school_compare/backend/tests/test_school_details.py

"""Regression tests for GET /api/schools/{urn}.

Schools with no performance rows (special post-16 institutions, sixth-form
centres, PRUs, brand-new schools) come back from the marts LEFT JOIN with
NaN in every numeric column. The endpoint must still serialize them — a NaN
that reaches Starlette's JSONResponse raises ValueError (allow_nan=False)
and the route 500s, which the frontend then renders as a 404.
"""

import numpy as np
import pandas as pd
import pytest
from fastapi.testclient import TestClient


def _no_results_school_df() -> pd.DataFrame:
    """One school row as produced by the marts query for a school with no
    performance data: GIAS/location fields partly populated, every
    results-linked column NaN (including year)."""
    return pd.DataFrame(
        [
            {
                "urn": 150275,
                "school_name": "West London Performing Arts Academy",
                "phase": "Secondary",
                "school_type": "Special post 16 institution",
                "trust_name": None,
                "religious_denomination": "Does not apply",
                "gender": None,
                "age_range": "16-25",
                "admissions_policy": None,
                "capacity": np.nan,
                "gias_total_pupils": np.nan,
                "headteacher_name": None,
                "website": None,
                "ofsted_grade": np.nan,
                "local_authority": "Ealing",
                "address": "268 Northfield Avenue, London, W5 4UB",
                "postcode": "W5 4UB",
                "latitude": 51.4986,
                "longitude": -0.3148,
                "year": np.nan,
                "total_pupils": np.nan,
                "eligible_pupils": np.nan,
                "rwm_expected_pct": np.nan,
            }
        ]
    )


@pytest.fixture()
def client(monkeypatch):
    from backend import app as app_module

    monkeypatch.setattr(app_module, "load_school_data", _no_results_school_df)
    monkeypatch.setattr(
        app_module, "get_supplementary_data", lambda db, urn: {}
    )
    return TestClient(app_module.app, raise_server_exceptions=False)


def test_school_without_performance_rows_returns_200(client):
    resp = client.get("/api/schools/150275")
    assert resp.status_code == 200, resp.text


def test_nan_gias_fields_serialize_as_null(client):
    info = client.get("/api/schools/150275").json()["school_info"]
    assert info["capacity"] is None
    assert info["total_pupils"] is None
    assert info["school_name"] == "West London Performing Arts Academy"


# ── Links out to the location layer ─────────────────────────────────────────
#
# School pages carried no link into the site at all: the only anchor on the
# template pointed at the school's own website, so ~27k pages received
# whatever authority the site had and sent it off-site. `places` is what the
# link module and the breadcrumb are built from.

def test_places_is_present_even_when_the_school_belongs_to_none(client):
    # This fixture's single school cannot clear any publish threshold, so the
    # honest answer is an empty list. The key must still be there: a missing
    # key and "no places" are different things to the page rendering it.
    body = client.get("/api/schools/150275").json()
    assert body["places"] == []


def test_places_names_only_pages_that_exist(monkeypatch):
    from backend import app as app_module
    from backend.places import MIN_SCHOOLS

    def _df():
        return pd.DataFrame([
            {
                "urn": 100000 + i,
                "school_name": f"Brentwood School {i}",
                "town": "Brentwood",
                "local_authority": "Essex",
                "postcode": "CM15 8AA",
                "phase": "Primary",
                "year": 202425,
                "rwm_expected_pct": 60.0,
                "attainment_8_score": np.nan,
                "ofsted_grade": 2.0,
                "ofsted_date": None,
            }
            for i in range(MIN_SCHOOLS)
        ])

    monkeypatch.setattr(app_module, "load_school_data", _df)
    monkeypatch.setattr(app_module, "get_supplementary_data", lambda db, urn: {})
    monkeypatch.setattr(app_module, "_place_registry", None)
    client = TestClient(app_module.app, raise_server_exceptions=False)

    places = client.get("/api/schools/100000").json()["places"]
    assert places, "a school in a published town must offer links"

    by_kind = {p["kind"]: p for p in places}
    assert by_kind["town"]["url"] == "/schools/brentwood"
    assert by_kind["authority"]["url"] == "/schools/authority/essex"

    # Every entry carries what the link text needs, and a count, so the anchor
    # can say what it leads to rather than "click here".
    for place in places:
        assert place["name"]
        assert place["count"] >= 1
        assert place["url"].startswith("/schools/")