You are given a string, s, and a list of words, words, that are all of the same length. Find all starting indices of substring(s) in s that is a concatenation of each word in words exactly once and without any intervening characters.
For example, given:
s: "barfoothefoobarman"
words: ["foo", "bar"]
You should return the indices: [0,9]
.
(order does not matter).
class Solution {
public:
vector<int> findSubstring(string s, vector<string>& words) {
unordered_map<string ,int> ump;
vector<int> res;
for(int i = 0; i < words.size(); i++)
ump[words[i]]++;
int s_size = s.size();
int words_size = words.size();
int word_size = words[0].size();
for(int i = 0; i < s_size - words_size * word_size + 1; i++){
unordered_map<string, int> book;
int j = 0;
for(; j < words_size; j++){
string tmp = s.substr(i + j * word_size, word_size);
if(ump.find(tmp) == ump.end())
break;
else{
book[tmp]++;
if(book[tmp] > ump[tmp])
break;
}
}
if(j == words_size)
res.push_back(i);
}
return res;
}
};