C++ 中字符串和字符数组声明的时间复杂度有什么区别？

Question

In the question link , I first tried it using a character array. This solution exceeded the time limit.

    char s[100000];
    cin >> s;
    int cnt[26];
    for(int i=0;i<26;++i)
        cnt[i]=0;
    for(int i=0;i<strlen(s);++i)
        cnt[s[i]-'a']++;
    int count =0;
    for(int i=0;i<26;++i)
        if(cnt[i]>0)
            count++;
    cout << count << endl;

But then I changed the above code to this:

    string s;
    cin >> s;
    int cnt[26];
    for(int i=0;i<26;++i)
        cnt[i]=0;
    for(int i=0;i<s.length();++i)
        cnt[s.at(i)-'a']++;
    int count =0;
    for(int i=0;i<26;++i)
        if(cnt[i]>0)
            count++;
    cout << count << endl;

And the code passed with ease. And also the first one exceeded the time limit of 1 sec, while the second one passed with time of execution of 0.04 sec. Why is there such a large difference in time of execution?

Answer 1

std::string stores its length separately, so s.size() is an instant operation.

for(int i=0;i<strlen(s);++i) calls strlen(s) on every iteration, and strlen loops through the entire string to the end to find the length of the string. So this innocent looking loop is actually O(n ² ) complexity.

Answer 2

The C strings do not store length explicily, they just have a null atbthe end. So to work out the length you need to iterate over the entire string character by character. 'sid::string::length' on the otherhand just returns the internally stored length.

You can easily speed up the C version though, by caching the length in a variable before the loop and using the cached length in the for statement.