Code & Func

第23天。

Given two strings text1 and text2, return the length of their longest common subsequence.

A subsequence of a string is a new string generated from the original string with some characters(can be none) deleted without changing the relative order of the remaining characters. (eg, “ace” is a subsequence of “abcde” while “aec” is not). A common subsequence of two strings is a subsequence that is common to both strings.

If there is no common subsequence, return 0.

Example 1:

1
Input: text1 = "abcde", text2 = "ace"
2
Output: 3
3
Explanation: The longest common subsequence is "ace" and its length is 3.

Example 2:

1
Input: text1 = "abc", text2 = "abc"
2
Output: 3
3
Explanation: The longest common subsequence is "abc" and its length is 3.

Example 3:

1
Input: text1 = "abc", text2 = "def"
2
Output: 0
3
Explanation: There is no such common subsequence, so the result is 0.

Constraints:

1 <= text1.length <= 1000
1 <= text2.length <= 1000
The input strings consist of lowercase English characters only.

这是一道比较经典的动态规划问题吧，它的动规方程为： $$ \begin{aligned} LCS(i,j) = \left{ \begin{array}{rcl}

& LCS[i-1, j-1] + 1 & ,{s1[i] = s2[j]} \ & max(LCS[i, j-1], LCS[i-1, j]) & ,{s1[i] \neq s2[j]}

\end{array} \right. \end{aligned} $$ 根据动规方程我们可以写出如下代码：

1
int longestCommonSubsequence(string text1, string text2) {
2
    vector<vector<int>> dp(text1.size() + 1, vector<int>(text2.size() + 1, 0));
3
    for(int i = 1;i < dp.size(); i++) {
4
        for(int j = 1;j < dp[0].size(); j++) {
5
            if (text1[i-1] == text2[j-1]) {
6
                dp[i][j] = dp[i-1][j-1] + 1;
7
            } else {
8
                dp[i][j] = max(dp[i-1][j], dp[i][j-1]);
9
            }
10
        }
11
    }
12
    return dp[text1.size()][text2.size()];
13
}

这里的空间复杂度可以继续进行优化，因为LCS[i,j]只与当前行和上一行有关系，所以可以优化成两个数组来做：

1
int longestCommonSubsequence(string text1, string text2) {
2

3
    int n = text1.size() + 1, m = text2.size() + 1;
4
    vector<int> dp1(m, 0);
5
    vector<int> dp2(m);
6

7
    for(int i = 1;i < n; i++) {
8
        dp1[0] = 0;
9
        for(int j = 1;j < m; j++) {
10
            if (text1[i-1] == text2[j-1]) {
11
                dp2[j] = dp1[j-1] + 1;
12
            } else {
13
                dp2[j] = max(dp1[j], dp2[j-1]);
14
            }
15
        }
4 collapsed lines
16
        swap(dp1, dp2);
17
    }
18
    return dp1[m-1];
19
}

再进一步的话，我们可以发现dp[i][j]只与 dp[i-1][j-1]，dp[i-1][j]，dp[i][j-1] 相关，如果我们只用一个数组的话，dp[i][j]与dp[i-1][j]其实存在同一个位置，而dp[i][j-1]是在同一行，所以我们只需要维护一个prev变量来保存dp[i-1][j-1]的值即可：

1
int longestCommonSubsequence(string text1, string text2) {
2
    int n = text1.size(), m = text2.size() + 1;
3
    vector<int> dp(m, 0);
4

5
    for(int i = 0;i < n; i++) {
6
        int prev = 0;
7
        for(int j = 1;j < m; j++) {
8
            int temp = prev;
9
            prev = dp[j];
10
            dp[j] = (text1[i] == text2[j-1]) ? (temp + 1) : (max(dp[j], dp[j-1]));
11
        }
12
    }
13
    return dp[m-1];
14
}