arbitrage/price_checker.py

import requests
import decimal
import time
import threading
import json
from flask import Flask, render_template, jsonify
from collections import deque  # 用于高效地存储固定数量的历史数据

# --- 配置部分 (与之前相同) ---
GATEIO_SPOT_PAIR = 'MUBARAK_USDT'
IN_TOKEN_ADDRESS_BSC = '0x55d398326f99059ff775485246999027b3197955'
OUT_TOKEN_ADDRESS_BSC = '0x5C85D6C6825aB4032337F11Ee92a72DF936b46F6'
AMOUNT_TO_QUERY_HUMAN = decimal.Decimal('1000')
PROXY_HOST = '127.0.0.1'
PROXY_PORT = '7890'
proxies = {
    'http': f'http://{PROXY_HOST}:{PROXY_PORT}',
    'https': f'http://{PROXY_HOST}:{PROXY_PORT}',
}
# proxies = None # 如果你不需要代理

decimal.getcontext().prec = 36


# --- 价格获取函数 (与之前相同, 省略以保持简洁) ---
def get_openocean_price_bsc(in_token_addr, out_token_addr, human_amount_in_decimal_for_request, gas_price=3):
    # ... (代码与之前相同)
    chain = 'bsc'
    url = f'https://open-api.openocean.finance/v4/{chain}/quote'
    params = {
        'inTokenAddress': in_token_addr,
        'outTokenAddress': out_token_addr,
        'amount': str(human_amount_in_decimal_for_request),
        'gasPrice': gas_price,
    }
    try:
        response = requests.get(url, params=params, proxies=proxies, timeout=10)
        response.raise_for_status()
        data = response.json()

        if data.get('code') == 200 and data.get('data'):
            api_data = data['data']
            if not (api_data.get('inToken') and api_data['inToken'].get('decimals') is not None and
                    api_data.get('outToken') and api_data['outToken'].get('decimals') is not None and
                    api_data.get('inAmount') is not None and api_data.get('outAmount') is not None):
                return {"error": "OO API响应中缺少必要的数据"}

            in_token_decimals = int(api_data['inToken']['decimals'])
            out_token_decimals = int(api_data['outToken']['decimals'])
            atomic_in_amount = decimal.Decimal(api_data['inAmount'])
            atomic_out_amount = decimal.Decimal(api_data['outAmount'])

            derived_human_in_amount = atomic_in_amount / (decimal.Decimal('10') ** in_token_decimals)
            derived_human_out_amount = atomic_out_amount / (decimal.Decimal('10') ** out_token_decimals)

            if derived_human_out_amount == 0:
                return {"error": "OO 计算的输出数量为零"}

            calculated_price = derived_human_in_amount / derived_human_out_amount
            return {"price_in_per_out": calculated_price}
        else:
            error_message = data.get('message', 'N/A') if data else '无响应数据'
            error_code = data.get('code', 'N/A') if data else 'N/A'
            return {"error": f"OO API 错误 - Code: {error_code}, Msg: {error_message}"}
    except requests.exceptions.RequestException as e:
        return {"error": f"OO请求失败: {e}"}
    except Exception as e:
        return {"error": f"OO意外错误: {e}"}


def get_gateio_spot_price(pair_symbol):
    # ... (代码与之前相同)
    url = f'https://api.gateio.ws/api/v4/spot/tickers'
    params = {'currency_pair': pair_symbol}
    try:
        response = requests.get(url, params=params, proxies=proxies, timeout=10)
        response.raise_for_status()
        data = response.json()
        if isinstance(data, list) and len(data) > 0:
            ticker_data = data[0]
            if ticker_data.get('currency_pair') == pair_symbol:
                last_price_str = ticker_data.get('last')
                if last_price_str:
                    return {"price_base_in_quote": decimal.Decimal(last_price_str)}
                else:
                    return {"error": "Gate未找到last price"}
            else:
                return {"error": f"Gate交易对不匹配"}
        else:
            return {"error": "Gate API数据格式错误"}
    except requests.exceptions.RequestException as e:
        return {"error": f"Gate请求失败: {e}"}
    except Exception as e:
        return {"error": f"Gate意外错误: {e}"}


app = Flask(__name__)

# --- 全局数据和历史记录 ---
MAX_HISTORY_POINTS = 60  # 图表上显示的最大数据点数量 (例如，过去5分钟，每5秒一个点)
latest_data = {
    "oo_price": None,
    "gate_price": None,
    "difference_percentage": None,
    "oo_error": None,
    "gate_error": None,
    "last_updated": None
}
# 使用 deque 来存储历史数据，它在添加和删除元素时具有 O(1) 的复杂度
historical_prices = deque(maxlen=MAX_HISTORY_POINTS)  # 存储 {'timestamp': ts, 'oo': price, 'gate': price}
historical_diff = deque(maxlen=MAX_HISTORY_POINTS)  # 存储 {'timestamp': ts, 'diff': percentage}

data_lock = threading.Lock()


def update_prices_periodically():
    global latest_data, historical_prices, historical_diff
    while True:
        fetch_timestamp_str = time.strftime("%H:%M:%S")  # 用于图表标签的简单时间戳

