group_fqcd_jr/app/service/promotion_performance.py

"""业绩曲线数据解析与图表生成。"""

from __future__ import annotations

import csv
import hashlib
from datetime import date, datetime
from decimal import Decimal
from pathlib import Path
from typing import Any


def parse_performance_file(path: str | Path) -> dict[str, Any]:
    source = Path(path)
    if source.suffix.lower() == ".csv":
        with source.open("r", encoding="utf-8-sig", newline="") as stream:
            rows = list(csv.DictReader(stream))
    elif source.suffix.lower() in {".xlsx", ".xlsm"}:
        from openpyxl import load_workbook  # type: ignore[import-untyped]

        # read_only 模式的工作簿会持有文件句柄，必须显式关闭；否则上传后的
        # 业绩文件一直被后端进程占用，运维侧无法移动或删除该文件。
        workbook = load_workbook(source, read_only=True, data_only=True)
        try:
            values = list(workbook.active.values)
        finally:
            workbook.close()
        if not values:
            return {"columns": [], "rows": [], "row_count": 0}
        headers = [str(value or "").strip() for value in values[0]]
        rows = [
            {headers[index]: value for index, value in enumerate(row) if index < len(headers)}
            for row in values[1:]
        ]
    else:
        raise ValueError("业绩数据只支持 CSV、XLSX 或 XLSM")
    normalized = [_normalize_row(row) for row in rows if any(row.values())]
    columns = list(normalized[0]) if normalized else []
    result = {"columns": columns, "rows": normalized, "row_count": len(normalized)}
    _validate_performance_data(result)
    result["summary"] = summarize_performance_data(result)
    return result


def summarize_performance_data(data: dict[str, Any]) -> dict[str, Any]:
    """从已校验的业绩数据中提取材料需要的业绩摘要。"""
    date_key = _find_key(data, ("date", "日期", "净值日期", "统计日期"))
    if date_key is None:
        raise ValueError("业绩数据必须包含日期列")
    dates = [_parse_date(row.get(date_key)) for row in data.get("rows", [])]
    valid_dates = [item for item in dates if item is not None]
    if not valid_dates:
        raise ValueError("业绩数据没有有效日期")
    first_date = valid_dates[0]
    latest_date = valid_dates[-1]
    product_key = _find_key(data, ("product_return", "产品收益率"))
    product_values = (
        [_number(row.get(product_key)) for row in data.get("rows", [])]
        if product_key is not None
        else []
    )
    valid_product_values = [value for value in product_values if value is not None]
    return {
        "initial_date": first_date.isoformat(),
        "as_of_date": latest_date.isoformat(),
        "history_months": (
            (latest_date.year - first_date.year) * 12
            + latest_date.month
            - first_date.month
        ),
        "product_return": (
            _format_percentage(valid_product_values[-1])
            if valid_product_values
            else None
        ),
        "max_drawdown": _format_percentage(_max_drawdown(valid_product_values)),
    }


def _max_drawdown(values: list[float]) -> float | None:
    if not values:
        return None
    # 业绩文件中的产品收益率是累计收益百分比，先还原为指数化净值再计算回撤。
    levels = [100 + value for value in values]
    peak = levels[0]
    drawdown = 0.0
    for level in levels:
        peak = max(peak, level)
        if peak:
            drawdown = min(drawdown, (level / peak - 1) * 100)
    return drawdown


def _format_percentage(value: float | None) -> str | None:
    if value is None:
        return None
    return f"{value:.2f}".rstrip("0").rstrip(".") + "%"


def create_performance_chart(
    data: dict[str, Any], output_path: str | Path, *, title: str
) -> str:
    import matplotlib
    import matplotlib.dates as mdates
    from matplotlib import font_manager

    matplotlib.use("Agg")
    import matplotlib.pyplot as plt

    # 优先使用部署机上的中文字体，避免图表标题和图例缺字。
    font_candidates = (
        Path(r"C:\Windows\Fonts\msyh.ttc"),
        Path(r"C:\Windows\Fonts\simhei.ttf"),
        Path("/usr/share/fonts/opentype/noto/NotoSansCJK-Regular.ttc"),
        Path("/usr/share/fonts/truetype/noto/NotoSansCJK-Regular.ttc"),
        Path("/usr/share/fonts/truetype/wqy/wqy-zenhei.ttc"),
    )
    chinese_font = next((item for item in font_candidates if item.exists()), None)
    if chinese_font is not None:
        font_manager.fontManager.addfont(str(chinese_font))
        font_name = font_manager.FontProperties(fname=str(chinese_font)).get_name()
        matplotlib.rcParams["font.family"] = font_name
    matplotlib.rcParams["axes.unicode_minus"] = False

