feat(HEL-463): 接入剩余行情数据到 datahub

扩展盘后正式集（涨跌停/人气/龙虎榜/板块日线）与盘中观察 API（报价/指数/分时），网站 bridge 按开关接入并回退旧链路；问天改为按数据依赖跟随开关，不再整栈强制旧路径。

Co-authored-by: Cursor <cursoragent@cursor.com>
Co-authored-by: multica-agent <github@multica.ai>

xiaobai-datahub/datahub/adapters/__init__.py

@@ -1,13 +1,13 @@
 from datahub.adapters.akshare import ADAPTER as akshare
-from datahub.adapters.eastmoney import ADAPTER as eastmoney
+from datahub.adapters.eastmoney import EastmoneyAdapter
 from datahub.adapters.ifind import ADAPTER as ifind
-from datahub.adapters.tencent import ADAPTER as tencent
+from datahub.adapters.tencent import TencentAdapter
 from datahub.adapters.ths import ADAPTER as ths
 from datahub.adapters.xgb import ADAPTER as xgb
 
 RESERVED = {
-    "eastmoney": eastmoney,
-    "tencent": tencent,
+    "eastmoney": EastmoneyAdapter(),
+    "tencent": TencentAdapter(),
     "ths": ths,
     "xgb": xgb,
     "akshare": akshare,

xiaobai-datahub/datahub/adapters/eastmoney.py

@@ -1,3 +1,250 @@
-from datahub.adapters.base import ReservedAdapter
+from __future__ import annotations
 
