QuantEngineByItz/tools/update_workbook_sector_insights.py

from __future__ import annotations

import json
from datetime import datetime
from pathlib import Path

from openpyxl import load_workbook
from openpyxl.chart import BarChart, LineChart, Reference
from openpyxl.styles import Font, PatternFill, Alignment
from openpyxl.utils import get_column_letter


ROOT = Path(__file__).resolve().parent.parent
INPUT_XLSX = ROOT / "GatherTradingData.xlsx"
OUTPUT_DIR = ROOT / "outputs" / "sector_insights_enhanced"
OUTPUT_XLSX = OUTPUT_DIR / "GatherTradingData_sector_insights.xlsx"
SECTOR_JSON = ROOT / "Temp" / "sector_trend_analysis_v1.json"
ETF_JSON = ROOT / "Temp" / "etf_representative_monitor_v1.json"


HEADER_FILL = PatternFill("solid", fgColor="1F4E78")
SUBHEADER_FILL = PatternFill("solid", fgColor="D9EAF7")
KPI_FILL = PatternFill("solid", fgColor="F3F7FB")
KPI_LABEL_FILL = PatternFill("solid", fgColor="E2F0D9")
KPI_VALUE_FILL = PatternFill("solid", fgColor="FFF2CC")
WHITE_FONT = Font(color="FFFFFF", bold=True)
BOLD_FONT = Font(bold=True)
TITLE_FONT = Font(size=14, bold=True)
NOTE_FONT = Font(italic=True, color="666666")


def load_json(path: Path) -> dict:
    return json.loads(path.read_text(encoding="utf-8"))


def remove_if_exists(wb, name: str) -> None:
    if name in wb.sheetnames:
        del wb[name]


def style_title(ws, title: str, subtitle: str | None = None, end_col: int = 8) -> None:
    ws.merge_cells(start_row=1, start_column=1, end_row=1, end_column=end_col)
    ws["A1"] = title
    ws["A1"].font = TITLE_FONT
    ws["A1"].fill = HEADER_FILL
    ws["A1"].font = WHITE_FONT
    ws["A1"].alignment = Alignment(horizontal="left")
    if subtitle:
        ws.merge_cells(start_row=2, start_column=1, end_row=2, end_column=end_col)
        ws["A2"] = subtitle
        ws["A2"].font = NOTE_FONT


def write_table(ws, start_row: int, start_col: int, headers: list[str], rows: list[list], header_fill=HEADER_FILL) -> int:
    for j, header in enumerate(headers, start=start_col):
        cell = ws.cell(start_row, j)
        cell.value = header
        cell.font = WHITE_FONT
        cell.fill = header_fill
        cell.alignment = Alignment(horizontal="center", vertical="center")
    for i, row in enumerate(rows, start=start_row + 1):
        for j, value in enumerate(row, start=start_col):
            cell = ws.cell(i, j)
            cell.value = value
            cell.alignment = Alignment(vertical="top")
    return start_row + len(rows)


def add_kpi_block(ws, start_row: int, items: list[tuple[str, object]]) -> int:
    ws.cell(start_row, 1).value = "KPI"
    ws.cell(start_row, 1).fill = SUBHEADER_FILL
    ws.cell(start_row, 1).font = BOLD_FONT
    row = start_row + 1
    for label, value in items:
        ws.cell(row, 1).value = label
        ws.cell(row, 1).fill = KPI_LABEL_FILL
        ws.cell(row, 1).font = BOLD_FONT
        ws.cell(row, 2).value = value
        ws.cell(row, 2).fill = KPI_VALUE_FILL
        row += 1
    return row


def set_col_widths(ws, widths: dict[str, int]) -> None:
    for col, width in widths.items():
        ws.column_dimensions[col].width = width


def style_sheet(ws) -> None:
    ws.freeze_panes = "A3"
    ws.sheet_view.showGridLines = False


def extract_sheet_rows(wb, sheet_name: str) -> tuple[list[str], list[list]]:
    ws = wb[sheet_name]
    first_row = [ws.cell(1, c).value for c in range(1, ws.max_column + 1)]
    second_row = [ws.cell(2, c).value for c in range(1, ws.max_column + 1)] if ws.max_row >= 2 else []
    use_second_row = (
        bool(second_row)
        and not any(v not in (None, "") for v in first_row)
        and any(v not in (None, "") for v in second_row)
    )
    headers = second_row if use_second_row else first_row
    start_row = 3 if use_second_row else 2
    rows: list[list] = []
    for r in range(start_row, ws.max_row + 1):
        row = [ws.cell(r, c).value for c in range(1, ws.max_column + 1)]
        if any(v is not None and v != "" for v in row):
            rows.append(row)
    return headers, rows


def build_portfolio_summary(wb) -> None:
    def display(value):
        return value if value not in (None, "") else "DATA_MISSING — 하네스 업데이트 필요"

