Persist map selections as derived datasets

backend/app/services/vector_operations_service.py

@@ -2,6 +2,7 @@ from __future__ import annotations
 
 import json
 import uuid
+from datetime import datetime, timezone
 from pathlib import Path
 from typing import Any
 
@@ -15,9 +16,11 @@ from sqlalchemy.orm import Session
 
 from app.core.errors import AppError
 from app.models import Area, Dataset
+from app.schemas.dataset import DatasetCreateResponse
 from app.schemas.operations import VectorOperationResult
 from app.services.geojson_service import parse_geojson_payload
 from app.services.storage_service import StorageService
+from app.services.vector_feature_service import VectorFeatureService
 
 
 class VectorOperationsService:
@@ -276,6 +279,125 @@ class VectorOperationsService:
             default_name="vector_intersect",
         )
 
+    @staticmethod
+    def derive_selection_dataset(
+        db: Session,
+        dataset_id: uuid.UUID,
+        bbox: dict[str, Any],
+        limit: int = 250,
+        output_name: str | None = None,
+    ) -> DatasetCreateResponse:
+        source_dataset = db.get(Dataset, dataset_id)
+        if not source_dataset:
+            raise AppError(code="DATASET_NOT_FOUND", message="Dataset not found", status_code=404)
+        VectorOperationsService._require_vector_dataset(source_dataset)
+
+        selection = VectorFeatureService.select_features_by_bbox(db, dataset_id=dataset_id, bbox=bbox, limit=limit)
+        if selection["feature_count"] <= 0:
+            raise AppError(
+                code="VECTOR_OPERATION_EMPTY_RESULT",
+                message="Selection produced no output features",
+                status_code=422,
+            )
+
+        feature_collection = VectorOperationsService._selection_geojson_for_derived_dataset(
+            selection["geojson"],
+            source_dataset_id=dataset_id,
+        )
+        derived_id = VectorOperationsService._persist_derived_dataset(
+            db=db,
+            source_dataset=source_dataset,
+            source_id=dataset_id,
+            operation="selection",
+            feature_collection=feature_collection,
+            output_name=output_name,
+            default_name="map_selection",
+            dataset_role="derived",
+            source_name="map_selection",
+            source_metadata={
+                "selection_bbox": selection["selection_bbox"],
+                "feature_count": selection["feature_count"],
+                "limit": selection["limit"],
+                "truncated": selection["truncated"],
+                "source_table": "vector_features",
+            },
+            provenance_metadata={
+                "operation": "map_bbox_selection",
+                "source_dataset_id": str(dataset_id),
+                "source_table": "vector_features",
+                "selection_bbox": selection["selection_bbox"],
+            },
+            metadata_extra={
+                "selection_bbox": selection["selection_bbox"],
+                "source_feature_count": selection["feature_count"],
+                "selection_limit": selection["limit"],
+                "selection_truncated": selection["truncated"],
+                "source_dataset_id": str(dataset_id),
+                "source_table": "vector_features",
+            },
+            persist_vector_features=True,
+        )
+        derived = db.get(Dataset, derived_id)
+        if not derived:
+            raise AppError(code="DATASET_NOT_FOUND", message="Derived dataset was not persisted", status_code=500)
+        metadata = derived.metadata_json or {}
+        return DatasetCreateResponse(
+            id=derived.id,
+            name=derived.name,
+            dataset_type=derived.dataset_type,
+            source=derived.source,
+            dataset_role=derived.dataset_role,
+            source_name=derived.source_name,
+            reference_layer_name=derived.reference_layer_name,
+            source_metadata=derived.source_metadata,
+            provenance_metadata=derived.provenance_metadata,
+            imported_at=derived.imported_at,
+            project_id=derived.project_id,
+            area_id=derived.area_id,
+            storage_path=derived.storage_path,
+            original_filename=derived.original_filename,
+            stored_filename=derived.stored_filename,
+            content_type=derived.content_type,
+            size_bytes=derived.size_bytes,
+            checksum_sha256=derived.checksum_sha256,
+            crs=derived.crs,
+            bounds_json=derived.bounds_json,
+            resolution_json=derived.resolution_json,
+            bands_json=derived.bands_json,
+            metadata_json=derived.metadata_json,
+            vector_summary=None,
+            status=derived.status,
+            derived_from_dataset_id=derived.derived_from_dataset_id,
+            created_at=derived.created_at,
+            feature_count=metadata.get("feature_count") if isinstance(metadata, dict) else None,
+        )
+
+    @staticmethod
+    def _selection_geojson_for_derived_dataset(payload: dict[str, Any], source_dataset_id: uuid.UUID) -> dict[str, Any]:
+        features = payload.get("features")
+        if payload.get("type") != "FeatureCollection" or not isinstance(features, list):
+            raise AppError(code="INVALID_GEOJSON", message="Selection payload must be a FeatureCollection", status_code=500)
+
+        output_features: list[dict[str, Any]] = []
+        for feature in features:
+            if not isinstance(feature, dict):
+                continue
+            properties = dict(feature.get("properties") or {})
+            source_vector_feature_id = properties.pop("vector_feature_id", feature.get("id"))
+            properties.pop("dataset_id", None)
+            properties["source_dataset_id"] = str(source_dataset_id)
+            if source_vector_feature_id is not None:
+                properties["source_vector_feature_id"] = str(source_vector_feature_id)
+            output_features.append(
+                {
+                    "type": "Feature",
+                    "geometry": feature.get("geometry"),
+                    "properties": properties,
+                }
+            )
+
+        return {"type": "FeatureCollection", "features": output_features}
+
     @staticmethod
     def _persist_derived_dataset(
         db: Session,
@@ -285,6 +407,12 @@ class VectorOperationsService:
         feature_collection: dict[str, Any],
         output_name: str | None,
         default_name: str,
+        dataset_role: str = "derived",
+        source_name: str | None = None,
+        source_metadata: dict[str, Any] | None = None,
+        provenance_metadata: dict[str, Any] | None = None,
+        metadata_extra: dict[str, Any] | None = None,
+        persist_vector_features: bool = False,
     ) -> uuid.UUID:
         derived_id = uuid.uuid4()
         output_name_value = f"{(output_name or default_name)}.geojson"
@@ -302,6 +430,8 @@ class VectorOperationsService:
         )
 
         metadata = parse_geojson_payload(json.dumps(feature_collection, ensure_ascii=False, separators=(",", ":")))
+        if metadata_extra:
+            metadata.update(metadata_extra)
         derived_dataset = Dataset(
             id=derived_id,
             project_id=source_dataset.project_id,
@@ -309,6 +439,11 @@ class VectorOperationsService:
             name=output_name_value,
             dataset_type="vector",
             source=f"operation:{operation}",
+            dataset_role=dataset_role,
+            source_name=source_name,
+            source_metadata=source_metadata,
+            provenance_metadata=provenance_metadata,
+            imported_at=datetime.now(timezone.utc),
             storage_path=storage_info["storage_path"],
             original_filename=storage_info["original_filename"],
             stored_filename=storage_info["stored_filename"],
@@ -326,4 +461,10 @@ class VectorOperationsService:
         db.add(derived_dataset)
         db.commit()
         db.refresh(derived_dataset)
+        if persist_vector_features:
+            VectorFeatureService.persist_geojson_features(
+                db=db,
+                dataset_id=derived_dataset.id,
+                payload=feature_collection,
+            )
         return derived_id