1 Commits
16 changed files with 1463 additions and 98 deletions

No files matched your search

+15 -1
View File
@@ -2,7 +2,21 @@
本项目遵循 [语义化版本](https://semver.org/lang/zh-CN/)。 本项目遵循 [语义化版本](https://semver.org/lang/zh-CN/)。
## [0.6.3] - 2026-10-18 ## [0.6.7] - 2026-10-22
### 新增
- `get_index_members` / `get_index_trade_stats` / `get_index_trade_stats_multi`:指数成分股与逐笔聚合统计(分钟级增量;北交所成分无笔数自动剔除)
- `sync_decimal_points` / `decimal_point_status`:权威价格小数位表运行时同步(报价标度不再依赖硬编码前缀规则),`start()` 自动装载
- `decimal_sync` 模块
### 修复
- **指数/ETF/LOF 逐笔的 `trade_count` 语义修正**:协议字段在指数上是「均价×1000」、ETF/LOF 上是「价格×1000」,**不是笔数**;新增 `TransactionRecord.trade_count_available`(判定见 `symbols.has_order_count`),CLI 对无笔数口径标的显示 `(-)` 而非假笔数
- 集合竞价委托(bs_flag=8)不再混入逐笔结果(会被覆盖、Σvol 虚高 20%~160%、不同主机条数不一致)
- 缓存管理器 / 连接池 / 分页调度若干健壮性修复
### 说明
- 0.6.4~0.6.6 为中间迭代,未单独发版记录;本次合并 PyPI sdist 与仓库,源码恢复双仓一致
- fix(pool): ExPool acquire 背压等待(Condition 排队,默认 30s 超时)替代 3x0.05s 快速失败;超时异常带池状态 - fix(pool): ExPool acquire 背压等待(Condition 排队,默认 30s 超时)替代 3x0.05s 快速失败;超时异常带池状态
- feat(api): 新增 ex_pool_cap 参数(默认 2),单 EX 服务器场景不再静默退化为 1 并发(fixes #1) - feat(api): 新增 ex_pool_cap 参数(默认 2),单 EX 服务器场景不再静默退化为 1 并发(fixes #1)
+15
View File
@@ -82,6 +82,21 @@ from pytdxdata import (
| `get_board_members(board, *, fields=None, count=100000)` | `list[SecurityQuote]` | 板块成分股报价(归一模型) | | `get_board_members(board, *, fields=None, count=100000)` | `list[SecurityQuote]` | 板块成分股报价(归一模型) |
| `get_board_of(symbols)` | `list[BelongBoard]` | 个股所属板块 | | `get_board_of(symbols)` | `list[BelongBoard]` | 个股所属板块 |
#### 指数成分与统计(v0.6.7)
| 方法 | 返回 | 说明 |
|------|------|------|
| `get_index_members(index, *, count=100000, fields=None)` | `list[SecurityQuote]` | 指数成分股含实时报价(MAC 0x122C,指数码→板块码映射) |
| `get_index_trade_stats(index, *, date=None, members=None, batch_size=50, concurrency=24, skip_errors=True)` | `IndexTradeStats` | 指数逐笔聚合统计(分钟级增量 vol/trade_count/amount;北交所成分无笔数自动剔除) |
| `get_index_trade_stats_multi(index, dates, *, concurrency=24, batch_size=50, skip_errors=True, on_progress=None)` | `dict[int, IndexTradeStats]` | 多日批量版(按日期键) |
#### 价格小数位(v0.6.7)
| 方法 | 返回 | 说明 |
|------|------|------|
| `sync_decimal_points(markets=("sh","sz"))` | `dict[str, int]` | 从证券清单同步权威价格小数位表(`start()` 已自动处理,此为手动刷新口) |
| `decimal_point_status()` | `dict` | 小数位表当前状态(source/stock/syncing,排障用) |
### 2.2 K线(旧接口,共 6 个) {#legacy-kline} ### 2.2 K线(旧接口,共 6 个) {#legacy-kline}
| 方法 | 返回 | 说明 | | 方法 | 返回 | 说明 |
+23 -6
View File
@@ -1,14 +1,24 @@
[project] [project]
name = "pytdxdata" name = "pytdxdata"
version = "0.6.3" version = "0.6.7"
description = "通达信(TongDaXin)行情数据客户端:动态连接池 + 双层TTL缓存 + 3连接并发分页,纯asyncio零运行时依赖,支持A股/港股/美股/期货/期权" description = "通达信(TongDaXin)行情数据客户端:动态连接池 + 双层TTL缓存 + 3连接并发分页,纯asyncio零运行时依赖,支持A股/港股/美股/期货/期权"
readme = "README.md" readme = "README.md"
requires-python = ">=3.11" requires-python = ">=3.11"
license = "MIT" license = "MIT"
license-files = ["LICENSE"] license-files = ["LICENSE"]
keywords = ["tdx", "trading", "stock", "kline", "quote", "通达信", "A股", "港股", "美股", "期货", "asyncio", "quant"] keywords = [
authors = [ "tdx",
{ name = "openbot-coder", email = "openbot-coder@users.noreply.github.com" } "trading",
"stock",
"kline",
"quote",
"通达信",
"A股",
"港股",
"美股",
"期货",
"asyncio",
"quant",
] ]
classifiers = [ classifiers = [
"Development Status :: 4 - Beta", "Development Status :: 4 - Beta",
@@ -24,6 +34,10 @@ classifiers = [
] ]
dependencies = [] dependencies = []
[[project.authors]]
name = "openbot-coder"
email = "openbot-coder@users.noreply.github.com"
[project.urls] [project.urls]
Homepage = "https://github.com/openbot-coder/pytdxdata" Homepage = "https://github.com/openbot-coder/pytdxdata"
Repository = "https://github.com/openbot-coder/pytdxdata" Repository = "https://github.com/openbot-coder/pytdxdata"
@@ -31,7 +45,11 @@ Documentation = "https://openbot-coder.github.io/pytdxdata/"
[project.optional-dependencies] [project.optional-dependencies]
uvloop = ["uvloop>=0.21"] uvloop = ["uvloop>=0.21"]
dev = ["pytest>=8.0", "pytest-asyncio>=0.24", "ruff>=0.6"] dev = [
"pytest>=8.0",
"pytest-asyncio>=0.24",
"ruff>=0.6",
]
[project.scripts] [project.scripts]
tdx = "pytdxdata.cli:main" tdx = "pytdxdata.cli:main"
@@ -59,4 +77,3 @@ docs = [
"ruff>=0.6", "ruff>=0.6",
"jieba>=0.42", "jieba>=0.42",
] ]
+5 -1
View File
@@ -7,16 +7,18 @@ from .models import (
Adjust, Adjust,
ExMarket, ExMarket,
FinanceRecord, FinanceRecord,
IndexTradeStats,
KlinePeriod, KlinePeriod,
Market, Market,
MinuteBar, MinuteBar,
MinuteTradeStat,
SecurityBar, SecurityBar,
SecurityQuote, SecurityQuote,
TransactionRecord, TransactionRecord,
XdxrRecord, XdxrRecord,
) )
__version__ = "0.6.2" __version__ = "0.6.7"
__all__ = [ __all__ = [
"TdxData", "TdxData",
"Market", "Market",
@@ -27,6 +29,8 @@ __all__ = [
"SecurityQuote", "SecurityQuote",
"TransactionRecord", "TransactionRecord",
"MinuteBar", "MinuteBar",
"MinuteTradeStat",
"IndexTradeStats",
"XdxrRecord", "XdxrRecord",
"FinanceRecord", "FinanceRecord",
"__version__", "__version__",
+524 -49
View File
@@ -23,11 +23,15 @@ from __future__ import annotations
import asyncio import asyncio
import logging import logging
import re import re
import time
import warnings import warnings
from collections import deque
from collections.abc import Callable
from datetime import date as _date, datetime from datetime import date as _date, datetime
from pathlib import Path from pathlib import Path
from typing import Sequence from typing import Sequence
from . import decimal_sync
from .cache.disk import DiskCache from .cache.disk import DiskCache
from .cache.key import CacheKey from .cache.key import CacheKey
from .cache.manager import CacheManager from .cache.manager import CacheManager
@@ -45,11 +49,13 @@ from .models import (
FinancialFileInfo, FinancialFileInfo,
FinancialRecord, FinancialRecord,
IndicatorSet, IndicatorSet,
IndexTradeStats,
IndustryInfo, IndustryInfo,
KlinePeriod, KlinePeriod,
Market, Market,
MemberQuote, MemberQuote,
MinuteBar, MinuteBar,
MinuteTradeStat,
SecurityBar, SecurityBar,
SecurityInfo, SecurityInfo,
SecurityQuote, SecurityQuote,
@@ -142,10 +148,13 @@ _KLINE_MAX_BARS: dict[KlinePeriod, int] = {
# 自定义字段报价(0x122B/0x122C)基础字段位:昨收/开/高/低/现价/量/量比/额 # 自定义字段报价(0x122B/0x122C)基础字段位:昨收/开/高/低/现价/量/量比/额
_QUOTE_BITS = [0x00, 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07] _QUOTE_BITS = [0x00, 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07]
_QUOTE_PAGE = 80 # 0x122B/0x122C 单次上限 80 只 _QUOTE_PAGE = 80 # 0x122B/0x122C 单次上限 80 只
_QUOTE_PIPE_CHUNK = 3 # 单连接流水线一次取几片(实测 3 片 + 每台 2 连接最稳)
_EX_PAGE = 1000 # EX 商品列表单页 _EX_PAGE = 1000 # EX 商品列表单页
_TICK_PAGE = 1000 # MAC 逐笔单页 _TICK_PAGE = 1000 # MAC 逐笔单页
_EX_TICK_PAGE = 1800 # 港股逐笔单页 _EX_TICK_PAGE = 1800 # 港股逐笔单页
_ALL_PAGES_CAP = 50 # count=None(全量)时的翻页上限 _ALL_PAGES_CAP = 50 # count=None(全量)时的翻页上限
# 集合竞价**委托**记录标记(0x122F 第 5 字段)。不是成交,见 _ticks_one 注释。
_BS_AUCTION_ORDER = 8
# 港股系市场:逐笔走 EX 专用 16B 协议(0x23FC/0x2406),其余扩展市场走 MAC 0x122F # 港股系市场:逐笔走 EX 专用 16B 协议(0x23FC/0x2406),其余扩展市场走 MAC 0x122F
_HK_MARKETS = frozenset({27, 31, 48, 49, 71, 98}) _HK_MARKETS = frozenset({27, 31, 48, 49, 71, 98})
@@ -230,7 +239,7 @@ class TdxData:
default_adjust: Adjust = Adjust.NONE, default_adjust: Adjust = Adjust.NONE,
max_inflight: int = 64, max_inflight: int = 64,
pool_min: int = 3, pool_min: int = 3,
pool_max: int = 12, pool_max: int = 24,
mac_pool_cap: int = 4, mac_pool_cap: int = 4,
std_pool_cap: int = 4, std_pool_cap: int = 4,
ex_pool_cap: int = 2, ex_pool_cap: int = 2,
@@ -279,6 +288,7 @@ class TdxData:
self._paginator_std = Paginator(self._std_pool) self._paginator_std = Paginator(self._std_pool)
self._paginator_mac = Paginator(self._mac_pool) self._paginator_mac = Paginator(self._mac_pool)
self._started = False self._started = False
self._decimal_task: asyncio.Task | None = None
async def start(self) -> None: async def start(self) -> None:
await self._std_pool.start() await self._std_pool.start()
@@ -286,8 +296,13 @@ class TdxData:
await self._ex_pool.start() await self._ex_pool.start()
await self._cache.start() await self._cache.start()
self._started = True self._started = True
# 权威价格小数位表:当日缓存命中则同步装表(无网络),否则挂后台同步。
# 不阻塞 start(),失败静默退回静态前缀规则。
self._decimal_task = decimal_sync.schedule(self)
async def close(self) -> None: async def close(self) -> None:
await decimal_sync.cancel(self._decimal_task)
self._decimal_task = None
await self._std_pool.close() await self._std_pool.close()
await self._mac_pool.close() await self._mac_pool.close()
await self._ex_pool.close() await self._ex_pool.close()
@@ -325,6 +340,39 @@ class TdxData:
out.extend(await self._universe_cn(mkt)) out.extend(await self._universe_cn(mkt))
return out return out
async def sync_decimal_points(self, markets: Sequence[str] = ("sh", "sz")