    daily_headers, daily_rows = extract_sheet_rows(wb, "daily_history")
    monthly_headers, monthly_rows = extract_sheet_rows(wb, "monthly_history")
    account_headers, account_rows = extract_sheet_rows(wb, "account_snapshot")

    ws = wb.create_sheet("portfolio_performance_summary")
    style_sheet(ws)
    style_title(
        ws,
        "포트폴리오 성과 요약",
        "내 자금의 일간/월간 추이와 최신 보유 비중을 함께 보는 요약 시트",
        end_col=10,
    )

    latest_daily = daily_rows[-1] if daily_rows else []
    latest_month = monthly_rows[-1] if monthly_rows else []
    latest_total_asset = latest_daily[1] if len(latest_daily) > 1 else None
    latest_peak_asset = latest_daily[2] if len(latest_daily) > 2 else None
    latest_mdd = latest_daily[3] if len(latest_daily) > 3 else None
    latest_month_total = latest_month[1] if len(latest_month) > 1 else None
    latest_month_return = latest_month[8] if len(latest_month) > 8 else None
    latest_ytd_return = latest_month[10] if len(latest_month) > 10 else None

    latest_capture = None
    if account_rows:
        latest_capture = account_rows[0][0]
        for row in account_rows:
            if row and row[0] and row[0] > latest_capture:
                latest_capture = row[0]

    latest_holdings = [r for r in account_rows if r and r[0] == latest_capture]
    holdings_sorted = sorted(
        latest_holdings,
        key=lambda r: (r[10] if len(r) > 10 and isinstance(r[10], (int, float)) else 0),
        reverse=True,
    )
    total_mv = sum(r[10] for r in holdings_sorted if len(r) > 10 and isinstance(r[10], (int, float)))
    total_cost = sum(r[8] for r in holdings_sorted if len(r) > 8 and isinstance(r[8], (int, float)))
    total_pl = sum(r[11] for r in holdings_sorted if len(r) > 11 and isinstance(r[11], (int, float)))

    items = [
        ("latest_daily_asset", display(latest_total_asset)),
        ("latest_peak_asset", display(latest_peak_asset)),
        ("latest_daily_mdd_pct", display(latest_mdd)),
        ("latest_month_total_asset", display(latest_month_total)),
        ("latest_month_return_pct", display(latest_month_return)),
        ("latest_ytd_return_pct", display(latest_ytd_return)),
        ("latest_capture", display(latest_capture)),
        ("latest_holdings_count", display(len(latest_holdings))),
        ("latest_holdings_market_value", display(total_mv)),
        ("latest_holdings_profit_loss", display(total_pl)),
    ]
    add_kpi_block(ws, 4, items)

    ws["D4"] = "Portfolio view"
    ws["D4"].fill = SUBHEADER_FILL
    ws["D4"].font = BOLD_FONT
    ws["D5"] = "일간/월간 자산 추이는 실제 계좌 스냅샷 기반입니다."
    ws["D6"] = "보유 비중 차트는 최신 스냅샷의 시장가치 기준입니다."
    ws["D7"] = "수익률이 음수여도 숨기지 않고 그대로 보여줍니다."

    ws["G4"] = "Top holdings"
    ws["G4"].fill = SUBHEADER_FILL
    ws["G4"].font = BOLD_FONT
    for i, row in enumerate(holdings_sorted[:10], start=5):
        name = row[4] if len(row) > 4 else ""
        mv = row[10] if len(row) > 10 else ""
        ws.cell(i, 7).value = f"{name} ({mv})"

    # Daily history chart helper
    daily_sheet = wb["daily_history"]
    daily_max = daily_sheet.max_row
    daily_chart = LineChart()
    daily_chart.title = "Daily Asset / MDD"
    daily_chart.y_axis.title = "KRW / %"
    daily_chart.x_axis.title = "Date"
    daily_chart.height = 7
    daily_chart.width = 13
    daily_data = Reference(daily_sheet, min_col=2, max_col=4, min_row=1, max_row=daily_max)
    daily_cats = Reference(daily_sheet, min_col=1, min_row=2, max_row=daily_max)
    daily_chart.add_data(daily_data, titles_from_data=True, from_rows=False)
    daily_chart.set_categories(daily_cats)
    daily_chart.style = 2
    ws.add_chart(daily_chart, "A13")

    # Monthly history chart
    monthly_sheet = wb["monthly_history"]
    monthly_max = monthly_sheet.max_row
    monthly_chart = LineChart()
    monthly_chart.title = "Monthly Return Trend"
    monthly_chart.y_axis.title = "%"
    monthly_chart.x_axis.title = "Month"
    monthly_chart.height = 7
    monthly_chart.width = 13
    monthly_data = Reference(monthly_sheet, min_col=8, max_col=11, min_row=1, max_row=monthly_max)
    monthly_cats = Reference(monthly_sheet, min_col=1, min_row=2, max_row=monthly_max)
    monthly_chart.add_data(monthly_data, titles_from_data=True, from_rows=False)
    monthly_chart.set_categories(monthly_cats)
    monthly_chart.style = 3
    ws.add_chart(monthly_chart, "G13")

    # Top holdings bar chart
    hold_chart = BarChart()
    hold_chart.type = "bar"
    hold_chart.title = "Top Holdings by Market Value"
    hold_chart.y_axis.title = "Holding"
    hold_chart.x_axis.title = "KRW"
    hold_chart.height = 8
    hold_chart.width = 13
    ws_hold = wb.create_sheet("_portfolio_holdings_helper")
    helper_headers = ["name", "market_value"]
    helper_rows = [[r[4], r[10]] for r in holdings_sorted[:10] if len(r) > 10]
    write_table(ws_hold, 1, 1, helper_headers, helper_rows)
    hold_data = Reference(ws_hold, min_col=2, min_row=1, max_row=1 + len(helper_rows))
    hold_cats = Reference(ws_hold, min_col=1, min_row=2, max_row=1 + len(helper_rows))
    hold_chart.add_data(hold_data, titles_from_data=True)
    hold_chart.set_categories(hold_cats)
    hold_chart.legend = None
    ws.add_chart(hold_chart, "A30")

    set_col_widths(ws, {"A": 22, "B": 18, "C": 18, "D": 24, "E": 18, "F": 18, "G": 26, "H": 26, "I": 18, "J": 18})


def build_portfolio_sector_exposure(wb) -> None:
    daily_headers, daily_rows = extract_sheet_rows(wb, "daily_history")
    account_headers, account_rows = extract_sheet_rows(wb, "account_snapshot")
    universe_headers, universe_rows = extract_sheet_rows(wb, "universe")

    account_dicts = [dict(zip(account_headers, row)) for row in account_rows if any(v not in (None, "") for v in row)]
    universe_dicts = [dict(zip(universe_headers, row)) for row in universe_rows if any(v not in (None, "") for v in row)]

    sector_map: dict[str, str] = {}
    for row in universe_dicts:
        ticker = str(row.get("Ticker", "") or "").zfill(6)
        sector = str(row.get("Sector", "") or "").strip()
        if ticker and sector:
            sector_map[ticker] = sector

    latest_capture = ""
    for row in account_dicts:
        cap = str(row.get("captured_at", "") or "")
        if cap and cap >= latest_capture:
            latest_capture = cap
    latest_rows = [r for r in account_dicts if str(r.get("captured_at", "") or "") == latest_capture]

    exposure: dict[str, dict[str, float]] = {}
    sector_holdings: dict[str, list[dict[str, object]]] = {}
    for row in latest_rows:
        ticker = str(row.get("ticker", "") or "").zfill(6)
        sector = sector_map.get(ticker, "미분류")
        mv = float(row.get("market_value", 0) or 0)
        pl = float(row.get("profit_loss", 0) or 0)
        cost = float(row.get("total_cost", 0) or 0)
        bucket = exposure.setdefault(sector, {"market_value": 0.0, "profit_loss": 0.0, "cost": 0.0, "count": 0.0})
        bucket["market_value"] += mv
        bucket["profit_loss"] += pl
        bucket["cost"] += cost
        bucket["count"] += 1
        sector_holdings.setdefault(sector, []).append({
            "ticker": ticker,
            "name": row.get("name", ""),
            "market_value": mv,
            "profit_loss": pl,
            "return_pct": row.get("return_pct", ""),
        })

    total_mv = sum(v["market_value"] for v in exposure.values()) or 1.0
    rows = []
    for sector, vals in sorted(exposure.items(), key=lambda kv: kv[1]["market_value"], reverse=True):
        pct = vals["market_value"] / total_mv * 100.0
        ret_pct = (vals["profit_loss"] / vals["cost"] * 100.0) if vals["cost"] else 0.0
        rows.append([sector, vals["count"], vals["market_value"], pct, vals["profit_loss"], ret_pct])

    ws = wb.create_sheet("portfolio_sector_exposure")
    style_sheet(ws)
    style_title(
        ws,
        "포트폴리오 섹터 노출",
        "최신 계좌 스냅샷 기준으로 섹터별 보유 시장가치와 손익률을 집계",
        end_col=8,
    )
    items = [
        ("latest_capture", latest_capture),
        ("sector_count", len(rows)),
        ("top_sector", rows[0][0] if rows else ""),
        ("top_sector_weight_pct", rows[0][3] if rows else 0),
        ("top3_sector_weight_pct", sum(r[3] for r in rows[:3]) if rows else 0),
        ("total_market_value", total_mv),
    ]
    add_kpi_block(ws, 4, items)
    headers = ["sector", "holding_count", "market_value", "weight_pct", "profit_loss", "return_pct"]
    write_table(ws, 4, 4, headers, rows)
    ws["D4"] = "Sector exposure"
    ws["D4"].fill = SUBHEADER_FILL
    ws["D4"].font = BOLD_FONT
    ws.freeze_panes = "A5"
    ws.column_dimensions["A"].width = 24
    ws.column_dimensions["B"].width = 14
    ws.column_dimensions["C"].width = 18
    ws.column_dimensions["D"].width = 24
    ws.column_dimensions["E"].width = 14
    ws.column_dimensions["F"].width = 14
    ws.column_dimensions["G"].width = 16
    ws.column_dimensions["H"].width = 14

