stock-analyzer-skill 1.1.0

package/scripts/fetchers/yfinance_kline.py

@@ -0,0 +1,66 @@
+"""Yahoo Finance K 线数据源（需要 yfinance 包）。"""
+import sys
+from pathlib import Path
+sys.path.insert(0, str(Path(__file__).resolve().parent.parent))
+
+from common import BaseFetcher
+
+try:
+    import yfinance as yf
+    HAS_YFINANCE = True
+except ImportError:
+    HAS_YFINANCE = False
+
+
+def _to_yf_symbol(code: str) -> str:
+    """转换为 yfinance 符号格式。"""
+    plain = code.lstrip("shszSHSZbjBJ").zfill(6)
+    if plain.startswith(("60", "68", "51", "56", "58")):
+        return f"{plain}.SS"
+    elif plain.startswith(("00", "30", "15", "16", "18")):
+        return f"{plain}.SZ"
+    elif plain.startswith(("43", "83", "87", "88", "92")):
+        return f"{plain}.BJ"
+    return code
+
+
+class YfinanceKlineFetcher(BaseFetcher):
+    """Yahoo Finance K 线数据源 (优先级 4) - 需要安装 yfinance 包。"""
+
+    def __init__(self):
+        super().__init__("yfinance_kline", priority=4)
+
+    def fetch(self, code: str, **kwargs) -> list | None:
+        if not HAS_YFINANCE:
+            return None
+        try:
+            scale = kwargs.get("scale", 240)
+            datalen = kwargs.get("datalen", 30)
+            symbol = _to_yf_symbol(code)
+
+            ticker = yf.Ticker(symbol)
+            if scale == 240:
+                df = ticker.history(period=f"{datalen}d")
+            elif scale == 60:
+                df = ticker.history(period="60d", interval="1h")
+            elif scale == 5:
+                df = ticker.history(period="5d", interval="5m")
+            else:
+                df = ticker.history(period=f"{datalen}d")
+
+            if df is None or df.empty:
+                return None
+
+            result = []
+            for idx, row in df.iterrows():
+                result.append({
+                    "day": str(idx)[:10],
+                    "open": str(round(row.get("Open", 0), 2)),
+                    "close": str(round(row.get("Close", 0), 2)),
+                    "high": str(round(row.get("High", 0), 2)),
+                    "low": str(round(row.get("Low", 0), 2)),
+                    "volume": str(int(row.get("Volume", 0))),
+                })
+            return result if result else None
+        except Exception:
+            return None

