def read_data(): H, W = map(int, input().split()) s = [input() for i in range(H)] return H, W, s def parse_board(H, W, s): reds = [False] * (H * W) start = -1 goal = -1 for h in range(H): sh = s[h] for w in range(W): c = sh[w] if c == 'S': start = h * W + w elif c == 'G': goal = h * W + w elif c == 'R': reds[h * W + w] = True return reds, start, goal def solve(H, W, s): next_move = [(), ()] next_move[0] = ((-2, -1), (-2, 1), (-1, -2), (-1, 2), (1, -2), (1, 2), (2, -1), (2, 1)) next_move[1] = ((-1, -1), (-1, 1), (1, -1), (1, 1)) reds, start, goal = parse_board(H, W, s) return wfs(start, goal, reds, H, W, next_move) def wfs(start, goal, reds, H, W, next_move): HW = H * W dist = [-1] * (HW * 2) goals = [goal, goal + HW] steps = 0 state = 0 frontiers = [start] dist[start] = 0 while frontiers: steps += 1 new_frontiers = [] for spos in frontiers: state, pos = divmod(spos, HW) h, w = divmod(pos, W) for dh, dw in next_move[state]: nh = h + dh nw = w + dw if 0 <= nh < H and 0 <= nw < W: npos = nh * W + nw new_state = 1 - state if reds[npos] else state nspos = new_state * HW + npos if dist[nspos] == -1: dist[nspos] = steps new_frontiers.append(nspos) if nspos in goals: return steps frontiers = new_frontiers return -1 if __name__ == '__main__': pars = read_data() print(solve(*pars))