    rows = data.get("rows", [])
    if not rows:
        raise ValueError("业绩曲线没有可用数据")
    date_key = _find_key(data, ("date", "日期", "净值日期", "统计日期"))
    series = (
        ("product_return", "产品", "#1f4e79"),
        ("benchmark_return", "业绩比较基准", "#7f8c8d"),
        ("manager_representative_return", "基金经理代表产品", "#b5651d"),
    )
    fig, axis = plt.subplots(figsize=(10, 4.8), dpi=160)
    plotted = 0
    for key, label, color in series:
        actual_key = _find_key(data, (key, _chinese_key(key)))
        if actual_key is None:
            continue
        values = [_number(row.get(actual_key)) for row in rows]
        dates = [_parse_date(row.get(date_key)) for row in rows]
        pairs = [
            (day, value)
            for day, value in zip(dates, values, strict=True)
            if day and value is not None
        ]
        if not pairs:
            continue
        axis.plot(
            [
                float(mdates.date2num(item[0]))  # type: ignore[no-untyped-call]
                for item in pairs
            ],
            [item[1] for item in pairs],
                  label=label, color=color, linewidth=1.8)
        plotted += 1
    if not plotted:
        plt.close(fig)
        raise ValueError("业绩曲线缺少可绘制序列")
    axis.set_title(title)
    axis.set_ylabel("累计收益 / 指数化净值")
    axis.xaxis_date()
    axis.grid(alpha=0.25)
    axis.legend()
    fig.autofmt_xdate()
    output = Path(output_path)
    output.parent.mkdir(parents=True, exist_ok=True)
    fig.savefig(output, bbox_inches="tight")
    plt.close(fig)
    return str(output)


def file_hash(payload: bytes) -> str:
    return hashlib.sha256(payload).hexdigest()


def _normalize_row(row: dict[Any, Any]) -> dict[str, Any]:
    # openpyxl 把日期单元格读成 datetime，写进 MySQL JSON 列会直接抛
    # "Object of type datetime is not JSON serializable"（表现为上传业绩文件 500）。
    # 这里统一转成 JSON 安全类型：日期转 YYYY-MM-DD 字符串（_parse_date 可再次解析），
    # Decimal 转 float，保证 CSV 和 XLSX 两种来源产出的结构一致。
    return {
        str(key).strip(): _json_safe(value)
        for key, value in row.items()
        if key is not None
    }


def _json_safe(value: Any) -> Any:
    if isinstance(value, datetime):
        return value.date().strftime("%Y-%m-%d")
    if isinstance(value, date):
        return value.strftime("%Y-%m-%d")
    if isinstance(value, Decimal):
        return float(value)
    return value


def _find_key(data: dict[str, Any], candidates: tuple[str, ...]) -> str | None:
    columns = {str(column).strip() for column in data.get("columns", [])}
    for candidate in candidates:
        if candidate in columns:
            return candidate
    return None


def _validate_performance_data(data: dict[str, Any]) -> None:
    if not data.get("rows"):
        raise ValueError("业绩数据没有有效数据行")
    date_key = _find_key(data, ("date", "日期", "净值日期", "统计日期"))
    if date_key is None:
        raise ValueError("业绩数据必须包含日期列")
    parsed_dates = [_parse_date(row.get(date_key)) for row in data["rows"]]
    if any(item is None for item in parsed_dates):
        raise ValueError("业绩数据存在无法识别的日期")
    valid_dates = [item for item in parsed_dates if item is not None]
    if len(set(valid_dates)) != len(valid_dates):
        raise ValueError("业绩数据存在重复日期")
    if valid_dates != sorted(valid_dates):
        raise ValueError("业绩数据必须按日期升序排列")
    series_keys = (
        ("product_return", "产品收益率"),
        ("benchmark_return", "业绩比较基准收益率"),
        ("manager_representative_return", "基金经理代表产品收益率"),
    )
    if not any(_find_key(data, pair) for pair in series_keys):
        raise ValueError("业绩数据至少需要一条可绘制曲线")


def _chinese_key(key: str) -> str:
    return {
        "product_return": "产品收益率",
        "benchmark_return": "业绩比较基准收益率",
        "manager_representative_return": "基金经理代表产品收益率",
    }.get(key, key)


def _number(value: object) -> float | None:
    try:
        return float(str(value).replace("%", "").replace(",", "").strip())
    except (TypeError, ValueError):
        return None


def _parse_date(value: object) -> date | None:
    if isinstance(value, datetime):
        return value.date()
    if isinstance(value, date):
        return value
    for fmt in ("%Y-%m-%d", "%Y/%m/%d", "%Y%m%d"):
        try:
            return datetime.strptime(str(value), fmt).date()
        except (TypeError, ValueError):
            continue
    return None