package/scripts/finance.py

@@ -0,0 +1,92 @@
+#!/usr/bin/env python3
+"""
+财务数据查询（多数据源自动切换）。
+数据源: 东方财富 → efinance → akshare
+用法:
+  finance.py SH600989                  # 单只，最近 4 季
+  finance.py -c SH600989,SZ000807      # 批量
+  finance.py -j SH600989               # JSON 输出
+  finance.py --sources                 # 显示可用数据源
+"""
+import sys
+import json
+from common import normalize_finance_code, parallel_map, err, DataError
+from data import get_finance
+
+
+def fetch(code: str, use_cache: bool = True) -> list:
+    """返回最近 4 季的财务数据（dict 列表，兼容旧接口）。"""
+    records = get_finance(code, use_cache=use_cache)
+    return [r.to_dict() for r in records]
+
+
+def render_table(records: list) -> str:
+    if not records:
+        return "(无数据)"
+    fields = [
+        ("eps", "每股收益"),
+        ("roe", "ROE%"),
+        ("revenue_yoy", "营收同比%"),
+        ("net_profit_yoy", "净利同比%"),
+        ("gross_margin", "毛利率%"),
+        ("net_margin", "净利率%"),
+        ("debt_ratio", "负债率%"),
+        ("bps", "每股净资产"),
+        ("ocf_per_share", "每股现金流"),
+    ]
+    lines = []
+    header = " | ".join(["报告期"] + [label for _, label in fields])
+    lines.append(header)
+    lines.append("-" * len(header))
+    for r in records:
+        period = r.get("report_date", "?")
+        values = [r.get(key, "-") for key, _ in fields]
+        lines.append(" | ".join([period] + [str(v)[:8] for v in values]))
+    return "\n".join(lines)
+
+
+def main():
+    if len(sys.argv) < 2:
+        err("用法: finance.py <代码> [-c codes] [-j] [--sources]")
+    args = sys.argv[1:]
+
+    if "--sources" in args:
+        from fetchers import get_finance_fetchers
+        fetchers = get_finance_fetchers()
+        print("可用财务数据源:")
+        for f in fetchers:
+            print(f"  - {f.name} (优先级 {f.priority})")
+        return
+
+    json_mode = "-j" in args
+    args = [a for a in args if a not in ("-j", "--sources")]
+    if not args:
+        err("缺少代码")
+
+    if args[0] == "-c":
+        codes = args[1].split(",")
+    else:
+        codes = [args[0]]
+
+    normalized_codes = [normalize_finance_code(c) for c in codes]
+
+    if len(normalized_codes) > 1:
+        results = parallel_map(fetch, normalized_codes, max_workers=4, timeout=30)
+        all_results = {k: v for k, v in results.items() if v}
+    else:
+        all_results = {normalized_codes[0]: fetch(normalized_codes[0])}
+
+    if json_mode:
+        print(json.dumps(all_results, ensure_ascii=False, indent=2))
+        return
+
+    for code, records in all_results.items():
+        print(f"\n=== {code} ===")
+        print(render_table(records))
+
+
+if __name__ == "__main__":
+    try:
+        main()
+    except DataError as e:
+        sys.exit(1)

package/scripts/init_pool.py

@@ -0,0 +1,105 @@
+#!/usr/bin/env python3
+"""
+首次安装初始化脚本 — 为每个板块拉取前 20 只股票。
+
+用法:
+  python3 scripts/init_pool.py              # 检测并初始化（已有数据则跳过）
+  python3 scripts/init_pool.py --force      # 强制重新初始化
+  python3 scripts/init_pool.py --top 30     # 每板块取 Top 30
+
+退出码始终为 0，不阻塞安装流程。
+"""
+
+import argparse
+import json
+import os
+import sys
+
+# 复用 refresh_pool 的核心逻辑
+sys.path.insert(0, os.path.dirname(__file__))
+from refresh_pool import (
+    POOL_FILE,
+    load_mapping,
+    load_current_pool,
+    refresh_pool,
+)
+
+# 初始化阈值：低于此值视为未初始化
+MIN_SECTORS = 10
+MIN_STOCKS = 100
+
+
+def is_pool_populated() -> tuple[bool, str]:
+    """检查股票池是否已有足够数据。返回 (是否已初始化, 描述)。"""
+    if not os.path.exists(POOL_FILE):
+        return False, "股票池文件不存在"
+
+    try:
+        with open(POOL_FILE, "r", encoding="utf-8") as f:
+            data = json.load(f)
+    except (json.JSONDecodeError, OSError):
+        return False, "股票池文件损坏"
+
+    sectors = {k: v for k, v in data.items() if not k.startswith("_")}
+    total_stocks = sum(len(v) for v in sectors.values())
+
+    if len(sectors) < MIN_SECTORS:
+        return False, f"仅 {len(sectors)} 个板块（需要 ≥{MIN_SECTORS}）"
+    if total_stocks < MIN_STOCKS:
+        return False, f"仅 {total_stocks} 只股票（需要 ≥{MIN_STOCKS}）"
+
+    return True, f"{len(sectors)} 个板块，{total_stocks} 只股票"
+
+
+def init_pool(top_n: int = 20, force: bool = False) -> bool:
+    """初始化股票池。返回是否实际执行了初始化。"""
+    # 检查是否已初始化
+    if not force:
+        populated, desc = is_pool_populated()
+        if populated:
+            print(f"✅ 股票池已存在（{desc}），跳过初始化")
+            print("   如需刷新，运行: python3 scripts/refresh_pool.py")
+            return False
+
+    # 检查 API Token
+    token = os.environ.get("EASTMONEY_API_TOKEN", "")
+    if not token:
+        print("⚠️  未设置 EASTMONEY_API_TOKEN，无法从东财 API 拉取数据")
+        print()
+        print("请按以下步骤操作：")
+        print("  1. 设置环境变量: export EASTMONEY_API_TOKEN=你的token")
+        print("     （东财 push2 API 的 ut 参数，可从网页版 F12 抓取）")
+        print("  2. 运行初始化: python3 scripts/init_pool.py --force")
+        print()
+        print("或直接运行: python3 scripts/refresh_pool.py")
+        return False
+
+    # 执行初始化
+    print(f"🚀 初始化股票池（每板块 Top {top_n}）...")
+    print()
+
+    try:
+        new_pool = refresh_pool(top_n=top_n, dry_run=False, show_diff=False)
+        total = sum(len(v) for v in new_pool.values())
+        print()
+        print(f"✅ 初始化完成: {len(new_pool)} 个板块，共 {total} 只股票")
+        return True
+    except Exception as e:
+        print(f"\n❌ 初始化失败: {e}", file=sys.stderr)
+        print("   可稍后重试: python3 scripts/init_pool.py --force")
+        return False
+
+
+def main():
+    parser = argparse.ArgumentParser(description="首次安装初始化股票池")
+    parser.add_argument("--force", "-f", action="store_true",
+                        help="强制重新初始化（忽略已有数据）")
+    parser.add_argument("--top", "-n", type=int, default=20,
+                        help="每板块取 Top N（默认 20）")
+    args = parser.parse_args()
+
+    init_pool(top_n=args.top, force=args.force)
+
+
+if __name__ == "__main__":
+    main()

