from datetime import timedelta, datetime
from collections import deque
from QuantConnect.Data.Custom.CBOE import *
from QuantConnect.Securities.Option import OptionStrategies

class MyAlgorithm(QCAlgorithm):
    
    def Initialize(self):
        
        self.SetStartDate(2020, 11, 1)  # Set Start Date
        self.SetEndDate(2020, 12, 31) #Set End Date
        self.SetCash(100000)  # Set Strategy Cash
        
        
        self.vix = self.AddData(CBOE, "VIX", Resolution.Daily).Symbol
        
        # Set number of coarse selectoin
        self.num_coarse_0 = 50
        self.num_coarse_1 = 5
        
        self.CandidateStockPool = {} # Record 5 equities we chosen
        self.Contracts = {} # Store the contracts we want to buy
        # self._Consolidators = {}
        
        self.AddUniverse(self.CoarseSelectionFunction)
        self.UniverseSettings.Resolution = Resolution.Daily
        self.Months = -1
        
        # parameters ------------------------------------------------------------
        self.DTE = 15  # target days till expiration
        self.OTM = 0  # target percentage OTM of put
        self.DaysBeforeExp = 2  # number of days before expiry to exit
        self.NumPosition = 0.1 # The number of capital for each order
        self.SpreadRatio = .01 # Calculate the strike price of option
        self.minPrice = 50 # Initialize the minimum price of stocks 
        # ------------------------------------------------------------------------
        
        self.AddEquity("SPY")
        # Buy credit spread 30 mins after market open
        self.Schedule.On(self.DateRules.EveryDay("SPY"), \
                         self.TimeRules.At(10, 1), \
                         self.BuyBullPutSpread)
        
        # Close call 30 mins before market close                
        self.Schedule.On(self.DateRules.EveryDay("SPY"), \
                         self.TimeRules.At(15, 31), \
                         self.ClosePosition)
                         
        # Plot our current open positions
        self.Schedule.On(self.DateRules.EveryDay(), \
                         self.TimeRules.At(17, 0), \
                         self.Plotting
                         ) 
    


    def FindContract(self, data, symbol):
        if symbol not in self.Contracts:
            if self.OptionFilter(data, symbol) != []:
                self.Contracts[symbol] = self.OptionFilter(data, symbol)
                self.Log("Found contract: {}".format(self.Contracts[symbol][0].ID, self.Contracts[symbol][1].ID))
                
    
        # Liquidate contracts
    def ClosePosition(self):
        for symbol in self.Contracts:
            if self.Contracts[symbol] != []:
                # liquidate if close to expiration date
                if (self.Contracts[symbol][0].ID.Date - self.Time) <= timedelta(self.DaysBeforeExp):
                    self.Log("Closed {}: too close to expiration".format(self.Contracts[symbol]))
                    self.Liquidate(self.Contracts[symbol][0])
                    self.Liquidate(self.Contracts[symbol][1])
                    self.Contracts[symbol] = []
    
    def OnData(self, data):
        
        # Liquidate all positions if VIX>35
        # if self.Securities[self.vix].Price > 35:
        #     self.Log("Current VIX is {}".format(self.Securities[self.vix].Price))
        #     self.Liquidate()
        #     return
            
        for candidate in self.CandidateStockPool.values():
            if candidate.buy:
                self.FindContract(data, candidate.symbol)


    
    # Find satisfied contracts by requirements
    def OptionFilter(self, data, symbol):
        try:
            UnderlyingPrice = data[symbol].Close
        except:
            return []
        
        OptionChain = self.OptionChainProvider.GetOptionContractList(symbol, data.Time)
        
        sort_by_expire = sorted([i for i in OptionChain if i.ID.OptionRight == OptionRight.Put\
                                                      and i.ID.StrikePrice <= UnderlyingPrice*(1+self.SpreadRatio)],\
                                                      key = lambda x: abs((x.ID.Date - self.Time).days - self.DTE), reverse = False)
        num = 0
        
        for x in sort_by_expire:
            num += 1
            expire_date = x.ID.Date
            filtered_contract = sorted([i for i in sort_by_expire if i.ID.Date == expire_date],\
                                                key = lambda x: x.ID.StrikePrice, reverse = True)
            if len(filtered_contract) >= 2:
                short_contract = filtered_contract[0]
                short_strike = short_contract.ID.StrikePrice
                long_strike = min(UnderlyingPrice, short_strike*(1-self.SpreadRatio))
                long_contract = list(filter(lambda x: x.ID.StrikePrice <= long_strike, filtered_contract))[0]
                self.Log("Current contract is {}".format(short_contract, long_contract))
                self.AddOptionContract(short_contract, Resolution.Minute)
                self.AddOptionContract(long_contract, Resolution.Minute)
                return [short_contract, long_contract]
            elif num >= 3:
                return []
        return []
        
    def BuyBullPutSpread(self):
        # If VIX is more than 35, liquidate all positions and return
        if self.Securities[self.vix].Price > 35:
            self.Log("Current VIX is {}, liquidate all positions".format(self.Securities[self.vix].Price))
            self.Liquidate()
            for symbol in self.Contracts:
                self.Contracts[symbol] = []
            return
        
        for symbol in self.Contracts:
            # if contract exists
            self.Log(self.Contracts[symbol])
            if self.Contracts[symbol] != []:
                
