@youhaozhao/cninfo-mcp 1.1.0 → 1.2.0

package/README.md

@@ -35,6 +35,7 @@
 ```
 查询 000888 的 2024 年报
 下载 688777 的年报
+查询 920185 的年报      # 北交所，新旧代码（如 835185）均可
 ```
 
 ## 系统要求
@@ -44,7 +45,7 @@
 
 ## 数据来源
 
-[巨潮资讯网](https://www.cninfo.com.cn) — 支持沪深两市及科创板
+[巨潮资讯网](https://www.cninfo.com.cn) — 支持沪深两市（主板、创业板、科创板）及北京证券交易所（北交所）
 
 ## Credits
 

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@youhaozhao/cninfo-mcp",
-  "version": "1.1.0",
+  "version": "1.2.0",
   "description": "MCP Server for querying and downloading Chinese listed companies' annual reports from CNINFO (巨潮资讯网)",
   "keywords": [
     "mcp",
@@ -45,8 +45,10 @@
   ],
   "files": [
     "bin/",
-    "python/",
-    "scripts/",
+    "python/mcp_server.py",
+    "python/spider.py",
+    "python/requirements.txt",
+    "scripts/install-python-deps.js",
     "README.md",
     "LICENSE"
   ]

package/python/spider.py

@@ -24,6 +24,13 @@ EARLIEST_DATE = "2001-01-01"
 PAGE_SIZE = 30
 # 翻页安全上限，防止异常情况下无限循环
 MAX_PAGES = 100
+# 巨潮搜索联想接口：把股票代码/简称解析为 orgId（北交所查询必需）
+TOP_SEARCH_URL = "http://www.cninfo.com.cn/new/information/topSearch/query"
+# 公告查询接口
+QUERY_URL = "http://www.cninfo.com.cn/new/hisAnnouncement/query"
+# 瞬时失败（网络抖动/限流）的重试次数与退避基数（秒）
+MAX_RETRIES = 3
+RETRY_BACKOFF = 1.0
 
 User_Agent = [
     "Mozilla/5.0 (compatible; MSIE 9.0; Windows NT 6.1; Win64; x64; Trident/5.0; .NET CLR 3.5.30729; .NET CLR 3.0.30729; .NET CLR 2.0.50727; Media Center PC 6.0)",
@@ -55,6 +62,85 @@ def _build_headers() -> dict:
     return headers
 
