Add Non Nullable DatasetConfig.ctl_dataset_id Field

.fides/db_dataset.yml

@@ -1025,6 +1025,9 @@ dataset:
     - name: connection_config_id
       data_categories: [system.operations]
       data_qualifier: aggregated.anonymized.unlinked_pseudonymized.pseudonymized.identified
+    - name: ctl_dataset_id
+      data_categories: [ system.operations ]
+      data_qualifier: aggregated.anonymized.unlinked_pseudonymized.pseudonymized.identified
     - name: created_at
       data_categories: [system.operations]
       data_qualifier: aggregated.anonymized.unlinked_pseudonymized.pseudonymized.identified

src/fides/api/ctl/migrations/versions/216cdc7944f1_add_datasetconfig_ctl_datasets_fk.py

@@ -0,0 +1,122 @@
+"""add datasetconfig.ctl_dataset_id fk
+
+Adding a FK to datasetconfig pointing to the ctl_datasets table.
+Also try to automigrate datasetconfig.datasets to the ctl_datasets row
+
+Revision ID: 216cdc7944f1
+Revises: 2fb48b0e268b
+Create Date: 2022-12-09 22:03:51.097585
+
+"""
+import json
+import uuid
+from typing import Any, Dict
+
+import sqlalchemy as sa
+from alembic import op
+from fideslang.models import Dataset
+from sqlalchemy import text
+
+# revision identifiers, used by Alembic.
+from sqlalchemy.exc import IntegrityError
+from sqlalchemy.sql.elements import TextClause
+
+revision = "216cdc7944f1"
+down_revision = "2fb48b0e268b"
+branch_labels = None
+depends_on = None
+
+AUTO_MIGRATED_STRING = "auto-migrated from datasetconfig.dataset"
+
+
+def upgrade():
+    # Schema migration - add a nullable datasetconfig.ctl_dataset_id field.  We will shortly make it non-nullable.
+    op.add_column(
+        "datasetconfig", sa.Column("ctl_dataset_id", sa.String(), nullable=True)
+    )
+    op.create_index(
+        op.f("ix_datasetconfig_ctl_dataset_id"),
+        "datasetconfig",
+        ["ctl_dataset_id"],
+        unique=False,
+    )
+    op.create_foreign_key(
+        "datasetconfig_ctl_dataset_id_fkey",
+        "datasetconfig",
+        "ctl_datasets",
+        ["ctl_dataset_id"],
+        ["id"],
+    )
+
+    # Data migration - automatically try to port datasetconfig.dataset -> ctl_datasets if possible.
+    bind = op.get_bind()
+    existing_datasetconfigs = bind.execute(
+        text("SELECT id, created_at, updated_at, dataset FROM datasetconfig;")
+    )
+    for row in existing_datasetconfigs:
+        dataset: Dict[str, Any] = row["dataset"]
+        fides_key: str = dataset["fides_key"]
+
+        insert_into_ctl_datasets_query: TextClause = text(
+            "INSERT INTO ctl_datasets (id, fides_key, organization_fides_key, name, description, meta, data_categories, "
+            "collections, data_qualifier, created_at, updated_at, joint_controller, retention, fides_meta, third_country_transfers, tags) "
+            "VALUES (:id, :fides_key, :organization_fides_key, :name, :description, :meta, :data_categories, :collections, "
+            ":data_qualifier, :created_at, :updated_at, :joint_controller, :retention, :fides_meta, :third_country_transfers, :tags)"
+        )
+
+        new_ctl_dataset_id: str = "ctl_" + str(uuid.uuid4())
+        # Stashing extra text into the "meta" column so we can use this to downgrade if needed
+        appended_meta: Dict = dataset["meta"] or {}
+        appended_meta["fides_source"] = AUTO_MIGRATED_STRING
+
+        validated_dataset: Dict = Dataset(
+            **dataset
+        ).dict()  # Validating before we store.
+        validated_dataset["id"] = new_ctl_dataset_id
+        validated_dataset["fides_key"] = fides_key
+        validated_dataset["collections"] = json.dumps(validated_dataset["collections"])
+        validated_dataset["meta"] = json.dumps(appended_meta)
+        validated_dataset["created_at"] = row["created_at"]
+        validated_dataset["updated_at"] = row["updated_at"]
+        validated_dataset["fides_meta"] = dataset.get("fides_meta") or dataset.get(
+            "fidesops_meta"
+        )
+
+        try:
+            bind.execute(
+                insert_into_ctl_datasets_query,
+                validated_dataset,
+            )
+        except IntegrityError as exc:
+            raise Exception(
+                f"Fides attempted to copy datasetconfig.datasets into their own ctl_datasets rows but got error: {exc}. "
+                f"Adjust fides_keys in ctl_datasets table to not conflict."
+            )
+
+        update_dataset_config_query: TextClause = text(
+            "UPDATE datasetconfig SET ctl_dataset_id= :new_ctl_dataset_id WHERE id= :datasetconfig_id"
+        )
+
+        bind.execute(
+            update_dataset_config_query,
+            {"new_ctl_dataset_id": new_ctl_dataset_id, "datasetconfig_id": row["id"]},
+        )
+
+
+def downgrade():
+    # Reverse schema migration
+    op.drop_constraint(
+        "datasetconfig_ctl_dataset_id_fkey", "datasetconfig", type_="foreignkey"
+    )
+    op.drop_index(op.f("ix_datasetconfig_ctl_dataset_id"), table_name="datasetconfig")
+    op.drop_column("datasetconfig", "ctl_dataset_id")
+
+    # Reverse data migration: remove ctl_datasets that were automatically created by the forward migration
+    bind = op.get_bind()
+    remove_automigrated_ctl_datasets_query: TextClause = text(
+        "DELETE FROM ctl_datasets WHERE meta->>'fides_source'= :automigration_string"
+    )
+    bind.execute(
+        remove_automigrated_ctl_datasets_query,
+        {"automigration_string": AUTO_MIGRATED_STRING},
+    )

