-                custom_request_field=None,
-            )
+        field = generate_field(
+            name="obj",
+            data_categories=["A.B.C"],
+            identity="identity",
+            data_type_name="object",
+            references=[],
+            is_pk=False,
+            length=0,
+            is_array=False,
+            sub_fields=[apt_no_sub_field],
+            return_all_elements=None,
+            read_only=False,
+            custom_request_field=None,
+        )
+        assert field
 
     def test_generate_read_only_scalar_field(self):
         field = generate_field(

tests/ops/util/test_dataset_yaml.py

@@ -20,7 +20,7 @@
 example_dataset_yaml = """dataset:
   - fides_key: xyz
     fidesops_meta:
-        after: [db1, db2, db3]        
+        after: [db1, db2, db3]
     name: xyz
     description: x
     collections:
@@ -29,11 +29,11 @@
             after: [a.b, c.d, e.f]
         fields:
           - name: city
-            data_categories: [user.contact.address.city] 
+            data_categories: [user.contact.address.city]
           - name: id
             data_categories: [system.operations]
             fidesops_meta:
-              primary_key: True  
+              primary_key: True
               data_type: integer
 """
 
@@ -63,7 +63,7 @@
                 data_type: string
           - name: submitter
             fidesops_meta:
-                data_type: string 
+                data_type: string
             data_categories: [user]
           - name: thumbnail
             fields:
@@ -293,32 +293,24 @@ def test_dataset_graph_connected_by_nested_fields():
 
 
 example_object_with_data_categories_nested_yaml = """dataset:
-  - fides_key: mongo_nested_test 
+  - fides_key: mongo_nested_test
     name: Mongo Example Nested Test Dataset
     description: Example of a Mongo dataset that has a data_category incorrectly declared at the object level
     collections:
       - name: photos
         fields:
           - name: thumbnail
-            data_categories: [user]    
+            data_categories: [user]
             fidesops_meta:
                 data_type: object
             fields:
               - name: photo_id
                 data_type: integer
               - name: name
-                data_categories: [user]    
+                data_categories: [user]
 """
 
 
-def test_object_data_category_validation():
-    """Test trying to validate object with data category specified"""
-    with pytest.raises(ValidationError):
-        Dataset.model_validate(
-            __to_dataset__(example_object_with_data_categories_nested_yaml)
-        )
-
-
 non_array_field_with_invalid_flag = """dataset:
   - fides_key: mongo_return_all_elements_test
     name: Mongo Return All Elements Test Dataset
@@ -334,7 +326,7 @@ def test_object_data_category_validation():
               - name: photo_id
                 data_type: integer
               - name: name
-                data_categories: [user]    
+                data_categories: [user]
 """
 
 
@@ -345,7 +337,7 @@ def test_return_all_elements_specified_on_non_array_field():
 
 
 skip_processing_yaml = """dataset:
-  - fides_key: a_dataset      
+  - fides_key: a_dataset
     name: a_dataset
     description: a description
     collections:
@@ -354,19 +346,19 @@ def test_return_all_elements_specified_on_non_array_field():
             skip_processing: True
         fields:
           - name: a_field
-            data_categories: [user.contact.address.city] 
+            data_categories: [user.contact.address.city]
           - name: id
             data_categories: [system.operations]
       - name: b_collection
         fields:
           - name: b_field
-            data_categories: [user.contact.address.city] 
+            data_categories: [user.contact.address.city]
           - name: id
             data_categories: [system.operations]
 """
 
 skip_processing_invalid_yaml = """dataset:
-  - fides_key: a_dataset      
+  - fides_key: a_dataset
     name: A Dataset
     description: a description
     collections:
@@ -375,13 +367,13 @@ def test_return_all_elements_specified_on_non_array_field():
             skip_processing: True
         fields:
           - name: a_field
-            data_categories: [user.contact.address.city] 
+            data_categories: [user.contact.address.city]
           - name: id
             data_categories: [system.operations]
       - name: b_collection
         fields:
           - name: b_field
-            data_categories: [user.contact.address.city] 
+            data_categories: [user.contact.address.city]
           - name: id
             data_categories: [system.operations]
             fides_meta: