3 Longest substring without repeating character

看了这题，其实我优点懵逼。。。没啥思路，也不知道第一次代码是参考了哪里写出来里的。。。但是看了code ganker的代码和解释后，豁然开朗，自己按照思路实现一遍就好。。。

第一次代码如下，用一个char[] 来记录是否已经出现过某个字母。。。

public class Solution {
    public int lengthOfLongestSubstring(String s) {
        int maxLen=0;
        int i=0;
        int j=0;
        int tag=0; // tag tells which loop did the program finally exit;
        int[] checklist= new int[256]; // in the book, the variable name is charMap, which is also good 
        while(j<s.length()){    // error: s is a String, should use s.length()
        	Arrays.fill(checklist, -1); // set all elem to my default value -1, means not appeared before, the non-negative value means the position of character: s.charAt(j)
        	for(j=i; j<s.length(); j++){
                if(checklist[s.charAt(j)]==-1){
                    checklist[s.charAt(j)]=j;
                    //System.out.println("checklist[s.charAt(j)]->"+(int)s.charAt(j)+"="+checklist[s.charAt(j)]);
                    tag=0; // tag tells which loop did the program finally exit; 0 means exiting here, the final char is NOT repeated
                }else{
                    maxLen=Math.max(j-i, maxLen);
                    i=checklist[s.charAt(j)]+1;
                    tag=1; // tag tells which loop did the program finally exit; 1 means exiting here, the final char is repeated
                    j++;
                    break;
                } 
                //System.out.println("tag:"+tag);
            }
        }
        // when exit while(), j==s.length-1, need to consider the last substring, two case: last character is repeating char or not
        if(tag==1) maxLen=Math.max(maxLen,j-1-i); 
        else if(tag==0) maxLen=Math.max(maxLen,j-1-i+1);
        return maxLen;
        
    }
}

今天写的代码：具体思路和解释见：http://blog.csdn.net/linhuanmars/article/details/19949159

public class Solution {
    public int lengthOfLongestSubstring(String s) {
        int max=0;
        int left=0;
        int right=0; // left and right are the side of valid window that scans over the substring
        Set<Character> set= new HashSet<>();
        while(right<s.length()){
            char ch=s.charAt(right);
            if(set.contains(ch)){
                max=Math.max(max, right-left);
                while(s.charAt(left)!=ch){ //如果但left==right，说明就是最后两个字母重复了
                    set.remove(s.charAt(left)); 
                    left++;
                }// 以上就是把left扫过的字母都移除了，重复的那个字母其实是不用移除了，因为right已经指中，在下一次while中会用到。
                left++; // ！！！通过上面的while只是找到了repeat char，但是left还得右移一位
            }
            else{
                set.add(ch);
            }
            right++;
        }
        
        return Math.max(max, right-left); // since the right will be s.length(), so do not need to +1 here
    }
}