    chart = BarChart()
    chart.type = "bar"
    chart.style = 10
    chart.title = "Sector Exposure by Market Value"
    chart.y_axis.title = "Sector"
    chart.x_axis.title = "KRW"
    chart.height = 8
    chart.width = 14
    data_ref = Reference(ws, min_col=6, min_row=4, max_row=4 + len(rows))
    cats = Reference(ws, min_col=4, min_row=5, max_row=4 + len(rows))
    chart.add_data(data_ref, titles_from_data=True)
    chart.set_categories(cats)
    chart.legend = None
    ws.add_chart(chart, "J4")

    detail_rows: list[list[object]] = []
    for sector, vals in sorted(exposure.items(), key=lambda kv: kv[1]["market_value"], reverse=True)[:5]:
        sector_mv = vals["market_value"] or 1.0
        holdings = sorted(sector_holdings.get(sector, []), key=lambda item: float(item.get("market_value", 0) or 0), reverse=True)[:3]
        for idx, holding in enumerate(holdings, start=1):
            mv = float(holding.get("market_value", 0) or 0)
            detail_rows.append([
                sector if idx == 1 else "",
                idx,
                holding.get("ticker", ""),
                holding.get("name", ""),
                mv,
                mv / sector_mv * 100.0,
                mv / total_mv * 100.0,
                holding.get("return_pct", ""),
            ])

    ws["A18"] = "Sector top holdings detail"
    ws["A18"].fill = SUBHEADER_FILL
    ws["A18"].font = BOLD_FONT
    write_table(
        ws,
        19,
        1,
        ["sector", "rank_in_sector", "ticker", "name", "market_value", "sector_weight_pct", "portfolio_weight_pct", "return_pct"],
        detail_rows,
    )
    ws.column_dimensions["I"].width = 18
    ws.column_dimensions["J"].width = 18
    ws.column_dimensions["K"].width = 18
    ws.column_dimensions["L"].width = 18
    ws.column_dimensions["M"].width = 18

    if detail_rows:
        detail_chart = BarChart()
        detail_chart.type = "bar"
        detail_chart.style = 11
        detail_chart.title = "Top Holdings Contribution"
        detail_chart.y_axis.title = "Holding"
        detail_chart.x_axis.title = "Portfolio Weight %"
        detail_chart.height = 8
        detail_chart.width = 14
        # Use the first 15 rows of the detail table for a readable chart.
        chart_end_row = 19 + min(len(detail_rows), 15)
        data_ref2 = Reference(ws, min_col=7, min_row=19, max_row=chart_end_row)
        cats2 = Reference(ws, min_col=4, min_row=20, max_row=chart_end_row)
        detail_chart.add_data(data_ref2, titles_from_data=True)
        detail_chart.set_categories(cats2)
        detail_chart.legend = None
        ws.add_chart(detail_chart, "J20")


def build_sector_summary(wb, data: dict) -> None:
    ws = wb.create_sheet("sector_trend_summary")
    style_sheet(ws)
    style_title(
        ws,
        "섹터 동향 분석 요약",
        "ETF 프록시, 스마트머니 유입, 수익률, 유동성 경고를 한 장에 요약한 시트",
        end_col=8,
    )
    summary = data.get("summary") or {}
    concentration = data.get("concentration") or {}
    items = [
        ("formula_id", data.get("formula_id", "")),
        ("gate", data.get("gate", "")),
        ("latest_snapshot_date", data.get("latest_snapshot_date", "")),
        ("previous_snapshot_date", data.get("previous_snapshot_date", "")),
        ("sector_count", data.get("sector_count", 0)),
        ("trend_posture", summary.get("trend_posture", "")),
        ("rising_count", summary.get("rising_count", 0)),
        ("fading_count", summary.get("fading_count", 0)),
        ("stable_count", summary.get("stable_count", 0)),
        ("etf_proxy_count", summary.get("etf_proxy_count", 0)),
        ("smart_money_inflow_count", summary.get("smart_money_inflow_count", 0)),
        ("smart_money_outflow_count", summary.get("smart_money_outflow_count", 0)),
        ("flow_aligned_count", summary.get("flow_aligned_count", 0)),
        ("flow_diverging_count", summary.get("flow_diverging_count", 0)),
    ]
    add_kpi_block(ws, 4, items)
    ws["D4"] = "Concentration"
    ws["D4"].fill = SUBHEADER_FILL
    ws["D4"].font = BOLD_FONT
    for idx, (label, value) in enumerate(
        [
            ("top_sector", concentration.get("top_sector", "")),
            ("top_sector_weight_pct", concentration.get("top_sector_weight_pct", 0)),
            ("top2_weight_pct", concentration.get("top2_weight_pct", 0)),
            ("concentration_gate", concentration.get("concentration_gate", "")),
        ],
        start=5,
    ):
        ws.cell(idx, 4).value = label
        ws.cell(idx, 4).fill = KPI_LABEL_FILL
        ws.cell(idx, 4).font = BOLD_FONT
        ws.cell(idx, 5).value = value
        ws.cell(idx, 5).fill = KPI_VALUE_FILL