) -> dict[str, int]:
"""从证券清单同步**权威价格小数位表**,返回统计。
报价接口返回的是整数,真实价格 = ``整数 / 10**小数位``。小数位不随报价下发,
但证券清单接口自带权威值 ``SecurityInfo.decimal_point`` —— 本方法把它灌进
:func:`~pytdxdata.protocol.commands.quotes.price_decimal_digits` 使用的运行时表,
从此价格标度不再依赖硬编码前缀规则(新增代码段也不会算错)。
``start()`` 已自动处理:当日缓存命中就直接装表,否则挂后台同步,**不用手动调**。
这个方法用于「立刻要权威表」「排障时强制刷新」。
参数:
markets: 拉哪些市场,默认 ``("sh", "sz")``。证券清单走 ``get_universe``,
其本身已有缓存层,重复调用只花一次网络。
返回:
``{"prefixes", "codes", "mixed_prefixes", "stock"}`` —— 已覆盖的三位前缀数、
例外代码数、出现混合小数位的前缀数、总条目数。
``mixed_prefixes`` > 0 说明某段代码混用了不同小数位(已自动退化为逐代码查表)。
用法::
async with TdxData() as td:
print(await td.sync_decimal_points()) # {'prefixes': 300+, 'codes': 0, ...}
q = await td.get_quotes("sh019742") # 110.228(4 位小数)
"""
return await decimal_sync.refresh(self, tuple(markets))
async def decimal_point_status(self) -> dict:
"""权威小数位表当前状态(``source`` / ``stock`` / ``syncing`` 等,排障用)。"""
return decimal_sync.status()
async def _universe_cn(self, market: int) -> list[SecurityInfo]: async def _universe_cn(self, market: int) -> list[SecurityInfo]:
key = CacheKey("list", market) key = CacheKey("list", market)
@@ -396,7 +444,9 @@ class TdxData:
# ================================================================== # # ================================================================== #
async def get_quotes(self, symbols: str | Sequence[str], *, async def get_quotes(self, symbols: str | Sequence[str], *,
fields: Sequence[int] | None = None fields: Sequence[int] | None = None,
concurrency: int | None = None,
pipeline: bool = True
) -> list[SecurityQuote]: ) -> list[SecurityQuote]:
"""报价,一次可跨市场、跨通道混合查询。 """报价,一次可跨市场、跨通道混合查询。
@@ -406,6 +456,11 @@ class TdxData:
fields: 可选自定义字段位(如 ``[0x00, 0x04, 0x38]``)。给了就走 MAC/EX 的 fields: 可选自定义字段位(如 ``[0x00, 0x04, 0x38]``)。给了就走 MAC/EX 的
0x122B 字段位通道(有 ``name``,无五档);不给时 A股走标准通道 0x122B 字段位通道(有 ``name``,无五档);不给时 A股走标准通道
(**有五档 bid/ask、买卖量、涨跌停价,无 name**),扩展市场走 0x122B。 (**有五档 bid/ask、买卖量、涨跌停价,无 name**),扩展市场走 0x122B。
concurrency: 连接并发数(默认取连接池上限)。协议硬上限 80 只/请求,
全市场约 5,200 只 A股需切 66 片;调小可降低对服务器的压力。
pipeline: 是否启用**单连接流水线**(默认开启)。开启后一条连接可一次发出
多个分片、只付一次往返,全市场 5,200 只 A股由 ~4s 降到 **~0.3s**;
个股或小批量不受影响。个别服务器不支持流水线时会自动降级为逐片。
返回: 返回:
``list[SecurityQuote]``,按入参顺序排列(各记录带 ``symbol``)。 ``list[SecurityQuote]``,按入参顺序排列(各记录带 ``symbol``)。
@@ -416,50 +471,155 @@ class TdxData:
bits = [int(b) for b in fields] if fields else None bits = [int(b) for b in fields] if fields else None
tag = "mac:" + _fields_tag(bits) if bits else "std" tag = "mac:" + _fields_tag(bits) if bits else "std"
key = CacheKey.quote_key([s.as_pair() for s in syms], tag=tag) key = CacheKey.quote_key([s.as_pair() for s in syms], tag=tag)
payload = await self._cache.get_or_fetch( quotes = await self._cache.get_or_fetch_obj(
key, lambda: self._fetch_quotes(syms, bits)) key,
quotes = unpack_payload(payload) lambda: self._fetch_quotes(syms, bits, concurrency, pipeline),
pack=pack_payload, unpack=unpack_payload)
order = {s.as_pair(): i for i, s in enumerate(syms)} order = {s.as_pair(): i for i, s in enumerate(syms)}
quotes.sort(key=lambda q: order.get((int(q.market), q.code), len(order))) quotes.sort(key=lambda q: order.get((int(q.market), q.code), len(order)))
return quotes return quotes
async def _fetch_quotes(self, syms: list[Symbol], bits: list[int] | None) -> bytes: async def _fetch_quotes(self, syms: list[Symbol], bits: list[int] | None,
concurrency: int | None = None,
pipeline: bool = True) -> list[SecurityQuote]:
out: list[SecurityQuote] = [] out: list[SecurityQuote] = []
cn = [s for s in syms if not s.is_ex] cn = [s for s in syms if not s.is_ex]
ex = [s for s in syms if s.is_ex] ex = [s for s in syms if s.is_ex]
if bits is None: if bits is None:
if cn: if cn:
out.extend(await self._quotes_std(cn)) out.extend(await self._quotes_std(
cn, concurrency=concurrency, pipeline=pipeline))
if ex: if ex:
rows = await self._execute_ex( rows = await self._execute_ex(
MacSymbolQuotesCmd([s.as_pair() for s in ex], None)) MacSymbolQuotesCmd([s.as_pair() for s in ex], None))
out.extend(_quote_from_member(m) for m in rows) out.extend(_quote_from_member(m) for m in rows)
else: else:
for i in range(0, len(syms), _QUOTE_PAGE): batches = [syms[i:i + _QUOTE_PAGE]
batch = syms[i:i + _QUOTE_PAGE] for i in range(0, len(syms), _QUOTE_PAGE)]
pairs = [s.as_pair() for s in batch] sem = asyncio.Semaphore(self._quote_concurrency(concurrency))
cn_pairs = [p for p, s in zip(pairs, batch) if not s.is_ex]
ex_pairs = [p for p, s in zip(pairs, batch) if s.is_ex]
if cn_pairs:
rows = await self._execute_mac(MacSymbolQuotesCmd(cn_pairs, bits))
out.extend(_quote_from_member(m) for m in rows)
if ex_pairs:
rows = await self._execute_ex(MacSymbolQuotesCmd(ex_pairs, bits))
out.extend(_quote_from_member(m) for m in rows)
return pack_payload(out)
async def _quotes_std(self, syms: list[Symbol]) -> list[SecurityQuote]: async def one(batch: list[Symbol]) -> list[SecurityQuote]:
"""A股标准通道五档报价(>80 只自动切批)。""" async with sem:
out: list[SecurityQuote] = [] pairs = [s.as_pair() for s in batch]
for i in range(0, len(syms), MAX_QUOTES): cn_pairs = [p for p, s in zip(pairs, batch) if not s.is_ex]
batch = syms[i:i + MAX_QUOTES] ex_pairs = [p for p, s in zip(pairs, batch) if s.is_ex]
cmd = GetSecurityQuotesCmd([s.as_pair() for s in batch]) got: list[SecurityQuote] = []
rows = await self._execute_standard(cmd) if cn_pairs:
rows = await self._execute_mac(
MacSymbolQuotesCmd(cn_pairs, bits))
got.extend(_quote_from_member(m) for m in rows)
if ex_pairs:
rows = await self._execute_ex(
MacSymbolQuotesCmd(ex_pairs, bits))
got.extend(_quote_from_member(m) for m in rows)
return got
if batches:
for got in await asyncio.gather(*(one(b) for b in batches)):
out.extend(got)
return out
def _quote_concurrency(self, requested: int | None) -> int:
"""连接并发数:显式指定优先,否则按「每台服务器 2 连接」估算。
实测(5,227 只 A股 / 66 片,11 台服务器):
========== ===== ====== ====== ======
并发 每台连接 流水深度 p50 max
========== ===== ====== ====== ======
11 1 6 455ms 464ms
12 1~2 3 457ms 523ms
**22** **2** **3** **325ms** **334ms**
33 3 2 328ms 345ms
========== ===== ====== ====== ======
结论:每台 2 连接 × 流水 3 片最稳;再多连接只是把同一台的队列搬个位置,
对整体无益反增抖动。池 ``per_server_cap`` 默认 4,2 连接仍留有余量。
"""
if requested is not None and requested > 0:
return max(1, int(requested))
n = len(self._std_pool.servers)
return max(1, min(self._std_pool.max_size, (n or 1) * 2))
async def _quotes_std(self, syms: list[Symbol], *,
concurrency: int | None = None,
pipeline: bool = True) -> list[SecurityQuote]:
"""A股标准通道五档报价(协议上限 80 只/请求,超出自动切批)。
切片多于 1 片时把各片**按轮转分给若干条连接**,每条连接内部用流水线
一次发出(1 次往返拿多片),不再逐片往返。
"""
if not syms:
return []
def stamp(rows: list[SecurityQuote]) -> list[SecurityQuote]:
for q in rows: for q in rows:
q.symbol = format_symbol(q.market, q.code) q.symbol = format_symbol(q.market, q.code)
out.extend(rows) return rows
batches = [syms[i:i + MAX_QUOTES]
for i in range(0, len(syms), MAX_QUOTES)]
if len(batches) == 1:
cmd = GetSecurityQuotesCmd([s.as_pair() for s in batches[0]])
return stamp(await self._execute_standard(cmd))
conc = min(self._quote_concurrency(concurrency), len(batches))
# 首次大批量:先把连接并发建好,否则 acquire 会在池锁里串行握手
# (22 路并发首次调用实测多花 5~6s)
await self._std_pool.warm(conc)
if not pipeline:
sem = asyncio.Semaphore(conc)
async def one(batch: list[Symbol]) -> list[SecurityQuote]:
async with sem:
cmd = GetSecurityQuotesCmd([s.as_pair() for s in batch])
return stamp(await self._execute_standard(cmd))
groups = await asyncio.gather(*(one(b) for b in batches))
return [q for g in groups for q in g]
# 工作窃取:连接做完一批就回来再取,快的多干、慢的少干。
# 平均分配(每台固定几片)会被最慢的一台拖住 —— 实测 11 台平均分
# 时 p50 在 400~580ms 之间抖,改窃取后波动收窄。
queue = deque(batches)
gate = asyncio.Lock()
async def worker() -> list[SecurityQuote]:
got: list[SecurityQuote] = []
while True:
async with gate:
if not queue:
return got
take = min(_QUOTE_PIPE_CHUNK, len(queue))
chunk = [queue.popleft() for _ in range(take)]
for rows in await self._execute_std_pipeline(chunk):
got.extend(stamp(rows))
out: list[SecurityQuote] = []
for part in await asyncio.gather(*(worker() for _ in range(conc))):
out.extend(part)
return out return out
async def _execute_std_pipeline(
self, batches: list[list[Symbol]]
) -> list[list[SecurityQuote]]:
"""一条连接流水线执行多片报价;节点不支持流水线时降级为逐片。"""
cmds = [GetSecurityQuotesCmd([s.as_pair() for s in b]) for b in batches]
pc = await self._std_pool.acquire()
try:
rows = await pc.execute_pipeline(cmds)
except Exception:
# 流水线中断后响应流已错配,连接必须废弃;改用逐片(每条独立借还)
await self._std_pool.release(pc, broken=True)
fallback: list[list[SecurityQuote]] = []
for cmd in cmds:
fallback.append(await self._execute_standard(cmd))
return fallback
await self._std_pool.release(pc)
return rows
