def action()

in online_attacks/online_algorithms/stochastic_virtual.py [0:0]

• 25 lines of code
• 12 McCabe index (conditional complexity)

    def action(self, value: float, index: int):
        if self.sampling_phase:
            self.R.append([value, index])
            self.R.sort(key=lambda tup: tup[0], reverse=True)  # sorts in place
            self.R = self.R[: self.k]

            if index >= self.threshold:
                self.sampling_phase = False
        else:
            k_value, k_index = self.R[-1]
            num_picked = len(self.S)
            num_left_to_pick = self.k - num_picked
            num_samples_left = self.N - index
            if num_samples_left <= num_left_to_pick and self.exhaust:
                # Just Pick the last samples to exhaust K
                self.S.append([value, index])
            elif value < k_value:
                # Don't pick or Update R
                pass
            elif value > k_value and k_index <= self.threshold:
                # Update and pick
                self.S.append([value, index])
                self.R.append([value, index])
                self.R.sort(key=lambda tup: tup[0], reverse=True)  # sorts in place
                self.R = self.R[: self.k]
            elif value > k_value and k_index > self.threshold:
                self.R.append([value, index])
                self.R.sort(key=lambda tup: tup[0], reverse=True)  # sorts in place
                self.R = self.R[: self.k]