-ADAPTER = ReservedAdapter("eastmoney")
+import json
+import time
+import urllib.error
+import urllib.parse
+import urllib.request
+from datetime import datetime
+from typing import Any
+
+from datahub.adapters.base import AdapterError, MarketAdapter
+from datahub.numbers import finite_number, round4
+
+EASTMONEY_INDEX_URL = "https://push2.eastmoney.com/api/qt/ulist.np/get"
+EASTMONEY_CLIST_URL = "https://push2.eastmoney.com/api/qt/clist/get"
+TRENDS_URL = "https://push2delay.eastmoney.com/api/qt/stock/trends2/get"
+BROWSER_UA = (
+    "Mozilla/5.0 (Windows NT 10.0; Win64; x64) "
+    "AppleWebKit/537.36 (KHTML, like Gecko) Chrome/138.0.0.0 Safari/537.36"
+)
+INDEX_SECIDS = {
+    "000001.SH": "1.000001",
+    "399001.SZ": "0.399001",
+    "399006.SZ": "0.399006",
+}
+
+
+class EastmoneyAdapter(MarketAdapter):
+    name = "eastmoney"
+
+    def __init__(self, timeout: int = 8) -> None:
+        self.timeout = timeout
+
+    def probe(self) -> dict[str, Any]:
+        started = time.perf_counter()
+        try:
+            rows = self.fetch_indices()
+            state = "ok" if len(rows) == 3 else "empty"
+        except AdapterError as exc:
+            return {
+                "provider": self.name,
+                "configured": True,
+                "state": "error",
+                "message": str(exc),
+                "latency_ms": round((time.perf_counter() - started) * 1000),
+            }
+        return {
+            "provider": self.name,
+            "configured": True,
+            "state": state,
+            "latency_ms": round((time.perf_counter() - started) * 1000),
+        }
+
+    def fetch(self, dataset: str, params: dict[str, Any]) -> list[dict[str, Any]]:
+        if dataset in {"indexes_quotes", "index_quotes"}:
+            return self.fetch_indices()
+        if dataset in {"quotes", "quotes_latest"}:
+            codes = params.get("codes") or []
+            if isinstance(codes, str):
+                codes = [item.strip() for item in codes.split(",") if item.strip()]
+            return self.fetch_quotes(list(codes))
+        raise AdapterError(f"{self.name} unsupported dataset: {dataset}")
+
+    def normalize(self, dataset: str, rows: list[dict[str, Any]]) -> list[dict[str, Any]]:
+        return list(rows)
+
+    def fetch_indices(self) -> list[dict[str, Any]]:
+        payload = self._get_json(
+            EASTMONEY_INDEX_URL,
+            {
+                "secids": "1.000001,0.399001,0.399006",
+                "fltt": "2",
+                "invt": "2",
+                "fields": "f12,f14,f2,f3,f4,f15,f16,f17,f18,f6,f124",
+            },
+            referer="https://quote.eastmoney.com/",
+        )
+        rows = list((payload.get("data") or {}).get("diff") or [])
+        result = []
+        for row in rows:
+            code = str(row.get("f12") or "")
+            if code not in {"000001", "399001", "399006"}:
+                continue
+            epoch = int(finite_number(row.get("f124")) or 0)
+            ts_code = f"{code}.SH" if code.startswith("0") and code == "000001" else f"{code}.SZ"
+            if code == "000001":
+                ts_code = "000001.SH"
+            result.append(
+                {
+                    "ts_code": ts_code,
+                    "code": code,
+                    "name": row.get("f14") or code,
+                    "price": round4(finite_number(row.get("f2"))),
+                    "pct_chg": round4(finite_number(row.get("f3"))),
+                    "change_amount": round4(finite_number(row.get("f4"))),
+                    "open": round4(finite_number(row.get("f17"))),
+                    "high": round4(finite_number(row.get("f15"))),
+                    "low": round4(finite_number(row.get("f16"))),
+                    "previous_close": round4(finite_number(row.get("f18"))),
+                    "amount": round4(finite_number(row.get("f6"))),
+                    "quote_time_epoch": epoch,
+                    "quote_time": (
+                        datetime.fromtimestamp(epoch).astimezone().isoformat(timespec="seconds")
+                        if epoch
+                        else ""
+                    ),
+                    "source": "eastmoney_push2",
+                }
+            )
+        if len(result) != 3:
+            raise AdapterError(f"Eastmoney returned {len(result)}/3 indices")
+        return result
+
+    def fetch_quotes(self, codes: list[str]) -> list[dict[str, Any]]:
+        # Eastmoney clist does not accept arbitrary code lists well; use ulist.np for batches.
+        secids = []
+        for code in codes:
+            ts = str(code or "").upper()
+            symbol = ts.split(".")[0]
+            if ts.endswith(".SH") or symbol.startswith(("5", "6", "9")):
+                secids.append(f"1.{symbol}")
+            else:
+                secids.append(f"0.{symbol}")
+        if not secids:
+            return []
+        payload = self._get_json(
+            EASTMONEY_INDEX_URL,
+            {
+                "secids": ",".join(secids[:60]),
