mitai-jinkendo/backend/tests/test_import_row_processing.py

"""Tests für CSV-Zeilenaggregation (import_row_processing)."""

from __future__ import annotations

import datetime as dt

import pytest

from csv_parser.import_row_processing import (
    aggregate_mapped_rows,
    resolve_import_row_processing,
    validate_import_row_processing,
)


def test_validate_rejects_unknown_aggregate():
    with pytest.raises(ValueError, match="ungültige Operation"):
        validate_import_row_processing(
            "nutrition",
            {"group_by": ["date"], "aggregates": {"kcal": "bogus"}},
            {"Kal": "date", "E": "kcal"},
        )


def test_validate_group_by_must_be_mapped():
    with pytest.raises(ValueError, match="keiner CSV-Spalte zugeordnet"):
        validate_import_row_processing(
            "nutrition",
            {"group_by": ["date"], "aggregates": {"kcal": "sum"}},
            {"Kal": "kcal"},  # date nicht gemappt
        )


def test_aggregate_mapped_rows_sums_same_group():
    d = dt.date(2024, 1, 15)
    rows = [
        {"date": d, "kcal": 500.0, "protein_g": 20},
        {"date": d, "kcal": 300.0, "protein_g": 15},
    ]
    spec = {"group_by": ["date"], "aggregates": {"kcal": "sum", "protein_g": "sum"}}
    out, err = aggregate_mapped_rows(rows, spec)
    assert err == []
    assert len(out) == 1
    assert out[0]["kcal"] == 800.0
    assert out[0]["protein_g"] == 35


def test_aggregate_mapped_rows_reject_second_group():
    d = dt.date(2024, 1, 15)
    rows = [
        {"date": d, "kcal": 100.0},
        {"date": d, "kcal": 200.0},
    ]
    spec = {
        "group_by": ["date"],
        "aggregates": {"kcal": "sum"},
        "multi_row_policy": "reject",
    }
    out, err = aggregate_mapped_rows(rows, spec)
    assert out == []
    assert len(err) == 1
    assert err[0].get("error") == "mehrere_zeilen_pro_schluessel"
    assert err[0].get("rows_in_group") == 2


def test_aggregate_mapped_rows_first_row_no_merge():
    d = dt.date(2024, 1, 15)
    rows = [
        {"date": d, "kcal": 100.0},
        {"date": d, "kcal": 999.0},
    ]
    spec = {
        "group_by": ["date"],
        "aggregates": {"kcal": "sum"},
        "multi_row_policy": "first_row",
    }
    out, err = aggregate_mapped_rows(rows, spec)
    assert err == []
    assert len(out) == 1
    assert out[0]["kcal"] == 100.0


def test_dedupe_identical_rows_before_group():
    d = dt.date(2024, 1, 15)
    rows = [
        {"date": d, "kcal": 50.0},
        {"date": d, "kcal": 50.0},
        {"date": d, "kcal": 50.0},
    ]
    spec = {
        "group_by": ["date"],
        "aggregates": {"kcal": "sum"},
        "dedupe_identical_rows": True,
    }
    out, err = aggregate_mapped_rows(rows, spec)
    assert err == []
    assert len(out) == 1
    assert out[0]["kcal"] == 50.0


def test_validate_multi_row_policy():
    with pytest.raises(ValueError, match="multi_row_policy"):
        validate_import_row_processing(
            "nutrition",
            {
                "group_by": ["date"],
                "aggregates": {"kcal": "sum"},
                "multi_row_policy": "nope",
            },
            {"D": "date", "K": "kcal"},
        )


def test_resolve_explicit_overrides_default():
    m = {
        "import_row_processing": {"group_by": ["date"], "aggregates": {"kcal": "mean"}},
    }
    spec = resolve_import_row_processing("nutrition", m)
    assert spec is not None
    assert spec["aggregates"]["kcal"] == "mean"


def test_resolve_empty_dict_falls_back_to_module_default():
    m: dict = {"import_row_processing": {}}
    spec = resolve_import_row_processing("nutrition", m)
    assert spec is not None
    assert spec["group_by"] == ["date"]
    assert spec["aggregates"]["kcal"] == "sum"


def test_resolve_none_uses_nutrition_default():
    spec = resolve_import_row_processing("nutrition", {})
    assert spec is not None
    assert "date" in (spec.get("group_by") or [])


def test_resolve_weight_default_uses_last():
    spec = resolve_import_row_processing("weight", {})
    assert spec is not None
    assert spec.get("aggregates", {}).get("weight") == "last"
    assert spec.get("aggregates", {}).get("note") == "last"


def test_validate_weight_import_row_processing_ok():
    validate_import_row_processing(
        "weight",
        {"group_by": ["date"], "aggregates": {"weight": "mean"}},
        {"D": "date", "W": "weight"},
    )


def test_resolve_vitals_baseline_default_uses_mean():
    spec = resolve_import_row_processing("vitals_baseline", {})
    assert spec is not None
    assert spec.get("aggregates", {}).get("resting_hr") == "mean"