from zipfile import ZipFile

import polars as pl

from pipeline.download.geolytix_retail_points import (
    read_latest_csv,
    select_latest_csv_name,
)


def test_select_latest_csv_ignores_previous_versions():
    names = [
        "README.txt",
        "geolytix_retailpoints_v41_202602.csv",
        "geolytix_retailpoints_v43_202603.csv",
        "Previous Versions/geolytix_retailpoints_v99_209901.csv",
    ]

    assert select_latest_csv_name(names) == "geolytix_retailpoints_v43_202603.csv"


def test_read_latest_csv_validates_required_columns(tmp_path):
    zip_path = tmp_path / "retail_points.zip"
    df = pl.DataFrame(
        {
            "id": [1],
            "retailer": ["Waitrose"],
            "fascia": ["Waitrose"],
            "store_name": ["Waitrose Test"],
            "postcode": ["SW1A 1AA"],
            "long_wgs": [-0.1],
            "lat_wgs": [51.5],
        }
    )

    with ZipFile(zip_path, "w") as zip_file:
        zip_file.writestr("geolytix_retailpoints_v1_202401.csv", "not,the,latest\n")
        with zip_file.open("geolytix_retailpoints_v2_202402.csv", "w") as csv_file:
            df.write_csv(csv_file)

    assert read_latest_csv(zip_path).to_dicts() == df.to_dicts()