    top_inflow = summary.get("top_inflow_sectors") or []
    outflow = summary.get("outflow_warning_sectors") or []
    ws["G4"] = "Top Inflow"
    ws["G4"].fill = SUBHEADER_FILL
    ws["G4"].font = BOLD_FONT
    for i, item in enumerate(top_inflow, start=5):
        ws.cell(i, 7).value = item
    ws["H4"] = "Outflow Warning"
    ws["H4"].fill = SUBHEADER_FILL
    ws["H4"].font = BOLD_FONT
    for i, item in enumerate(outflow, start=5):
        ws.cell(i, 8).value = item

    ws["A20"] = "Notes"
    ws["A20"].fill = SUBHEADER_FILL
    ws["A20"].font = BOLD_FONT
    ws["A21"] = "섹터별 ETF 프록시와 스마트머니 방향이 다르면 매수 근거를 보수적으로 해석해야 합니다."
    ws["A21"].alignment = Alignment(wrap_text=True)
    ws["A22"] = "데이터 결측은 하네스 업데이트가 필요합니다."
    ws["A22"].alignment = Alignment(wrap_text=True)

    chart = LineChart()
    chart.title = "Average Sector Score / Breadth Trend"
    chart.y_axis.title = "Score / Count"
    chart.x_axis.title = "Snapshot"
    chart.height = 7.5
    chart.width = 13
    timeline_sheet = wb["sector_trend_timeline"]
    max_row = timeline_sheet.max_row
    data_ref = Reference(timeline_sheet, min_col=13, min_row=4, max_row=max_row, max_col=17)
    cats = Reference(timeline_sheet, min_col=12, min_row=5, max_row=max_row)
    chart.add_data(data_ref, titles_from_data=True, from_rows=False)
    chart.set_categories(cats)
    chart.style = 2
    ws.add_chart(chart, "G12")

    set_col_widths(ws, {"A": 28, "B": 18, "C": 16, "D": 24, "E": 16, "F": 18, "G": 24, "H": 24})


def build_sector_analysis(wb, data: dict) -> None:
    ws = wb.create_sheet("sector_trend_analysis")
    style_sheet(ws)
    style_title(
        ws,
        "섹터 동향 분석",
        "섹터별 ETF 프록시, 스마트머니 유입, 수익률, 유동성 방향을 함께 보는 상세 시트",
        end_col=18,
    )
    headers = [
        "sector", "proxy_ticker", "proxy_name", "proxy_type", "etf_code",
        "etf_execution_use", "etf_liquidity_score", "etf_liquidity_status", "etf_nav_risk",
        "proxy_confidence", "rank", "rank_delta_w1", "rank_delta_w2", "sector_score",
        "score_delta", "sector_ret5d", "sector_ret20d", "etf_return_5d", "etf_return_20d",
        "sector_etf_ret_gap_5d", "sector_etf_ret_gap_20d", "smart_money_5d_krw_raw",
        "smart_money_20d_krw_raw", "smart_money_direction", "liquidity_direction",
        "flow_alignment_state", "momentum_state", "concentration_weight_pct"
    ]
    rows = []
    for row in data.get("rows") or []:
        rows.append([row.get(h, "") for h in headers])
    write_table(ws, 4, 1, headers, rows)
    ws.auto_filter.ref = f"A4:{get_column_letter(len(headers))}{4 + len(rows)}"
    ws.freeze_panes = "A5"
    for col in ["F", "G", "H", "I", "J", "K", "L", "M", "N", "O", "P", "Q", "R", "S", "T", "U", "V", "W", "X", "Y", "Z", "AA", "AB"]:
        ws.column_dimensions[col].width = 16
    ws.column_dimensions["C"].width = 18
    ws.column_dimensions["A"].width = 16
    ws.column_dimensions["B"].width = 12
    ws.column_dimensions["D"].width = 12
    ws.column_dimensions["E"].width = 12
    ws.column_dimensions["J"].width = 14
    ws.column_dimensions["P"].width = 12
    ws.column_dimensions["Q"].width = 12
    ws.column_dimensions["AA"].width = 18
    ws.column_dimensions["AB"].width = 18

