Add Flemish bathymetry partition workflow
GeoIntel CI / docs-smoke (push) Canceled after 0s
GeoIntel CI / contract-smoke (push) Canceled after 0s

This commit is contained in:
Codex
2026-07-17 16:15:29 +02:00
parent 8ef42a23e5
commit 132c4feed8
34 changed files with 1862 additions and 27 deletions
@@ -15,8 +15,10 @@ from shapely.geometry import Point, box, mapping
from app.core.config import Settings, get_settings
from app.core.errors import AppError
from app.models import Area, Dataset, Project
from app.models import Area, Dataset, DatasetVersion, Project
from app.schemas.bathymetry import (
BathymetryPartitionFinalizeRequest,
BathymetryPartitionFinalizationResult,
BathymetryProfileAcquireRequest,
BathymetryProfileAcquisitionResult,
BathymetrySourceRead,
@@ -70,16 +72,16 @@ class BathymetryProfileAcquisitionService:
"vertical_reference": "LAT",
"horizontal_crs": "bronafhankelijk; expliciet per WCS-respons",
"native_resolution": "20 x 20 m",
"integration_status": "available_not_integrated",
"integration_status": "probe_only",
"acquisition_supported": False,
"configured": False,
"service_url": "https://bathy.agentschapmdk.be/spatialfusionserver/services/ows/wcs/WCS_Public",
"service_url": "https://bathy.agentschapmdk.be/spatialfusionserver/services/ows/wcs/EL_wcs",
"catalog_url": "https://www.vlaanderen.be/datavindplaats/catalogus/dieptemodel-van-de-zeebodem-belgisch-continentaal-plat-noordzee",
"attribution": "Agentschap Maritieme Dienstverlening en Kust",
"license_note": "Zie de officiële datasetmetadata en gebruiksvoorwaarden.",
"limitation_message": (
"Niet operationeel in GeoIntel totdat WCS, maritieme begrenzing, tegels, LAT-semantiek en "
"servercertificaten in een live integratietest zijn gevalideerd."
"Alleen een read-only GetCapabilities-probe is beschikbaar. Rasteracquisitie blijft uit totdat "
"WCS, maritieme begrenzing, tegels, LAT-semantiek en servercertificaten live zijn gevalideerd."
),
},
{
@@ -404,6 +406,8 @@ class BathymetryProfileAcquisitionService:
raw_features: list[dict[str, Any]],
scope_geometry,
watercourse_names: dict[int, dict[str, str | None]],
*,
partition_properties: dict[str, Any] | None = None,
) -> tuple[dict[str, Any], dict[str, Any]]:
normalized: list[dict[str, Any]] = []
dates: list[str] = []
@@ -465,6 +469,7 @@ class BathymetryProfileAcquisitionService:
"provider": BathymetryProfileAcquisitionService.PROVIDER,
"measurement_semantics": "historical_cross_section_profile_point",
"vertical_reference": "document-specific",
**(partition_properties or {}),
},
"geometry": mapping(point),
}
@@ -487,6 +492,17 @@ class BathymetryProfileAcquisitionService:
},
)
@staticmethod
def _municipality_name(area: Area | None) -> str | None:
if area is None:
return None
normalized = str(area.name or "").strip()
prefix = "Gemeente "
if not normalized.casefold().startswith(prefix.casefold()):
return None
municipality = normalized[len(prefix) :].split(" - ", 1)[0].strip()
return municipality or None
@staticmethod
def _cached_dataset(db, project_id: UUID, filename: str) -> Dataset | None:
return (
@@ -541,6 +557,8 @@ class BathymetryProfileAcquisitionService:
scope_geometry = BathymetryProfileAcquisitionService._scope_geometry(
db, project_id, payload.area_id, bbox_values
)
area = db.get(Area, payload.area_id) if payload.area_id else None
municipality = BathymetryProfileAcquisitionService._municipality_name(area)
exact_bbox = tuple(float(value) for value in scope_geometry.bounds)
request_identity = {
"provider": BathymetryProfileAcquisitionService.PROVIDER,
@@ -568,7 +586,16 @@ class BathymetryProfileAcquisitionService:
vhag_codes, resolved_settings, opener
)
feature_collection, summary = BathymetryProfileAcquisitionService._normalize_features(
raw_features, scope_geometry, names
raw_features,
scope_geometry,
names,
partition_properties={
"partition_area_id": str(area.id),
"partition_area_name": area.name,
**({"municipality": municipality} if municipality else {}),
}
if area
else None,
)
if summary["profile_count"] == 0:
raise AppError(
@@ -585,6 +612,11 @@ class BathymetryProfileAcquisitionService:
"theme": "bathymetry",
"layer_name": "bathymetry_profiles",
"coverage_scope": "municipality" if payload.area_id else "bounded_selection",
"partition_area_id": str(area.id) if area else None,
"partition_area_name": area.name if area else None,
"municipality": municipality,
"partitioned_source_audit": False,
"regional_partitions_complete": False,
"bbox_epsg4326": list(exact_bbox),
**summary,
"selection_aggregation": {
@@ -669,3 +701,160 @@ class BathymetryProfileAcquisitionService:
)
persisted = db.get(Dataset, dataset.id)
return BathymetryProfileAcquisitionService._result(persisted, reused=False)
@staticmethod
def finalize_partitions(
db,
project_id: UUID,
payload: BathymetryPartitionFinalizeRequest,
) -> dict[str, Any]:
if not db.get(Project, project_id):
raise AppError(code="PROJECT_NOT_FOUND", message="Project not found", status_code=404)
expected_area_ids = set(payload.expected_area_ids)
dataset_ids = set(payload.dataset_ids)
no_profile_area_ids = set(payload.no_profile_area_ids)