 
+def _post_json(url: str, data: dict) -> dict:
+    """POST 请求并解析 JSON，仅对可重试的瞬时失败按指数退避重试。
+
+    巨潮接口在批量/高频访问下会偶发超时或限流，导致本可成功的查询失败。
+    可重试：网络异常（超时/连接错误）、5xx、429，以及空/截断响应导致的
+    JSON 解析失败（requests 会抛 JSONDecodeError，实测多为限流时返回空体）。
+    不可重试：4xx（除 429）等客户端错误，快速失败，避免无谓的退避等待。
+    重试用尽后抛出最后一次异常，交由调用方的 try/except 记录并降级。
+    """
+    last_exc = None
+    for attempt in range(MAX_RETRIES):
+        try:
+            resp = requests.post(
+                url, headers=_build_headers(), data=data, timeout=30
+            )
+            resp.raise_for_status()
+            return resp.json()
+        except requests.exceptions.HTTPError as e:
+            status = e.response.status_code if e.response is not None else None
+            # 4xx（429 除外）不会自愈，立即失败，不浪费退避等待
+            if status is not None and status != 429 and 400 <= status < 500:
+                raise
+            last_exc = e
+        except requests.exceptions.RequestException as e:
+            # 网络异常 + JSONDecodeError（空/截断响应，多为瞬时限流）
+            last_exc = e
+        if attempt < MAX_RETRIES - 1:
+            time.sleep(RETRY_BACKOFF * (2**attempt) + random.random())
+    if last_exc is not None:
+        raise last_exc
+    raise RuntimeError(f"_post_json 未执行任何请求（MAX_RETRIES={MAX_RETRIES}）")
+
+
+def _query_announcements(query: dict) -> list:
+    """调用公告查询接口并返回 announcements 列表（带重试）。"""
+    result = _post_json(QUERY_URL, query)
+    if result and "announcements" in result and result["announcements"]:
+        return result["announcements"]
+    return []
+
+
+def _is_bse_code(stock_code) -> bool:
+    """判断是否为北交所代码。
+
+    北交所代码段：4xxxxx / 8xxxxx（原新三板平移），以及 92xxxx 标准段
+    （920，2024-04 起启用，预留 920-929）。这里特意只匹配 92 而非整个 9
+    开头，以排除沪市 B 股 900xxx，避免为其多发一次无效的北交所查询。
+    """
+    digits = re.sub(r"\D", "", str(stock_code or ""))
+    return digits[:1] in ("4", "8") or digits[:2] == "92"
+
+
+def _resolve_org_id(stock_code) -> Optional[tuple]:
+    """通过巨潮搜索联想接口把股票代码解析为 (code, orgId)。
+
+    北交所的 hisAnnouncement 接口不接受 searchkey 或裸代码，必须以
+    stock="代码,orgId" 的形式查询，因此需要先解析 orgId。
+    优先返回 code 完全等于输入的条目；找不到精确匹配则取第一条
+    （同一公司新旧代码共用同一 orgId）。无结果返回 None。
+    """
+    try:
+        hits = _post_json(TOP_SEARCH_URL, {"keyWord": stock_code, "maxNum": 10})
+    except Exception as e:
+        logger.warning("orgId 解析失败（%s）: %s", stock_code, e)
+        return None
+
+    if not isinstance(hits, list) or not hits:
+        return None
+
+    target = re.sub(r"\D", "", str(stock_code or ""))
+    for it in hits:
+        if str(it.get("code")) == target and it.get("orgId"):
+            return str(it.get("code")), str(it.get("orgId"))
+    first = hits[0]
+    if first.get("orgId"):
+        return str(first.get("code")), str(first.get("orgId"))
+    return None
+
+
 def _date_range(start_date: str) -> str:
     """构造查询时间区间，结束日期取当天，避免硬编码过期。"""
     datetime.datetime.strptime(start_date, "%Y-%m-%d")
@@ -125,7 +211,6 @@ def _is_annual_report_title(
 
 # 深市 年度报告
 def szseAnnual(page, stock):
-    query_path = "http://www.cninfo.com.cn/new/hisAnnouncement/query"
     query = {
         "pageNum": page,  # 页码
         "pageSize": PAGE_SIZE,
@@ -140,18 +225,11 @@ def szseAnnual(page, stock):
         "seDate": _date_range(EARLIEST_DATE),  # 时间区间
     }
 
-    namelist = requests.post(
-        query_path, headers=_build_headers(), data=query, timeout=30
-    )
-    result = namelist.json()
-    if result and "announcements" in result and result["announcements"]:
-        return result["announcements"]
-    return []
+    return _query_announcements(query)
 
 
 # 沪市 年度报告
 def sseAnnual(page, stock):
-    query_path = "http://www.cninfo.com.cn/new/hisAnnouncement/query"
     query = {
         "pageNum": page,  # 页码
         "pageSize": PAGE_SIZE,
@@ -166,18 +244,35 @@ def sseAnnual(page, stock):
         "seDate": _date_range(EARLIEST_DATE),  # 时间区间
     }
 