src/fides/api/ctl/migrations/versions/9c6f62e4c9da_make_datasetconfig_ctl_datasets_non_nullable.py

@@ -0,0 +1,28 @@
+"""make datasetconfig.ctl_datasets non-nullable
+
+Revision ID: 9c6f62e4c9da
+Revises: 216cdc7944f1
+Create Date: 2022-12-09 23:56:13.022119
+
+"""
+import sqlalchemy as sa
+from alembic import op
+
+# revision identifiers, used by Alembic.
+revision = "9c6f62e4c9da"
+down_revision = "216cdc7944f1"
+branch_labels = None
+depends_on = None
+
+
+def upgrade():
+    """Followup migration to make datasetconfig.ctl_dataset_id non nullable"""
+    op.alter_column(
+        "datasetconfig", "ctl_dataset_id", existing_type=sa.VARCHAR(), nullable=False
+    )
+
+
+def downgrade():
+    op.alter_column(
+        "datasetconfig", "ctl_dataset_id", existing_type=sa.VARCHAR(), nullable=True
+    )

src/fides/api/ctl/sql_models.py

@@ -6,6 +6,7 @@
 
 from typing import Dict
 
+from fideslang.models import Dataset as FideslangDataset
 from sqlalchemy import (
     ARRAY,
     BOOLEAN,
@@ -19,6 +20,7 @@
     type_coerce,
 )
 from sqlalchemy.dialects.postgresql import BYTEA
+from sqlalchemy.orm import Session
 from sqlalchemy.sql import func
 from sqlalchemy.sql.sqltypes import DateTime
 
@@ -200,6 +202,16 @@ class Dataset(Base, FidesBase):
     retention = Column(String)
     third_country_transfers = Column(ARRAY(String))
 
+    @classmethod
+    def create_from_dataset_dict(cls, db: Session, dataset: dict) -> "Dataset":
+        """Add a method to create directly using a synchronous session"""
+        validated_dataset: FideslangDataset = FideslangDataset(**dataset)
+        ctl_dataset = cls(**validated_dataset.dict())
+        db.add(ctl_dataset)
+        db.commit()
+        db.refresh(ctl_dataset)
+        return ctl_dataset
+
 
