Skip to main content

QOS行情API Python SDK

PyPI version Python versions

官方Python SDK for QOS行情API,支持港股、美股、A股和加密货币的实时行情数据,K线数据,历史K线数据,直接提供了港股API,美股API,A股API,加密货币API。

功能特性

  • 完整实现所有7个HTTP接口和11个WebSocket命令
  • 强类型检查(基于Pydantic)
  • 同步HTTP客户端和异步WebSocket客户端
  • 自动重连和心跳维护
  • 完善的错误处理机制

安装

pip install qos-api

注册API KEY

注册api key

快速开始

初始化客户端

注册api key

from qos_api import QOSClient

client = QOSClient(api_key="您的API_KEY")

HTTP接口使用示例

获取品种基础信息

# 获取股票基础信息
info = client.get_instrument_info(["US:AAPL", "HK:00700", "SH:600519"])
print(info)

获取实时行情快照

# 获取实时行情
snapshot = client.get_snapshot(["US:TSLA", "HK:09988"])
print(snapshot)

获取K线数据

from qos_api.constants import KLineType

# 获取日K线
kline = client.get_kline(
    codes=["SH:600519"],
    ktype=KLineType.DAY.value,
    count=10
)
print(kline)

WebSocket接口使用示例

实时行情订阅

注册api key

import asyncio
from qos_api import QOSClient

async def handle_snapshot(snapshot):
    print(f"行情更新: {snapshot}")

async def pull_snapshot_periodically(client):
    """每10秒主动请求一次快照"""
    try:
        while True:
            data = await client.request_snapshot(["US:AAPL"])
            print(f"手动拉取快照数据: {data}")
            await asyncio.sleep(10)
    except asyncio.CancelledError:
        print("快照任务已取消")
        raise

async def main():
    client = QOSClient(api_key="您的API_KEY")
    client.register_callback("S", handle_snapshot)
    await client.connect_ws()
    await client.subscribe_snapshot(["US:AAPL", "HK:700"])

    # 启动后台任务
    snapshot_task = asyncio.create_task(pull_snapshot_periodically(client))

    try:
        while True:
            await asyncio.sleep(1)
    except KeyboardInterrupt:
        print("收到退出信号,正在清理...")
        snapshot_task.cancel()
        await asyncio.gather(snapshot_task, return_exceptions=True)
        await client.close()
        print("已优雅退出")

if __name__ == "__main__":
    try:
        asyncio.run(main())
    except KeyboardInterrupt:
        # 再次捕获,防止 asyncio.run 抛出错误
        print("主程序强制退出")

请求实时数据

注册api key

import asyncio
from qos_api import QOSClient
from qos_api.constants import KLineType

async def request_realtime_data():
    client = QOSClient(api_key="您的API_KEY")
    await client.connect_ws()
    
    # 请求实时行情
    snapshot = await client.request_snapshot(["US:AAPL"])
    print(snapshot)
    
    # 请求K线
    kline = await client.request_kline(
        codes=["HK:700"],
        ktype=KLineType.DAY.value,
        count=10
    )
    print(kline)

asyncio.run(request_realtime_data())

完整API参考

HTTP接口

方法 描述 对应API
get_instrument_info(codes) 获取品种基础信息 GET /instrument-info
get_snapshot(codes) 获取实时行情快照 GET /snapshot
get_depth(codes) 获取盘口深度 GET /depth
get_trades(codes, count) 获取逐笔成交 GET /trade
get_kline(codes, ktype, count, adjust) 获取K线数据 GET /kline
get_history_kline(codes, ktype, end_time, count, adjust) 获取历史K线 GET /history

WebSocket接口

订阅管理

方法 描述 对应命令
subscribe_snapshot(codes) 订阅实时快照 S
unsubscribe_snapshot(codes) 取消订阅快照 SC
subscribe_trades(codes) 订阅逐笔成交 T
unsubscribe_trades(codes) 取消订阅逐笔 TC
subscribe_depth(codes) 订阅盘口数据 D
unsubscribe_depth(codes) 取消订阅盘口 DC
subscribe_kline(codes, ktype) 订阅K线数据 K
unsubscribe_kline(codes, ktype) 取消订阅K线 KC

数据请求

方法 描述 对应命令
request_snapshot(codes) 请求实时快照 RS
request_trades(codes, count) 请求逐笔成交 RT
request_depth(codes) 请求盘口数据 RD
request_kline(codes, ktype, count) 请求K线数据 RK
request_history_kline(codes, ktype, end_time, count) 请求历史K线 RH
request_instrument_info(codes) 请求品种信息 RI

连接管理

方法 描述
connect_ws() 建立WebSocket连接
disconnect_ws() 断开WebSocket连接
heartbeat() 发送心跳包
register_callback(data_type, callback) 注册数据回调

数据模型

所有返回数据都使用Pydantic模型,主要模型包括:

  • InstrumentInfo: 品种基础信息
  • QuoteSnapshot: 行情快照
  • MarketDepth: 盘口深度
  • TradeTick: 逐笔成交
  • KLine: K线数据

错误处理

所有异常都继承自 QOSAPIError:

from qos_api import QOSAPIError

try:
    data = client.get_snapshot(["US:AAPL"])
except QOSAPIError as e:
    print(f"API错误: {e.code} - {e.message}")

限制说明

  1. 默认每个连接最多订阅10个品种
  2. HTTP请求频率限制为每分钟10次
  3. WebSocket消息间隔需大于1秒

技术支持

License

MIT

Metadata

Release files for qos-api 0.2.1

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for qos-api 0.2.1
File Size Uploaded Provenance
qos_api-0.2.1.tar.gz 12.4 kB Without provenance Details

Built distribution (wheel)

Table of built distributions (wheels) for qos-api 0.2.1
File Interpreter ABI Platform
qos_api-0.2.1-py3-none-any.whl Python 3 none any Details

Total release size: 24.8 kB

Release files / qos_api-0.2.1.tar.gz

Download URL qos_api-0.2.1.tar.gz
Size 12.4 kB
Tags Source
SHA-256 checksum
How to use checksums
3298e586a63e43f34a18249de6ec65cb9101483146b5d7299440b77551374712
BLAKE2b-256 checksum
How to use checksums
be01daed6afbb445635b605cbba64d8018bbe751818cee36ae7f0d5594c07660
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/6.1.0 CPython/3.11.2

Release files / qos_api-0.2.1-py3-none-any.whl

Download URL qos_api-0.2.1-py3-none-any.whl
Size 12.5 kB
Tags Python 3
SHA-256 checksum
How to use checksums
c2441352c3ffffb95bc4b5f999729e6ec8a3c5d4b809e63b3d289932fa3a76a6
BLAKE2b-256 checksum
How to use checksums
056cd9525809545c3af00be1374a507b8b7863f4e68c45558236d8a1f3d77ce4
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/6.1.0 CPython/3.11.2

Security for qos-api 0.2.1

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

No provenance found

None of the files in this release were uploaded with provenance metadata.

A note on package security

PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.

To learn more about protecting yourself from malicious actors, visit our package security documentation.

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page