    chart = BarChart()
    chart.type = "bar"
    chart.style = 10
    chart.title = "Sector 20D Return by Sector"
    chart.y_axis.title = "Sector"
    chart.x_axis.title = "20D Return"
    chart.height = 8
    chart.width = 14
    data_ref = Reference(ws, min_col=17, min_row=4, max_row=4 + len(rows))
    cats = Reference(ws, min_col=1, min_row=5, max_row=4 + len(rows))
    chart.add_data(data_ref, titles_from_data=True)
    chart.set_categories(cats)
    chart.legend = None
    ws.add_chart(chart, "AD4")


def build_sector_timeline(wb, data: dict, source_data: dict | None = None) -> None:
    ws = wb.create_sheet("sector_trend_timeline")
    style_sheet(ws)
    style_title(ws, "섹터 시계열", "최근 스냅샷 기준 경향성 추세", end_col=10)
    headers = [
        "snapshot_date", "sector_count", "avg_sector_score", "top_sector", "top_sector_score",
        "positive_breadth_count", "liquidity_warn_count", "net_smart_money_5d_krw",
        "top_sector_rank", "top_sector_smart_money_5d_krw"
    ]
    rows = []
    for row in data.get("timeline") or []:
        parsed_date = row.get("snapshot_date", "")
        if isinstance(parsed_date, str) and parsed_date:
            try:
                parsed_date = datetime.fromisoformat(parsed_date.replace("Z", "+00:00")).date()
            except Exception:
                pass
        rows.append([
            parsed_date,
            row.get("sector_count", ""),
            row.get("avg_sector_score", ""),
            row.get("top_sector", ""),
            row.get("top_sector_score", ""),
            row.get("positive_breadth_count", ""),
            row.get("liquidity_warn_count", ""),
            row.get("net_smart_money_5d_krw", ""),
            row.get("top_sector_rank", ""),
            row.get("top_sector_smart_money_5d_krw", ""),
        ])
    write_table(ws, 4, 1, headers, rows)
    helper_headers = [
        "snapshot_date", "avg_sector_score", "top_sector_score",
        "positive_breadth_count", "liquidity_warn_count", "net_smart_money_5d_krw"
    ]
    helper_rows = []
    for row in rows:
        helper_rows.append([row[0], row[2], row[4], row[5], row[6], row[7]])
    write_table(ws, 4, 12, helper_headers, helper_rows)
    ws.freeze_panes = "A5"
    ws.column_dimensions["A"].width = 14
    ws.column_dimensions["B"].width = 12
    ws.column_dimensions["C"].width = 14
    ws.column_dimensions["D"].width = 16
    ws.column_dimensions["E"].width = 14
    ws.column_dimensions["F"].width = 16
    ws.column_dimensions["G"].width = 16
    ws.column_dimensions["H"].width = 18
    ws.column_dimensions["I"].width = 14
    ws.column_dimensions["J"].width = 18

    chart = LineChart()
    chart.title = "Trend Score / Breadth / Liquidity"
    chart.y_axis.title = "Count / Score"
    chart.x_axis.title = "Snapshot"
    chart.height = 8
    chart.width = 15
    data_ref = Reference(ws, min_col=13, max_col=17, min_row=4, max_row=4 + len(helper_rows))
    cats = Reference(ws, min_col=12, min_row=5, max_row=4 + len(helper_rows))
    chart.add_data(data_ref, titles_from_data=True, from_rows=False)
    chart.set_categories(cats)
    chart.style = 3
    ws.add_chart(chart, "L4")

    history_rows = []
    if isinstance(source_data, dict):
        history_rows = source_data.get("sector_flow_history") or []
    if not history_rows:
        history_rows = data.get("timeline_history") or data.get("history") or []
    if isinstance(history_rows, list) and history_rows:
        history_by_sector: dict[str, list[dict[str, object]]] = {}
        for item in history_rows:
            if not isinstance(item, dict):
                continue
            sector = str(item.get("Sector") or "").strip()
            if not sector:
                continue
            history_by_sector.setdefault(sector, []).append(item)

        top_sectors = []
        for row in rows[:3]:
            if len(row) > 3 and row[3]:
                top_sectors.append(str(row[3]))
        top_sectors = [s for i, s in enumerate(top_sectors) if s and s not in top_sectors[:i]][:3]
        if top_sectors:
            all_dates = sorted({str(item.get("Snapshot_Date") or "") for item in history_rows if str(item.get("Snapshot_Date") or "")})
            recent_dates = all_dates[-8:]

