Add map area selection extraction

backend/app/api/routes/datasets.py

@@ -24,6 +24,9 @@ from app.schemas import (
     VectorBufferRequest,
     VectorClipRequest,
     VectorIntersectRequest,
+    VectorSelectionBBox,
+    VectorSelectionRequest,
+    VectorSelectionResponse,
 )
 from app.schemas.job import JobCreate
 from app.schemas.dataset import DatasetCreateResponse
@@ -31,6 +34,7 @@ from app.schemas.operations import VectorOperationResult
 from app.services.job_service import JobService
 from app.services.raster_operations_service import RasterOperationsService
 from app.services.vector_operations_service import VectorOperationsService
+from app.services.vector_feature_service import VectorFeatureService
 from app.services.dataset_service import DatasetService
 from app.utils.response import envelope
 
@@ -180,6 +184,27 @@ def vector_stats(
     return envelope(VectorOperationsService.stats(db, dataset_id))
 
 
+@router.post("/datasets/{dataset_id}/vector/select", response_model=dict)
+def select_vector_features(
+    project_id: UUID,
+    dataset_id: UUID,
+    payload: VectorSelectionRequest,
+    db: Session = Depends(get_db),
+):
+    dataset = DatasetService.get_dataset(db, dataset_id)
+    if dataset.project_id != project_id:
+        raise HTTPException(status_code=404, detail="Dataset not found")
+    if dataset.dataset_type not in {"vector", "geojson"}:
+        raise AppError(code="DATASET_NOT_VECTOR", message="Area selection requires a vector dataset", status_code=400)
+    result = VectorFeatureService.select_features_by_bbox(
+        db,
+        dataset_id=dataset_id,
+        bbox=payload.bbox.model_dump(),
+        limit=payload.limit,
+    )
+    return envelope(VectorSelectionResponse(**result).model_dump())
+
+
 @router.post("/datasets/{dataset_id}/vector/clip", status_code=201, response_model=dict)
 def clip_vector_dataset(
     project_id: UUID,

backend/app/schemas/__init__.py

@@ -71,6 +71,9 @@ from .operations import (
     VectorIntersectRequest,
     VectorOperationRequest,
     VectorOperationResult,
+    VectorSelectionBBox,
+    VectorSelectionRequest,
+    VectorSelectionResponse,
     VectorStatsRequest,
     VectorStatsResponse,
 )
@@ -122,6 +125,9 @@ __all__ = [
     "VectorIntersectRequest",
     "VectorOperationRequest",
     "VectorOperationResult",
+    "VectorSelectionBBox",
+    "VectorSelectionRequest",
+    "VectorSelectionResponse",
     "RasterClipRequest",
     "RasterStatsResponse",
     "RasterReprojectRequest",

backend/app/schemas/operations.py

@@ -1,6 +1,6 @@
 from __future__ import annotations
 
-from pydantic import BaseModel
+from pydantic import BaseModel, Field, field_validator
 
 
 class VectorOperationResult(BaseModel):
@@ -191,3 +191,31 @@ class VectorStatsResponse(BaseModel):
     geometry_type_summary: dict[str, int]
     bounds_json: dict | None
     crs: str | None = None
+
+
+class VectorSelectionBBox(BaseModel):
+    min_x: float
+    min_y: float
+    max_x: float
+    max_y: float
+    crs: str = "EPSG:4326"
+
+    @field_validator("crs")
+    @classmethod
+    def validate_crs(cls, value: str) -> str:
+        if value.upper() != "EPSG:4326":
+            raise ValueError("Only EPSG:4326 bbox selection is supported")
+        return "EPSG:4326"
+
+
+class VectorSelectionRequest(BaseModel):
+    bbox: VectorSelectionBBox
+    limit: int = Field(default=100, ge=1, le=1000)
+
+
+class VectorSelectionResponse(BaseModel):
+    selection_bbox: VectorSelectionBBox
+    feature_count: int
+    limit: int
+    truncated: bool
+    geojson: dict

backend/app/services/vector_feature_service.py

@@ -3,7 +3,10 @@ from __future__ import annotations
 from typing import Any
 from uuid import UUID
 
+from geoalchemy2.functions import ST_Intersects, ST_MakeEnvelope
 from geoalchemy2.shape import from_shape
+from geoalchemy2.shape import to_shape
+from shapely.geometry import mapping
 from shapely.geometry import shape
 from shapely.validation import make_valid
 