+                "fltt": "2",
+                "invt": "2",
+                "fields": "f12,f14,f2,f3,f4,f15,f16,f17,f18,f5,f6,f8,f124",
+            },
+            referer="https://quote.eastmoney.com/",
+        )
+        rows = list((payload.get("data") or {}).get("diff") or [])
+        result = []
+        for row in rows:
+            symbol = str(row.get("f12") or "")
+            if not symbol:
+                continue
+            ts_code = f"{symbol}.SH" if symbol.startswith(("5", "6", "9")) else f"{symbol}.SZ"
+            epoch = int(finite_number(row.get("f124")) or 0)
+            result.append(
+                {
+                    "ts_code": ts_code,
+                    "name": row.get("f14") or symbol,
+                    "price": round4(finite_number(row.get("f2"))),
+                    "pct_chg": round4(finite_number(row.get("f3"))),
+                    "change_amount": round4(finite_number(row.get("f4"))),
+                    "open": round4(finite_number(row.get("f17"))),
+                    "high": round4(finite_number(row.get("f15"))),
+                    "low": round4(finite_number(row.get("f16"))),
+                    "previous_close": round4(finite_number(row.get("f18"))),
+                    "volume": round4(finite_number(row.get("f5"))),
+                    "amount": round4(finite_number(row.get("f6"))),
+                    "turnover_rate": round4(finite_number(row.get("f8"))),
+                    "quote_time_epoch": epoch,
+                    "quote_time": (
+                        datetime.fromtimestamp(epoch).astimezone().isoformat(timespec="seconds")
+                        if epoch
+                        else ""
+                    ),
+                    "source": "eastmoney_push2",
+                }
+            )
+        return result
+
+    def fetch_intraday(self, ts_code: str) -> dict[str, Any]:
+        code = str(ts_code or "").upper()
+        if code in INDEX_SECIDS:
+            secid = INDEX_SECIDS[code]
+            entity = "index"
+            identifier = code
+        else:
+            symbol = code.split(".")[0]
+            market = "1" if symbol.startswith(("5", "6", "9")) else "0"
+            secid = f"{market}.{symbol}"
+            entity = "stock"
+            identifier = symbol
+        payload = self._get_json(
+            TRENDS_URL,
+            {
+                "secid": secid,
+                "fields1": "f1,f2,f3,f4,f5,f6,f7,f8,f9,f10,f11,f12,f13",
+                "fields2": "f51,f52,f53,f54,f55,f56,f57,f58",
+                "iscr": "0",
+                "ndays": "1",
+            },
+            referer="https://quote.eastmoney.com/",
+        )
+        data = payload.get("data") or {}
+        points = []
+        for raw in data.get("trends") or []:
+            point = _parse_trend(raw)
+            if point:
+                points.append(point)
+        if not points:
+            raise AdapterError("No intraday chart data returned")
+        return {
+            "entity_type": entity,
+            "identifier": identifier,
+            "ts_code": code if "." in code else f"{identifier}.{'SH' if identifier.startswith(('5','6','9')) else 'SZ'}",
+            "name": str(data.get("name") or ""),
+            "code": str(data.get("code") or identifier),
+            "trade_date": points[-1]["date"],
+            "previous_close": round4(finite_number(data.get("preClose"))),
+            "points": points,
+            "source": "eastmoney_trends2",
+        }
+
+    def _get_json(self, url: str, params: dict[str, str], referer: str) -> dict[str, Any]:
+        request_url = f"{url}?{urllib.parse.urlencode(params)}"
+        request = urllib.request.Request(
+            request_url,
+            headers={
+                "Accept": "application/json,text/plain,*/*",
+                "User-Agent": BROWSER_UA,
+                "Referer": referer,
+            },
+            method="GET",
+        )
+        try:
+            with urllib.request.urlopen(request, timeout=self.timeout) as response:
+                return json.loads(response.read().decode("utf-8"))
+        except Exception as exc:
+            raise AdapterError(f"eastmoney request failed: {exc}") from exc
+
+
+def _parse_trend(raw: Any) -> dict[str, Any] | None:
+    text = str(raw or "")
+    parts = text.split(",")
+    if len(parts) < 8:
+        return None
+    stamp = parts[0]
+    try:
+        when = datetime.strptime(stamp, "%Y-%m-%d %H:%M")
+    except ValueError:
+        return None
+    return {
+        "time": when.strftime("%H:%M"),
+        "date": when.strftime("%Y-%m-%d"),
+        "open": round4(finite_number(parts[1])),
+        "close": round4(finite_number(parts[2])),
+        "high": round4(finite_number(parts[3])),
+        "low": round4(finite_number(parts[4])),
+        "avg_price": round4(finite_number(parts[7] if len(parts) > 7 else parts[2])),
+        "volume": round4(finite_number(parts[5])),
+        "amount": round4(finite_number(parts[6])),
+    }