areas: dict[UUID, Area] = {}
for area_id in expected_area_ids:
area = db.get(Area, area_id)
if area is None or area.project_id != project_id:
raise AppError(
code="BATHYMETRY_PARTITION_AREA_INVALID",
message="Every expected partition Area must belong to the project",
details={"area_id": str(area_id)},
status_code=400,
)
if BathymetryProfileAcquisitionService._municipality_name(area) is None:
raise AppError(
code="BATHYMETRY_PARTITION_AREA_INVALID",
message="Bathymetry partitions must use persisted municipality Areas",
details={"area_id": str(area_id), "area_name": area.name},
status_code=400,
)
areas[area_id] = area
if not no_profile_area_ids.issubset(expected_area_ids):
raise AppError(
code="BATHYMETRY_PARTITION_MANIFEST_INVALID",
message="No-profile partitions must be part of the expected Area set",
status_code=400,
)
datasets: list[Dataset] = []
data_area_ids: set[UUID] = set()
for dataset_id in payload.dataset_ids:
dataset = db.get(Dataset, dataset_id)
if (
dataset is None
or dataset.project_id != project_id
or dataset.source_name != BathymetryProfileAcquisitionService.PROVIDER
or dataset.status != "ready"
or dataset.area_id not in expected_area_ids
):
raise AppError(
code="BATHYMETRY_PARTITION_DATASET_INVALID",
message="Every partition Dataset must be a ready VHA profile Dataset scoped to an expected Area",
details={"dataset_id": str(dataset_id)},
status_code=400,
)
if dataset.area_id in data_area_ids:
raise AppError(
code="BATHYMETRY_PARTITION_DATASET_DUPLICATE",
message="A complete manifest may reference only one profile Dataset per Area",
details={"area_id": str(dataset.area_id)},
status_code=400,
)
data_area_ids.add(dataset.area_id)
datasets.append(dataset)
accounted_area_ids = data_area_ids.union(no_profile_area_ids)
if accounted_area_ids != expected_area_ids:
raise AppError(
code="BATHYMETRY_PARTITION_MANIFEST_INCOMPLETE",
message="Every expected Area must have one ready Dataset or an explicit no-profile result",
details={
"missing_area_ids": sorted(str(value) for value in expected_area_ids - accounted_area_ids),
"unexpected_area_ids": sorted(str(value) for value in accounted_area_ids - expected_area_ids),
},
status_code=400,
)
profile_count = 0
document_count = 0
structured_depth_count = 0
dates_min: list[str] = []
dates_max: list[str] = []
shared_metadata = {
"partition_scope_key": payload.partition_scope_key,
"partition_count": len(expected_area_ids),
"data_partition_count": len(datasets),
"no_profile_partition_count": len(no_profile_area_ids),
"partition_manifest_sha256": payload.manifest_sha256,
"partition_manifest_observed_at": payload.observed_at.isoformat(),
"partitioned_source_audit": True,
"regional_partitions_complete": True,
}
shared_provenance = {
"partition_manifest_sha256": payload.manifest_sha256,
"partition_manifest_observed_at": payload.observed_at.isoformat(),
"partition_scope_key": payload.partition_scope_key,
"regional_partitions_complete": True,
"no_profile_area_ids": sorted(str(value) for value in no_profile_area_ids),
}
for dataset in datasets:
source_metadata = dict(dataset.source_metadata or {})
provenance_metadata = dict(dataset.provenance_metadata or {})
profile_count += int(source_metadata.get("profile_count") or 0)
document_count += int(source_metadata.get("document_count") or 0)
structured_depth_count += int(source_metadata.get("structured_depth_count") or 0)
if source_metadata.get("measurement_date_min"):
dates_min.append(str(source_metadata["measurement_date_min"]))
if source_metadata.get("measurement_date_max"):
dates_max.append(str(source_metadata["measurement_date_max"]))
area = areas[dataset.area_id]
source_metadata.update(
{
**shared_metadata,
"coverage_scope": payload.partition_scope_key,
"partition_area_id": str(area.id),
"partition_area_name": area.name,
"municipality": BathymetryProfileAcquisitionService._municipality_name(area),
}
)
provenance_metadata.update(shared_provenance)
dataset.source_metadata = source_metadata
dataset.provenance_metadata = provenance_metadata
if dataset_ids:
versions = (
db.query(DatasetVersion)
.filter(DatasetVersion.dataset_id.in_(dataset_ids))
.all()
)
for version in versions:
version.source_metadata = dict(
next(dataset.source_metadata for dataset in datasets if dataset.id == version.dataset_id)
)
version.provenance_metadata = dict(
next(dataset.provenance_metadata for dataset in datasets if dataset.id == version.dataset_id)
)
db.commit()
return BathymetryPartitionFinalizationResult(
partition_scope_key=payload.partition_scope_key,
regional_partitions_complete=True,
partition_count=len(expected_area_ids),
data_partition_count=len(datasets),
no_profile_partition_count=len(no_profile_area_ids),
profile_count=profile_count,
document_count=document_count,
structured_depth_count=structured_depth_count,
measurement_date_min=min(dates_min) if dates_min else None,
measurement_date_max=max(dates_max) if dates_max else None,
dataset_ids=payload.dataset_ids,
manifest_sha256=payload.manifest_sha256,
observed_at=payload.observed_at,
limitation_message=BathymetryProfileAcquisitionService.LIMITATION,
).model_dump(mode="json")