-    namelist = requests.post(
-        query_path, headers=_build_headers(), data=query, timeout=30
-    )
-    result = namelist.json()
-    if result and "announcements" in result and result["announcements"]:
-        return result["announcements"]
-    return []
+    return _query_announcements(query)
+
+
+# 北交所 年度报告
+def bseAnnual(page, stock):
+    """北交所年报查询。
+
+    stock 形如 "代码,orgId"，由 _resolve_org_id 解析得到。北交所必须
+    通过 plate=bj + stock="代码,orgId" 查询，searchkey/裸代码均返回空。
+    """
+    query = {
+        "pageNum": page,  # 页码
+        "pageSize": PAGE_SIZE,
+        "tabName": "fulltext",
+        "column": "bj",  # 北交所
+        "stock": stock,  # 必须为 "代码,orgId"
+        "searchkey": "",
+        "secid": "",
+        "plate": "bj",
+        "category": "category_ndbg_szsh",  # 年度报告
+        "trade": "",
+        "seDate": _date_range(EARLIEST_DATE),  # 时间区间
+    }
+
+    return _query_announcements(query)
 
 
 # 深市 招股
 def szseStock(page, stock):
-    query_path = "http://www.cninfo.com.cn/new/hisAnnouncement/query"
     query = {
         "pageNum": page,  # 页码
         "pageSize": PAGE_SIZE,
@@ -192,18 +287,11 @@ def szseStock(page, stock):
         "seDate": _date_range(EARLIEST_DATE),  # 时间区间
     }
 
-    namelist = requests.post(
-        query_path, headers=_build_headers(), data=query, timeout=30
-    )
-    result = namelist.json()
-    if result and "announcements" in result and result["announcements"]:
-        return result["announcements"]
-    return []
+    return _query_announcements(query)
 
 
 # 沪市 招股
 def sseStock(page, stock):
-    query_path = "http://www.cninfo.com.cn/new/hisAnnouncement/query"
     query = {
         "pageNum": page,  # 页码
         "pageSize": PAGE_SIZE,
@@ -218,13 +306,7 @@ def sseStock(page, stock):
         "seDate": _date_range(EARLIEST_DATE),  # 时间区间
     }
 
-    namelist = requests.post(
-        query_path, headers=_build_headers(), data=query, timeout=30
-    )
-    result = namelist.json()
-    if result and "announcements" in result and result["announcements"]:
-        return result["announcements"]
-    return []
+    return _query_announcements(query)
 
 
 def Download(
@@ -362,6 +444,18 @@ def query_annual_reports(stock_code, year=None):
     except Exception as e:
         logger.warning("深市年报查询失败: %s", e)
 
+    # 查询北交所（代码以 4/8/9 开头）。北交所接口必须用 orgId，
+    # 故先解析 orgId 再以 stock="代码,orgId" 翻页查询。
+    if _is_bse_code(stock_code):
+        try:
+            resolved = _resolve_org_id(stock_code)
+            if resolved:
+                code, org_id = resolved
+                announcements_bse = _paginate(bseAnnual, f"{code},{org_id}")
+                all_announcements.extend(announcements_bse)
+        except Exception as e:
+            logger.warning("北交所年报查询失败: %s", e)
+
     # 按年份过滤
     if year:
         year_expr = re.escape(str(year))