async def get_snapshot(self, symbols: str | Sequence[str]) -> list[SymbolSnapshot]: async def get_snapshot(self, symbols: str | Sequence[str]) -> list[SymbolSnapshot]:
"""个股特征快照(当日/换手/活跃度等,仅 A股)。""" """个股特征快照(当日/换手/活跃度等,仅 A股)。"""
syms = parse_symbols(symbols) syms = parse_symbols(symbols)
@@ -795,12 +955,30 @@ class TdxData:
参数: 参数:
date: ``None``=当日,``YYYYMMDD``=历史某日。 date: ``None``=当日,``YYYYMMDD``=历史某日。
start: 起始偏移。 start: 起始偏移,**从最新一笔往回数**(``0``=当天最后一条)。
``start=1000`` 表示跳过最近 1000 条、从更早的位置开始。
count: ``None``=尽量取全量(自动翻页,上限 50 页);给了就取这么多条。 count: ``None``=尽量取全量(自动翻页,上限 50 页);给了就取这么多条。
返回: 返回:
``list[TransactionRecord]``,按入参标的顺序拼接(A股含 ``trade_count``, ``list[TransactionRecord]``,按入参标的顺序拼接,**每条标的内部按时间升序**。
港股为 HH:MM 时间 + ``fields["zengcang"]`` 增仓)。 注意服务端是按约 3 秒聚合的切片,``vol``/``trade_count`` 是切片增量
而非当日累计。
A股含 ``trade_count``,港股为 HH:MM 时间 + ``fields["zengcang"]`` 增仓。
**集合竞价委托(``bs_flag=8``)不会出现在结果里**:它们不是成交,``vol``
是累计撮合量(会被后续记录覆盖),计入会让 Σvol 比真值虚高 20%~160%,
且不同主机收录的条数不同 → 同一查询两次调用结果不一致。
**指数(``sh000016`` 等)的两个字段语义与个股不同**:
- ``vol`` 单位是**百元**(成交额/100),不是手。
- ``trade_count`` 是**当时均价 × 1000**,不是笔数。
**ETF / LOF 同样没有笔数**:``trade_count`` 是「价格 × 1000」(货币 ETF 恒为 0),
这类记录的 ``trade_count_available`` 为 ``False`` 且 ``trade_count`` 已置 0。
判定规则见 :func:`~pytdxdata.symbols.has_order_count`。
详见 :class:`~pytdxdata.models.TransactionRecord` 的类注释。
""" """
syms = parse_symbols(symbols) syms = parse_symbols(symbols)
sem = asyncio.Semaphore(max(1, concurrency)) sem = asyncio.Semaphore(max(1, concurrency))
@@ -821,18 +999,19 @@ class TdxData:
if sym.is_ex: if sym.is_ex:
rows = await self._ticks_ex(sym, date=date, start=start, count=count) rows = await self._ticks_ex(sym, date=date, start=start, count=count)
else: else:
def make_cmd(s: int, c: int): rows = await self._ticks_cn_serial(sym, date=date, start=start,
return MacTransactionCmd(sym.market, sym.code, date or 0, s, c) count=count)
# ★ 过滤集合竞价**委托**记录(bs_flag=8)。必须在逐页拉取**全部结束之后**做:
spec = PageSpec( # 分页靠 `len(rows) < page_size` 判终止,若在 parse_response 里过滤,
page_size=_TICK_PAGE, # 满页会被剥短成"短页" → 提前终止 → 丢数据。
total=None if count is None else start + count, # 实测(2026-10-21 sh600000):b8 是 09:15~09:25 / 14:57~15:00 竞价段的
min_page=100, # 撮合快照,vol 是**累计值**(可被后续记录覆盖),不是成交增量;
stop_when_short=True, # 计入后 Σvol 比真值虚高 20%~160%,且**不同主机的 b8 集合不同**
max_pages=_ALL_PAGES_CAP if count is None else None, # (干净主机 0 条、脏主机 135 条)→ 同一查询两次调用结果不一致。
start_offset=start, # 剔除后:Σvol 与快照 vol 精确相等(比值 1.0000),且跨主机逐条一致。
) rows = [r for r in rows if r.bs_flag != _BS_AUCTION_ORDER]
rows = await self._fetch_paginated_list("mac", make_cmd, spec) # 排序放在 pack 之前,让缓存直接存入正确顺序,避免每次命中都重排。
rows.sort(key=lambda r: r.time)
return pack_payload(rows) return pack_payload(rows)
records = unpack_payload(await self._cache.get_or_fetch(key, fetch)) records = unpack_payload(await self._cache.get_or_fetch(key, fetch))
@@ -840,6 +1019,54 @@ class TdxData:
r.symbol = sym.text r.symbol = sym.text
return records return records
async def _ticks_cn_serial(self, sym: Symbol, *, date: int | None, start: int,
count: int | None) -> list[TransactionRecord]:
"""A股逐笔(0x122F)**单连接串行游标**翻页。
为什么不用并发分页器:``0x122F`` 的 ``start`` 是「从**最新一笔**往回数的偏移」,
是个**相对锚点**,不是稳定的绝对序号。而池里各主机对集合竞价委托(``bs_flag=8``)
的收录**不一致**(实测干净主机 0 条、脏主机 135 条),同一 ``start`` 在不同主机上
落到**不同的真实成交**上。并发分页 = 各页来自不同主机 → 页边界既**重叠**
(sh600000 冗余 116 条)又**漏条**(13:03 附近丢 59 条),Σvol 偏移 −1.2%。
逐笔单日仅 5~6 页(1000 条/页),串行开销可忽略、换来精确一致,
故这里固定**单连接**、按 ``start`` 递增(= 向更早方向)推进。
返回:按页序累加后再**整体翻转**,得到时间升序的完整列表。
"""
chunks: list[list[TransactionRecord]] = []
pos = start
remaining = count
pc = await self._mac_pool.acquire()
try:
for _ in range(_ALL_PAGES_CAP):
n = _TICK_PAGE if remaining is None else min(_TICK_PAGE, remaining)
if n <= 0:
break
rows = await pc.execute(
MacTransactionCmd(sym.market, sym.code, date or 0, pos, n))
if not rows:
break
chunks.append(list(rows))
pos += len(rows)
if remaining is not None:
remaining -= len(rows)
if remaining <= 0:
break
# 短页 = 已到最早一笔
if len(rows) < n:
break
except Exception:
await self._mac_pool.release(pc, broken=True)
raise
await self._mac_pool.release(pc)
out: list[TransactionRecord] = []
for chunk in reversed(chunks):
out.extend(chunk)
for r in out:
r.symbol = sym.text
return out
async def _ticks_ex(self, sym: Symbol, *, date: int | None, start: int, async def _ticks_ex(self, sym: Symbol, *, date: int | None, start: int,
count: int | None) -> list[TransactionRecord]: count: int | None) -> list[TransactionRecord]:
from .protocol.commands.ex_ext import ( from .protocol.commands.ex_ext import (
@@ -848,7 +1075,7 @@ class TdxData:
) )
hk = sym.market in _HK_MARKETS hk = sym.market in _HK_MARKETS
page = _EX_TICK_PAGE if hk else _TICK_PAGE page = _EX_TICK_PAGE if hk else _TICK_PAGE
out: list[TransactionRecord] = [] chunks: list[list[TransactionRecord]] = []
pos = start pos = start
remaining = count remaining = count
for _ in range(_ALL_PAGES_CAP): for _ in range(_ALL_PAGES_CAP):
@@ -860,22 +1087,27 @@ class TdxData:
if date else if date else
GetExTransactionDataCmd(sym.market, sym.code, pos, n)) GetExTransactionDataCmd(sym.market, sym.code, pos, n))
rows = await self._execute_ex(cmd) rows = await self._execute_ex(cmd)
out.extend(self._ex_tick(sym, r) for r in rows) chunk = [self._ex_tick(sym, r) for r in rows]
else: else:
cmd = MacTransactionCmd(sym.market, sym.code, date or 0, pos, n) cmd = MacTransactionCmd(sym.market, sym.code, date or 0, pos, n)
rows = await self._execute_ex(cmd) rows = await self._execute_ex(cmd)
for r in rows: for r in rows:
r.symbol = sym.text r.symbol = sym.text
out.extend(rows) chunk = list(rows)
if not rows: if not chunk:
break break
pos += len(rows) # EX 通道的 pos 同样是"从最新往回数",逐页倒序累加 → 最后整体翻转。
chunks.append(chunk)
pos += len(chunk)
if remaining is not None: if remaining is not None:
remaining -= len(rows) remaining -= len(chunk)
if remaining <= 0: if remaining <= 0:
break break
if len(rows) < n: if len(chunk) < n:
break break
out: list[TransactionRecord] = []
for chunk in reversed(chunks):
out.extend(chunk)
return out return out
@staticmethod @staticmethod
@@ -892,6 +1124,7 @@ class TdxData:
trade_count=0, trade_count=0,
bs_flag=int(row.get("direction", 0) or 0), bs_flag=int(row.get("direction", 0) or 0),
fields=fields, fields=fields,
trade_count_available=False, # 扩展市场逐笔不带笔数字段
) )
async def get_minutes( async def get_minutes(
@@ -1221,6 +1454,248 @@ class TdxData:
result.sort(key=lambda x: x["change_pct"], reverse=True) result.sort(key=lambda x: x["change_pct"], reverse=True)
return result[:top_n] return result[:top_n]
# ================================================================== #
# 七之二、指数成分成交汇总(成分股逐笔 → 每分钟成交量 / 笔数)
# ================================================================== #
@staticmethod
def _index_pair(index: str) -> tuple[int, str]:
"""解析指数标的 → ``(市场号, 6 位代码)``。
与 :func:`~pytdxdata.symbols.parse_symbol` 的关键区别:**裸 6 位数字
一律按指数语义解析**,不做「``000xxx`` → 深市股票」的推断 ——
在指数接口里 ``000001`` 是**上证指数**,不是平安银行 ``sz000001``。
判定规则(与通达信板块码一致)::
000xxx / 999xxx / 880xxx -> 沪市(上证指数 / A股指数 / 板块指数)
399xxx -> 深市(深证成指 / 创业板指)
899xxx -> 北交所(北证50)
显式前缀 ``sh``/``sz``/``bj`` 优先,例如 ``sz399001``、``sh000300``。
"""
s = str(index).strip().lower().replace(".", "").replace(":", "")
if not s:
raise ValueError("指数标的不能为空")
prefix_table = {"sh": 1, "sz": 0, "bj": 2}
if s[:2] in prefix_table:
code = s[2:]
if not (code.isdigit() and len(code) == 6):
raise ValueError(f"非法指数标的: {index!r}(前缀后需 6 位数字)")
return prefix_table[s[:2]], code
if not (s.isdigit() and len(s) == 6):
raise ValueError(f"非法指数标的: {index!r}(需 6 位代码或 sh/sz/bj 前缀)")
if s.startswith(("000", "999", "880")):
return 1, s
if s.startswith("399"):
return 0, s
if s.startswith("899"):
return 2, s
raise ValueError(f"无法判定指数市场: {index!r}(不是 A股指数代码段)")
async def get_index_members(self, index: str, *, count: int = 100000,
fields: Sequence[int] | None = None
) -> list[SecurityQuote]:
"""指数**成分股**(含实时报价)。
走 MAC ``0x122C`` 板块成分命令,指数代码映射为服务端板块码:``000300`` →
31300、``399001`` → 30001、``899050`` → 32050(见 :meth:`_convert_board_code`)。
实测(2026-10-22):``sh000001`` 2230 只(沪市全部)、``sh000300`` 300 只、
``sh000016`` 50 只、``sh000905`` 500 只、``sh000852`` 1000 只、
``sz399006`` 100 只、``bj899050`` 50 只。
参数:
index: 指数代码,可裸写或带前缀(``"000300"`` / ``"sh000300"``)。
count: 最多取的成分数(分页上限,默认足够覆盖全部)。
fields: 附加报价字段位(同 :meth:`get_board_members`)。
注意:返回的是**当前**成分,指数调仓后历史日期不适用。
⚠️ **综合指数的成员名单不完整**(实测 2026-10-22):``sh000001`` 只返回
2230 只(2190 A + 40 B),而当日沪市 A 股实有 2318 只;缺的 128 只
(多为近期科创板/主板新股)当日成交 282 亿,正好对上「指数日线 amount −
成员日线 amount」的 297 亿(残差 4.9%)。因此
:meth:`get_index_trade_stats` 对**综合指数**会系统性低估约 4.6%;
样本指数 ``sh000016``(50 只)/ ``sh000300``(300 只)实测比值
**1.000000**,不受影响。要精确汇总综指成交,请自行用全市场股票列表。
用法::
members = await td.get_index_members("000300")
syms = [m.symbol for m in members] # ['sz000001', ...]
