pipeline/tests/test_worldbank.py

import pandas as pd

from isora_hf.worldbank import (
    ISORA_TO_WB_CODE,
    _fy_label_to_year,
    classification_year,
    income_group_for_years,
    wb_code,
)


def test_code_mapping_covers_imf_specific_codes():
    assert wb_code("KOS") == "XKX"
    assert wb_code("UAE") == "ARE"
    assert wb_code("FRA") == "FRA"
    assert set(ISORA_TO_WB_CODE) == {"KOS", "UAE"}


def test_fy_labels_and_alignment_rule():
    assert _fy_label_to_year("FY25") == 2025
    assert _fy_label_to_year("FY89") == 1989
    assert _fy_label_to_year("2025") is None
    assert classification_year(2022) == 2024  # FY24 groups are based on 2022 GNI


def test_income_group_lookup_uses_same_year_gni():
    history = pd.DataFrame(
        [
            {"wb_code": "XKX", "wb_fiscal_year": 2020, "income_group": "Lower middle income"},
            {"wb_code": "XKX", "wb_fiscal_year": 2024, "income_group": "Upper middle income"},
        ]
    )
    keys = pd.DataFrame(
        {"jurisdiction_code": ["KOS", "KOS", "FRA"], "fiscal_year": [2018, 2022, 2022]}
    )
    got = income_group_for_years(history, keys).tolist()
    assert got[0] == "Lower middle income" and got[1] == "Upper middle income"
    assert pd.isna(got[2])