# =============================================================================
# QWHPI — Quebec Weekly Housing Price Index
# Author  : Simon-Pierre Boucher
# Contact : contact@spboucher.ai
# File    : engine/tests/test_smoke_pipeline.py
# Purpose : CI smoke test — synthetic market through stage 1 + stage 2;
#           the recovered index must track the simulated price path.
# =============================================================================
"""End-to-end smoke test on a fully synthetic market (no data files)."""

from __future__ import annotations

import datetime as dt
import sys
from pathlib import Path

sys.path.insert(0, str(Path(__file__).resolve().parents[1] / "src"))

import numpy as np
import polars as pl

from qwhpi.hierarchy import cell_deviation, stage1_residualize, week_grid

RNG = np.random.default_rng(3)


def simulate(n_weeks: int = 60, per_week: int = 150) -> tuple[pl.DataFrame, np.ndarray]:
    """Synthetic province with a known weekly log price path."""
    path = np.cumsum(RNG.normal(0.005, 0.003, n_weeks))  # clear boom
    rows = []
    start = dt.date(2021, 1, 4)
    for w in range(n_weeks):
        week = start + dt.timedelta(weeks=w)
        for i in range(per_week):
            fa = float(np.exp(RNG.normal(4.7, 0.35)))
            age_yrs = int(RNG.integers(0, 90))
            ptype = ["unifamilial", "condo", "plex"][int(RNG.integers(0, 3))]
            base = 12.0 + 0.6 * (np.log(fa) - 4.7) - 0.002 * age_yrs \
                + {"unifamilial": 0.0, "condo": -0.15, "plex": 0.2}[ptype]
            lp = base + path[w] + RNG.normal(0, 0.25)
            rows.append({
                "id": f"{w}-{i}",
                "log_amount": lp,
                "week_str": week.isoformat(),
                "propertyType": ptype,
                "log_fa": np.log(fa),
                "fa_missing": False,
                "age_bin": f"bin{age_yrs // 20}",
                "building_type": "single-story",
                "loc_fine": f"Z{int(RNG.integers(0, 6))}",
                "region_code": "03" if RNG.random() < 0.5 else "06",
                "geo_code": "23027",
                "municipality": "Québec",
            })
    return pl.DataFrame(rows), path


def test_pipeline_recovers_simulated_path():
    df, truth = simulate()
    s1 = stage1_residualize(df)
    # province path per type ~ truth (up to a constant)
    uni = s1.paths.filter(pl.col("property_type") == "unifamilial").sort("week")
    est = uni["delta"].to_numpy()
    est = est - est.mean()
    tr = truth[-len(est):] - truth[-len(est):].mean()
    corr = np.corrcoef(est, tr)[0, 1]
    assert corr > 0.9, f"stage-1 path corr {corr:.3f}"

    # stage 2: a region with no true deviation should stay near zero
    grid = week_grid(s1.residuals)
    cell = s1.residuals.filter(
        (pl.col("region_code") == "03")
        & (pl.col("propertyType") == "unifamilial")).rename({"resid": "dev"})
    dev = cell_deviation(cell.select("week_str", "dev"), grid,
                         float(cell["dev"].var()))
    assert np.abs(dev.fit.smoothed).max() < 0.05