        oo_price_data = get_openocean_price_bsc(
            IN_TOKEN_ADDRESS_BSC,
            OUT_TOKEN_ADDRESS_BSC,
            AMOUNT_TO_QUERY_HUMAN
        )
        gate_price_data = get_gateio_spot_price(GATEIO_SPOT_PAIR)

        current_oo_price_decimal = None
        current_gate_price_decimal = None
        current_oo_error = None
        current_gate_error = None
        diff_percentage_val = None
        diff_percentage_str = "N/A"

        if "error" not in oo_price_data:
            current_oo_price_decimal = oo_price_data['price_in_per_out']
        else:
            current_oo_error = oo_price_data['error']

        if "error" not in gate_price_data:
            current_gate_price_decimal = gate_price_data['price_base_in_quote']
        else:
            current_gate_error = gate_price_data['error']

        # 只有当两个价格都有效时，才计算价差并记录历史
        if current_oo_price_decimal is not None and current_gate_price_decimal is not None:
            # 将Decimal转换为浮点数以便存储和图表绘制 (Chart.js 通常使用JS number)
            oo_price_float = float(current_oo_price_decimal)
            gate_price_float = float(current_gate_price_decimal)

            historical_prices.append({
                "timestamp": fetch_timestamp_str,
                "oo": oo_price_float,
                "gate": gate_price_float
            })

            if current_gate_price_decimal > 0:
                difference = current_oo_price_decimal - current_gate_price_decimal
                percentage_diff_decimal = (difference / current_gate_price_decimal) * 100
                diff_percentage_val = float(percentage_diff_decimal)  # 用于图表
                diff_percentage_str = f"{percentage_diff_decimal:+.4f}%"  # 用于显示
                historical_diff.append({
                    "timestamp": fetch_timestamp_str,
                    "diff": diff_percentage_val
                })
            else:
                diff_percentage_str = "Gate价格为0"
                # 对于价差图表，如果gate价格为0，可以记录一个特殊值或跳过
                historical_diff.append({
                    "timestamp": fetch_timestamp_str,
                    "diff": None  # 或者一个非常大/小的值来表示无效
                })
        else:
            # 如果其中一个价格无效，我们仍然可以尝试记录有效的那个，或者都标记为None
            # 这里我们选择如果任一价格无效，价格历史点也记录None，价差历史点也记录None
            historical_prices.append({
                "timestamp": fetch_timestamp_str,
                "oo": float(current_oo_price_decimal) if current_oo_price_decimal else None,
                "gate": float(current_gate_price_decimal) if current_gate_price_decimal else None
            })
            historical_diff.append({
                "timestamp": fetch_timestamp_str,
                "diff": None
            })

        with data_lock:
            latest_data["oo_price"] = str(current_oo_price_decimal) if current_oo_price_decimal is not None else None
            latest_data["gate_price"] = str(
                current_gate_price_decimal) if current_gate_price_decimal is not None else None
            latest_data["difference_percentage"] = diff_percentage_str
            latest_data["oo_error"] = current_oo_error
            latest_data["gate_error"] = current_gate_error
            latest_data["last_updated"] = time.strftime("%Y-%m-%d %H:%M:%S")

        log_oo = f"{current_oo_price_decimal:.6f}" if current_oo_price_decimal else "N/A"
        log_gate = f"{current_gate_price_decimal:.6f}" if current_gate_price_decimal else "N/A"
        print(
            f"Data updated at {fetch_timestamp_str}: OO: {log_oo}, Gate: {log_gate}, Diff: {diff_percentage_str}, Errors: OO: {current_oo_error}, Gate: {current_gate_error}")

        time.sleep(5)  # 更新频率


@app.route('/')
def index():
    return render_template('index.html')


@app.route('/data')
def get_data():
    with data_lock:
        # 准备图表需要的数据
        # 对于价格图表
        price_chart_labels = [item['timestamp'] for item in historical_prices]
        oo_price_values = [item['oo'] for item in historical_prices]
        gate_price_values = [item['gate'] for item in historical_prices]

        # 对于价差图表
        diff_chart_labels = [item['timestamp'] for item in historical_diff]
        diff_values = [item['diff'] for item in historical_diff]

        # 将最新的数据点加入 (如果历史记录已满，deque会自动移除旧的)
        # 这里的数据是 latest_data 和 historical_data 的组合
        # Chart.js需要完整的历史数据来初始化，然后可以增量更新
        # 或者，前端每次都获取完整的历史窗口

        # 为了简单起见，前端将定期获取完整的 /data，其中包含历史和当前
        response_data = {
            "current": latest_data,
            "history": {
                "prices": {
                    "labels": list(price_chart_labels),  # deque 不是直接 JSON 可序列化的
                    "oo": list(oo_price_values),
                    "gate": list(gate_price_values)
                },
                "difference": {
                    "labels": list(diff_chart_labels),
                    "values": list(diff_values)
                }
            }
        }
        return jsonify(response_data)


if __name__ == "__main__":
    price_updater_thread = threading.Thread(target=update_prices_periodically, daemon=True)
    price_updater_thread.start()
    app.run(debug=True, host='0.0.0.0', port=5000, use_reloader=False)