package/scripts/kline.py

@@ -0,0 +1,62 @@
+#!/usr/bin/env python3
+"""
+K 线数据查询（多数据源自动切换）。
+数据源: 新浪 → 东方财富 → 腾讯 → efinance → akshare → tushare → 通达信 → baostock → yfinance
+用法:
+  kline.py sh600989                       # 日 K，30 根
+  kline.py sh600989 5 48                  # 5 分钟 K，48 根
+  kline.py sh600989 240 30 -j             # JSON
+  kline.py --sources                      # 显示可用数据源
+"""
+import sys
+import json
+from common import normalize_quote_code, err, DataError
+from data import get_kline
+
+
+def fetch(symbol: str, scale: int, datalen: int, use_cache: bool = True) -> list:
+    """获取 K 线数据，返回 dict 列表（兼容旧接口）。"""
+    bars = get_kline(symbol, scale, datalen, use_cache=use_cache)
+    return [b.to_dict() for b in bars]
+
+
+def render_table(records: list) -> str:
+    if not records:
+        return "(无数据)"
+    lines = []
+    for d in records:
+        lines.append(f"{d['day']} | O:{d['open']:>7} H:{d['high']:>7} L:{d['low']:>7} C:{d['close']:>7} V:{d['volume']:>12}")
+    return "\n".join(lines)
+
+
+def main():
+    args = sys.argv[1:]
+
+    if "--sources" in args:
+        from fetchers import get_kline_fetchers
+        fetchers = get_kline_fetchers()
+        print("可用 K 线数据源:")
+        for f in fetchers:
+            print(f"  - {f.name} (优先级 {f.priority})")
+        return
+
+    json_mode = "-j" in args
+    args = [a for a in args if a not in ("-j", "--sources")]
+    if not args:
+        err("用法: kline.py <symbol> [scale=240] [datalen=30] [-j]")
+    symbol = normalize_quote_code(args[0])
+    scale = int(args[1]) if len(args) > 1 else 240
+    datalen = int(args[2]) if len(args) > 2 else 30
+
+    records = fetch(symbol, scale, datalen)
+    if json_mode:
+        print(json.dumps(records, ensure_ascii=False, indent=2))
+    else:
+        print(render_table(records))
+
+
+if __name__ == "__main__":
+    try:
+        main()
+    except DataError as e:
+        sys.exit(1)

