#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""probe_unified_schema.py — 探查方案A 新权威数据层 schema (LocalUnifiedProvider 实现依赖)。

输出 constituent_unified / dbbardata('d') / bs_adjust_factor / valuation_baostock 的
列结构 + 样本 + 覆盖, 供 spec §6 使用层 provider 实现参考。
"""
import sqlite3
from pathlib import Path

import pandas as pd

BASE = Path(r"C:\sanguo_vnpy_v2")
DB = BASE / "data" / "quant_trading.db"
VAL_DIR = BASE / "data" / "valuation_baostock"


def section(title):
    print(f"\n===== {title} =====")


# 1. constituent_unified
section("constituent_unified schema")
c = sqlite3.connect(str(DB))
try:
    info = c.execute("PRAGMA table_info(constituent_unified)").fetchall()
    print("columns:", [(r[1], r[2]) for r in info])
    print("count/index:",
          c.execute("SELECT COUNT(*), COUNT(DISTINCT index_code) FROM constituent_unified").fetchone())
    print("per-index (total / in_current / was_removed):")
    for row in c.execute("SELECT index_code, COUNT(*), SUM(in_current), SUM(was_removed) "
                         "FROM constituent_unified GROUP BY index_code ORDER BY index_code"):
        print("  ", row)
    print("sample 300:")
    for row in c.execute("SELECT * FROM constituent_unified WHERE index_code LIKE '%300%' LIMIT 3"):
        print("  ", row)
finally:
    c.close()

# 2. dbbardata('d') raw 个股 + ETF
section("dbbardata('d') sample")
c = sqlite3.connect(str(DB))
try:
    print("columns:", [r[1] for r in c.execute("PRAGMA table_info(dbbardata)").fetchall()])
    for sym, lbl in [("600519", "个股在市"), ("000005", "退市"), ("510300", "ETF")]:
        r = c.execute("SELECT COUNT(*), MIN(datetime), MAX(datetime) FROM dbbardata "
                      "WHERE symbol=? AND interval='d'", (sym,)).fetchone()
        print(f"  {sym}({lbl}): {r}")
    print("sample 600519 last 3:")
    for row in c.execute("SELECT symbol,exchange,datetime,volume,turnover,open_price,close_price "
                         "FROM dbbardata WHERE symbol='600519' AND interval='d' "
                         "ORDER BY datetime DESC LIMIT 3"):
        print("  ", row)
finally:
    c.close()

# 3. bs_adjust_factor
section("bs_adjust_factor schema")
c = sqlite3.connect(str(DB))
try:
    tabs = [r[0] for r in c.execute("SELECT name FROM sqlite_master WHERE type='table' "
                                    "AND name LIKE '%adjust%'").fetchall()]
    print("adjust tables:", tabs)
    if "bs_adjust_factor" in tabs:
        print("columns:", [r[1] for r in c.execute("PRAGMA table_info(bs_adjust_factor)").fetchall()])
        print("count:", c.execute("SELECT COUNT(*) FROM bs_adjust_factor").fetchone())
        print("sample 600519:")
        for row in c.execute("SELECT * FROM bs_adjust_factor WHERE code LIKE '%600519%' LIMIT 3"):
            print("  ", row)
finally:
    c.close()

# 4. valuation_baostock parquet
section("valuation_baostock parquet")
print("years:", sorted(p.name for p in VAL_DIR.glob("*.parquet")) if VAL_DIR.exists() else "DIR MISSING")
p2026 = VAL_DIR / "2026.parquet"
if p2026.exists():
    df = pd.read_parquet(p2026)
    print("columns:", list(df.columns))
    print("shape:", df.shape)
    print("sample 600519:")
    sub = df[df["symbol"].astype(str).str.contains("600519")] if "symbol" in df.columns else df.head(0)
    print(sub.head(3).to_dict("records"))

# 5. static akshare 三表 + valuation (market_cap/total_share 来源 + 文件格式)
from collections import Counter
section("static akshare (格式 + market_cap/total_share 来源)")
for sub in ["valuation", "balance", "income"]:
    d = BASE / "data" / "static" / sub
    if not d.exists():
        print(f"{sub}: DIR MISSING")
        continue
    allf = list(d.iterdir())
    exts = Counter(p.suffix for p in allf)
    print(f"{sub}: {len(allf)} files, ext分布={dict(exts)}")
    matches = sorted([p for p in allf if "600519" in p.name])
    if not matches:
        print("  600519 无文件")
        continue
    f0 = matches[0]
    with open(f0, "rb") as fh:
        head = fh.read(8)
    print(f"  600519 file={f0.name} magic={head[:4]!r}")
    if head[:4] == b"PAR1":
        df = pd.read_parquet(f0)
        print(f"  parquet cols({len(df.columns)}):", list(df.columns)[:20])
        if sub == "valuation":
            hit = [c for c in df.columns if any(k in str(c) for k in ("市值", "股本"))]
            print("  市值/股本列:", hit, "sample:", df.head(1).to_dict("records"))
        if sub == "balance":
            hit = [c for c in df.columns if any(k in str(c) for k in ("TOTAL_SHARES", "TOTAL_SHARE", "总股本", "实收资本"))]
            print("  股本相关列:", hit)
    else:
        print(f"  非 parquet (magic={head[:4]!r})")

# 6. 复权因子构造验证 (600519 foreAdjustFactor 语义)
section("复权因子 foreAdjustFactor 语义验证")
c = sqlite3.connect(str(DB))
try:
    rows = c.execute("SELECT dividOperateDate, foreAdjustFactor FROM bs_adjust_factor "
                     "WHERE code='sh.600519' ORDER BY dividOperateDate").fetchall()
    print("600519 除权事件数:", len(rows))
    print("前3:", rows[:3])
    print("后3:", rows[-3:])
    print("语义推断: foreAdjustFactor 递增=累计前复权因子; 最新事件后=1.0; "
          "qfq[t]=raw[t]*factor[t]")
finally:
    c.close()

print("\nPROBE DONE")