package/python/bridge.py

@@ -1,161 +0,0 @@
-#!/usr/bin/env python3
-"""CLI bridge that exposes cninfo spider functions as JSON in/out for Node.js."""
-
-import json
-import os
-import sys
-import traceback
-from typing import Any, Optional
-
-sys.path.insert(0, os.path.dirname(os.path.abspath(__file__)))
-
-from spider import (  # noqa: E402
-    download_annual_reports,
-    download_prospectus,
-    query_annual_reports,
-    query_prospectus,
-    saving_path,
-)
-
-
-BASE_URL = "https://static.cninfo.com.cn/"
-
-
-def _format_reports(reports: list[dict]) -> list[dict]:
-    formatted = []
-    for report in reports:
-        adj = report.get("adjunctUrl", "")
-        formatted.append(
-            {
-                "announcementTitle": report.get("announcementTitle", ""),
-                "announcementTime": report.get("announcementTime", ""),
-                "secCode": report.get("secCode", ""),
-                "secName": report.get("secName", ""),
-                "adjunctUrl": BASE_URL + adj if adj else "",
-            }
-        )
-    return formatted
-
-
-def _require_stock_code(payload: dict) -> str:
-    stock_code = (payload.get("stock_code") or "").strip()
-    if not stock_code:
-        raise ValueError("stock_code is required")
-    return stock_code
-
-
-def _optional_year(payload: dict) -> Optional[int]:
-    year = payload.get("year")
-    if year is None:
-        return None
-    return int(year)
-
-
-def _resolve_save_path(payload: dict) -> str:
-    save_path = payload.get("save_path")
-    return save_path if save_path else saving_path
-
-
-def action_query_annual_reports(payload: dict) -> dict:
-    stock_code = _require_stock_code(payload)
-    year = _optional_year(payload)
-    reports = query_annual_reports(stock_code, year)
-    suffix = f" for year {year}" if year else ""
-    if not reports:
-        return {
-            "success": False,
-            "stock_code": stock_code,
-            "year": year,
-            "count": 0,
-            "reports": [],
-            "message": f"No annual reports found for stock {stock_code}{suffix}",
-        }
-    return {
-        "success": True,
-        "stock_code": stock_code,
-        "year": year,
-        "count": len(reports),
-        "reports": _format_reports(reports),
-        "message": f"Found {len(reports)} annual report(s){suffix}",
-    }
-
-
-def action_download_annual_reports(payload: dict) -> dict:
-    stock_code = _require_stock_code(payload)
-    year = _optional_year(payload)
-    output_dir = _resolve_save_path(payload)
-    os.makedirs(output_dir, exist_ok=True)
-    result = download_annual_reports(stock_code, year, save_path=output_dir)
-    result["stock_code"] = stock_code
-    result["year"] = year
-    return result
-
-
-def action_query_prospectus(payload: dict) -> dict:
-    stock_code = _require_stock_code(payload)
-    reports = query_prospectus(stock_code)
-    if not reports:
-        return {
-            "success": False,
-            "stock_code": stock_code,
-            "count": 0,
-            "reports": [],
-            "message": f"No prospectus found for stock {stock_code}",
-        }
-    return {
-        "success": True,
-        "stock_code": stock_code,
-        "count": len(reports),
-        "reports": _format_reports(reports),
-        "message": f"Found {len(reports)} prospectus document(s)",
-    }
-
-
-def action_download_prospectus(payload: dict) -> dict:
-    stock_code = _require_stock_code(payload)
-    output_dir = _resolve_save_path(payload)
-    os.makedirs(output_dir, exist_ok=True)
-    result = download_prospectus(stock_code, save_path=output_dir)
-    result["stock_code"] = stock_code
-    return result
-
-
-ACTIONS = {
-    "query_annual_reports": action_query_annual_reports,
-    "download_annual_reports": action_download_annual_reports,
-    "query_prospectus": action_query_prospectus,
-    "download_prospectus": action_download_prospectus,
-}
-
-
-def main() -> int:
-    if len(sys.argv) < 3:
-        sys.stderr.write("usage: bridge.py <action> <json-payload>\n")
-        return 2
-
-    action = sys.argv[1]
-    raw_payload = sys.argv[2]
-
-    handler = ACTIONS.get(action)
-    if handler is None:
-        sys.stdout.write(json.dumps({"ok": False, "error": f"Unknown action: {action}"}))
-        return 1
-
-    try:
-        payload = json.loads(raw_payload) if raw_payload else {}
-    except json.JSONDecodeError as exc:
-        sys.stdout.write(json.dumps({"ok": False, "error": f"Invalid JSON payload: {exc}"}))
-        return 1
-
-    try:
-        result: Any = handler(payload)
-        sys.stdout.write(json.dumps({"ok": True, "payload": result}, ensure_ascii=False))
-        return 0
-    except Exception as exc:
-        sys.stderr.write(traceback.format_exc())
-        sys.stdout.write(json.dumps({"ok": False, "error": str(exc)}))
-        return 1
-
-
-if __name__ == "__main__":
-    raise SystemExit(main())