            score_start = 12
            score_headers = ["snapshot_date"] + [f"{sector}_score" for sector in top_sectors]
            score_rows = []
            for snapshot_date in recent_dates:
                row_vals = [snapshot_date]
                for sector in top_sectors:
                    series = sorted(history_by_sector.get(sector, []), key=lambda r: str(r.get("Snapshot_Date") or ""))
                    match = next((r for r in series if str(r.get("Snapshot_Date") or "") == snapshot_date), {})
                    row_vals.append(match.get("Sector_Score", ""))
                score_rows.append(row_vals)
            write_table(ws, 4, score_start, score_headers, score_rows)
            ws.column_dimensions[get_column_letter(score_start)].width = 14
            for offset in range(1, len(score_headers)):
                ws.column_dimensions[get_column_letter(score_start + offset)].width = 16

            score_chart = LineChart()
            score_chart.title = "Top Sector Score Trend"
            score_chart.y_axis.title = "Score"
            score_chart.x_axis.title = "Snapshot"
            score_chart.height = 8
            score_chart.width = 15
            score_chart.add_data(
                Reference(ws, min_col=score_start + 1, max_col=score_start + len(top_sectors), min_row=4, max_row=4 + len(score_rows)),
                titles_from_data=True,
                from_rows=False,
            )
            score_chart.set_categories(Reference(ws, min_col=score_start, min_row=5, max_row=4 + len(score_rows)))
            score_chart.style = 2
            ws.add_chart(score_chart, "L20")

            money_start = 20
            money_headers = ["snapshot_date"] + [f"{sector}_smart_money" for sector in top_sectors]
            money_rows = []
            for snapshot_date in recent_dates:
                row_vals = [snapshot_date]
                for sector in top_sectors:
                    series = sorted(history_by_sector.get(sector, []), key=lambda r: str(r.get("Snapshot_Date") or ""))
                    match = next((r for r in series if str(r.get("Snapshot_Date") or "") == snapshot_date), {})
                    row_vals.append(match.get("SmartMoney_5D_KRW", ""))
                money_rows.append(row_vals)
            write_table(ws, 4, money_start, money_headers, money_rows)
            ws.column_dimensions[get_column_letter(money_start)].width = 14
            for offset in range(1, len(money_headers)):
                ws.column_dimensions[get_column_letter(money_start + offset)].width = 18

            money_chart = LineChart()
            money_chart.title = "Top Sector Smart Money Trend"
            money_chart.y_axis.title = "KRW"
            money_chart.x_axis.title = "Snapshot"
            money_chart.height = 8
            money_chart.width = 15
            money_chart.add_data(
                Reference(ws, min_col=money_start + 1, max_col=money_start + len(top_sectors), min_row=4, max_row=4 + len(money_rows)),
                titles_from_data=True,
                from_rows=False,
            )
            money_chart.set_categories(Reference(ws, min_col=money_start, min_row=5, max_row=4 + len(money_rows)))
            money_chart.style = 3
            ws.add_chart(money_chart, "L36")


def build_etf_summary(wb, data: dict) -> None:
    ws = wb.create_sheet("etf_representative_summary")
    style_sheet(ws)
    style_title(
        ws,
        "ETF 대표 종목 요약",
        "ETF 구성비중 우선, 부족분은 유동성 우선 후보로 보강한 3종목 바스켓 요약",
        end_col=8,
    )
    summary = data.get("summary") or {}
    items = [
        ("formula_id", data.get("formula_id", "")),
        ("gate", data.get("gate", "")),
        ("etf_sector_count", data.get("etf_sector_count", 0)),
        ("tracked_count", data.get("tracked_count", 0)),
        ("complete_basket_count", summary.get("complete_basket_count", 0)),
        ("partial_basket_count", summary.get("partial_basket_count", 0)),
        ("basket_missing_total", summary.get("basket_missing_total", 0)),
        ("weighted_basis_count", summary.get("weighted_basis_count", 0)),
        ("fallback_basis_count", summary.get("fallback_basis_count", 0)),
        ("selected_sector_count", summary.get("selected_sector_count", 0)),
    ]
    add_kpi_block(ws, 4, items)
    ws["D4"] = "Representative principle"
    ws["D4"].fill = SUBHEADER_FILL
    ws["D4"].font = BOLD_FONT
    ws["D5"] = "1) ETF constituent weight first"
    ws["D6"] = "2) Missing slots filled with same-sector live candidates"
    ws["D7"] = "3) Missing data stays explicit as DATA_MISSING"
    ws["D8"] = "4) Minimum 3 names per sector basket"
    ws["G4"] = "Top reps"
    ws["G4"].fill = SUBHEADER_FILL
    ws["G4"].font = BOLD_FONT
    for i, item in enumerate(summary.get("top_rep_names") or [], start=5):
        ws.cell(i, 7).value = item
    set_col_widths(ws, {"A": 28, "B": 18, "C": 16, "D": 30, "E": 18, "F": 18, "G": 24, "H": 24})


