feat: 港股数据补采2——hk_fundamentals_collector基本面采集(腾讯行情hk前缀,PE/PB/市值,复用A股解析模式,港股pb在field43)

2026-08-14 21:58:52 +08:00
parent ad5208c6cd
commit 1710d9821a
1 changed files with 89 additions and 0 deletions
@@ -0,0 +1,89 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+"""hk_fundamentals_collector.py — 港股通基本面采集（PE/PB/市值，阶段2数据补采）
+
+数据源：腾讯批量行情 hk 前缀（qt.gtimg.cn/q=hk00700,...）。
+复用 fundamentals_full_refresh 的解析模式，港股字段位置：
+  pe=parts[39]  pb=parts[43]（A股是46，港股是43）  总市值=parts[44]  流通市值=parts[45]（亿）
+写入：stock_fundamentals 表（与A股同表复用，code=5位港股代码，eps=price/pe）。
+调度：每日盘后（随A股基本面刷新节奏）。
+限速：批量100只/批，批间0.5s。
+"""
+import sqlite3
+import sys
+import time
+import urllib.request
+
+DB_PATH = "/home/hmo/MoFin/data/mofin.db"
+UA = "Mozilla/5.0"
+BATCH = 100
+
+
+def fetch_hk_batch(codes):
+    """腾讯批量行情（hk前缀，每批100只），返回 {code: {pe,pb,price,mcap_total,mcap_flow}}"""
+    if not codes:
+        return {}
+    results = {}
+    symbols = [f"hk{c}" for c in codes]
+    url = f"http://qt.gtimg.cn/q={','.join(symbols)}"
+    try:
+        req = urllib.request.Request(url, headers={"User-Agent": UA})
+        opener = urllib.request.build_opener(urllib.request.ProxyHandler({}))
+        with opener.open(req, timeout=15) as r:
+            text = r.read().decode("gbk", errors="ignore")
+        for line in text.strip().split("\n"):
+            if "~" not in line:
+                continue
+            parts = line.split("~")
+            if len(parts) < 46:
+                continue
+            code = parts[2]          # 港股代码（00700）
+            price = float(parts[3]) if parts[3] else 0
+            pe = float(parts[39]) if parts[39] else 0
+            pb = float(parts[43]) if parts[43] else 0          # 港股 PB 在 43（A股在46）
+            mcap_t = float(parts[44]) if parts[44] else 0       # 总市值（亿HKD）
+            mcap_f = float(parts[45]) if parts[45] else 0       # 流通市值（亿HKD）
+            if price > 0 and code:
+                results[code] = {"price": price, "pe": pe, "pb": pb,
+                                 "mcap_total": mcap_t, "mcap_flow": mcap_f}
+    except Exception as e:
+        print(f"  batch fetch error: {str(e)[:60]}", flush=True)
+    return results
+
+
+def main():
+    conn = sqlite3.connect(DB_PATH, timeout=30)
+    conn.execute("PRAGMA busy_timeout=30000")
+    codes = [r[0] for r in conn.execute(
+        "SELECT code FROM hk_connect_stocks WHERE is_active=1 ORDER BY code").fetchall()]
+    print(f"港股通基本面采集：{len(codes)} 只（腾讯行情 PE/PB/市值）", flush=True)
+
+    ok = 0
+    t0 = time.time()
+    for i in range(0, len(codes), BATCH):
+        batch = codes[i:i + BATCH]
+        data = fetch_hk_batch(batch)
+        for code, d in data.items():
+            eps = round(d["price"] / d["pe"], 2) if d["pe"] > 0 else 0
+            conn.execute(
+                "INSERT OR REPLACE INTO stock_fundamentals (code, pe, pb, eps, mcap_total, mcap_flow, updated_at) "
+                "VALUES (?,?,?,?,?,?,datetime('now','localtime'))",
+                (code, d["pe"], d["pb"], eps, d["mcap_total"], d["mcap_flow"]))
+            ok += 1
+        conn.commit()
+        print(f"  [{i+len(batch)}/{len(codes)}] 本批{len(data)}只 | {time.time()-t0:.0f}s", flush=True)
+        time.sleep(0.5)
+
+    n = conn.execute("SELECT COUNT(*) FROM stock_fundamentals WHERE length(code)=5").fetchone()[0]
+    sample = conn.execute(
+        "SELECT code, pe, pb, mcap_total FROM stock_fundamentals WHERE code IN ('00700','09988','01810')").fetchall()
+    conn.close()
+    print(f"\n完成：{ok} 只入 stock_fundamentals（港股 {n} 条），耗时{time.time()-t0:.0f}s", flush=True)
+    print("样例（code/pe/pb/总市值亿HKD）:", flush=True)
+    for r in sample:
+        print(f"  {r[0]}: pe={r[1]} pb={r[2]} mcap={r[3]:.0f}亿", flush=True)
+    return 0
+
+
+if __name__ == "__main__":
+    sys.exit(main())