所以我编写了自己的ArrayList实现(又名C++中的向量),并在其中包含了几种算法。现在我的合并排序方法似乎泄漏了内存,但是我已经逐行检查了代码,跟踪了删除的分配,一切似乎都进展顺利!
我应该注意,我有一个 ArrayList 中每个方法的测试脚本,我遇到了崩溃,然后我尝试删除合并排序测试,然后繁荣,不再崩溃。但有趣的是...它并不总是崩溃,它有时会工作,它会崩溃其他人。
这两种方法的代码如下:
快速变量枚举:
array = 支持数组列表的数组
size = 跟踪数组大小的整数。
sort = 指示列表是否已排序的布尔值
/**
* Runs merge sort on this ArrayList<T>. Interface function to the central,
* recursive, merge sort function.
*
* Runs in O(nlogn) time. However it consumes extra memory.
*/
template<class T>
void ArrayList<T>::mergeSort() {
T* temp = mergeSort(array, size);
delete [] array;
array = temp;
sorted = true;
}
/**
* Runs merge sort on the passed in array. Recursive.
*
* Runs in O(nlogn) time. However it consumes extra memory.
*
* @param array the array to sort.
* @param arraySize the size of the array that is to be sorted.
* @return the sorted array.
*/
template<class T>
T* ArrayList<T>::mergeSort(T* array, int arraySize) {
T* returnArray;
//If the array is more than one element.
if (arraySize > 1) {
int size1 = arraySize / 2;
int size2 = arraySize - size1;
T* array1;
T* array2;
//Recurse.
array1 = mergeSort(array, size1);
array2 = mergeSort(array + size1, size2);
//Allocate memory for return array.
returnArray = new T[arraySize];
//Loop through all elements in returnArray.
int i = 0, j = 0, k = 0;
while (i < arraySize) {
//Place the lesser of two elements in returnArray.
if ((array1[j] <= array2[k] && j < size1)
|| k == size2) {
returnArray[i] = array1[j];
j++;
}
else {
returnArray[i] = array2[k];
k++;
}
i++;
}
//Free the memory allocated in the recursive calls.
delete [] array1;
delete [] array2;
array1 = 0;
array2 = 0;
}
//If one element is in the passed array.
else {
//Allocate memory for new array, and assign passed value to it.
//This is done so delete can be called in the calling function.
returnArray = new T[1];
returnArray[0] = array[0];
}
return returnArray;
}
您正在访问array1 [ j ]
,然后再检查它是否j < size1
。如果j >= size1
则访问该索引处的数组是非法的。它可能不会总是崩溃,具体取决于堆中事物的内存布局,但有时会崩溃。您的支票应该是这样的:
if (((j < size1) && (array1[j] <= array2[k])) || k == size2) {
...