package/scripts/monitor.py

@@ -0,0 +1,107 @@
+#!/usr/bin/env python3
+"""
+数据源健康检查和缓存监控。
+用法:
+  python3 scripts/monitor.py              # 完整健康检查
+  python3 scripts/monitor.py --cache      # 缓存状态
+  python3 scripts/monitor.py --sources    # 数据源状态
+  python3 scripts/monitor.py --cleanup    # 清理过期缓存
+"""
+import argparse
+import json
+import os
+import sys
+import time
+from pathlib import Path
+
+sys.path.insert(0, str(Path(__file__).resolve().parent))
+from common import CACHE_DIR, cache_cleanup
+
+
+def check_cache_status():
+    """检查缓存目录状态。"""
+    if not CACHE_DIR.exists():
+        print("缓存目录不存在")
+        return
+
+    files = list(CACHE_DIR.glob("*.cache"))
+    total_size = sum(f.stat().st_size for f in files)
+    expired = sum(1 for f in files if time.time() - f.stat().st_mtime > 21600)
+
+    print(f"缓存文件数: {len(files)}")
+    print(f"缓存总大小: {total_size / 1024:.1f} KB")
+    print(f"过期文件数: {expired}")
+
+    # 按前缀统计
+    prefixes = {}
+    for f in files:
+        prefix = f.name.split("_")[0] if "_" in f.name else "other"
+        prefixes[prefix] = prefixes.get(prefix, 0) + 1
+    if prefixes:
+        print("\n按类型分布:")
+        for prefix, count in sorted(prefixes.items()):
+            print(f"  {prefix}: {count} 个")
+
+
+def check_sources():
+    """检查数据源可用性。"""
+    from fetchers import get_quote_fetchers, get_kline_fetchers, get_finance_fetchers
+
+    print("=== 行情数据源 ===")
+    for f in get_quote_fetchers():
+        status = "✅ 可用" if f.is_available() else "❌ 熔断"
+        print(f"  {f.name} (优先级 {f.priority}) - {status}")
+
+    print("\n=== K线数据源 ===")
+    for f in get_kline_fetchers():
+        status = "✅ 可用" if f.is_available() else "❌ 熔断"
+        print(f"  {f.name} (优先级 {f.priority}) - {status}")
+
+    print("\n=== 财务数据源 ===")
+    for f in get_finance_fetchers():
+        status = "✅ 可用" if f.is_available() else "❌ 熔断"
+        print(f"  {f.name} (优先级 {f.priority}) - {status}")
+
+
+def run_health_check():
+    """完整健康检查。"""
+    print("🔍 stock-analyzer-skill 健康检查\n")
+
+    print("--- 缓存状态 ---")
+    check_cache_status()
+
+    print("\n--- 数据源状态 ---")
+    check_sources()
+
+    print("\n--- 行业阈值配置 ---")
+    thresholds_path = Path(__file__).resolve().parent / "data" / "industry_thresholds.json"
+    if thresholds_path.exists():
+        thresholds = json.loads(thresholds_path.read_text(encoding="utf-8"))
+        industries = [k for k in thresholds.keys() if not k.startswith("_")]
+        print(f"  已配置 {len(industries)} 个行业: {', '.join(industries)}")
+    else:
+        print("  ⚠ 行业阈值配置文件不存在")
+
+    print("\n✅ 健康检查完成")
+
+
+def main():
+    parser = argparse.ArgumentParser(description="数据源健康检查和缓存监控")
+    parser.add_argument("--cache", action="store_true", help="显示缓存状态")
+    parser.add_argument("--sources", action="store_true", help="显示数据源状态")
+    parser.add_argument("--cleanup", action="store_true", help="清理过期缓存")
+    args = parser.parse_args()
+
+    if args.cache:
+        check_cache_status()
+    elif args.sources:
+        check_sources()
+    elif args.cleanup:
+        cleaned = cache_cleanup()
+        print(f"已清理 {cleaned} 个过期缓存文件")
+    else:
+        run_health_check()
+
+
+if __name__ == "__main__":
+    main()