u

                break

    return y_eval

@njit

def evaluate_spline_singlex(x, y, b, c, d, x_eval):

    if xi <= x_eval[0] :

       return y[0]

    if x[-1] <= x_eval :

       return y[-1]

    #

    i=len(x)//2

    while True :

       if x_eval < x[i] :

          i=i//2

          if i==0 :

             return y[0]

       elif x[i+1]<x_eval :

          i=(i+len(x)-1)//2

          if i==len(x)-1 :

             return y[-1]

       else :

          #if x[i] <= x_eval <= x[i+1]: 

          dx = x_eval - x[i]

          return y[i] + b[i]*dx + c[i]*dx**2 + d[i]*dx**3

    return 

@njit

def evaluate_spline_derivative(x, b, c, d, x_eval):

    n = len(x) - 1

        self.y = np.asarray(y)

        self.b, self.c, self.d = _compute_coeffs(self.x, self.y)

    def get_vectors(self) :

        return [self.x,self.y,self.b,self.c,self.d]

    def __call__(self, x_eval):

        x_eval = np.asarray(x_eval)

        return _evaluate(self.x, self.y, self.b, self.c, self.d, x_eval)