linux合并排序,合并排序C++代码

合并排序的单向链表实现并不是最快的方法，对于100万个数，合并排序单向链表的实现用时15秒，比合并排序的数组实现(1.2秒)和快速排序的数组实现(0.4秒)都要慢。但是有些应用不得不适用链表，而快速排序法由于涉及太多的random access，用链表实现显然会非常的慢。所以单向链表的合并排序的应用也是很广泛的。

以下是我的C/C++代码：

// merge sort on linked list

// by Qingxing Zhang, Dec 26, 2009. prep for google interview

#include

#include

#include

#define N 1000000

using namespace std;

struct node

{

int data;

node *next;

};

node *insert(node *head, int data)

{

node *tmp = new node;

tmp->data = data;

tmp->next = head;

return tmp;

}

node *remove(node *head)

{

if(head!=NULL)

{

node *tmp = head;

head = head->next;

delete tmp;

}

return head;

}

node *reverse(node *head)

{

node *newHead = NULL;

while(head!=NULL)

{

node *tmp = head->next;

head->next = newHead;

newHead = head;

head = tmp;

}

return newHead;

}

node *mix(node *left, node *right)

{

node *result = NULL;

while((left!=NULL)&&(right!=NULL))

{

if(left->data < right->data)

{

result = insert(result, left->data);

left = remove(left);

}

else

{

result = insert(result, right->data);

right = remove(right);

}

}

while(left!=NULL)

{

result = insert(result, left->data);

left = remove(left);

}

while(right!=NULL)

{

result = insert(result,right->data);

right = remove(right);

}

result = reverse(result);

return result;

}

node *merge(node *head, int n)

{

if(n==1)

return head;

// divide into 2 lists

node *right = head;

node *left = head;

for(int i=0;i

right = right->next;

node *tmp = right;

right = right->next;

tmp->next = NULL;

// do merge sort on both lists

left = merge(left,n/2);

right = merge(right,n-n/2);

// mix left and right

return mix(left,right);

}

int main()

{

clock_t start,end;

double diff;

node *head = NULL;

srand ( (unsigned int)time(NULL) );

for(int i=0;i

head = insert(head,rand());

start = clock();

head = merge(head,N);

end = clock();

diff = end - start;

cout << "Merge sort for linked list: " << diff/CLOCKS_PER_SEC << " seconds..." << endl;

//while(head!=NULL)

//{

// cout << head->data << endl;

// head = head->next;

//}

return 0;

}