 # Evaluation
 class Evaluation(Base):

src/fides/api/ops/api/v1/endpoints/dataset_endpoints.py

@@ -1,4 +1,4 @@
-from typing import List
+from typing import Callable, List
 
 import yaml
 from fastapi import Depends, HTTPException, Request
@@ -22,6 +22,7 @@
     HTTP_422_UNPROCESSABLE_ENTITY,
 )
 
+from fides.api.ctl.sql_models import Dataset as CtlDataset  # type: ignore[attr-defined]
 from fides.api.ops.api import deps
 from fides.api.ops.api.v1.scope_registry import (
     DATASET_CREATE_OR_UPDATE,
@@ -30,6 +31,7 @@
 )
 from fides.api.ops.api.v1.urn_registry import (
     DATASET_BY_KEY,
+    DATASET_CONFIGS,
     DATASET_VALIDATE,
     DATASETS,
     V1_URL_PREFIX,
@@ -51,6 +53,7 @@
 from fides.api.ops.schemas.api import BulkUpdateFailed
 from fides.api.ops.schemas.dataset import (
     BulkPutDataset,
+    DatasetConfigCtlDataset,
     DatasetTraversalDetails,
     ValidateDatasetResponse,
     validate_data_categories_against_db,
@@ -157,6 +160,64 @@ def validate_dataset(
     )
 
 
+@router.patch(
+    DATASET_CONFIGS,
+    dependencies=[Security(verify_oauth_client, scopes=[DATASET_CREATE_OR_UPDATE])],
+    status_code=HTTP_200_OK,
+    response_model=BulkPutDataset,
+)
+def patch_dataset_configs(
+    dataset_pairs: conlist(DatasetConfigCtlDataset, max_items=50),  # type: ignore
+    db: Session = Depends(deps.get_db),
+    connection_config: ConnectionConfig = Depends(_get_connection_config),
+) -> BulkPutDataset:
+    """
+    Endpoint to create or update DatasetConfigs by passing in pairs of:
+    1) A DatasetConfig fides_key
+    2) The corresponding CtlDataset fides_key which stores the bulk of the actual dataset
+
+    Currently this endpoint looks up the ctl dataset and writes its contents back to the DatasetConfig.dataset
+    field for backwards compatibility but soon DatasetConfig.dataset will go away.
+
+    """
+    created_or_updated: List[Dataset] = []
+    failed: List[BulkUpdateFailed] = []
+    logger.info("Starting bulk upsert for {} Dataset Configs", len(dataset_pairs))
+
+    for dataset_pair in dataset_pairs:
+        logger.info(
+            "Finding ctl_dataset with key '{}'", dataset_pair.ctl_dataset_fides_key
+        )
+        ctl_dataset: CtlDataset = (
+            db.query(CtlDataset)
+            .filter_by(fides_key=dataset_pair.ctl_dataset_fides_key)
+            .first()
+        )
+        fetched_dataset: Dataset = Dataset.from_orm(ctl_dataset)
+
+        data = {
+            "connection_config_id": connection_config.id,
+            "fides_key": dataset_pair.fides_key,
+            "dataset": fetched_dataset.dict(),
+            "ctl_dataset_id": ctl_dataset.id,
+        }
+
+        create_or_update_dataset(
+            connection_config,
+            created_or_updated,
+            data,
+            fetched_dataset,
+            db,
+            failed,
+            DatasetConfig.create_or_update,
+        )
+
+    return BulkPutDataset(
+        succeeded=created_or_updated,
+        failed=failed,
+    )
+
+
 @router.patch(
     DATASETS,
     dependencies=[Security(verify_oauth_client, scopes=[DATASET_CREATE_OR_UPDATE])],
@@ -172,10 +233,10 @@ def patch_datasets(
     Given a list of dataset elements, create or update corresponding Dataset objects
     or report failure
 
-    Use for bulk creating and/or updating datasets.
+    Use for bulk creating and/or updating DatasetConfig resources.
 