xiaobai-datahub/datahub/adapters/tencent.py

@@ -1,3 +1,99 @@
-from datahub.adapters.base import ReservedAdapter
+from __future__ import annotations
 
-ADAPTER = ReservedAdapter("tencent")
+import time
+import urllib.error
+import urllib.request
+from datetime import datetime
+from typing import Any
+
+from datahub.adapters.base import AdapterError, MarketAdapter
+from datahub.numbers import finite_number, round4
+
+TENCENT_INDEX_URL = "https://qt.gtimg.cn/q=sh000001,sz399001,sz399006"
+BROWSER_UA = (
+    "Mozilla/5.0 (Windows NT 10.0; Win64; x64) "
+    "AppleWebKit/537.36 (KHTML, like Gecko) Chrome/138.0.0.0 Safari/537.36"
+)
+
+
+class TencentAdapter(MarketAdapter):
+    name = "tencent"
+
+    def __init__(self, timeout: int = 8) -> None:
+        self.timeout = timeout
+
+    def probe(self) -> dict[str, Any]:
+        started = time.perf_counter()
+        try:
+            rows = self.fetch_indices()
+            state = "ok" if len(rows) == 3 else "empty"
+        except AdapterError as exc:
+            return {
+                "provider": self.name,
+                "configured": True,
+                "state": "error",
+                "message": str(exc),
+                "latency_ms": round((time.perf_counter() - started) * 1000),
+            }
+        return {
+            "provider": self.name,
+            "configured": True,
+            "state": state,
+            "latency_ms": round((time.perf_counter() - started) * 1000),
+        }
+
+    def fetch(self, dataset: str, params: dict[str, Any]) -> list[dict[str, Any]]:
+        if dataset in {"indexes_quotes", "index_quotes"}:
+            return self.fetch_indices()
+        raise AdapterError(f"{self.name} unsupported dataset: {dataset}")
+
+    def normalize(self, dataset: str, rows: list[dict[str, Any]]) -> list[dict[str, Any]]:
+        return list(rows)
+
+    def fetch_indices(self) -> list[dict[str, Any]]:
+        request = urllib.request.Request(
+            TENCENT_INDEX_URL,
+            headers={"User-Agent": BROWSER_UA, "Referer": "https://gu.qq.com/"},
+            method="GET",
+        )
+        try:
+            with urllib.request.urlopen(request, timeout=self.timeout) as response:
+                raw = response.read().decode("gb18030", errors="ignore")
+        except Exception as exc:
+            raise AdapterError(f"tencent request failed: {exc}") from exc
+        result = []
+        for line in raw.splitlines():
+            if '="' not in line:
+                continue
+            fields = line.split('="', 1)[1].rsplit('";', 1)[0].split("~")
+            if len(fields) < 38:
+                continue
+            code = fields[2]
+            if code not in {"000001", "399001", "399006"}:
+                continue
+            try:
+                quote_time = datetime.strptime(fields[30], "%Y%m%d%H%M%S").astimezone()
+            except ValueError as exc:
+                raise AdapterError(f"Tencent invalid quote time for {code}") from exc
+            ts_code = "000001.SH" if code == "000001" else f"{code}.SZ"
+            result.append(
+                {
+                    "ts_code": ts_code,
+                    "code": code,
+                    "name": fields[1] or code,
+                    "price": round4(finite_number(fields[3])),
+                    "pct_chg": round4(finite_number(fields[32])),
+                    "change_amount": round4(finite_number(fields[31])),
+                    "open": round4(finite_number(fields[5])),
+                    "high": round4(finite_number(fields[33])),
+                    "low": round4(finite_number(fields[34])),
+                    "previous_close": round4(finite_number(fields[4])),
+                    "amount": round4(finite_number(fields[37]) * 10000),
+                    "quote_time_epoch": int(quote_time.timestamp()),
+                    "quote_time": quote_time.isoformat(timespec="seconds"),
+                    "source": "tencent_qt",
+                }
+            )
+        if len(result) != 3:
+            raise AdapterError(f"Tencent returned {len(result)}/3 indices")
+        return result

