geointel/scripts/provision_mol_bathymetry_profiles.py

"""Provision official VHA cross-section profile points for the persisted Mol Area.

The operator uses only canonical GeoIntel API endpoints. The backend queries the
official VHA service, clips points against the exact persisted Area geometry,
stores the GeoJSON artifact and persists VectorFeature rows through DatasetService.
"""

from __future__ import annotations

import argparse
import json
import os
from typing import Any, Iterable

import requests


DEFAULT_API_URL = "http://127.0.0.1:8000"
DEFAULT_PROJECT_NAME = "Kempen Regional Workbench"
DEFAULT_AREA_FRAGMENT = "Gemeente Mol"


def parse_args() -> argparse.Namespace:
    parser = argparse.ArgumentParser(description="Provision official VHA cross-section profiles for Mol.")
    parser.add_argument("--base-url", default=os.environ.get("GEOINTEL_INTERNAL_API_URL", DEFAULT_API_URL))
    parser.add_argument("--project-name", default=DEFAULT_PROJECT_NAME)
    parser.add_argument("--area-name", default=DEFAULT_AREA_FRAGMENT)
    parser.add_argument("--timeout", type=int, default=900)
    parser.add_argument("--force", action="store_true")
    return parser.parse_args()


def unwrap(response: requests.Response) -> Any:
    response.raise_for_status()
    payload = response.json()
    if not isinstance(payload, dict) or "data" not in payload:
        raise RuntimeError(f"Non-canonical API response from {response.url}")
    return payload["data"]


def coordinates(geometry: dict[str, Any]) -> Iterable[tuple[float, float]]:
    def walk(value: Any):
        if isinstance(value, list) and len(value) >= 2 and all(isinstance(item, (int, float)) for item in value[:2]):
            yield float(value[0]), float(value[1])
            return
        if isinstance(value, list):
            for child in value:
                yield from walk(child)

    yield from walk(geometry.get("coordinates", []))


def geometry_bbox(geometry: dict[str, Any]) -> dict[str, float | str]:
    points = list(coordinates(geometry))
    if not points:
        raise RuntimeError("Persisted Area geometry contains no coordinates")
    xs = [point[0] for point in points]
    ys = [point[1] for point in points]
    return {
        "min_x": min(xs),
        "min_y": min(ys),
        "max_x": max(xs),
        "max_y": max(ys),
        "crs": "EPSG:4326",
    }


def main() -> int:
    args = parse_args()
    base_url = args.base_url.rstrip("/")
    session = requests.Session()
    session.headers.update({"User-Agent": "GeoIntel-Bathymetry-Operator/1.0"})

    projects = unwrap(session.get(f"{base_url}/api/v1/projects", params={"limit": 200, "offset": 0}, timeout=60))["items"]
    project = next((item for item in projects if item["name"] == args.project_name), None)
    if project is None:
        raise RuntimeError(f"Project {args.project_name!r} was not found")

    areas = unwrap(
        session.get(
            f"{base_url}/api/v1/projects/{project['id']}/areas",
            params={"limit": 200, "offset": 0},
            timeout=60,
        )
    )["items"]
    fragment = args.area_name.casefold()
    area = next((item for item in areas if fragment in item["name"].casefold()), None)
    if area is None:
        raise RuntimeError(f"Area containing {args.area_name!r} was not found")
    bbox = geometry_bbox(area["geometry"])

    job = unwrap(
        session.post(
            f"{base_url}/api/v1/projects/{project['id']}/datasets/bathymetry/profiles/acquire",
            json={
                "bbox": bbox,
                "area_id": area["id"],
                "force_refresh": args.force,
            },
            timeout=args.timeout,
        )
    )
    if job.get("status") != "success" or not job.get("output_dataset_id"):
        raise RuntimeError(f"Bathymetry profile acquisition failed: {job.get('error_message') or job}")
    result = job.get("result_json") or {}
    dataset_id = job["output_dataset_id"]
    selection = unwrap(
        session.post(
            f"{base_url}/api/v1/projects/{project['id']}/datasets/{dataset_id}/vector/select",
            json={"bbox": bbox, "area_id": area["id"], "limit": 1000},
            timeout=args.timeout,
        )
    )
    if selection.get("total_feature_count") != result.get("profile_count"):
        raise RuntimeError(
            "Persisted vector selection count does not match the exact acquired profile count "
            f"({selection.get('total_feature_count')} != {result.get('profile_count')})"
        )
    metrics = {
        item["metric_key"]: item
        for item in (selection.get("summary") or {}).get("metrics", [])
        if isinstance(item, dict) and item.get("metric_key")
    }
    if metrics.get("profile_count", {}).get("metric_value") != result.get("profile_count"):
        raise RuntimeError("Bathymetry profile summary does not expose the persisted profile count")

    print(
        json.dumps(
            {
                "status": "ok",
                "project_id": project["id"],
                "area_id": area["id"],
                "area_name": area["name"],
                "bbox": bbox,
                "dataset_id": dataset_id,
                "reused": bool(result.get("reused")),
                "profile_count": result.get("profile_count"),
                "document_count": result.get("document_count"),
                "structured_depth_count": result.get("structured_depth_count"),
                "watercourse_count": result.get("watercourse_count"),
                "measurement_date_min": result.get("measurement_date_min"),
                "measurement_date_max": result.get("measurement_date_max"),
                "metrics": list(metrics.values()),
                "limitation_message": result.get("limitation_message"),
            },
            ensure_ascii=False,
            indent=2,
        )
    )
    return 0


if __name__ == "__main__":
    raise SystemExit(main())