"""
_market, code = self._index_pair(index)
board_code = self._convert_board_code(code)
bits = [int(b) for b in fields] if fields else list(_QUOTE_BITS)
items: list[SecurityQuote] = []
start = 0
while start < count:
batch = await self._execute_mac(MacBoardMembersQuotesCmd(
board_code, sort_type=0, start=start,
page_size=_QUOTE_PAGE, bits=bits))
if not batch:
break
items.extend(_quote_from_member(m) for m in batch)
start += len(batch)
if len(batch) < _QUOTE_PAGE:
break
return items[:count]
async def get_index_trade_stats(
self,
index: str,
date: int | None = None,
*,
concurrency: int = 24,
batch_size: int = 50,
members: Sequence[str] | None = None,
skip_errors: bool = True,
on_progress: Callable[[int, int], None] | None = None,
) -> IndexTradeStats:
"""按指数**全部成分股**的逐笔汇总出每分钟成交量与成交笔数。
为什么必须绕这一圈:**通达信协议里指数本身没有成交笔数** —— 指数逐笔的
``trade_count`` 是「当时均价 × 1000」,不是笔数(见
:class:`TransactionRecord` 类注释)。所以「指数笔数」只能用成分股逐笔累加。
口径校验:``total_amount`` 是**成分股**合计。对「成分 = 指数覆盖面」的指数
它与指数快照 ``amount`` 完全相等(``sh000016`` 661.34 亿、``sh000300``
2373.98 亿,比值 1.0000);但深市 ``399xxx`` 的指数 ``amount`` 是**全板块**
口径(``sz399001`` = 深市全部股票 5081 亿、``sz399006`` = 创业板全部),
两者不可直接比。详见 :class:`IndexTradeStats`。
参数:
index: 指数代码(``"000300"`` / ``"sh000300"``,规则见 :meth:`_index_pair`)。
date: ``None`` = 当日;``YYYYMMDD`` = 历史某日(通达信逐笔仅保留最近
若干交易日,实测 2026-10-22 可回溯到 10-14,更早返回空)。
concurrency: 逐笔并发数(受 MAC 连接池上限约束,池满自动排队)。
batch_size: 每批处理的成分数 —— 批内并发拉取、批间串行聚合,
用于压住内存峰值(全量 tick 不驻留)。
members: 直接给成分股列表(跳过 :meth:`get_index_members`,
多日循环时可复用一次成分请求)。
skip_errors: 单只成分失败是否忽略(默认忽略并计入 ``failed_members``)。
on_progress: 进度回调 ``(已处理, 总数)``,每批结束调用一次。
返回:
:class:`IndexTradeStats`;``minutes`` 是**每分钟增量**,
「当日截止该分钟累计」用 ``.cumulative()`` / ``.cum_trade_count()``。
性能(实测 2026-10-22,盘中):
====================== ========== ============
指数 成分数 耗时
====================== ========== ============
``sh000016`` 上证50 50 ~5 s
``sh000300`` 沪深300 300 ~10 s
``sh000001`` 上证指数 2230 ~60-90 s
====================== ========== ============
**北交所成分没有笔数**(实测 ``trade_count`` 恒 0),此时
``trade_count_available`` 为 ``False``,只有成交量可用。
用法::
st = await td.get_index_trade_stats("000300")
for hhmm, cum in st.cum_trade_count():
print(hhmm, cum)
"""
started = time.perf_counter()
market, code = self._index_pair(index)
idx_text = format_symbol(market, code)
if members is None:
members = [m.symbol for m in await self.get_index_members(code)]
syms = [s for s in members if s]
total = len(syms)
if total == 0:
raise ValueError(f"指数 {index!r} 没有取到成分股")
vol_by_min: dict[str, float] = {}
tc_by_min: dict[str, int] = {}
amt_by_min: dict[str, float] = {}
active: set[str] = set()
with_tc: set[str] = set()
failed = 0
step = max(1, int(batch_size))
done = 0
for i in range(0, total, step):
batch = syms[i:i + step]
try:
ticks = await self.get_ticks(batch, date=date, concurrency=concurrency)
except Exception:
if not skip_errors:
raise
# 整批失败 → 退化为逐只重试,避免一只坏股拖垮一批
ticks = []
for one in batch:
try:
ticks.extend(await self.get_ticks(one, date=date))
except Exception:
failed += 1
for t in ticks:
hhmm = t.time[:5]
vol_by_min[hhmm] = vol_by_min.get(hhmm, 0.0) + t.vol
# 成分里若混入 ETF/LOF(协议无笔数),不能把占位值累加进笔数
if getattr(t, "trade_count_available", True):
tc_by_min[hhmm] = tc_by_min.get(hhmm, 0) + t.trade_count
amt_by_min[hhmm] = amt_by_min.get(hhmm, 0.0) + t.price * t.vol * 100.0
active.add(t.symbol)
if getattr(t, "trade_count_available", True) and t.trade_count:
with_tc.add(t.symbol)
done += len(batch)
if on_progress is not None:
on_progress(min(done, total), total)
minutes = [
MinuteTradeStat(time=k, volume=vol_by_min[k],
trade_count=tc_by_min[k], amount=amt_by_min[k])
for k in sorted(vol_by_min)
]
return IndexTradeStats(
index=idx_text,
date=date,
member_count=total,
active_members=len(active),
trade_count_members=len(with_tc),
minutes=minutes,
total_volume=sum(vol_by_min.values()),
total_trade_count=sum(tc_by_min.values()),
total_amount=sum(amt_by_min.values()),
failed_members=failed,
elapsed=time.perf_counter() - started,
)
async def get_index_trade_stats_multi(
self,
index: str,
dates: Sequence[int | None],
*,
concurrency: int = 24,
batch_size: int = 50,
skip_errors: bool = True,
on_progress: Callable[[int, int], None] | None = None,
) -> list[IndexTradeStats]:
"""多日批量版 :meth:`get_index_trade_stats`(成分只取一次)。
用于「近 N 日同时点累计平均笔数」这类基准曲线:一次拿到 N 天数据,
避免每天重复请求成分表。
用法::
# 最近 5 个交易日的笔数曲线(日期自己给,节假日请自行跳过)
stats = await td.get_index_trade_stats_multi(
"000300", [20261016, 20261019, 20261020, 20261021, 20261022])
"""
market, code = self._index_pair(index)
members = [m.symbol for m in await self.get_index_members(code)]
out: list[IndexTradeStats] = []
for d in dates:
out.append(await self.get_index_trade_stats(
code, d, concurrency=concurrency, batch_size=batch_size,
members=members, skip_errors=skip_errors,
on_progress=on_progress))
return out
# ================================================================== # # ================================================================== #
# 八、服务器文件(原始字节 + 解析) # 八、服务器文件(原始字节 + 解析)
# ================================================================== # # ================================================================== #
+32
View File
@@ -90,6 +90,38 @@ class CacheManager:
# 确保清理锁(即使 fetcher 异常也清理,避免无限增长) # 确保清理锁(即使 fetcher 异常也清理,避免无限增长)
self._locks.pop(ks, None) self._locks.pop(ks, None)
async def get_or_fetch_obj(
self,
key: CacheKey,
fetcher: Callable[[], Awaitable[Any]],
*,
pack: Callable[[Any], bytes],
unpack: Callable[[bytes], Any],
) -> Any:
"""同 :meth:`get_or_fetch`,但**回源时直接返回新取到的对象**。
用于「一次取回上千条记录」的场景(如全市场报价):常规路径对冷数据
会先 ``pack`` 再立刻 ``unpack`` 一次,纯属往返浪费 —— 5,200 条报价实测
白烧约 90ms。这里命中缓存才解序列化,回源则把对象原样返回,只在写缓存时打包。
"""
ks = key.to_str()
cached = await self.get(key)
if cached is not None:
return unpack(cached)
lock = self._locks.setdefault(ks, asyncio.Lock())
try:
async with lock:
cached = await self.get(key)
if cached is not None:
return unpack(cached)
async with self._semaphore:
obj = await fetcher()
await self.set(key, pack(obj))
return obj
finally:
self._locks.pop(ks, None)
async def invalidate(self, category: str, *, market: int | None = None, async def invalidate(self, category: str, *, market: int | None = None,
code: str | None = None) -> int: code: str | None = None) -> int:
"""按前缀批量失效,返回失效条数。""" """按前缀批量失效,返回失效条数。"""
+112 -13
View File
@@ -8,6 +8,7 @@
tdx transactions sz000001 --count 50 # 逐笔成交 tdx transactions sz000001 --count 50 # 逐笔成交
tdx minute sz000001 # 分时(今日/历史--date) tdx minute sz000001 # 分时(今日/历史--date)
tdx index 000001 --count 30 # 指数K线 tdx index 000001 --count 30 # 指数K线
tdx index-stats 000300 --top 10 # 指数成分成交汇总(每分钟成交量/笔数)
tdx board list [--type 1] # 板块列表(0全部/1行业/2概念) tdx board list [--type 1] # 板块列表(0全部/1行业/2概念)
tdx board members 881001 --count 20 # 板块成分 tdx board members 881001 --count 20 # 板块成分
tdx board ranking [--top 10] # 板块涨跌幅排行 tdx board ranking [--top 10] # 板块涨跌幅排行
@@ -37,6 +38,7 @@ from .api import TdxData
from dataclasses import asdict from dataclasses import asdict
from .models import Adjust, ExMarket, KlinePeriod, Market from .models import Adjust, ExMarket, KlinePeriod, Market
from .protocol.commands.quotes import price_decimal_digits
PERIODS = { PERIODS = {
"min1": KlinePeriod.MIN_1, "min5": KlinePeriod.MIN_5, "min15": KlinePeriod.MIN_15, "min1": KlinePeriod.MIN_1, "min5": KlinePeriod.MIN_5, "min15": KlinePeriod.MIN_15,
@@ -106,6 +108,40 @@ def _pct(price: float, pre: float) -> str:
return "%+.2f%%" % ((price / pre - 1) * 100) return "%+.2f%%" % ((price / pre - 1) * 100)
def _price_fmt(dp: int) -> str:
"""按品种小数位生成右对齐格式串。
股票 2 位、**可转债/ETF 3 位**、债券/回购 4 位。固定写 ``%8.2f`` 会把
可转债的 128.891 打成 128.89(少一位有效报价)。
"""
dp = max(0, min(int(dp or 2), 4))
return "%%%d.%df" % (6 + dp, dp)
# 可转债后缀(解码 4 位,但交易所报价步长 0.001 元,第 4 位恒为 0)
_CB_PREFIX = {1: ("110", "111", "113", "118"), 0: ("123", "127", "128")}
def _display_dp(market: int, code: str) -> int:
"""报价**显示**小数位。
注意 ``price_decimal_digits`` 给的是**解码位数**(原始整数 / 10**dp),
可转债是 4 位;但真实报价是 0.001 元的步长,第 4 位恒为 0,
按 4 位打印只会多一个 ``0``(``128.8920``)。
"""
dp = price_decimal_digits(market, code)
if str(code).startswith(_CB_PREFIX.get(int(market), ())):
return 3
return dp
def _dp_of(sym) -> int:
"""Symbol -> 价格小数位(非 A 股/指数一律 2 位)。"""
if getattr(sym, "kind", "") == "cn":
return _display_dp(sym.market, sym.code)
return 2
# ------------------------------------------------------------------ # # ------------------------------------------------------------------ #
# 各子命令 # 各子命令
# ------------------------------------------------------------------ # # ------------------------------------------------------------------ #
@@ -119,24 +155,31 @@ async def cmd_quotes(td: TdxData, symbols: list[str]) -> None:
def _print_quote(q) -> None: def _print_quote(q) -> None:
pct = (q.price / q.pre_close - 1) * 100 if q.pre_close else 0 pct = (q.price / q.pre_close - 1) * 100 if q.pre_close else 0
mkt = {31: "港股", 74: "美股"}.get(q.market, "") mkt = {31: "港股", 74: "美股"}.get(q.market, "")
if q.market in (0, 1):
dp = _display_dp(q.market, q.code)
else:
dp = getattr(q, "decimal_point", 2) or 2
print("=" * 58) print("=" * 58)
label = "%s %s" % (mkt, q.code) if mkt else "%s %s" % (q.code, q.server_time or datetime.now().strftime("%H:%M:%S")) label = "%s %s" % (mkt, q.code) if mkt else "%s %s" % (q.code, q.server_time or datetime.now().strftime("%H:%M:%S"))
print(" %s" % label) print(" %s" % label)
print(" 最新 %-10.2f 涨跌 %+8.2f (%+.2f%%)" % (q.price, q.price - q.pre_close, pct)) print((" 最新 %-10.*f 涨跌 %+8.*f (%+.2f%%)" % (dp, q.price, dp, q.price - q.pre_close, pct)))
print(" 今开 %-10.2f 最高 %.2f 最低 %.2f" % (q.open, q.high, q.low)) print((" 今开 %-10.*f 最高 %.*f 最低 %.*f" % (dp, q.open, dp, q.high, dp, q.low)))
