"""
China's green sanctuaries study: analysis pipeline.

Reads the raw World Bank "Forest area (% of land area)" series for China
(indicator AG.LND.FRST.ZS, sourced from FAO), cleans it into a continuous
yearly series, derives milestone-decade bars, and forecasts forest cover
percent through 2030 with a Holt linear-trend model.

Output: ../../src/data/china-green/clean.json  (consumed by the static site)

Run:
    cd analysis
    .venv/bin/python china-green/pipeline.py

Source: World Bank, "Forest area (% of land area)" (AG.LND.FRST.ZS) for China,
        data sourced from the Food and Agriculture Organization (FAO).
        https://data.worldbank.org/indicator/AG.LND.FRST.ZS?locations=CN
"""

import json
from pathlib import Path

import pandas as pd

import sys
sys.path.insert(0, str(Path(__file__).resolve().parent.parent))
from _lib import holt_forecast, write_clean

HERE = Path(__file__).parent
RAW = HERE / "raw" / "china-forest-area.json"
OUT = HERE.parent.parent / "src" / "data" / "china-green" / "clean.json"

FORECAST_TO = 2030
MILESTONE_YEARS = (1990, 2000, 2010, 2020)


def load() -> pd.Series:
    """Parse the World Bank JSON array [metadata, data[]] into a yearly series."""
    payload = json.loads(RAW.read_text())
    rows = payload[1]
    records = [
        (int(r["date"]), float(r["value"]))
        for r in rows
        if r["value"] is not None
    ]
    df = pd.DataFrame(records, columns=["year", "value"]).sort_values("year")
    return df.set_index("year")["value"]


def nearest_year(series: pd.Series, target: int) -> int:
    return int(min(series.index, key=lambda y: abs(y - target)))


def main() -> None:
    series = load()
    first_year = int(series.index[0])
    latest_year = int(series.index[-1])
    first_pct = round(float(series.loc[first_year]), 2)
    forest_pct = round(float(series.loc[latest_year]), 2)

    fc = holt_forecast(series.values, latest_year, FORECAST_TO, ndigits=2, scale=1.0)

    # --- Milestone-decade bars ---
    base_year = nearest_year(series, MILESTONE_YEARS[0])
    base_pct = float(series.loc[base_year])
    bars = []
    for target in MILESTONE_YEARS:
        yr = nearest_year(series, target)
        pct = round(float(series.loc[yr]), 2)
        if yr == base_year:
            note = "baseline"
        else:
            note = f"+{round(pct - base_pct, 1)} pts since {base_year}"
        bars.append({"label": str(yr), "value": pct, "note": note})

    payload = {
        "meta": {
            "source": "World Bank — Forest area (% of land area), China (data sourced from FAO)",
            "sourceUrl": "https://data.worldbank.org/indicator/AG.LND.FRST.ZS?locations=CN",
            "item": "Forest area as a percent of land area, China (AG.LND.FRST.ZS)",
            "asOf": f"{latest_year} (latest year in the World Bank release)",
            "unitNote": "Values shown as percent of total land area under forest.",
            "method": "Holt linear-trend exponential smoothing; 80% band from residual std scaled by sqrt(horizon).",
        },
        "headline": {
            "latestYear": latest_year,
            "forestPct": forest_pct,
            "firstYear": first_year,
            "firstPct": first_pct,
            "forecastYear": FORECAST_TO,
            "forecastPct": fc["mean"][-1],
            "gainPts": round(forest_pct - first_pct, 1),
        },
        "series": [
            {"year": int(yr), "value": round(float(v), 2)}
            for yr, v in series.items()
        ],
        "forecast": fc,
        "bars": bars,
    }

    write_clean(OUT, payload)
    print(f"wrote {OUT.relative_to(HERE.parent.parent)}")
    print(
        f"  China {first_year}: {first_pct}% -> {latest_year}: {forest_pct}% "
        f"(+{payload['headline']['gainPts']} pts)"
    )
    print(
        f"  forecast {FORECAST_TO}: {fc['mean'][-1]}% "
        f"(80% {fc['lower'][-1]}–{fc['upper'][-1]})"
    )


if __name__ == "__main__":
    main()