import json
import pytest
from pathlib import Path
import pandas as pd

import src.artifact_1.main as main


def load_test_data():
    test_data_path = Path(__file__).parent / 'data' / 'test_timing_data.json'
    if not test_data_path.exists():
        # fallback synthetic data
        return [
            {"expires_at_dist_hours": 48.0, "delta_t": 1.2, "pinned": True, "unpinned": False},
            {"expires_at_dist_hours": 12.0, "delta_t": -0.5, "pinned": False, "unpinned": True},
            {"expires_at_dist_hours": 5.0, "delta_t": 0.3, "pinned": False, "unpinned": True}
        ]
    with open(test_data_path, encoding='utf-8') as f:
        return json.load(f)


@pytest.fixture(scope='module')
def sample_data():
    return load_test_data()


def test_analyze_timing_data_basic(sample_data):
    result = main.analyze_timing_data(sample_data)
    assert isinstance(result, dict)
    assert 'negative_dt_count' in result
    assert 'total_count' in result
    assert result['total_count'] == len(sample_data)
    assert result['negative_dt_count'] <= result['total_count']


def test_analyze_timing_data_near_expiry_threshold(sample_data):
    result = main.analyze_timing_data(sample_data)
    threshold = result.get('recommended_near_expiry_threshold')
    assert isinstance(threshold, (int, float))
    assert 0 <= threshold <= 72


def test_analyze_timing_data_invalid_input():
    # wrong type
    with pytest.raises((ValueError, TypeError)):
        main.analyze_timing_data('not a list')
    # missing keys
    bad_data = [{"delta_t": 1.0}]
    with pytest.raises((KeyError, ValueError)):
        main.analyze_timing_data(bad_data)


def test_analyze_timing_data_pandas_consistency(sample_data):
    # Ensure that conversion to pandas does not break the structure assumptions
    df = pd.DataFrame(sample_data)
    assert all(col in df.columns for col in ['expires_at_dist_hours', 'delta_t'])
    result = main.analyze_timing_data(sample_data)
    assert isinstance(result.get('mean_delta_t'), (float, int))


def test_analyze_timing_data_negative_dt_flag(sample_data):
    has_negative = any(d['delta_t'] < 0 for d in sample_data)
    result = main.analyze_timing_data(sample_data)
    if has_negative:
        assert result['negative_dt_count'] > 0
    else:
        assert result['negative_dt_count'] == 0