def build_etf_monitor(wb, data: dict) -> None:
    ws = wb.create_sheet("etf_representative_monitor")
    style_sheet(ws)
    style_title(
        ws,
        "ETF 대표 종목 모니터",
        "섹터별 3종목 바스켓과 선택 근거, 커버리지, 품질 상태를 추적",
        end_col=18,
    )
    headers = [
        "sector", "etf_proxy_ticker", "etf_proxy_name", "etf_proxy_type", "sector_rank",
        "sector_score", "sector_smart_money_5d_krw", "sector_ret20d", "representative_count",
        "representative_ticker", "representative_name", "representative_basis",
        "representative_basis_detail", "constituent_weight", "basket_quality_state",
        "basket_coverage_pct", "basket_state", "basket_buy_review_count",
        "basket_track_count", "basket_watch_count", "basket_caution_count",
        "basket_aligned_count", "basket_missing_count", "basket_real_count",
        "selection_source", "selection_score", "monitor_reason"
    ]
    rows = []
    for row in data.get("rows") or []:
        rows.append([row.get(h, "") for h in headers])
    write_table(ws, 4, 1, headers, rows)
    ws.auto_filter.ref = f"A4:{get_column_letter(len(headers))}{4 + len(rows)}"
    ws.freeze_panes = "A5"
    for col, width in {"A": 18, "B": 12, "C": 16, "D": 12, "E": 12, "F": 12, "G": 18, "H": 12,
                       "I": 14, "J": 12, "K": 18, "L": 18, "M": 24, "N": 14, "O": 14,
                       "P": 14, "Q": 14, "R": 14, "S": 14, "T": 14, "U": 14, "V": 14,
                       "W": 14, "X": 18, "Y": 14, "Z": 12, "AA": 24}.items():
        ws.column_dimensions[col].width = width

    chart = BarChart()
    chart.type = "bar"
    chart.style = 10
    chart.title = "Basket Coverage by Sector"
    chart.y_axis.title = "Sector"
    chart.x_axis.title = "Coverage %"
    chart.height = 8
    chart.width = 14
    data_ref = Reference(ws, min_col=16, min_row=4, max_row=4 + len(rows))
    cats = Reference(ws, min_col=1, min_row=5, max_row=4 + len(rows))
    chart.add_data(data_ref, titles_from_data=True)
    chart.set_categories(cats)
    chart.legend = None
    ws.add_chart(chart, "AC4")


def main() -> None:
    if not INPUT_XLSX.exists():
        raise FileNotFoundError(INPUT_XLSX)
    if not SECTOR_JSON.exists():
        raise FileNotFoundError(SECTOR_JSON)
    if not ETF_JSON.exists():
        raise FileNotFoundError(ETF_JSON)
    raw_json_path = ROOT / "GatherTradingData.json"

    sector = load_json(SECTOR_JSON)
    etf = load_json(ETF_JSON)
    raw_data = load_json(raw_json_path) if raw_json_path.exists() else {}
    raw_source = raw_data.get("data", {}) if isinstance(raw_data.get("data"), dict) else {}

    wb = load_workbook(INPUT_XLSX)
    for name in [
        "portfolio_performance_summary",
        "portfolio_sector_exposure",
        "_portfolio_holdings_helper",
        "sector_trend_summary",
        "sector_trend_analysis",
        "sector_trend_timeline",
        "etf_representative_summary",
        "etf_representative_monitor",
    ]:
        remove_if_exists(wb, name)

    # Build data sheets first so summary sheets can reference the timeline sheet.
    build_portfolio_summary(wb)
    build_portfolio_sector_exposure(wb)
    build_sector_timeline(wb, sector, raw_source)
    build_sector_analysis(wb, sector)
    build_sector_summary(wb, sector)
    build_etf_monitor(wb, etf)
    build_etf_summary(wb, etf)

    # Put summary sheets near the front.
    order = [
        "settings",
        "portfolio_performance_summary",
        "portfolio_sector_exposure",
        "sector_trend_summary",
        "sector_trend_analysis",
        "sector_trend_timeline",
        "etf_representative_summary",
        "etf_representative_monitor",
    ]
    existing = [s for s in wb.sheetnames if s not in order]
    wb._sheets = [wb[s] for s in order if s in wb.sheetnames] + [wb[s] for s in existing]
    if "_portfolio_holdings_helper" in wb.sheetnames:
        wb["_portfolio_holdings_helper"].sheet_state = "hidden"
    wb.active = wb.sheetnames.index("sector_trend_summary")

    OUTPUT_DIR.mkdir(parents=True, exist_ok=True)
    wb.save(OUTPUT_XLSX)
    print(f"saved {OUTPUT_XLSX}")
    print("sheets", wb.sheetnames[:10])


if __name__ == "__main__":
    main()