def solve(a, b, c):
    #what is the maximum possible value of x that guarantees that x+y<C?
    x_val_guarantee_success = c-b
    #and what is the probability of getting that x value?
    p_guaranteed_success = max([min([x_val_guarantee_success,a])/a,0])
    #so it might be that we never get that value, but we could do.
    middle_group_lbound = max([x_val_guarantee_success,0])
    #next; what is the minimum possible value of x that guarantees x+y>=C, failure?
    #that'll be when x>=C even without considering Y
    middle_group_ubound = min([c,a])
    p_guaranteed_failure = max([0,(a-c)/a])
    #now, the probability of sitting somewhere in the middle
    p_contingent_on_y = 1-p_guaranteed_success-p_guaranteed_failure
    #now, the probability of the limits on each side
    p_lower = (c-middle_group_lbound)/b
    #ok and now at the upper limit, i.e., x<C
    #proportion of possible values of y for which x+y<C
    #y_value_needed_at_p_higher<-C-min(x_val_guarantee_failure,A)
    p_higher = (c-middle_group_ubound)/b

    p_total  = p_guaranteed_success+sum([p_lower,p_higher])/2*p_contingent_on_y+p_guaranteed_failure*0


    if(p_total==1):
        retval = "1/1"
    else:
        retval = str(Fraction(p_total).limit_denominator(max_denominator=10000000000))
        
    return(retval)