技术问询：如何聚合OrderBook档位并调整价格间隔至0.50

核心逻辑说明 #

• 原tick间隔为0.05，目标间隔0.5，意味着每10个连续的原始档位需要合并为1个新档位
• 合并规则：
  • 价格：将原始档位价格向下取整到最近的0.5倍数（例如1800.25→1800.0，1800.6→1800.5）
  • 数量：同一价格区间内所有档位的挂单数量累加
• 注意：Bybit WebSocket推送的是增量更新数据，所以必须先维护一份完整的本地订单簿快照，再基于增量更新本地数据，最后对本地订单簿执行合并操作

修改后的完整代码 #

from time import sleep
from pybit import usdt_perpetual

# 初始化本地订单簿，存储最新的买盘和卖盘原始数据
local_orderbook = {
    "b": {},  # 买盘：key为原始价格字符串，value为挂单数量
    "a": {}   # 卖盘：key为原始价格字符串，value为挂单数量
}
TARGET_TICK = 0.5  # 目标档位间隔
ORIGINAL_TICK = 0.05  # 平台原始tick间隔

def round_price(price):
    """将价格向下取整到最近的TARGET_TICK倍数"""
    price_float = float(price)
    return round(price_float - (price_float % TARGET_TICK), 1)

def merge_orderbook(side_data):
    """合并指定方向（买/卖）的订单簿档位"""
    merged = {}
    for price_str, qty in side_data.items():
        rounded_price = round_price(price_str)
        qty_float = float(qty)
        
        if rounded_price not in merged:
            merged[rounded_price] = 0.0
        merged[rounded_price] += qty_float
        
        # 若累加后数量为0，移除该档位（对应增量更新中的撤单）
        if merged[rounded_price] <= 0:
            del merged[rounded_price]
    
    # 按订单簿规则排序：买盘降序，卖盘升序
    is_bid = "b" in side_data
    sorted_merged = dict(sorted(merged.items(), reverse=is_bid))
    return sorted_merged

def handle_message(msg):
    global local_orderbook
    data = msg.get("data", [])
    if not data:
        return
    
    for item in data:
        action = item.get("action")
        bids = item.get("b", [])
        asks = item.get("a", [])
        
        if action == "snapshot":
            # 快照数据直接覆盖本地订单簿
            local_orderbook["b"] = {price: qty for price, qty, _ in bids}
            local_orderbook["a"] = {price: qty for price, qty, _ in asks}
        elif action == "update":
            # 增量更新：更新或删除本地档位
            for price, qty, _ in bids:
                if float(qty) > 0:
                    local_orderbook["b"][price] = qty
                else:
                    local_orderbook["b"].pop(price, None)
            for price, qty, _ in asks:
                if float(qty) > 0:
                    local_orderbook["a"][price] = qty
                else:
                    local_orderbook["a"].pop(price, None)
    
    # 合并档位并输出结果
    merged_bids = merge_orderbook(local_orderbook["b"])
    merged_asks = merge_orderbook(local_orderbook["a"])
    print("合并后买盘（0.5间隔）：", merged_bids)
    print("合并后卖盘（0.5间隔）：", merged_asks)

ws_linear = usdt_perpetual.WebSocket(
    test=False,
    ping_interval=30,
    ping_timeout=10,
    domain="bybit"
)

ws_linear.orderbook_25_stream(
    handle_message, "ETHUSDT"
)

while True:
    sleep(1)

关键代码说明 #

• 本地订单簿维护：用local_orderbook字典实时同步Bybit的订单簿状态，确保增量更新能准确反映最新挂单情况
• 价格取整函数：round_price保证所有原始档位被映射到目标间隔的基准价格上，实现档位归并
• 合并函数：merge_orderbook完成数量累加、无效档位清理和排序，输出符合交易习惯的合并后订单簿
• 消息处理：区分快照和增量更新，快照用于初始化本地数据，增量用于实时更新，避免数据不同步

内容的提问来源于stack exchange，提问作者xabe