xiaobai-datahub/datahub/adapters/tushare.py

@@ -11,8 +11,12 @@ from datahub.normalize import (
     normalize_auction,
     normalize_calendar,
     normalize_daily,
+    normalize_dragon_tiger,
     normalize_index_daily,
+    normalize_limit_event,
     normalize_moneyflow,
+    normalize_popularity,
+    normalize_sector_daily,
     normalize_stock,
     normalize_valuation,
 )
@@ -31,6 +35,21 @@ TUSHARE_FIELDS = {
         "buy_lg_amount,sell_lg_amount,buy_elg_amount,sell_elg_amount,net_mf_amount"
     ),
     "stk_auction": "ts_code,trade_date,vol,price,amount,pre_close,turnover_rate,volume_ratio,float_share",
+    "limit_list_d": (
+        "trade_date,ts_code,industry,name,close,pct_chg,amount,limit_amount,"
+        "float_mv,total_mv,turnover_ratio,fd_amount,first_time,last_time,"
+        "open_times,up_stat,limit_times,limit_type"
+    ),
+    "ths_hot": "ts_code,ts_name,hot,rank,pct_change,current_price,concept,data_type,trade_date",
+    "dc_hot": "ts_code,ts_name,rank,pct_change,current_price,hot,concept,data_type,trade_date",
+    "hm_detail": "trade_date,ts_code,ts_name,buy_amount,sell_amount,net_amount,hm_name,hm_orgs,tag",
+    "hm_list": "name,desc,orgs",
+    "top_list": "trade_date,ts_code,name,pct_change,reason",
+    "top_inst": "trade_date,ts_code,exalter,buy,buy_rate,sell,sell_rate,net_buy,side,reason",
+    "ths_index": "ts_code,name,count,exchange,list_date,type",
+    "ths_daily": "ts_code,trade_date,open,high,low,close,pre_close,pct_change,vol,turnover_rate",
+    "dc_index": "ts_code,trade_date,name,open,high,low,close,pre_close,pct_change,vol,amount,turnover_rate",
+    "sw_daily": "ts_code,trade_date,name,open,high,low,close,pct_change,vol,amount",
 }
 
 DATASET_API = {
@@ -42,12 +61,15 @@ DATASET_API = {
     "index_daily": "index_daily",
     "moneyflow": "moneyflow",
     "auction": "stk_auction",
+    "limit_events": "limit_list_d",
+    "popularity": "ths_hot",
+    "dragon_tiger": "hm_detail",
+    "sector_daily": "ths_daily",
 }
 
-# Website actual index usage: market cards / 90-day charts (SH/SZ/CYB) plus
-# screener 沪深300 benchmark (lookback up to 260 trading days).
 WEBSITE_INDEX_CODES = ("000001.SH", "399001.SZ", "399006.SZ", "000300.SH")
 DEFAULT_INDEX_CODES = WEBSITE_INDEX_CODES
+LIMIT_TYPES = ("U", "D", "Z")
 
 
 class TushareAdapter(MarketAdapter):
@@ -85,6 +107,14 @@ class TushareAdapter(MarketAdapter):
         }
 
     def fetch(self, dataset: str, params: dict[str, Any]) -> list[dict[str, Any]]:
+        if dataset == "limit_events":
+            return self.fetch_limit_events(str(params.get("trade_date") or ""))
+        if dataset == "popularity":
+            return self.fetch_popularity(str(params.get("trade_date") or ""))
+        if dataset == "dragon_tiger":
+            return self.fetch_dragon_tiger(str(params.get("trade_date") or ""))
+        if dataset == "sector_daily":
+            return self.fetch_sector_daily(str(params.get("trade_date") or ""))
         api_name = DATASET_API.get(dataset, dataset)
         fields = TUSHARE_FIELDS.get(api_name, "")
         query_params = dict(params)
@@ -93,10 +123,67 @@ class TushareAdapter(MarketAdapter):
         if api_name == "trade_cal" and "exchange" not in query_params:
             query_params["exchange"] = "SSE"
         if api_name == "index_daily" and "ts_code" not in query_params:
-            # Caller typically loops codes; a missing code would pull nothing useful.
             query_params.setdefault("ts_code", DEFAULT_INDEX_CODES[0])
         return self._query(api_name, query_params, fields)
 
