原题:
Given a
pattern
and a stringstr
, find ifstr
follows the same pattern.Here follow means a full match, such that there is a bijection between a letter in
pattern
and a non-empty word instr
.Example 1:
Input: pattern = "abba", str = "dog cat cat dog" Output: trueExample 2:
Input:pattern = "abba", str = "dog cat cat fish" Output: falseExample 3:
Input: pattern = "aaaa", str = "dog cat cat dog" Output: falseExample 4:
Input: pattern = "abba", str = "dog dog dog dog" Output: falseNotes:
You may assumepattern
contains only lowercase letters, andstr
contains lowercase letters that may be separated by a single space.
刚看完这道题我是懵逼的,想了想我还是懵逼的,写完代码测试时还是懵逼的,这真是简单题吗?先说下思路,为了能让类似abba这种pattern中相同的元素相同并且可以根据目标string更改ab的值,我打算用string指针vector来存储这种关系。首先建立查找用的string指针数组,因为题中说了限小写字母,所以只要26个就够了,可以挨个申请地址空间也可以像我一样再建立26个string的实空间然后依次指向实空间。之后就可以向vector中push了,比如abba我就依次push:patt[a-97], patt[b-97]...。之后就是读入目标串,在处理第n个单词时,如果vector第n个元素是空串就赋值为当前单词并n++,如果不为空则与当前单词比较,相同则n++,不同则返回false。在调试时发现如果单词长于pattern会溢出,于是每次循环都要比较一下n和vector的size,后来发现如果pattern多于单词数会出现假true的问题,所以循环后又加了一个n和vector的size的比较。最后一个发现的问题就是如果abba中a和b是相同的单词,比如“dog dog dog dog”也会出现假true的现象,所以最后用hash map比较了一下26个真string的值,如果有相同的会返回false。
比较啰嗦但是速度很好,不过存储因为string数组的关系还可以在降低:
Success
Runtime: 0 ms, faster than 100.00% of C++ online submissions for Word Pattern.
Memory Usage: 8.5 MB, less than 84.50% of C++ online submissions for Word Pattern.
代码:
class Solution {
public:
bool wordPattern(string pattern, string str) {
vector<string*> lib;
string* patt[26]={nullptr};
string realpatt[26]={""};
for(int i=0;i<26;i++){
patt[i]=&realpatt[i];
}
for(int i=0;i<pattern.size();i++){
lib.push_back(patt[pattern[i]-97]);
}
int n=0;
for(int i=0;i<str.size();i++){
if(n>=lib.size()){return false;}
string s="";
while(str[i]!=' '&&i<str.size()){
s+=str[i];
i++;
}
if(*lib[n]==""){*lib[n]=s;n++;continue;}
else if(*lib[n]==s){n++;continue;}
else {return false;}
}
if(n!=lib.size()){return false;}
map<string,int> lastcheck;
for(int i=0;i<26;i++){
if(realpatt[i]==""){continue;}
else{
if(lastcheck.count(realpatt[i])){return false;}
else{lastcheck.insert(map<string, int>::value_type (realpatt[i], i));}
}
}
return true;
}
};
之后看了看submission中的一些结果,感觉双map也能解决这个问题而且代码更简洁,但是memory似乎会占用更多。这里贴一下结果和代码:
Success
Runtime: 4 ms, faster than 78.40% of C++ online submissions for Word Pattern.
Memory Usage: 8.9 MB, less than 10.62% of C++ online submissions for Word Pattern.
class Solution {
public:
bool wordPattern(string pattern, string str) {
map<char, int> p2i;
map<string, int> w2i;
istringstream in(str);
int i = 0, n = pattern.size();
for (string word; in >> word; ++i) {
if (i == n || p2i[pattern[i]] != w2i[word])
return false;
p2i[pattern[i]] = w2i[word] = i + 1;
}
return i == n;
}
};
代码原地址:
https://leetcode.com/problems/word-pattern/discuss/73409/Short-C%2B%2B-read-words-on-the-fly