print(" 昨收 %-10.2f 成交量 %s 成交额 %s" % (q.pre_close, _fmt_num(q.vol), _fmt_num(q.amount))) print((" 昨收 %-10.*f 成交量 %s 成交额 %s" % (dp, q.pre_close, _fmt_num(q.vol), _fmt_num(q.amount))))
if q.bid and q.ask: if q.bid and q.ask:
print(" " + "-" * 56) print(" " + "-" * 56)
pf = _price_fmt(dp)
for i in range(max(len(q.bid), len(q.ask))): for i in range(max(len(q.bid), len(q.ask))):
b = q.bid[i] if i < len(q.bid) else (0, 0) b = q.bid[i] if i < len(q.bid) else (0, 0)
a = q.ask[i] if i < len(q.ask) else (0, 0) a = q.ask[i] if i < len(q.ask) else (0, 0)
print(" 买%-1d %8.2f %10s | 卖%-1d %8.2f %10s" % ( print((" 买%-1d " + pf + " %10s | 卖%-1d " + pf + " %10s") % (
i + 1, b[0], _fmt_num(b[1]), i + 1, a[0], _fmt_num(a[1]))) i + 1, b[0], _fmt_num(b[1]), i + 1, a[0], _fmt_num(a[1])))
print() print()
async def cmd_kline(td: TdxData, sym: str, period: KlinePeriod, count: int, async def cmd_kline(td: TdxData, sym: str, period: KlinePeriod, count: int,
start: int, adjust: Adjust) -> None: start: int, adjust: Adjust) -> None:
s = parse_symbol(sym)
pf = _price_fmt(_dp_of(s))
bars = await td.get_bars(sym, period, start=start, count=count, adjust=adjust) bars = await td.get_bars(sym, period, start=start, count=count, adjust=adjust)
print("%s K线 (%s, %d根):" % (sym, period.name, len(bars))) print("%s K线 (%s, %d根):" % (sym, period.name, len(bars)))
print(" %-12s %8s %8s %8s %8s %12s" % ("日期", "开盘", "最高", "最低", "收盘", "成交量")) print(" %-12s %8s %8s %8s %8s %12s" % ("日期", "开盘", "最高", "最低", "收盘", "成交量"))
@@ -144,30 +187,38 @@ async def cmd_kline(td: TdxData, sym: str, period: KlinePeriod, count: int,
dt = "%04d-%02d-%02d" % (b.year, b.month, b.day) dt = "%04d-%02d-%02d" % (b.year, b.month, b.day)
if b.hour: if b.hour:
dt += " %02d:%02d" % (b.hour, b.minute) dt += " %02d:%02d" % (b.hour, b.minute)
print(" %-12s %8.2f %8.2f %8.2f %8.2f %12s" % ( print((" %-12s " + " ".join([pf] * 4) + " %12s") % (
dt, b.open, b.high, b.low, b.close, _fmt_num(b.vol))) dt, b.open, b.high, b.low, b.close, _fmt_num(b.vol)))
async def cmd_transactions(td: TdxData, sym: str, count: int, date: int | None) -> None: async def cmd_transactions(td: TdxData, sym: str, count: int, date: int | None) -> None:
s = parse_symbol(sym) s = parse_symbol(sym)
tx = await td.get_transactions(Market(s.market), s.code, date=date, count=count) pf = _price_fmt(_dp_of(s))
tx = await td.get_ticks(sym, date=date, count=count)
print("%s 逐笔成交 (%d条):" % (sym, len(tx))) print("%s 逐笔成交 (%d条):" % (sym, len(tx)))
print(" 时间 价格 成交量 方向") print(" 时间 价格 成交量 方向")
for t in tx: for t in tx:
bs = "买" if getattr(t, "bs_flag", 0) == 1 else "卖" if getattr(t, "bs_flag", 0) == 2 else "-" bs = "买" if getattr(t, "bs_flag", 0) == 1 else "卖" if getattr(t, "bs_flag", 0) == 2 else "-"
tc = getattr(t, "trade_count", 0) if not getattr(t, "trade_count_available", True):
print(" %-10s %8.2f %10s %s%s" % (t.time, t.price, _fmt_num(t.vol), bs, tc = " (-)" # 指数/ETF/LOF:协议不提供笔数
" (%d笔)" % tc if tc else "")) elif getattr(t, "trade_count", 0):
tc = " (%d笔)" % t.trade_count
else:
tc = ""
print((" %-10s " + pf + " %10s %s%s") % (
t.time, t.price, _fmt_num(t.vol), bs, tc))
async def cmd_minute(td: TdxData, sym: str, date: int | None) -> None: async def cmd_minute(td: TdxData, sym: str, date: int | None) -> None:
s = parse_symbol(sym) s = parse_symbol(sym)
bars = await td.get_minute(Market(s.market), s.code, date=date) pf = _price_fmt(_dp_of(s))
bars = await td.get_minutes(sym, date=date)
print("%s 分时 (%d条):" % (sym, len(bars))) print("%s 分时 (%d条):" % (sym, len(bars)))
print(" 时间 价格 均价 成交量") print(" 时间 价格 均价 成交量")
for i, b in enumerate(bars): for b in bars:
print(" %02d:%02d %8.2f %8.2f %10s" % ( # 用 bar 自带时间:早期按序号推算会打成 00:00~03:59
i // 60, i % 60, b.price, b.avg_price, _fmt_num(b.vol))) print((" %02d:%02d " + pf + " " + pf + " %10s") % (
b.hour, b.minute, b.price, b.avg_price, _fmt_num(b.vol)))
async def cmd_board(td: TdxData, args) -> None: async def cmd_board(td: TdxData, args) -> None:
@@ -204,6 +255,43 @@ async def cmd_board(td: TdxData, args) -> None:
print(" %-12s %+8.2f%%" % (r.get("name", ""), r.get("change_pct", 0))) print(" %-12s %+8.2f%%" % (r.get("name", ""), r.get("change_pct", 0)))
async def cmd_index_stats(td: TdxData, index: str, date: int | None,
show_all: bool, top: int, concurrency: int) -> None:
"""指数成分成交汇总:每分钟成交量 / 笔数(笔数=成分股逐笔累加)。"""
total = 0
def on_progress(done: int, all_: int) -> None:
nonlocal total
total = all_
print("\r 拉取成分逐笔 %d/%d ..." % (done, all_), end="", file=sys.stderr,
flush=True)
st = await td.get_index_trade_stats(
index, date, concurrency=concurrency, on_progress=on_progress)
if total:
print("\r" + " " * 40 + "\r", end="", file=sys.stderr, flush=True)
print("%s 成分成交汇总" % st.index)
print(" 成分 %d 只 / 有成交 %d 只 / 有笔数 %d 只%s" % (
st.member_count, st.active_members, st.trade_count_members,
"" if st.trade_count_available else " ★该指数无笔数口径"))
print(" 成交量 %s 手 成交笔数 %s 笔 成交额 %.2f 亿 耗时 %.1fs" % (
_fmt_num(st.total_volume),
_fmt_num(st.total_trade_count) if st.trade_count_available else "-",
st.total_amount / 1e8, st.elapsed))
if st.failed_members:
print(" (%d 只成分逐笔拉取失败,已跳过)" % st.failed_members)
print()
print(" 时间 累计成交量 累计笔数")
rows = st.cumulative()
if not show_all:
rows = rows[-top:]
for m in rows:
print(" %-6s %12s %12s" % (
m.time, _fmt_num(m.volume),
_fmt_num(m.trade_count) if st.trade_count_available else "-"))
async def cmd_info(td: TdxData, sym: str) -> None: async def cmd_info(td: TdxData, sym: str) -> None:
s = parse_symbol(sym) s = parse_symbol(sym)
info = await td.get_symbol_info(Market(s.market), s.code) info = await td.get_symbol_info(Market(s.market), s.code)
@@ -390,6 +478,17 @@ def main() -> None:
ix.set_defaults(func=lambda td, a: cmd_kline( ix.set_defaults(func=lambda td, a: cmd_kline(
td, a.code, KlinePeriod.DAY, a.count, 0, Adjust.NONE)) td, a.code, KlinePeriod.DAY, a.count, 0, Adjust.NONE))
ist = add("index-stats", "指数成分成交汇总(每分钟成交量/笔数)")
ist.add_argument("index", help="指数代码,如 000300 / 000001 / sh000016")
ist.add_argument("--date", type=str, default=None, help="YYYYMMDD,缺省=当日")
ist.add_argument("--top", type=int, default=10, help="只显示最后 N 分钟(默认10)")
ist.add_argument("--all", dest="show_all", action="store_true",
help="显示全部分钟")
ist.add_argument("--concurrency", type=int, default=24, help="逐笔并发数")
ist.set_defaults(func=lambda td, a: cmd_index_stats(
td, a.index, int(a.date) if a.date else None,
a.show_all, a.top, a.concurrency))
b = add("board", "板块") b = add("board", "板块")
b.add_argument("action", choices=["list", "members", "ranking", "change"]) b.add_argument("action", choices=["list", "members", "ranking", "change"])
b.add_argument("symbol", nargs="?", default="") b.add_argument("symbol", nargs="?", default="")
+13
View File
@@ -33,6 +33,7 @@ _SERVERS_FILE = Path(__file__).resolve().parent / "servers.json" # 包内清单
# 当日测速缓存(结构变更时递增,旧版本缓存自动失效) # 当日测速缓存(结构变更时递增,旧版本缓存自动失效)
PROBE_CACHE_VERSION = 1 PROBE_CACHE_VERSION = 1
_PROBE_CACHE_NAME = "hosts-today.json" _PROBE_CACHE_NAME = "hosts-today.json"
_DECIMAL_CACHE_NAME = "decimal-points.json"
# ---------- 包内清单(只读基线) ---------- # ---------- 包内清单(只读基线) ----------
@@ -97,6 +98,18 @@ def probe_cache_file() -> Path:
return probe_cache_dir() / _PROBE_CACHE_NAME return probe_cache_dir() / _PROBE_CACHE_NAME
def decimal_cache_file() -> Path:
"""权威价格小数位表缓存路径(每次调用重新解析,便于测试改环境变量后生效)。"""
return probe_cache_dir() / _DECIMAL_CACHE_NAME
def decimal_sync_disabled() -> bool:
"""是否已通过 ``PYPYTDX_DATA_DISABLE_DECIMAL_SYNC`` 关闭小数位后台同步。"""
return os.environ.get("PYPYTDX_DATA_DISABLE_DECIMAL_SYNC", "").strip().lower() in {
"1", "true", "yes", "on",
}
def probe_disabled() -> bool: def probe_disabled() -> bool:
"""是否已通过 ``PYPYTDX_DATA_DISABLE_PROBE`` 关闭当日测速缓存。""" """是否已通过 ``PYPYTDX_DATA_DISABLE_PROBE`` 关闭当日测速缓存。"""
return os.environ.get("PYPYTDX_DATA_DISABLE_PROBE", "").strip().lower() in { return os.environ.get("PYPYTDX_DATA_DISABLE_PROBE", "").strip().lower() in {
+223
View File
@@ -0,0 +1,223 @@
"""权威价格小数位表的同步与缓存(对用户无感)。
为什么需要它
------------
``get_quotes`` 解析报价时,协议给的是**整数**,真实价格 = ``整数 / 10**小数位``。
小数位在协议里**不是**随报价一起下发的,只能靠推断(见
:func:`pytdxdata.protocol.commands.quotes.price_decimal_digits`)。
但证券清单接口(``get_universe``,底层 ``GetSecurityListCmd``)**自带**这个字段:
``SecurityInfo.decimal_point``,同时还带 ``pre_close``。这就是权威来源。本模块把它
灌进 ``price_decimal_digits`` 使用的运行时表,使价格标度不再依赖硬编码前缀规则。
行为
----
- **当日首次启动**检查缓存文件(``decimal-points.json``):命中就直接装表(纯内存,
无网络、约 10 ms);未命中则挂一个**后台 task** 拉证券清单并写缓存,
当天不再重复。
- 后台 task 不阻塞 ``start()``,不打印、不抛异常(取消除外),失败只落 debug 日志。
- 未同步成功时 ``price_decimal_digits`` 静默退回静态前缀规则,行为与 0.6.5 一致。
- 证券清单本身已被 ``cache/`` 层缓存,因此同步的实际网络开销只在**首次**
(约 26 + 25 个分页请求)。
开关
----
- ``PYPYTDX_DATA_DISABLE_DECIMAL_SYNC=1`` 关闭后台同步(CI / 测试用,保证确定性)。
- ``PYPYTDX_DATA_CACHE_DIR`` 覆盖缓存目录(测试指向临时目录)。
"""
from __future__ import annotations
import asyncio
import json
import logging
import os
from datetime import date
from pathlib import Path
from typing import TYPE_CHECKING, Any
from .config import decimal_cache_file, decimal_sync_disabled
from .protocol.commands.quotes import (
clear_decimal_map,
decimal_map_source,
decimal_map_stats,
install_decimal_map,
)
if TYPE_CHECKING: # pragma: no cover
from .api import TdxData
log = logging.getLogger(__name__)
#: 同步哪些市场(只对 A 股有意义:扩展市场报价本身不含标度歧义)
DEFAULT_MARKETS: tuple[str, ...] = ("sh", "sz")
_CN_NAME = {"sh": 1, "sz": 0}
_state: dict[str, Any] = {"loaded": "", "task": None}
# ---------------------------------------------------------------------- #
# 磁盘缓存
# ---------------------------------------------------------------------- #
def cache_file() -> Path:
"""缓存文件路径(每次重新解析,便于测试改环境变量)。"""
return decimal_cache_file()
def _read_cache(path: Path | None = None) -> dict[str, dict[str, int]] | None:
"""读缓存 → ``{"1": {code: dp}}``;缺失 / 损坏 → ``None``。"""
target = Path(path) if path is not None else cache_file()
try:
raw = json.loads(target.read_text(encoding="utf-8"))
except (OSError, ValueError):
return None
if not isinstance(raw, dict):
return None
if raw.get("date") != date.today().isoformat():
return None
markets = raw.get("markets")
if not isinstance(markets, dict) or not markets:
return None
out: dict[str, dict[str, int]] = {}
for k, v in markets.items():
if not isinstance(v, dict):
continue
try:
out[str(int(k))] = {str(c): int(d) for c, d in v.items() if int(d) > 0}
except (TypeError, ValueError):
continue
return out or None
def load_cached(path: Path | None = None) -> int:
"""把当日缓存装进权威表;成功返回覆盖条目数,否则 0。"""
if decimal_sync_disabled():
return 0
markets = _read_cache(path)
if not markets:
return 0
items = [(int(m), c, d) for m, codes in markets.items() for c, d in codes.items()]
stats = install_decimal_map(items, source="cache")
_state["loaded"] = date.today().isoformat()
log.debug("小数位表已从缓存加载: %s", stats)
return int(stats["stock"])
def _write_cache(markets: dict[str, dict[str, int]], path: Path | None = None) -> None:
"""原子写缓存(tmp + ``os.replace``),失败只落 debug 日志。"""
target = Path(path) if path is not None else cache_file()
payload = {"date": date.today().isoformat(), "markets": markets}
tmp = target.with_suffix(".tmp")
try:
target.parent.mkdir(parents=True, exist_ok=True)
tmp.write_text(
json.dumps(payload, ensure_ascii=False, separators=(",", ":")),
encoding="utf-8",
)
os.replace(tmp, target)
except OSError as e: # pragma: no cover - 磁盘不可写等
log.debug("小数位表缓存写入失败: %s", e)
try:
tmp.unlink(missing_ok=True)
except OSError:
pass
# ---------------------------------------------------------------------- #
# 同步
# ---------------------------------------------------------------------- #
async def refresh(td: "TdxData", markets: tuple[str, ...] = DEFAULT_MARKETS,
*, path: Path | None = None) -> dict[str, int]:
"""拉证券清单 → 装表 + 写缓存。返回统计。
证券清单走 ``get_universe``(含 ``cache/`` 层缓存),只有首次真正发请求。
"""
tables: dict[str, dict[str, int]] = {}
items: list[tuple[int, str, int]] = []
for name in markets:
m = _CN_NAME.get(name)
if m is None:
continue
rows = await td.get_universe(name)
table: dict[str, int] = {}
for it in rows:
dp = int(getattr(it, "decimal_point", 0) or 0)
code = str(getattr(it, "code", "") or "").strip().rstrip("\x00")
if dp <= 0 or not code:
continue
table[code] = dp
items.append((m, code, dp))
if table:
tables[str(m)] = table
stats = install_decimal_map(items, source="universe")
_state["loaded"] = date.today().isoformat()
if tables:
_write_cache(tables, path)
log.debug("小数位表同步完成: %s", stats)
return stats
async def _run(td: "TdxData") -> dict[str, int]:
try:
return await refresh(td)
except asyncio.CancelledError:
raise
except Exception as e: # pragma: no cover - 网络/解析异常都不该影响用户
log.debug("小数位表后台同步失败: %s", e)
return {}
def schedule(td: "TdxData") -> asyncio.Task | None:
"""[内部] 需要时把同步挂到后台并返回 Task(调用方持有 / 取消)。
- 已关闭(``PYPYTDX_DATA_DISABLE_DECIMAL_SYNC=1``)→ ``None``
- 当日缓存命中 → 立即装表(同步、无网络),返回 ``None``
- 否则创建后台 task;进程内单例,重复调用复用同一个 task
"""
if decimal_sync_disabled():
return None
if decimal_map_stats()["stock"] and _state["loaded"] == date.today().isoformat():
return None
if load_cached():
return None
task = _state.get("task")
if task is not None and not task.done():
return task
task = asyncio.get_running_loop().create_task(_run(td), name="pytdxdata-decimal-sync")
_state["task"] = task
return task
async def cancel(task: asyncio.Task | None) -> None:
"""取消并等待后台同步 task(供 ``TdxData.close()`` 调用)。"""
if task is None or task.done():
return
task.cancel()
try:
await task
except (asyncio.CancelledError, Exception): # noqa: BLE001 - 关闭路径不抛
pass
def status() -> dict[str, Any]:
"""当前状态快照(排障 / 测试用)。"""
task = _state.get("task")
return {
"loaded": _state["loaded"],
"source": decimal_map_source(),
"syncing": bool(task is not None and not task.done()),
**decimal_map_stats(),
}
def reset() -> None:
"""清空内存状态(测试用;不动磁盘缓存)。"""
task = _state.get("task")
if task is not None and not task.done():
task.cancel()
_state["task"] = None
_state["loaded"] = ""
clear_decimal_map()
+180 -8
View File
@@ -117,8 +117,8 @@ class SecurityBar:
high: float high: float
low: float low: float
close: float close: float
vol: float # 股票=股 / 转债=手(与vipdoc一致) vol: float # 股票=股 / 转债=手(1 手=10 张,与 vipdoc 一致)
amount: float # 元 amount: float # 元。⚠️ amount/vol 只在股票上等于均价;转债还要再 /10(手=10张)
year: int year: int
month: int month: int
day: int day: int
@@ -228,23 +228,72 @@ class SecurityInfo:
@dataclass(slots=True) @dataclass(slots=True)
class TransactionRecord: class TransactionRecord:
"""逐笔成交(跨市场统一模型;A股 MAC 通道含 trade_count=成交笔数)""" """逐笔成交(跨市场统一模型;A股 MAC 通道含 trade_count=成交笔数)。
注意(实测 2026-10-21):
- 服务端是按**约 3 秒聚合**返回的切片,不是交易所原始逐笔成交。
- ``vol`` / ``trade_count`` 都是**该切片自身的增量**,不是当日累计值。
- ``trade_count`` 为**有符号**值:集合竞价段(``bs_flag=8``)会是负数
(实测 sh600000 09:15 段为 -874),不要对全表直接求 max/mean。
- 返回列表已按时间升序,但同秒多条时顺序不保证。
- ``bs_flag=8`` 的**竞价委托记录已被 ``get_ticks()`` 过滤掉**,不会出现在结果里。
**指数语义完全不同**(``is_cn_index`` 为真时,如 ``sh000016``):
- ``vol`` 单位是**百元**(= 成交额 / 100),不是「手」。实测
``快照amount / Σtick.vol = 100.0002``,且 ``Σtick.vol`` 与分时 Σvol **逐位相等**。
想和成分股成交量对比必须先换算,直接比会得到 ~26 倍的假比例。
- ``trade_count`` **不是笔数**,而是**当时均价 × 1000**(实测 tc/1000 与分时
``avg_price`` 的差 < 1.1)。指数没有「成交笔数」概念。
- ``bs_flag`` 语义不成立(实测返回 2,而模型只定义 0=卖 / 1=买)。
**场内基金同样没有笔数**(ETF / 多数 LOF,见
:func:`~pytdxdata.symbols.has_order_count`):
- ``trade_count`` 是 ``价格 × 1000``(实测 ``sh510310`` 收盘 tc=3994,
当时价 3.989 → 3989,偏差 σ≈2;货币 ETF ``sh511990`` 恒为 0)。
- ``vol`` 仍是**手**,``bs_flag`` 正常。
- 这类标的的 :attr:`trade_count_available` 为 ``False``,``trade_count`` 已被
置 ``0``,不要拿去求和/求均值。
``trade_count_available=False`` 表示该品种**协议上就不带笔数**,
与「本切片恰好 0 笔」不是一回事。
**可转债(110/111/113/118/123/127/128/132)三样都是真的**(实测 2026-10-22,
16 只 A股可转债当日 + 8 只历史日全部自洽):
- ``vol`` 单位是**手**(1 手 = 10 张),与股票逐笔一致;``trade_count`` 是**真实成交笔数**
(``sh111000`` 3631 条切片 / 20443 笔,可转债在
:func:`~pytdxdata.symbols.has_order_count` 里判为**可用**)。
- ``Σvol`` 与**日线** ``SecurityBar.vol`` **1:1 相等**(比值 1.000000)——
这点和股票相反:股票逐笔是「手」、日线是「股」,Σ比值是 **0.01**。
- ⚠️ 但**日线** ``amount / vol`` **不等于**均价,而是均价的 **10 倍**(``vol`` 的「手」= 10 张)。
要算可转债均价须用 ``amount / vol / 10``;股票则是 ``amount / vol`` 直接就对。
实测:``sz127025`` 逐笔 VWAP 105.9063 vs ``amount/vol`` 1059.0957,比值 10.0003。
"""
market: int market: int
code: str code: str
time: str # HH:MM:SS(扩展市场为 HH:MM) time: str # HH:MM:SS(扩展市场为 HH:MM)
price: float price: float
vol: float # 手(MAC逐笔) vol: float # 股票/转债=手;指数=百元(成交额/100),见类注释
trade_count: int # 成交笔数(MAC=逐笔记录字段;标准当日逐笔协议也有但需单独解析) trade_count: int # 股票=笔数增量;指数/基金=占位值,见类注释
bs_flag: int # 1=主动买 0=主动卖 bs_flag: int # 1=主动买 0=主动卖 2=中性 5=盘后固定价(指数语义不成立)
symbol: str = "" symbol: str = ""
fields: dict[str, float] = field(default_factory=dict) # 扩展市场额外字段(增仓等) fields: dict[str, float] = field(default_factory=dict) # 扩展市场额外字段(增仓等)
trade_count_available: bool = True # 指数/ETF/LOF 为 False(协议无笔数字段)
@dataclass(slots=True) @dataclass(slots=True)
class MinuteBar: class MinuteBar:
"""分时(跨市场统一模型;A股 MAC 0x122D / 扩展市场 0x122D 同构)""" """分时(跨市场统一模型;A股 MAC 0x122D / 扩展市场 0x122D 同构)。
``hour``/``minute`` 是**真实时钟**(``0 <= hour < 24``),跨日的分时请用
:attr:`datetime` / :attr:`trading_date`,不要自己拿 ``hour`` 拼日期。
"""
price: float price: float
vol: float vol: float # 手(股票/转债同);Σ分时vol 与当日日线 vol 同口径
avg_price: float = 0.0 # 均价(协议第2字段) avg_price: float = 0.0 # 均价(协议第2字段)
hour: int = 0 hour: int = 0
minute: int = 0 minute: int = 0
@@ -253,6 +302,50 @@ class MinuteBar:
code: str = "" code: str = ""
date: int = 0 # YYYYMMDD date: int = 0 # YYYYMMDD
fields: dict[str, float] = field(default_factory=dict) # 扩展市场额外字段(动量等) fields: dict[str, float] = field(default_factory=dict) # 扩展市场额外字段(动量等)
day_offset: int = 0 # 相对 date 的自然日偏移:美股凌晨段为 +1
def __post_init__(self) -> None:
"""美股(EX 通道)分时时间归一化。
EX 分时的 ``minutes`` 字段是**自开盘起的连续分钟数**,不是「当天的分钟数」:
实测 ``usAAPL`` 返回 ``21:30`` ~ ``27:59``(390 条里 240 条的 ``hour`` ≥ 24)。
``27:59`` 实际是次日 ``03:59`` 北京时间。直接 ``minutes // 60`` 会得到
``hour=27`` 这种**非法时钟值**,任何 ``datetime(...)`` 拼接都会抛异常,
按小时聚合也会把凌晨段错误地归到同一天。
这里拆成「真实小时 + 日偏移」:``hour = (minutes // 60) % 24``,
``day_offset = (minutes // 60) // 24``。A股/港股/期货的 ``hour`` 本就合法
(A股合成时间 09:30~15:00、港股 09:30~16:00),**零行为变化**。
"""
if self.hour >= 24:
self.day_offset += self.hour // 24
self.hour %= 24
@property
def datetime(self) -> datetime:
"""该分时点的时刻(北京时间,已含跨日偏移)。"""
base = date(self.date // 10000, self.date // 100 % 100, self.date % 100)
d = base + timedelta(days=self.day_offset)
return datetime(d.year, d.month, d.day, self.hour, self.minute)
@property
def is_us(self) -> bool:
"""是否美股(EX 通道 ``US_STOCK``)。"""
return self.market in _US_EX_MARKETS
@property
def trading_date(self) -> str:
"""该分时点归属的交易日,``YYYY-MM-DD``。
美股:一个交易日横跨两个北京自然日(21:30~次日 04:00),凌晨段
(北京 hour < 12)归属**前一天**的美东交易日,与
:attr:`SecurityBar.trading_date` 口径一致。A股/港股/期货即 ``date`` 本身。
按交易日聚合分时(合成日线、日内统计)**必须用它**。
"""
base = self.datetime
if self.is_us and base.hour < _US_DAY_SPLIT_HOUR:
base -= timedelta(days=1)
return base.date().isoformat()
@dataclass(slots=True) @dataclass(slots=True)
@@ -494,3 +587,82 @@ class FinancialRecord:
market: int # 0=深 1=沪 market: int # 0=深 1=沪
report_date: int # 报告期 YYYYMMDD report_date: int # 报告期 YYYYMMDD
fields: list[float] = field(default_factory=list) fields: list[float] = field(default_factory=list)
# ===== 指数成分成交汇总(TdxData.get_index_trade_stats) =====
@dataclass(slots=True)
class MinuteTradeStat:
"""指数**全部成分股**汇总后的单分钟成交**增量**(不是累计)。
累计曲线用 :meth:`IndexTradeStats.cumulative`,不要自己再对同一字段反复累加。
"""
time: str # HH:MM
volume: float # 成交量合计(手,1 手 = 100 股)
trade_count: int # 成交笔数合计
amount: float # 成交额估算(元)= Σ(逐笔价 × 手 × 100)
@dataclass(slots=True)
class IndexTradeStats:
