C++ qsort() 函數調用時實參與形參不兼容的問題解決

《劍指OFFER》刷題筆記 —— 撲克牌順子

LL今天心情特別好,因為他去買了一副撲克牌,發現里面居然有2個大王,2個小王(一副牌原本是54張^_^)...他隨機從中抽出了5張牌,想測測自己的手氣,看看能不能抽到順子,如果抽到的話,他決定去買體育彩票,嘿嘿！！“紅心A,黑桃3,小王,大王,方片5”,“Oh My God!”不是順子.....LL不高興了,他想了想,決定大\小 王可以看成任何數字,並且A看作1,J為11,Q為12,K為13。上面的5張牌就可以變成“1,2,3,4,5”(大小王分別看作2和4),“So Lucky!”。LL決定去買體育彩票啦。 現在,要求你使用這幅牌模擬上面的過程,然后告訴我們LL的運氣如何， 如果牌能組成順子就輸出true，否則就輸出false。為了方便起見,你可以認為大小王是0。

 

ERROR:  VS 的錯誤提示 —— int (Solution::*) (const void* a, const void* b) 類型的實參與 int (_cdecl::*) (const void* , const void* ) 類型的形參不兼容

class Solution {
public: 
　　bool IsContinuous(vector<int> numbers) {
        if (numbers.size() != 5)
            return false;
        //qsort(&numbers, numbers.size(), sizeof(int), compare);
        qsort(&numbers[0], numbers.size(), sizeof(int), compare);
        //sort(numbers.begin(), numbers.end());
        
        int numZeros = 0;
        int diff = 0;
        for (int i = 0; i < numbers.size() - 1; ++i) {
            if (numbers[i] == 0)
                ++ numZeros;
            else {
                if (numbers[i] == numbers[i + 1])
                    return false;
                diff += numbers[i + 1] - numbers[i] - 1;
            }
        }
        return (numZeros < diff) ? false : true;
    }
　　
　 int compare(const void *a, const void *b) {
　　  return (*(int*)a - *(int*)b);
　 }

};

 

CORRECT:

int compare(const void *a, const void *b) {
　　return (*(int*)a - *(int*)b);
}

class Solution {
public: 
　　bool IsContinuous(vector<int> numbers) {
        if (numbers.size() != 5)
            return false;
        //qsort(&numbers, numbers.size(), sizeof(int), compare);
        qsort(&numbers[0], numbers.size(), sizeof(int), compare);
        //sort(numbers.begin(), numbers.end());
        
        int numZeros = 0;
        int diff = 0;
        for (int i = 0; i < numbers.size() - 1; ++i) {
            if (numbers[i] == 0)
                ++ numZeros;
            else {
                if (numbers[i] == numbers[i + 1])
                    return false;
                diff += numbers[i + 1] - numbers[i] - 1;
            }
        }
        return (numZeros < diff) ? false : true;
    }
　　
};

 

注意：

C++ 排序請使用 sort()，不要使用 C 語言中的 qsort()；

轉一下 https://stackoverflow.com/questions/12308243/trying-to-use-qsort-with-vector 的大佬評論：

First of all, DON'T.

If you just want to muck about, you can replace iterators with actual pointers:

qsort(&numbers[0], numbers.size(), sizeof(int), compvar);

Apart from not doing all the work std::sort does, there is one unexpected thing about qsort. It is slower.

1. sort (myvector1.begin(), myvector1.end());

2. sort (myvector2.begin(), myvector2.end(), myfunction);

3. sort (myvector3.begin(), myvector3.end(), myobject);

4. qsort(&myvector4[0], myvector4.size(), sizeof(int), cmyfunction);

4 is the slowest, followed by 2 (function pointer passed to std::sort). 1 and 3 (default and functor) are the fastest (compiled with gnu's g++ with -O3 flag).