@@ -12,6 +15,117 @@ from app.models import VectorFeature
 
 
 class VectorFeatureService:
+    @staticmethod
+    def _normalize_selection_bbox(bbox: dict[str, Any]) -> dict[str, float | str]:
+        try:
+            min_x = float(bbox["min_x"])
+            min_y = float(bbox["min_y"])
+            max_x = float(bbox["max_x"])
+            max_y = float(bbox["max_y"])
+        except (KeyError, TypeError, ValueError) as exc:
+            raise AppError(
+                code="INVALID_SELECTION_BBOX",
+                message="Selection bbox must include numeric min_x, min_y, max_x and max_y values",
+                status_code=400,
+            ) from exc
+
+        crs = str(bbox.get("crs") or "EPSG:4326").upper()
+        if crs != "EPSG:4326":
+            raise AppError(
+                code="UNSUPPORTED_SELECTION_CRS",
+                message="Map selection currently supports EPSG:4326 bbox coordinates only",
+                details={"crs": crs},
+                status_code=400,
+            )
+        if min_x >= max_x or min_y >= max_y:
+            raise AppError(
+                code="INVALID_SELECTION_BBOX",
+                message="Selection bbox must have min_x < max_x and min_y < max_y",
+                status_code=400,
+            )
+        if min_x < -180 or max_x > 180 or min_y < -90 or max_y > 90:
+            raise AppError(
+                code="INVALID_SELECTION_BBOX",
+                message="Selection bbox is outside EPSG:4326 longitude/latitude bounds",
+                status_code=400,
+            )
+
+        return {"min_x": min_x, "min_y": min_y, "max_x": max_x, "max_y": max_y, "crs": "EPSG:4326"}
+
+    @staticmethod
+    def _row_to_geojson_feature(row: VectorFeature) -> dict[str, Any]:
+        geometry_value = row.geometry
+        try:
+            geometry = geometry_value if hasattr(geometry_value, "__geo_interface__") else to_shape(geometry_value)
+        except Exception as exc:
+            raise AppError(
+                code="INVALID_VECTOR_FEATURE_GEOMETRY",
+                message="Persisted vector feature geometry could not be converted to GeoJSON",
+                details={"vector_feature_id": str(row.id)},
+                status_code=500,
+            ) from exc
+
+        properties = dict(row.properties_json or {})
+        properties.update(
+            {
+                "vector_feature_id": str(row.id),
+                "dataset_id": str(row.dataset_id),
+                "source_feature_id": row.source_feature_id,
+                "feature_class": row.feature_class,
+            }
+        )
+
+        return {
+            "type": "Feature",
+            "id": str(row.id),
+            "geometry": mapping(geometry),
+            "properties": properties,
+        }
+
+    @staticmethod
+    def select_features_by_bbox(
+        db,
+        dataset_id: UUID,
+        bbox: dict[str, Any],
+        limit: int = 100,
+    ) -> dict[str, Any]:
+        normalized_bbox = VectorFeatureService._normalize_selection_bbox(bbox)
+        safe_limit = max(1, min(int(limit), 1000))
+
+        rows = (
+            db.query(VectorFeature)
+            .filter(VectorFeature.dataset_id == dataset_id)
+            .filter(
+                ST_Intersects(
+                    VectorFeature.geometry,
+                    ST_MakeEnvelope(
+                        normalized_bbox["min_x"],
+                        normalized_bbox["min_y"],
+                        normalized_bbox["max_x"],
+                        normalized_bbox["max_y"],
+                        4326,
+                    ),
+                )
+            )
+            .order_by(VectorFeature.created_at.asc())
+            .limit(safe_limit + 1)
+            .all()
+        )
+        truncated = len(rows) > safe_limit
+        selected_rows = rows[:safe_limit]
+        features = [VectorFeatureService._row_to_geojson_feature(row) for row in selected_rows]
+
+        return {
+            "selection_bbox": normalized_bbox,
+            "feature_count": len(features),
+            "limit": safe_limit,
+            "truncated": truncated,
+            "geojson": {
+                "type": "FeatureCollection",
+                "features": features,
+            },
+        }
+
     @staticmethod
     def persist_geojson_features(
         db,