from __future__ import annotations

import pandas as pd

from web_core.constants import TREND_BEAR, TREND_BULL, TREND_NEUTRAL
from web_core.ui.training_ui import build_learning_window_rows


def _make_analyzed(start: str = "2025-01-01", periods: int = 420) -> pd.DataFrame:
    idx = pd.date_range(start, periods=periods, freq="D", tz="UTC")
    closes = [100.0 + (i * 0.2) for i in range(periods)]
    classifications_cycle = ["real_bull", "fake", "real_bear", "fake", "real_bull"]
    trend_cycle = [TREND_BULL, TREND_BULL, TREND_BEAR, TREND_BEAR, TREND_NEUTRAL]
    classifications = [classifications_cycle[i % len(classifications_cycle)] for i in range(periods)]
    trends = [trend_cycle[i % len(trend_cycle)] for i in range(periods)]
    return pd.DataFrame({"Close": closes, "classification": classifications, "trend_state": trends}, index=idx)


def test_build_learning_window_rows_includes_standard_windows() -> None:
    analyzed = _make_analyzed()
    rows = build_learning_window_rows(analyzed)

    assert list(rows["Window"]) == ["1M", "3M", "6M", "1Y"]
    assert set(rows.columns) == {
        "Window",
        "Bars",
        "Price Change %",
        "Real Bull Bars",
        "Real Bear Bars",
        "Fake Bars",
        "Trend Flips",
        "What this says",
    }


def test_build_learning_window_rows_fallbacks_with_short_history() -> None:
    analyzed = _make_analyzed(periods=10)
    rows = build_learning_window_rows(analyzed)

    assert len(rows) == 1
    assert rows.iloc[0]["Window"] == "All data"
    assert int(rows.iloc[0]["Bars"]) == 10