我最近完成了一个解决模拟魔方的程序,但是,我的解决方案总是很长,因为程序通常会循环多个单层旋转以找到某种颜色。
如果您不知道Rubik的立方体符号是什么,基本上单个字母表示层(U)顺时针旋转90度,带撇号的字母是90度逆时针旋转(U'),并且后跟2的字母是180转(U2)。
我想要做的是在我的解决方案中使用任何[U,U] s并将它们转换为[U2],然后取任何[U,U,U]并使它们成为[U'] s 。基本上将解决方案压缩到更小的尺寸
我将旋转字母作为字符串存储在字符串向量中,所以我正在考虑制作一个这样的循环
for (int i = 0; i < vector.size(); i++)
{
//and then check if there are similarities
if (vector[i] == "U" && vector[i+1] == "U" && vector[i+2] == "U")
{
//Replace the first string with U' and delete the other 2 from the vector
}
else if (vector[i] == "U" && vector[i+1] == "U" && vector[i+2] != "U")
{
//Replace with U2
}
//Etc.
}
我只是想知道是否有人有更优雅的解决方案,或者你是否能看到我的缺陷(我还没有实施)
谢谢!
答案 0 :(得分:1)
这可以通过坚持algorithms library来执行搜索来完成,从而避免担心边界条件。
我编写了一个帮助函数,使用std::search_n
搜索你,找到连续的U
。
template<typename RanIter>
std::pair<RanIter, RanIter> do_search(RanIter first, RanIter last)
{
// first look for 2 consecutive Us
auto it = std::search_n(first, last, 2, "U");
if(it == last) {
// nothing to replace, bail
return std::make_pair(last, last);
}
// check if we have 3 consecutive Us
if(std::distance(it, last) >= 3) {
// there are at least 3 elements in the range, it's ok to check
if(*std::next(it, 2) == "U") {
return std::make_pair(it, std::next(it, 3));
}
}
return std::make_pair(it, std::next(it, 2));
}
辅助函数返回一对迭代器,指示要替换的元素范围。如果该对的第一个元素等于向量的end()
,则不再有要替换的元素。否则,根据对元素之间的距离是2还是3,我们分别用"U2"
或"U'"
替换范围。
while(first != vec.end()) {
auto result = do_search(first, vec.end());
first = result.first;
if(first == vec.end()) {
break;
}
auto dist = std::distance(first, result.second);
if(dist == 3) {
first = vec.insert(first, "U'");
} else {
first = vec.insert(first, "U2");
}
// advance first to the first element to be removed and erase them
std::advance(first, 1);
first = vec.erase(first, std::next(first, dist));
}
答案 1 :(得分:0)
//as pointed out by chris if i = vector.size() - 2 you're in deep trouble
for (int i = 0; i < (vector.size()-2); i++)
{
//and then check if there are similarities
if (vector[i] == "U" && vector[i+1] == "U" && vector[i+2] == "U")
{
//Replace the first string with U' and delete the other 2 from the vector
}
else if (vector[i] == "U" && vector[i+1] == "U" && vector[i+2] != "U")
{
//Replace with U2
}
//Etc.
}
修复了它。
答案 2 :(得分:0)
for (int i = 0; i < vector.size(); i++)
{
//and then check if there are similarities
if (vector[i] == "U" )
{
if ( (i < vector.size() - 1) && (vector[i+1] == "U") )
{
if ( (i < vector.size() - 2) && (vector[i+2] == "U") )
{
// Replace the first string with U'
// and delete the other 2 from the vector
}
else
{
//Replace with U2
}
}
else
{
// Just a plain U
}
}
else if {...}
//Etc.
}
答案 3 :(得分:0)
感谢您的建议pipja,chris和Sahu。我实际上最终做到这一点的方式是这样的:
for (int i = 0; i < solution.size(); i++)
{
if (solution[i] == solution[i+1] && solution[i+1] == solution[i+2]) //Check if there are 3 in a row
{
string temp = solution[i];
//Check if it's legal to compress this (ie it isn't a triplet of [U']s
bool isLegal = true;
for (int i = 0; i < temp.length(); i++)
{
if (temp[i] == '\'' || temp[i] == '2')
{
isLegal = false;
}
}
if (isLegal == false)
{
continue;
}
temp+="'"; //Add an apostrophe
//Replace and delete the others
solution[i] = temp;
solution.erase(solution.begin()+i+1);
solution.erase(solution.begin()+i+2);
}
else if (solution[i] == solution[i+1] && solution[i+1] != solution[i+2]) //Repeat for if it's 2 in a row
{
string temp = solution[i];
bool isLegal = true;
for (int i = 0; i < temp.length(); i++)
{
if (temp[i] == '\'' || temp[i] == '2')
{
isLegal = false;
}
}
if (isLegal == false)
{
continue;
}
temp+="2";
solution[i] = temp;
solution.erase(solution.begin()+i+1);
}
else //This else solves the problem mentioned by Sahu
{
continue;
}
}
我对它很满意,它不是最优雅的解决方案,但我认为它很有效