How do you buy and sell the roll with RollOver Feed

[adamb032]

hey - ive got the basic RollOver feed example (https://www.backtrader.com/docu/data-rollover/rolling-futures-over/) working as expected and I can verify the rolls are occurring when i want.

however I'd like to simulate actually trading the roll, assuming its a long only model, so in my next() function

def next(self):
        txt = list()
        txt.append('{}'.format(self.data0._name))
        # Internal knowledge ... current expiration in use is in _d
        txt.append('{}'.format(self.data0._d._name))
        txt.append('{}'.format(self.data.datetime.date()))
        txt.append('{}'.format(self.data.datetime.date().strftime('%a')))
        txt.append('{}'.format(self.data.open[0]))
        txt.append('{}'.format(self.data.high[0]))
        txt.append('{}'.format(self.data.low[0]))
        txt.append('{}'.format(self.data.close[0]))
        self.logger.info(', '.join(txt))

        if not self.position:
            self.logger.info(f"{self.curr_future} :: [{self.data.datetime.date(0)}] SUBMITTED Buy MOC T+1 ref_px:{self.data.close[0]:.2f}")
            self.buy()
        else:
            '''we will need to roll the position so we check if tomorrow is roll date'''
            if self._check_is_roll_date(self.data.datetime.date(), 
                                         self.curr_future):
        
                self.logger.info(f"[{self.data.datetime.date(0)}] I SHOULD ROLL HERE!")
                self.sell()
                self.buy()

so initially i buy on next bar data.close[0] to establish the initial long position

the hiccup i get is on the roll (selling near, buying far). when i issue the self.sell() order its going to look at data.close[0]. however because of the way RollOver works, this will be the close of the next maturity in sequence. thats fine for the buy order, however the sell i need to sell from the previous maturity, but I don't see anywhere this is referenced in the RollOver class.

Since RollOver is not backadjusting the prices but just mechanically creating a dataframe of on a given date, what is my active maturity close, it makes simulating the roll trade hard to understand.