"""某个指数「全部成分股逐笔」汇总出的成交曲线。
为什么需要它:**通达信协议里指数本身没有成交笔数** —— 指数逐笔
(``sh000001``)的 ``trade_count`` 字段实际是「当时均价 × 1000」(见
:class:`TransactionRecord`)。所以「指数笔数」唯一可行的口径是
**把成分股的逐笔笔数累加**,本类即该口径。
口径校验(2026-10-22 实测):
- ``total_amount`` 是**成分股**合计。对「成分恰好就是指数覆盖面」的指数,它与
指数快照 ``amount`` **完全相等**:``sh000016`` 661.34 亿 = 661.34 亿、
``sh000300`` 2373.98 亿 = 2373.98 亿(比值 1.0000)。
- **但深市 ``399xxx`` 的指数成交额字段是「全板块」口径**,不是样本股:
实测 ``sz399001`` 快照 5081.40 亿 = 深市 2908 只股票合计 5081.01 亿、
``sz399006`` 2568.60 亿 = 创业板全部股票合计。所以拿深市指数 ``amount``
与 ``total_amount`` 比会得到 ~0.43 这类无意义比值——不是数据错,是口径不同。
- ``total_volume`` 单位是**手**(与指数逐笔的「百元」单位不同,勿混用)。
- 北交所成分(``bj`` 前缀)**没有笔数字段**(实测 ``trade_count`` 恒为 0):
此时 ``trade_count_members = 0``、:attr:`trade_count_available` 为 ``False``,
只有成交量可用。``sh``/``sz`` 成分均正常返回笔数。
``minutes`` 只包含**实际有成交**的分钟(午休 11:30~13:00 之间无桶),
画累计曲线时请按 ``time`` 自行补齐或直接连线。
"""
index: str # 规范化指数标的(如 sh000300)
date: int | None # None=当日;否则 YYYYMMDD
member_count: int # 指数成分数
active_members: int # 实际有逐笔成交的成分数(停牌/无成交不计)
trade_count_members: int # 返回了笔数的成分数(北交所为 0)
minutes: list[MinuteTradeStat] = field(default_factory=list)
total_volume: float = 0.0
total_trade_count: int = 0
total_amount: float = 0.0
failed_members: int = 0 # 逐笔拉取失败的成分数(skip_errors=True 时)
elapsed: float = 0.0 # 本次汇总耗时(秒)
@property
def trade_count_available(self) -> bool:
"""是否有笔数口径(北交所成分为 ``False``,此时 ``trade_count`` 全 0)。"""
return self.trade_count_members > 0
def cumulative(self) -> list[MinuteTradeStat]:
"""累计曲线:当日**截止该分钟**的累计量,时间升序。
这是「当日截止该分钟累计成交笔数」的直接数据源。
"""
out: list[MinuteTradeStat] = []
vol = 0.0
tc = 0
amt = 0.0
for m in self.minutes:
vol += m.volume
tc += m.trade_count
amt += m.amount
out.append(MinuteTradeStat(time=m.time, volume=vol,
trade_count=tc, amount=amt))
return out
def cum_trade_count(self) -> list[tuple[str, int]]:
"""``[(HH:MM, 累计笔数)]``,画「截止该分钟累计笔数」曲线用。"""
return [(m.time, m.trade_count) for m in self.cumulative()]
+54
View File
@@ -3,6 +3,7 @@ from __future__ import annotations
import asyncio import asyncio
import time import time
from collections.abc import Sequence
from typing import Generic, TypeVar from typing import Generic, TypeVar
from ..protocol.commands.base import BaseCommand from ..protocol.commands.base import BaseCommand
@@ -22,6 +23,7 @@ class TdxConnection(Generic[T]):
self._reader: asyncio.StreamReader | None = None self._reader: asyncio.StreamReader | None = None
self._writer: asyncio.StreamWriter | None = None self._writer: asyncio.StreamWriter | None = None
self._last_result: T | None = None self._last_result: T | None = None
self._pipeline_result: list[T] = []
@property @property
def alive(self) -> bool: def alive(self) -> bool:
@@ -107,6 +109,58 @@ class TdxConnection(Generic[T]):
body = decompress_body(header, raw_body) body = decompress_body(header, raw_body)
self._last_result: T = cmd.parse_response(body) self._last_result: T = cmd.parse_response(body)
async def execute_pipeline(
self, cmds: Sequence[BaseCommand[T]], *, timeout: float | None = None
) -> list[T]:
"""单连接**流水线**:一次性发出全部请求,再按序读取全部响应。
通达信协议是自描述长度帧(16 字节帧头 + body),且服务器**按序**应答,
因此可以在同一 TCP 连接上连续下发多条请求、省掉 ``(N-1)`` 次往返。
实测 66 个「80 只报价」请求:逐条 7,372ms → 流水线 526ms(14 倍)。
与 :meth:`execute` 的关键差异:``execute`` 逐条严格配对,失败可小范围重试;
流水线一旦中途失败,**响应流已错配**,剩余响应无法再与请求对齐,
故此处强制断开连接(上层应换连接重试)。
参数:
cmds: 同一连接的请求序列(通常为同一类命令的不同分片)。
timeout: 整体超时(默认 ``self.timeout * 2``,流水线耗时随条数增长)。
返回:
与 ``cmds`` **等长且同序**的结果列表。
"""
if not cmds:
return []
if not self.alive:
raise ConnectionError(f"连接未建立或已关闭: {self.host}")
assert self._reader is not None and self._writer is not None
budget = timeout if timeout is not None else self.timeout * 2
try:
try:
await asyncio.wait_for(self._io_pipeline(cmds), budget)
except asyncio.TimeoutError:
self._force_close()
raise ConnectionError(
f"流水线超时 {self.host} ({budget:.1f}s, {len(cmds)} 条)") from None
except (ConnectionError, OSError) as e:
self._force_close()
raise ConnectionError(f"流水线错误 {self.host}: {e}") from e
except (ConnectionError, OSError):
raise
return self._pipeline_result
async def _io_pipeline(self, cmds: Sequence[BaseCommand[T]]) -> None:
assert self._reader is not None and self._writer is not None
self._writer.write(b"".join(c.build_request() for c in cmds))
await self._writer.drain()
results: list[T] = []
for cmd in cmds:
header_buf = await self._recv_exact(HEADER_SIZE)
header = parse_header(header_buf)
raw_body = await self._recv_exact(header.zipsize)
results.append(cmd.parse_response(decompress_body(header, raw_body)))
self._pipeline_result: list[T] = results
async def _send_setup(self) -> None: async def _send_setup(self) -> None:
"""按序发送3条握手命令并丢弃响应。""" """按序发送3条握手命令并丢弃响应。"""
assert self._reader is not None and self._writer is not None assert self._reader is not None and self._writer is not None
+45
View File
@@ -59,6 +59,11 @@ class PooledConnection:
self.last_used = time.monotonic() self.last_used = time.monotonic()
return await self.conn.execute(cmd) return await self.conn.execute(cmd)
async def execute_pipeline(self, cmds, *, timeout: float | None = None):
"""在同一条连接上流水线执行多条命令(见 TdxConnection.execute_pipeline)。"""
self.last_used = time.monotonic()
return await self.conn.execute_pipeline(cmds, timeout=timeout)
async def close(self) -> None: async def close(self) -> None:
self.alive = False self.alive = False
try: try:
@@ -273,6 +278,46 @@ class DynamicPool:
] ]
return random.choice(alive) if alive else None return random.choice(alive) if alive else None
async def warm(self, n: int) -> None:
"""并发预建连接至 ``n`` 条(最多到 ``max_size``)。
:meth:`acquire` 建连时持有池锁,**一次只能建一条**——22 路并发首次
调用会排队 22×握手时间(实测全市场首拉多花 5~6s)。大批量请求前先调
本方法把连接并发建好,首拉只多付一次握手。
失败静默忽略:缺的连接后续由 ``acquire`` 按需补齐。
"""
if self._closed:
return
need = min(n, self.max_size) - self._total
if need <= 0:
return
# 用局部计数挑 host,避免并发建连时反复选中同一台
counts = dict(self._server_count)
ranked = self.health.rank(self.servers)
hosts: list[str] = []
for _ in range(need):
host = next((h for h in ranked
if not self.health.in_cooldown(h)
and counts.get(h, 0) < self.per_server_cap), None)
if host is None:
break
hosts.append(host)
counts[host] = counts.get(host, 0) + 1
if not hosts:
return
results = await asyncio.gather(
*(self._new_conn(h) for h in hosts), return_exceptions=True)
async with self._lock:
for pc in results:
if isinstance(pc, PooledConnection):
self._total += 1
self.stats.total = self._total
if not self._closed and pc.alive:
self._idle.append(pc)
async def release(self, pc: PooledConnection, *, broken: bool = False) -> None: async def release(self, pc: PooledConnection, *, broken: bool = False) -> None:
if self.stats.borrowed > 0: if self.stats.borrowed > 0:
self.stats.borrowed -= 1 self.stats.borrowed -= 1
+146 -11
View File
@@ -12,20 +12,155 @@ from .base import BaseCommand
MAX_QUOTES = 80 MAX_QUOTES = 80
def price_decimal_digits(market: int, code: str) -> int: # ====================================================================== #
"""推断报价有效小数位:股票2位(/100)、指数/ETF/基金/转债/国债3位(/1000)。""" # 权威小数位表(由证券清单 SecurityInfo.decimal_point 灌入)
code = (code or "").strip().rstrip("\x00") # ====================================================================== #
if market == 1: # SH # 查找顺序:**权威表 → 静态前缀规则**。权威表来自 ``get_universe`` 返回的
if code.startswith(("5", "000", "8")): # ``SecurityInfo.decimal_point``(协议自带,见 protocol/commands/security_list.py),
return 3 # 由 ``pytdxdata.decimal_sync`` 在后台每日同步一次;未同步时静默退回静态规则,
return 2 # 行为与 0.6.5 完全一致。
if market == 0: # SZ _DECIMAL_PREFIX: dict[tuple[int, str], int] = {}
if code.startswith("1"): _DECIMAL_CODE: dict[tuple[int, str], int] = {}
return 3 _DECIMAL_MIXED: set[tuple[int, str]] = set()
return 2 _DECIMAL_SOURCE: str = ""
def install_decimal_map(
items: "list[tuple[int, str, int]] | tuple[tuple[int, str, int], ...]",
*,
source: str = "",
) -> dict[str, int]:
"""用证券清单灌入权威小数位表。
``items`` 为 ``(market, code, decimal_point)`` 序列(直接来自
``SecurityInfo``)。按**三位代码前缀**归并成一个极小的表(全量 5 万只标的
只产生数百个前缀);若某个前缀出现**不同**小数位,则该前缀退化为逐代码查表,
只把这些例外代码存进 ``_DECIMAL_CODE``——这样即使将来交易所新增代码段,
也不会因为前缀规则过时而算错价格。
返回统计:``{"prefixes", "codes", "mixed_prefixes", "stock"}``。
"""
global _DECIMAL_SOURCE
votes: dict[tuple[int, str], dict[int, int]] = {}
per_code: dict[tuple[int, str], int] = {}
for market, code, dp in items:
c = str(code or "").strip().rstrip("\x00")
if len(c) < 3:
continue
m = int(market)
if m not in (0, 1):
continue
d = int(dp)
if d <= 0:
continue
per_code[(m, c)] = d
key = (m, c[:3])
bucket = votes.setdefault(key, {})
bucket[d] = bucket.get(d, 0) + 1
_DECIMAL_PREFIX.clear()
_DECIMAL_CODE.clear()
_DECIMAL_MIXED.clear()
for key, bucket in votes.items():
if len(bucket) == 1:
_DECIMAL_PREFIX[key] = next(iter(bucket))
else:
_DECIMAL_MIXED.add(key)
for (m, c), d in per_code.items():
if (m, c[:3]) in _DECIMAL_MIXED:
_DECIMAL_CODE[(m, c)] = d
_DECIMAL_SOURCE = source or "universe"
return decimal_map_stats()
def clear_decimal_map() -> None:
"""清空权威表(回到纯静态前缀规则;测试与排障用)。"""
global _DECIMAL_SOURCE
_DECIMAL_PREFIX.clear()
_DECIMAL_CODE.clear()
_DECIMAL_MIXED.clear()
_DECIMAL_SOURCE = ""
def decimal_map_stats() -> dict[str, int]:
"""权威表状态:已覆盖前缀数 / 例外代码数 / 混合前缀数 / 总条目数。"""
return {
"prefixes": len(_DECIMAL_PREFIX),
"codes": len(_DECIMAL_CODE),
"mixed_prefixes": len(_DECIMAL_MIXED),
"stock": len(_DECIMAL_PREFIX) + len(_DECIMAL_CODE),
}
def decimal_map_source() -> str:
"""权威表来源标记(``universe`` / ``cache`` / ``""``=未加载)。"""
return _DECIMAL_SOURCE
def _decimal_digits_static(market: int, code: str) -> int:
"""静态前缀规则(**兜底**,权威表缺失时使用)。"""
if market == 1: # 沪市
if code.startswith("5"):
return 3 # 基金/ETF/LOF(50/51/52/56/58 系)
if code.startswith("900"):
return 3 # B股
if code.startswith(("000", "36", "88", "99", "7", "6")):
return 2 # 指数/板块/优先股/配售/股票(60/601/603/605/688)
