geointel/backend/tests/test_rc4_national_scope_operator.py

from __future__ import annotations

import json
import sys
import zipfile
from pathlib import Path

import pytest
from shapely.geometry import box, mapping, shape


ROOT = Path(__file__).parents[2]
SCRIPTS = ROOT / "scripts"
if str(SCRIPTS) not in sys.path:
    sys.path.insert(0, str(SCRIPTS))

import provision_belgium_north_sea_scope as operator  # noqa: E402


def marine_feature(identifier: str, geometry):
    return {
        "type": "Feature",
        "id": identifier,
        "geometry": mapping(geometry),
        "properties": {"MarineReportingUnitId": identifier},
    }


def test_marine_legal_scopes_are_derived_from_official_reporting_units() -> None:
    reporting_units = {
        "type": "FeatureCollection",
        "features": [
            marine_feature("ANS-BE-MS-1", box(0, 0, 10, 10)),
            marine_feature("ANS-BE-AA-CW", box(0, 0, 2, 2)),
            marine_feature("ANS-BE-AA-TEW", box(0, 2, 2, 4)),
            marine_feature("ANS-BE-AA-OFFSHORE", box(2, 0, 10, 10)),
        ],
    }

    payload = operator.derive_marine_scope_payload(reporting_units)
    by_zone = {
        feature["properties"]["coverage_zone"]: feature
        for feature in payload["features"]
    }

    assert set(by_zone) == {
        "belgian_north_sea",
        "territorial_sea",
        "exclusive_economic_zone",
        "continental_shelf",
    }
    assert shape(by_zone["territorial_sea"]["geometry"]).area == pytest.approx(8.0)
    assert shape(by_zone["exclusive_economic_zone"]["geometry"]).equals(
        shape(by_zone["continental_shelf"]["geometry"])
    )
    assert (
        by_zone["exclusive_economic_zone"]["properties"]["legal_domain"]
        != by_zone["continental_shelf"]["properties"]["legal_domain"]
    )
    assert by_zone["territorial_sea"]["properties"]["derived_from_reporting_unit_ids"] == [
        "ANS-BE-AA-CW",
        "ANS-BE-AA-TEW",
    ]


def test_marine_scope_derivation_fails_when_a_required_unit_is_missing() -> None:
    with pytest.raises(RuntimeError, match="ANS-BE-AA-CW"):
        operator.derive_marine_scope_payload(
            {
                "type": "FeatureCollection",
                "features": [
                    marine_feature("ANS-BE-MS-1", box(0, 0, 10, 10)),
                    marine_feature("ANS-BE-AA-TEW", box(0, 2, 2, 4)),
                    marine_feature("ANS-BE-AA-OFFSHORE", box(2, 0, 10, 10)),
                ],
            }
        )


def test_archive_extraction_accepts_one_safe_geopackage_and_rejects_traversal(tmp_path: Path) -> None:
    archive = tmp_path / "adminvector.zip"
    with zipfile.ZipFile(archive, "w") as handle:
        handle.writestr("release/adminvector.gpkg", b"sqlite-bytes")

    result = operator.extract_single_geopackage(archive, tmp_path / "output")
    assert result.read_bytes() == b"sqlite-bytes"

    unsafe = tmp_path / "unsafe.zip"
    with zipfile.ZipFile(unsafe, "w") as handle:
        handle.writestr("../adminvector.gpkg", b"unsafe")
    with pytest.raises(RuntimeError, match="unsafe"):
        operator.extract_single_geopackage(unsafe, tmp_path / "unsafe-output")


class FakeResponse:
    def __init__(self, payload):
        self.payload = payload
        self.content = json.dumps(payload).encode("utf-8")

    def raise_for_status(self):
        return None

    def json(self):
        return self.payload


class FakeSession:
    def __init__(self, pages):
        self.pages = list(pages)
        self.calls = []

    def get(self, url, params, timeout):
        self.calls.append({"url": url, "params": params, "timeout": timeout})
        return FakeResponse(self.pages.pop(0))


def test_wfs_fetch_is_allowlisted_paginated_and_complete() -> None:
    pages = [
        {
            "type": "FeatureCollection",
            "numberMatched": 2,
            "features": [{"type": "Feature", "id": "unit.1", "geometry": None, "properties": {}}],
        },
        {
            "type": "FeatureCollection",
            "numberMatched": 2,
            "features": [{"type": "Feature", "id": "unit.2", "geometry": None, "properties": {}}],
        },
    ]
    session = FakeSession(pages)
    payload = operator.fetch_wfs_layer(
        session,
        service_url=operator.RBINS_MRU_WFS_URL,
        layer_name=operator.RBINS_MRU_LAYER,
        timeout=30,
        page_size=1,
    )

    assert [feature["id"] for feature in payload["features"]] == ["unit.1", "unit.2"]
    assert [call["params"]["startIndex"] for call in session.calls] == [0, 1]
    assert all(call["params"]["srsName"] == "EPSG:4326" for call in session.calls)

    with pytest.raises(RuntimeError, match="allowlist"):
        operator.fetch_wfs_layer(
            FakeSession([]),
            service_url=operator.RBINS_MSP_WFS_URL,
            layer_name="untrusted:layer",
            timeout=30,
        )


def test_operator_is_packaged_and_guarded_by_readiness() -> None:
    dockerfile = (ROOT / "deploy" / "unraid" / "Dockerfile.all-in-one").read_text(encoding="utf-8")
    readiness = (ROOT / "scripts" / "run_readiness_check.sh").read_text(encoding="utf-8")
    source = (ROOT / "scripts" / "provision_belgium_north_sea_scope.py").read_text(encoding="utf-8")

    assert "COPY scripts/provision_belgium_north_sea_scope.py /app/scripts/" in dockerfile
    assert "py_compile scripts/provision_belgium_north_sea_scope.py" in readiness
    assert "/datasets/upload" in source
    assert "from app.models" not in source
    assert "INSERT INTO vector_features" not in source