from datetime import datetime
import talib.abstract as ta                   # TA-Lib 抽象接口（可直接对 DataFrame 使用）
import pandas_ta as pta                       # pandas_ta 指标库，这里用到 cti
from freqtrade.persistence import Trade
from freqtrade.strategy.interface import IStrategy
from pandas import DataFrame
from freqtrade.strategy import DecimalParameter, IntParameter
from functools import reduce                  # 组合多个条件用（逻辑或）

class SampleStrategy1(IStrategy):
    # ROI 表：持仓 0 分钟起要求 2% 盈利即可触发 ROI 卖出；持仓满 120 分钟后只要 1%
    minimal_roi = {
        "0": 0.02,
        "120": 0.01
    }

    timeframe = '5m'                          # 策略运行在 5 分钟线
    process_only_new_candles = True           # 只在新 K 线计算，省 CPU
    startup_candle_count = 120                # 至少准备 120 根 K，用于指标热身
    position_adjustment_enable = True         # 启用“仓位调整”（分批加减仓）

    # 订单类型设定
    order_types = {
        'entry': 'market',                    # 入场用市价
        'exit': 'market',                     # 出场用市价
        'emergency_exit': 'market',
        'force_entry': 'market',
        'force_exit': "market",
        'stoploss': 'market',                 # 止损用市价
        'stoploss_on_exchange': False,        # 不在交易所挂实体止损单（本地管）
        # 下面两项仅在 stoploss_on_exchange=True 时生效，这里相当于预留
        'stoploss_on_exchange_interval': 60,
        'stoploss_on_exchange_market_ratio': 0.99
    }

    stoploss = -0.13                          # 全局硬止损：-13%

    # —— 超参定义（可被 Hyperopt 优化）——
    is_optimize_32 = True
    buy_rsi_fast_32 = IntParameter(20, 70, default=46, space='buy', optimize=is_optimize_32)
    buy_rsi_32 = IntParameter(15, 50, default=19, space='buy', optimize=is_optimize_32)
    buy_sma15_32 = DecimalParameter(0.900, 1, default=0.942, decimals=3, space='buy', optimize=is_optimize_32)
    buy_cti_32 = DecimalParameter(-1, 0, default=-0.86, decimals=2, space='buy', optimize=is_optimize_32)

    # —— 初始下单金额控制：只用建议仓位的 70% 开仓 —— 
    def custom_stake_amount(self, pair: str, current_time: datetime, current_rate: float,
                            proposed_stake: float, min_stake: float, max_stake: float,
                            leverage: float, entry_tag: str, side: str,
                            **kwargs) -> float:
        return proposed_stake * 0.7           # 预留 30% 弹药以便后续加仓

    # —— 仓位调整（分批/加减仓）——
    def adjust_trade_position(self, trade: Trade, current_time: datetime,
                              current_rate: float, current_profit: float,
                              min_stake: float, max_stake: float,
                              current_entry_rate: float, current_exit_rate: float,
                              current_entry_profit: float, current_exit_profit: float,
                              **kwargs) -> float:

        count_of_exits = trade.nr_of_successful_exits
        # 条件1：若当前浮盈 ≥ 2%，且从未分批减仓过 → 减仓 50%
        if current_profit >= 0.02 and count_of_exits == 0:
            return -0.5 * trade.stake_amount  # 负值=减仓金额（按初始 stake 的一半）

        count_of_entries = trade.nr_of_successful_entries
        filled_entries = trade.select_filled_orders(trade.entry_side)  # 历史入场订单
        initial_stake = filled_entries[0].cost                         # 第一次入场的花费（70%）

        # 条件2：若浮亏 < -5% 且仅入场过一次 → 补仓到满仓（把剩余 30% 补上）
        if current_profit < -0.05 and count_of_entries == 1 :
            return initial_stake / 0.7 * 0.3   # 初始 70% 反推总额，再乘 30% 得到补仓额
        return None                             # 其他情况不调整

    # —— 计算技术指标 —— 
    def populate_indicators(self, dataframe: DataFrame, metadata: dict) -> DataFrame:

        dataframe['sma_15'] = ta.SMA(dataframe, timeperiod=15)              # 15 期简单均线
        dataframe['cti'] = pta.cti(dataframe["close"], length=20)           # CTI(20)：相关性趋势指标，负值偏超卖
        dataframe['rsi'] = ta.RSI(dataframe, timeperiod=14)                 # 标准 RSI(14)
        dataframe['rsi_fast'] = ta.RSI(dataframe, timeperiod=4)             # 快 RSI(4)：更敏感
        dataframe['rsi_slow'] = ta.RSI(dataframe, timeperiod=20)            # 慢 RSI(20)：趋势性

        # STOCHF：快速随机指标，这里只用 fastk（0–100）
        stoch_fast = ta.STOCHF(dataframe, 5, 3, 0, 3, 0)                    # 建议改成关键词参数以免歧义
        dataframe['fastk'] = stoch_fast['fastk']

        return dataframe

    # —— 入场条件 —— 
    def populate_entry_trend(self, dataframe: DataFrame, metadata: dict) -> DataFrame:

        conditions = []
        dataframe.loc[:, 'enter_tag'] = ''     # 记录触发的买入标签

        # buy_1 组合信号：
        buy_1 = (
                (dataframe['rsi_slow'] < dataframe['rsi_slow'].shift(1)) &      # 慢 RSI 走低（捕捉回撤）
                (dataframe['rsi_fast'] < self.buy_rsi_fast_32.value) &          # 快 RSI 低于阈值（短期超卖）
                (dataframe['rsi'] > self.buy_rsi_32.value) &                    # 但整体 RSI 不能太弱
                (dataframe['close'] < dataframe['sma_15'] * self.buy_sma15_32.value) &  # 价格低于 SMA15 的某比例
                (dataframe['cti'] < self.buy_cti_32.value)                      # CTI 足够负，倾向反弹
        )

        conditions.append(buy_1)
        dataframe.loc[buy_1, 'enter_tag'] += 'buy_1'  # 打标签方便回测分析

        # 任一买入条件满足 → 标记 enter_long = 1（Freqtrade 会据此下单）
        if conditions:
            dataframe.loc[
                reduce(lambda x, y: x | y, conditions),
                'enter_long'] = 1

        return dataframe

    # —— 出场条件 —— 
    def populate_exit_trend(self, dataframe: DataFrame, metadata: dict) -> DataFrame:
#         dataframe.loc[(), ['exit_long', 'exit_tag']] = (0, 'long_out')  # 示例：可自定义出场信号
        dataframe['exit_long'] = 0
        dataframe['exit_tag'] = ''            # 这里不主动给卖出信号
                                             # 实际卖出由 ROI 表 / 全局止损 / adjust_trade_position 决定
        return dataframe