return 4 # 债券(01x/1xx/2xx)、回购(204xxx)、ABS
if market == 0: # 深市
if code.startswith(("080", "158", "159", "160", "161", "162", "163",
"164", "165", "166", "167", "168", "169",
"180", "181")):
return 3 # 基金/ETF/LOF/配债
if code.startswith(("000", "001", "002", "003", "070", "140",
"200", "201", "300", "301", "302", "395", "399")):
return 2 # 股票/指数/B股/优先股
return 4 # 债券(10x-14x/19x)、回购(131xxx)、ABS(50x/52x/56x)
return 2 return 2
def price_decimal_digits(market: int, code: str) -> int:
"""报价小数位:**协议原始整数除以 ``10**该值``** 才是真实价格。
查找顺序:
1. **权威表**(:func:`install_decimal_map` 灌入)—— 来自证券清单里协议自带的
``SecurityInfo.decimal_point``,由 ``pytdxdata.decimal_sync`` 每日后台同步一次。
2. **静态前缀规则**(:func:`_decimal_digits_static`)—— 权威表未就绪时兜底。
四档:股票/指数/板块 ``2``、基金/ETF/B股 ``3``、债券/回购 ``4``。
为什么还要静态规则:``get_quotes`` 解析是**纯同步、无网络**的路径,解析那一刻
可能还没同步过证券清单;而现网首次安装/离线环境也必须能算出价格。
权威表与静态规则的一致性:2026-10-22 拉取全量 51,982 只(沪 27,520 + 深 24,462)
逐条比对,静态规则与 ``decimal_point`` **一一吻合、零例外**——因为全量清单里
没有任何一个三位前缀出现混合小数位。因此两者并存不会产生口径冲突。
历史修正(两次都是同一类错误:把某段前缀想当然地归错档):
- **2026-10-21**:沪市 ``000``/``8`` 被判成 3 位 → 上证指数/上证50/科创50/
板块指数**价格少 10 倍**(3713 读成 371)。它们实际是 2 位。
- **2026-10-22**:债券与回购全线偏差 —— 沪市 ``010/018/019/020/1xx/2xx``
与深市 ``10x/11x/12x/13x/14x/19x/50x/52x/56x`` 实际是 **4 位**,原先
沪市落到 2 位(**少 100 倍**)、深市落到 3 位(**少 10 倍**)。实测例:
``sh019742`` 110.228 读成 11022.8、``sh204001`` 1.385 读成 138.5、
``sz131810`` 1.38 读成 13.8。同时沪市 B 股 ``900xxx`` 是 3 位(原先 2 位)。
实证方法:拿 ``pre_close``(快照口径)除以上一交易日**日线收盘**(K线口径,
走 ``/1000`` 独立路径),比值即标度误差 —— 1.0 正确、10 少 10 倍、100 少 100 倍。
"""
code = (code or "").strip().rstrip("\x00")
m = int(market)
if m in (0, 1):
hit = _DECIMAL_PREFIX.get((m, code[:3]))
if hit is not None:
return hit
hit = _DECIMAL_CODE.get((m, code))
if hit is not None:
return hit
return _decimal_digits_static(m, code)
def format_server_time(raw: int) -> str: def format_server_time(raw: int) -> str:
"""服务器时间编码:小时 + 百万分之一小时。14999212 -> 14:59:57.163""" """服务器时间编码:小时 + 百万分之一小时。14999212 -> 14:59:57.163"""
hours, fractional_hour = divmod(raw, 1_000_000) hours, fractional_hour = divmod(raw, 1_000_000)
+19 -6
View File
@@ -5,6 +5,7 @@ import struct
from ..._binary import unpack_from from ..._binary import unpack_from
from ...models import TransactionRecord from ...models import TransactionRecord
from ...symbols import has_order_count
from ..datetime_ import get_time from ..datetime_ import get_time
from ..mac_frame import build_mac_request from ..mac_frame import build_mac_request
from ..price import get_price from ..price import get_price
@@ -32,6 +33,8 @@ class GetTransactionDataCmd(BaseCommand[list[TransactionRecord]]):
pos = 2 pos = 2
last_price = 0 last_price = 0
records: list[TransactionRecord] = [] records: list[TransactionRecord] = []
code = self.code.decode("utf-8")
has_tc = has_order_count(self.market, code)
for _ in range(num): for _ in range(num):
hour, minute, pos = get_time(body, pos) hour, minute, pos = get_time(body, pos)
price_diff, pos = get_price(body, pos) price_diff, pos = get_price(body, pos)
@@ -41,10 +44,12 @@ class GetTransactionDataCmd(BaseCommand[list[TransactionRecord]]):
_unknown, pos = get_price(body, pos) _unknown, pos = get_price(body, pos)
last_price += price_diff last_price += price_diff
records.append(TransactionRecord( records.append(TransactionRecord(
market=self.market, code=self.code.decode("utf-8"), market=self.market, code=code,
time=f"{hour:02d}:{minute:02d}:00", time=f"{hour:02d}:{minute:02d}:00",
price=last_price / 100.0, vol=float(vol), price=last_price / 100.0, vol=float(vol),
trade_count=int(num_orders), bs_flag=bs_flag, # 指数/ETF/LOF 该字段是「价格×1000」占位,不是笔数
trade_count=int(num_orders) if has_tc else 0,
bs_flag=bs_flag, trade_count_available=has_tc,
)) ))
return records return records
@@ -68,6 +73,7 @@ class GetHistoryTransactionDataCmd(BaseCommand[list[TransactionRecord]]):
pos = 6 # 2(num) + 4(skip) pos = 6 # 2(num) + 4(skip)
last_price = 0 last_price = 0
records: list[TransactionRecord] = [] records: list[TransactionRecord] = []
code = self.code.decode("utf-8")
for _ in range(num): for _ in range(num):
hour, minute, pos = get_time(body, pos) hour, minute, pos = get_time(body, pos)
price_diff, pos = get_price(body, pos) price_diff, pos = get_price(body, pos)
@@ -76,10 +82,11 @@ class GetHistoryTransactionDataCmd(BaseCommand[list[TransactionRecord]]):
_unknown, pos = get_price(body, pos) _unknown, pos = get_price(body, pos)
last_price += price_diff last_price += price_diff
records.append(TransactionRecord( records.append(TransactionRecord(
market=self.market, code=self.code.decode("utf-8"), market=self.market, code=code,
time=f"{hour:02d}:{minute:02d}:00", time=f"{hour:02d}:{minute:02d}:00",
price=last_price / 100.0, vol=float(vol), price=last_price / 100.0, vol=float(vol),
trade_count=0, bs_flag=bs_flag, # 历史无笔数 trade_count=0, bs_flag=bs_flag,
trade_count_available=False, # 历史通道整条链路都没有笔数字段
)) ))
return records return records
@@ -108,11 +115,15 @@ class MacTransactionCmd(BaseCommand[list[TransactionRecord]]):
def parse_response(self, body: bytes) -> list[TransactionRecord]: def parse_response(self, body: bytes) -> list[TransactionRecord]:
(count,) = unpack_from("<H", body, 29, "mac transaction count") (count,) = unpack_from("<H", body, 29, "mac transaction count")
records: list[TransactionRecord] = [] records: list[TransactionRecord] = []
has_tc = has_order_count(self.market, self.code)
for i in range(count): for i in range(count):
offset = 39 + i * 18 offset = 39 + i * 18
try: try:
# trade_count 必须按**有符号** int32 解析:集合竞价段(bs_flag=8)
# 该字段实为负值,按 uint32 会读出 42 亿量级的脏数据。
# 实测(2026-10-21 sh600000):09:15:03 原始 96fcffff = -874。
time_sec, price, volume, trade_count, bs_flag = unpack_from( time_sec, price, volume, trade_count, bs_flag = unpack_from(
"<IfIIH", body, offset, "mac transaction item" "<IfIiH", body, offset, "mac transaction item"
) )
except ValueError: except ValueError:
break break
@@ -121,6 +132,8 @@ class MacTransactionCmd(BaseCommand[list[TransactionRecord]]):
market=self.market, code=self.code, market=self.market, code=self.code,
time=f"{h:02d}:{m:02d}:{s:02d}", time=f"{h:02d}:{m:02d}:{s:02d}",
price=price, vol=float(volume), price=price, vol=float(volume),
trade_count=int(trade_count), bs_flag=bs_flag, # 指数/ETF/LOF 该字段是「价格×1000」占位,不是笔数
trade_count=int(trade_count) if has_tc else 0,
bs_flag=bs_flag, trade_count_available=has_tc,
)) ))
return records return records
+7 -3
View File
@@ -27,6 +27,7 @@ class PageSpec:
ordered: bool = True # 按 start 升序拼接 ordered: bool = True # 按 start 升序拼接
min_page: int = 100 # 规避标准K线 count<100 空响应坑 min_page: int = 100 # 规避标准K线 count<100 空响应坑
start_offset: int = 0 # 起始偏移(从该偏移开始分页,避免拉取历史全量) start_offset: int = 0 # 起始偏移(从该偏移开始分页,避免拉取历史全量)
newest_first: bool = False # 服务端 start 从最新往回数(0x122F 逐笔即如此)
@dataclass(slots=True) @dataclass(slots=True)
@@ -94,7 +95,7 @@ class Paginator:
if s.pc is not None: if s.pc is not None:
pc, s.pc = s.pc, None pc, s.pc = s.pc, None
await self.pool.release(pc) await self.pool.release(pc)
return self._assemble(out, spec.ordered) return self._assemble(out, spec.ordered, spec.newest_first)
def _plan_starts(self, spec: PageSpec) -> list[tuple[int, int]]: def _plan_starts(self, spec: PageSpec) -> list[tuple[int, int]]:
"""切页:返回 [(start, count), ...]。""" """切页:返回 [(start, count), ...]。"""
@@ -158,9 +159,12 @@ class Paginator:
return None, False return None, False
return None, False return None, False
def _assemble(self, results: list[PageResult], ordered: bool) -> list[Any]: def _assemble(self, results: list[PageResult], ordered: bool,
newest_first: bool = False) -> list[Any]:
# newest_first=True:start=0 是最新一页,start 越大越旧 → 反向拼才对得上时间序。
# 单页内部服务端已按时间升序,所以只翻页序,不翻页内元素。
if ordered: if ordered:
results.sort(key=lambda r: r.start) results.sort(key=lambda r: r.start, reverse=newest_first)
out: list[Any] = [] out: list[Any] = []
for r in results: for r in results:
out.extend(r.items) out.extend(r.items)
+50
View File
@@ -238,3 +238,53 @@ def is_cn_index(market: int, code: str) -> bool:
or (code.startswith("000") and int(market) == int(Market.SH)) or (code.startswith("000") and int(market) == int(Market.SH))
or (code.startswith("399") and int(market) == int(Market.SZ)) or (code.startswith("399") and int(market) == int(Market.SZ))
) )
def has_order_count(market: int, code: str) -> bool:
"""该标的的逐笔 ``trade_count`` 是否是**真实成交笔数**。
通达信对**指数**和**多数场内基金**不提供笔数,但那 4 字节里仍塞了值:
- 指数:``当时均价 × 1000``(见 :class:`~pytdxdata.models.TransactionRecord`)
- ETF / LOF:``价格 × 1000``;货币 ETF(如 ``sh511990``)恒为 ``0``
直接当笔数用会得到 3994 这种荒谬值(实测 ``sh510310``)。
实测覆盖(2026-10-22,按代码前缀各取样 1~7 只):
================== ========================== ==========
品种 前缀 笔数可用
================== ========================== ==========
沪股票/科创 600/601/603/605/688/689 是
沪 LOF 501/502/506 是
沪 ETF/LOF 51/52/56/58 **否**
沪指数 000/88 否
沪可转债 110/111/113/118 是
沪 EB(可交换债) 132 是
沪逆回购 204 是
深股票 000/001/002/003/300/301 是
深 ETF 159 **否**
深 LOF/基金 15/16/17/18 **否**
深指数 399 否
深可转债 123/127/128 是
================== ========================== ==========
``501x/502x/506x``(沪市 LOF)实测**有**真实笔数,所以不能用「沪市 5 开头
一律当基金」的粗规则;必须按上面这组前缀精确判定。
可转债(沪 110/111/113/118/132、深 123/127/128)**走的是兜底分支**——
它们既不在沪市 ``51/52/56/58`` 里,也不在深市 ``159/15/16/17/18`` 里,
所以两个分支都落到「是」。这不是巧合而是设计:兜底按「**排除法**」认定
基金/指数,其余一律当普通品种。实测验证(2026-10-22,16 只当日 + 8 只历史日):
笔数合计与日线量级吻合(``sh111000`` 3631 条切片 / 20443 笔,
``sz128136`` 1939 条 / 14593 笔),且 ``Σvol`` 与日线 ``vol`` 比值 **1.000000**。
"""
code = str(code)
if is_cn_index(market, code):
return False
m = int(market)
if m == int(Market.SH):
return code[:2] not in ("51", "52", "56", "58")
if m == int(Market.SZ):
return not (code[:3] == "159" or code[:2] in ("15", "16", "17", "18"))
return True