LintCode 1186: Encode and Decode TinyURL

1186. Encode and Decode TinyURL

TinyURL is a URL shortening service where you enter a URL such as https://lintcode.com/problems/design-tinyurl and it returns a short URL such as http://tinyurl.com/4e9iAk.

Design the encode and decode methods for the TinyURL service. There is no restriction on how your encode/decode algorithm should work. You just need to ensure that a URL can be encoded to a tiny URL and the tiny URL can be decoded to the original URL.

Example
Example 1:

Input：“https://lintcode.com/problems/design-tinyurl”
Output：http://tinyurl.com/4e9iAk
Explanation：encode and decode by your own algorithm.
Example 2:

Input：“https://lintcode.com/problems/solution”
Output：http://tinyurl.com/5d7fiu
Explanation：encode and decode by your own algorithm.

解法1：Base62
为什么要是base62，而不是base256(ASCII码个数)呢？这里问题在于用base256的话，string2Id好转，但是Id/256又id%256的话可能会落在ASCII里面某个非数字非字母的字符。

1. 全局自增id。
2. 用两个map, longUrl2id和id2longUrl，存放id和longUrl的对应关系。
3. shortKey仅仅是对tinyURL的主体部分而言。注意shortKey2Id()和id2ShortKey()必须能够互相编码和解码，不然就对不上了。id2L
4. encode(longUrl)时，先看longUrl是不是在longUrl2Id里面已经有对应id了，如果有id了，直接调用id2ShortKey(id)就可以了，否则id++，并保存longUrl2Id和id2longUrl的关系，然后调用id2ShortKey(id)把shortKey算出来。
5. decode(shortUrl)时，先把shortUrl的主题部分shortKey找出来，然后用shortKey2Id(shortKey)算出id，然后通过id2longUrl把longUrl找出来。
   注意：该解法有个缺点就是当gId变得很大时会整数溢出。可能换了long long会好些，但gId变得足够大的时候还是会溢出。

代码同步更新在
https://github.com/luqian2017/Algorithm

class Solution {
public:
    string encode(string &longUrl) {
       if (longUrl2id.find(longUrl) != longUrl2id.end()) {
           return "http://tiny.url/" + id2ShortKey(longUrl2id[longUrl]);
       }
       gId++;
       longUrl2id[longUrl] = gId;
       id2longUrl[gId] = longUrl;
       cout<<"http://tiny.url/" + id2ShortKey(gId)<<endl;
       return "http://tiny.url/" + id2ShortKey(gId);
    }

    
    string decode(string shortUrl) {
        const string tinyUrlHeader = "http://tiny.url/";
        int pos = tinyUrlHeader.size();

        long long id = shortKey2Id(shortUrl.substr(pos));
        return id2longUrl[id];
    }

private:
    map<string, long long> longUrl2id;
    map<long long, string> id2longUrl;
    
    long long gId = 0;
    long long shortKey2Id(string shortKey) {
        long long id = 0;

        for (int i = 0; i < min(6, (int)shortKey.size()); ++i) {
            if ('a' <= shortKey[i] && shortKey[i] <= 'z') {
                id = id * 62 + shortKey[i] - 'a';
            } else if ('A' <= shortKey[i] && shortKey[i] <= 'Z') {
                id = id * 62 + shortKey[i] - 'A' + 26;
            } else if ('0' <= shortKey[i] && shortKey[i] <= '9') {
                id = id * 62 + shortKey[i] - '0' + 52;
            }
        }
        cout<<"id="<<id<<endl;
        return id;
    }
    
    string id2ShortKey(long long id) {
        string shortKey;
        string charSet = "abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789";
        while(id) {
            shortKey += charSet[id % 62];
            id = id / 62;
        }
        while(shortKey.size() < 6) {
            shortKey += 'a';
        }
        reverse(shortKey.begin(), shortKey.end());
        return shortKey;
    }

};
// Your Codec object will be instantiated and called as such:
// Codec codec = new Codec();
// codec.decode(codec.encode(url));

解法2：
参考了https://www.cnblogs.com/grandyang/p/6562209.html
这个解法我觉得比较好。因为没有用id而是用一个随机数，所以没有溢出问题。

代码如下：

class Solution {
public:
    Solution() {
        dict = "0123456789abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ";
        short2long.clear();
        long2short.clear();
        srand(time(NULL));
    }

    // Encodes a URL to a shortened URL.
    string encode(string longUrl) {
        if (long2short.count(longUrl)) {
            return "http://tinyurl.com/" + long2short[longUrl];
        }
        int idx = 0;
        string randStr;
        for (int i = 0; i < 6; ++i) randStr.push_back(dict[rand() % 62]);
        while (short2long.count(randStr)) {
            randStr[idx] = dict[rand() % 62];
            idx = (idx + 1) % 5;
        }
        short2long[randStr] = longUrl;
        long2short[longUrl] = randStr;
        return "http://tinyurl.com/" + randStr;
    }

    // Decodes a shortened URL to its original URL.
    string decode(string shortUrl) {
        string randStr = shortUrl.substr(shortUrl.find_last_of("/") + 1);
        return short2long.count(randStr) ? short2long[randStr] : shortUrl;
    }
    
private:
    unordered_map<string, string> short2long, long2short;
    string dict;
};