                [sell, buy] = self.Contracts[symbol]
                # if the ticker hasnt't been invested
                if not self.Portfolio[sell].Invested and not self.Portfolio[buy].Invested:
                    # self.Buy(OptionStrategies.BullPutSpread(sell.Symbol.Value, sell.ID.StrikePrice, buy.ID.StrikePrice, sell.ID.Date))
                    # self.MarketOrder(sell, -1)
                    # self.MarketOrder(buy, 1)
                    self.Buy(buy, 1)
                    self.Sell(sell, 1)
                    self.Log('Put Credit Spread {} bought'.format(self.Contracts[symbol][0], self.Contracts[symbol][1]))
                    return
        
        return
                
    def OnOrderEvent(self, orderEvent):
        self.Log(str(orderEvent))
        
    def Plotting(self):
        l = list(self.Contracts.keys())
        num = len(l)
        self.Plot('# of Open Positions', 'Positions', num)
        self.Plot('VIX', 'VIX', self.Securities[self.vix].Price)

    def CoarseSelectionFunction(self, coarse):
        
        # Pass data to the indicator for its update
        if self.CandidateStockPool != {}:
            for sec in coarse:
                symbol = sec.Symbol
                
                if symbol in self.CandidateStockPool:
                    input = Input_Class(sec.EndTime, sec.AdjustedPrice)
                    self.CandidateStockPool[symbol].Update(input)
                    
        #1. If it isn't time to update data, return the previous symbols 
        #2. Update self.lastMonth with current month to make sure only process once per month
        if self.Time.month == self.Months:
            return Universe.Unchanged
        self.Months = self.Time.month
        
        # Prefilter the coarse in 300 stocks by DollarVolume
        prefilter = sorted([x for x in coarse if x.DollarVolume > 10000000 and x.Price > self.minPrice], \
                    key = lambda x: x.DollarVolume, reverse = True)[:self.num_coarse_0]
        
        dict = {}
        # Iterate through the coarse and update indicators
        for sec in prefilter:
            dict[sec.Symbol] = SymbolInd(sec.Symbol)
            indicators = dict[sec.Symbol]
            
            try:
                history = self.History(sec.Symbol, 60, Resolution.Daily).loc[sec.Symbol]
            except:
                dict.pop(sec.Symbol)
                continue
            
            for tuple in history.itertuples():
                indicators.Update(tuple)
        
        # Get the stocks in uptrends
        coarse_list = list(filter(lambda x: x.is_uptrend, dict.values()))
        coarse_list = [x for x in coarse_list if x.RSI.IsReady and x.MACD.IsReady]
        
        # Choose final stocks, sorted by price improvement. 
        # Check SymbolInd class for details
        final_stocks = sorted([x for x in coarse_list if (x.rsi <= 70) and \
                                                  (x.macd > 0)], \
                          key = lambda x: x.scale, reverse = True)[:self.num_coarse_1]
                          
        final_symbols = [x.symbol for x in final_stocks]
        
        # Remove the old stocks in the CandidateStockPool list
        for sym in list(self.CandidateStockPool.keys()):
            if sym not in final_symbols:
                self.CandidateStockPool.pop(sym)
                self.RemoveSecurity(sym)
                # self.SubscriptionManager.RemoveConsolidator(sym, self._Consolidators[sym])
                # self._Consolidators.pop(sym)
        
        # Add new candidates to the CandidateStockPool list
        for x in final_stocks:
            if x.symbol not in self.CandidateStockPool:
                equity = self.AddEquity(x.symbol, Resolution.Minute)
                equity.SetDataNormalizationMode(DataNormalizationMode.Raw)
                symbol = equity.Symbol 
                self.CandidateStockPool[symbol] = x
                # consolidator = self.ResolveConsolidator(symbol, Resolution.Daily)
                # self.RegisterSymbolInd(symbol, self.CandidateStockPool[symbol], Resolution.Daily)
                # self._Consolidators[symbol] = consolidator
                
        return Universe.Unchanged

class Input_Class(object):
    def __init__(self, time, close):
        self.EndTime = time
        self.Close = close
        
class SymbolInd:
    """
    Record every indicator for each active security, which is in our dictionary
    self.coarse_dict
    """
    def __init__(self, symbol):
        # Initialize all indicators and parameters
        self.symbol = symbol
        # self.tolerance = 1
        self.MA_10 = SimpleMovingAverage(10)
        self.MA_20 = SimpleMovingAverage(20)
        self.Volatility = StandardDeviation(14)
        self.MACD = MovingAverageConvergenceDivergence(symbol, 12, 26, 9)
        self.RSI = RelativeStrengthIndex(symbol, 14)
        # The period of price improvement
        self.window = RollingWindow[float](30)
        
        self.queue = deque(maxlen = 60)
        self.buy = False
        self.is_uptrend = False
        self.scale = 0 # Quantify the uptrend
        self.difference = None # Diff between current close price and 2 weeks ago
        self.macd = None
        self.rsi = None
        self.open_date = None # Record opening date
        self.open_price = None # Record open price
        
        self.IsReady = False

    def Update(self, input):
        try:
            close = input.Close
            time = input.EndTime
        except:
            close = input.close
            time = input.Index
            
        self.queue.appendleft(close)
        count = len(self.queue)
        self.IsReady = count == self.queue.maxlen
        
        # Update all indicators using current data
        self.window.Add(close)

        if self.window.IsReady:
            # Calculate the price improvements in a specific period
            self.difference = self.window[0] - self.window[self.window.Count - 1]
            self.is_uptrend = self.difference >= 0

        if self.Volatility.Update(time, close):
            self.vol = self.Volatility.Current.Value

            if self.is_uptrend:
                self.scale = self.difference/self.vol

        if self.MA_10.Update(time, close) and self.MA_20.Update(time, close):
            ma_10 = self.MA_10.Current.Value
            ma_20 = self.MA_20.Current.Value

            # Check if it is a pull back as the entry point
            if ma_10 > close:
                self.buy = True
            else:
                self.buy = False

        if self.MACD.Update(time, close):
            self.macd = self.MACD.Signal.Current.Value

        if self.RSI.Update(time, close):
            self.rsi = self.RSI.Current.Value