+    def fetch_limit_events(self, trade_date: str) -> list[dict[str, Any]]:
+        rows: list[dict[str, Any]] = []
+        for limit_type in LIMIT_TYPES:
+            part = self._query(
+                "limit_list_d",
+                {"trade_date": trade_date, "limit_type": limit_type},
+                TUSHARE_FIELDS["limit_list_d"],
+            )
+            for row in part:
+                row = dict(row)
+                row.setdefault("limit_type", limit_type)
+                rows.append(row)
+        return rows
+
+    def fetch_popularity(self, trade_date: str) -> list[dict[str, Any]]:
+        rows: list[dict[str, Any]] = []
+        for api_name, source in (("ths_hot", "ths"), ("dc_hot", "dc")):
+            for row in self._query(api_name, {"trade_date": trade_date}, TUSHARE_FIELDS[api_name]):
+                item = dict(row)
+                item["source"] = source
+                item.setdefault("trade_date", trade_date)
+                rows.append(item)
+        return rows
+
+    def fetch_dragon_tiger(self, trade_date: str) -> list[dict[str, Any]]:
+        details = self._query("hm_detail", {"trade_date": trade_date}, TUSHARE_FIELDS["hm_detail"])
+        top_rows = self._query("top_list", {"trade_date": trade_date}, TUSHARE_FIELDS["top_list"])
+        context = {
+            str(row.get("ts_code") or ""): row
+            for row in top_rows
+            if str(row.get("ts_code") or "")
+        }
+        rows: list[dict[str, Any]] = []
+        for row in details:
+            item = dict(row)
+            stock = context.get(str(item.get("ts_code") or ""), {})
+            if item.get("pct_change") is None and stock.get("pct_change") is not None:
+                item["pct_change"] = stock.get("pct_change")
+            if not item.get("reason") and stock.get("reason"):
+                item["reason"] = stock.get("reason")
+            if not item.get("ts_name") and stock.get("name"):
+                item["ts_name"] = stock.get("name")
+            rows.append(item)
+        return rows
+
+    def fetch_sector_daily(self, trade_date: str) -> list[dict[str, Any]]:
+        rows: list[dict[str, Any]] = []
+        for api_name, family in (("ths_daily", "ths"), ("dc_index", "dc"), ("sw_daily", "sw")):
+            try:
+                part = self._query(api_name, {"trade_date": trade_date}, TUSHARE_FIELDS[api_name])
+            except AdapterError:
+                part = []
+            for row in part:
+                item = dict(row)
+                item["family"] = family
+                rows.append(item)
+        return rows
+
     def fetch_index_daily(self, trade_date: str, codes: tuple[str, ...] = DEFAULT_INDEX_CODES) -> list[dict[str, Any]]:
         rows: list[dict[str, Any]] = []
         for ts_code in codes:
@@ -104,6 +191,17 @@ class TushareAdapter(MarketAdapter):
         return rows
 
     def normalize(self, dataset: str, rows: list[dict[str, Any]]) -> list[dict[str, Any]]:
+        if dataset in {"limit_events", "limit_list_d"}:
+            return [normalize_limit_event(row) for row in rows]
+        if dataset == "popularity":
+            return [normalize_popularity(row, source=str(row.get("source") or "")) for row in rows]
+        if dataset == "dragon_tiger":
+            return [normalize_dragon_tiger(row) for row in rows]
+        if dataset == "sector_daily":
+            return [
+                normalize_sector_daily(row, family=str(row.get("family") or "ths"))
+                for row in rows
+            ]
         mapping = {
             "calendar": normalize_calendar,
             "trade_cal": normalize_calendar,
@@ -148,12 +246,11 @@ class TushareAdapter(MarketAdapter):
         try:
             with urllib.request.urlopen(request, timeout=self.timeout) as response:
                 result = json.loads(response.read().decode("utf-8"))
-        except json.JSONDecodeError:
-            raise AdapterError("Tushare returned invalid json") from None
-        except (urllib.error.URLError, TimeoutError) as exc:
-            raise AdapterError(f"Tushare request failed: {exc}") from exc
-        if result.get("code") != 0:
-            raise AdapterError(result.get("msg") or "Tushare returned an unknown error")
+        except (urllib.error.URLError, TimeoutError, json.JSONDecodeError) as exc:
+            raise AdapterError(f"Tushare 请求失败: {exc}") from exc
+        if result.get("code") not in (0, "0", None):
+            raise AdapterError(str(result.get("msg") or f"Tushare error {result.get('code')}"))
         data = result.get("data") or {}
-        columns = data.get("fields") or []
-        return [dict(zip(columns, item)) for item in data.get("items") or []]
+        items = data.get("items") or []
+        fields_list = data.get("fields") or (fields.split(",") if fields else [])
+        return [dict(zip(fields_list, item)) for item in items]