-    If the fides_key for a given dataset exists, it will be treated as an update.
-    Otherwise, a new dataset will be created.
+    If the fides_key for a given DatasetConfig exists, it will be treated as an update.
+    Otherwise, a new DatasetConfig will be created.
     """
 
     created_or_updated: List[Dataset] = []
@@ -198,7 +259,13 @@ def patch_datasets(
             "dataset": dataset.dict(),
         }
         create_or_update_dataset(
-            connection_config, created_or_updated, data, dataset, db, failed
+            connection_config,
+            created_or_updated,
+            data,
+            dataset,
+            db,
+            failed,
+            DatasetConfig.upsert_with_ctl_dataset,
         )
     return BulkPutDataset(
         succeeded=created_or_updated,
@@ -249,9 +316,10 @@ async def patch_yaml_datasets(
                 connection_config,
                 created_or_updated,
                 data,
-                yaml_request_body,
+                Dataset(**dataset),
                 db,
                 failed,
+                DatasetConfig.upsert_with_ctl_dataset,
             )
     return BulkPutDataset(
         succeeded=created_or_updated,
@@ -263,16 +331,17 @@ def create_or_update_dataset(
     connection_config: ConnectionConfig,
     created_or_updated: List[Dataset],
     data: dict,
-    dataset: dict,
+    dataset: Dataset,
     db: Session,
     failed: List[BulkUpdateFailed],
+    create_method: Callable,
 ) -> None:
     try:
         if connection_config.connection_type == ConnectionType.saas:
             _validate_saas_dataset(connection_config, dataset)  # type: ignore
         # Try to find an existing DatasetConfig matching the given connection & key
-        dataset_config = DatasetConfig.create_or_update(db, data=data)
-        created_or_updated.append(dataset_config.dataset)
+        dataset_config = create_method(db, data=data)
+        created_or_updated.append(dataset_config.ctl_dataset)
     except (
         SaaSConfigNotFoundException,
         ValidationError,
@@ -340,7 +409,7 @@ def get_datasets(
     params: Params = Depends(),
     connection_config: ConnectionConfig = Depends(_get_connection_config),
 ) -> AbstractPage[Dataset]:
-    """Returns all datasets in the database."""
+    """Returns all DatasetConfig datasets in the database."""
 
     logger.info(
         "Finding all datasets for connection '{}' with pagination params {}",
@@ -357,7 +426,7 @@ def get_datasets(
     # paginated query is handled by paginate()
     paginated_results = paginate(dataset_configs, params=params)
     paginated_results.items = [  # type: ignore
-        dataset_config.dataset for dataset_config in paginated_results.items  # type: ignore
+        dataset_config.ctl_dataset for dataset_config in paginated_results.items  # type: ignore
     ]
     return paginated_results
 
@@ -389,7 +458,7 @@ def get_dataset(
             status_code=HTTP_404_NOT_FOUND,
             detail=f"No dataset with fides_key '{fides_key}' and connection key {connection_config.key}'",
         )
-    return dataset_config.dataset
+    return dataset_config.ctl_dataset
 
 
 @router.delete(
@@ -403,7 +472,7 @@ def delete_dataset(
     db: Session = Depends(deps.get_db),
     connection_config: ConnectionConfig = Depends(_get_connection_config),
 ) -> None:
-    """Removes the dataset based on the given key."""
+    """Removes the DatasetConfig based on the given key."""
 
     logger.info(
         "Finding dataset '{}' for connection '{}'", fides_key, connection_config.key

src/fides/api/ops/api/v1/endpoints/saas_config_endpoints.py

@@ -334,6 +334,7 @@ def instantiate_connection_from_template(
         template_values.instance_key,
         saas_connector_type,
     )
+
     return SaasConnectionTemplateResponse(
-        connection=connection_config, dataset=dataset_config.dataset
+        connection=connection_config, dataset=dataset_config.ctl_dataset
     )