# --- Do not remove these libs --- freqtrade backtesting --strategy SmoothScalp --timerange 20210110-20210410
from freqtrade.strategy.interface import IStrategy
from typing import Dict, List
from functools import reduce
from pandas import DataFrame
# --------------------------------
import talib.abstract as ta
import freqtrade.vendor.qtpylib.indicators as qtpylib
from typing import Dict, List
from functools import reduce
from pandas import DataFrame, DatetimeIndex, merge
# --------------------------------
import talib.abstract as ta
import freqtrade.vendor.qtpylib.indicators as qtpylib
import numpy  # noqa
' use 15 open trade, unlimited stake. \n\npairlist setting:\n\n"pairlists": [\n        {\n            "method": "VolumePairList",\n            "number_assets": 50,\n            "sort_key": "quoteVolume",\n            "refresh_period": 1800\n        }\n    ],\n\n'

class HansenSmaOffsetV1(IStrategy):
    INTERFACE_VERSION = 3
    timeframe = '15m'
    #I haven't found the optimal ROI yet
    minimal_roi = {'0': 10}
    stoploss = -99

    def populate_indicators(self, dataframe: DataFrame, metadata: dict) -> DataFrame:
        dataframe['smau1'] = ta.SMA(dataframe['close'], timeperiod=20) + 0.05 * ta.SMA(dataframe['close'], timeperiod=20)
        dataframe['smad1'] = ta.SMA(dataframe['close'], timeperiod=20) - 0.05 * ta.SMA(dataframe['close'], timeperiod=20)
        dataframe['hclose'] = (dataframe['open'] + dataframe['high'] + dataframe['low'] + dataframe['close']) / 4
        dataframe['hopen'] = (dataframe['open'].shift(2) + dataframe['close'].shift(2)) / 2
        dataframe['hhigh'] = dataframe[['open', 'close', 'high']].max(axis=1)
        dataframe['hlow'] = dataframe[['open', 'close', 'low']].min(axis=1)
        dataframe['emac'] = ta.SMA(dataframe['hclose'], timeperiod=6)
        dataframe['emao'] = ta.SMA(dataframe['hopen'], timeperiod=6)
        return dataframe

    def populate_entry_trend(self, dataframe: DataFrame, metadata: dict) -> DataFrame:
        dataframe.loc[(dataframe['high'] < dataframe['smad1']) & (dataframe['hopen'] < dataframe['hclose']), 'entry'] = 1
        return dataframe

    def populate_exit_trend(self, dataframe: DataFrame, metadata: dict) -> DataFrame:
        dataframe.loc[(dataframe['low'] > dataframe['smau1']) & (dataframe['hopen'] > dataframe['hclose']), 'exit'] = 1
        return dataframe