优先级队列堆实现

时间:2012-05-15 06:08:22

标签: c++ data-structures priority-queue

我正在编写代码来使用堆实现优先级队列。当我按照特定顺序8 10 4 3 7 6 9 5将这些优先级输入到队列中时,一旦我开始用get_front()函数弹出它们,我就会收到错误。

问题是swap_with_parent()函数的while循环中get_front()函数的断言失败。该参数以某种方式增长大于数组中的项目数many_items。我会发布整个代码,如果有人能发现问题我会很感激,如果你能告诉我的话。我提前为缺乏评论而道歉,我希望我已经清楚地了解了我的情况。

// INVARIANT for the PriorityQueue Class:
//   1. The member variable many_items is the number of items in the
//      PriorityQueue.
//   2. The items themselves are stored in the member variable heap,
//      which is a partially filled array organized to follow the usual
//      heap storage rules from Chapter 11 of the class notes.
// NOTE: Private helper functions are implemented at the bottom of this
// file along with their precondition/postcondition contracts.

#include <assert.h>    // Provides assert function
#include <iomanip>   // Provides setw
#include <iostream>  // Provides cin, cout
#include <math.h>      // Provides log2
#include "pqueue2.h"

using namespace std;

PriorityQueue::PriorityQueue( )
{
    heap[CAPACITY];
    many_items=0;
}

void PriorityQueue::insert(const Item& entry, unsigned int priority)
{
    if(many_items==0)
    {
        heap[many_items].data= entry;
        heap[many_items].priority= priority;
        many_items++;
    }
    else
    {
        heap[many_items].data= entry;
        heap[many_items].priority= priority;
        unsigned int i= many_items;
        many_items++;
        while(parent_priority(i)<priority)
        {
            swap_with_parent(i);
            i=parent_index(i);
        }
    }
}

PriorityQueue::Item PriorityQueue::get_front( )
{
    assert(many_items>0);
    if(many_items==1)
    {
        Item front_value=heap[0].data;
        many_items--;
        return front_value;
    }
    else
    {
        Item front_value=heap[0].data;
        heap[0]=heap[many_items-1];
        unsigned int priority= heap[many_items-1].priority;
        unsigned int k=0;
        while( (k<many_items) && !is_leaf(k) && big_child_priority(k)>priority)
        {
            unsigned int j=big_child_index(k);
            swap_with_parent(big_child_index(k));
            k= j; 
        }
        many_items--;
        return front_value;
    }
}

bool PriorityQueue::is_leaf(size_t i) const
// Precondition: (i < many_items)
// Postcondition: If heap[i] has no children in the heap, then the function
// returns true. Otherwise the function returns false.
{    
    if(((2*i)+1)>many_items)
        return 1;
    else 
        return 0;
}

size_t PriorityQueue::parent_index(size_t i) const
// Precondition: (i > 0) && (i < many_items)
// Postcondition: The return value is the index of the parent of heap[i].
{
    //assert( /*(i>0) && */(i<many_items));
    return (i-1)/2;
}

unsigned int PriorityQueue::parent_priority(size_t i) const
// Precondition: (i > 0) && (i < many_items)
// Postcondition: The return value is the priority of the parent of heap[i].
{
    return heap[ (i-1)/2].priority;
}

size_t PriorityQueue::big_child_index(size_t i) const
// Precondition: !is_leaf(i)
// Postcondition: The return value is the index of one of heap[i]'s children.
// This is the child with the larger priority.
{
    assert(!is_leaf(i));
    if((2*i)+2<many_items)
    {
        if(heap[(2*i)+1].priority>heap[(2*i)+2].priority)
        {
            return (2*i)+1;
        }
        else
        {
            return (2*i)+2;
        }
    }
    else
    {
        return(2*i)+1;
    }
}

unsigned int PriorityQueue::big_child_priority(size_t i) const
// Precondition: !is_leaf(i)
// Postcondition: The return value heap[big_child_index(i)].priority
{
    return heap[big_child_index(i)].priority;
}

void PriorityQueue::swap_with_parent(size_t i)
// Precondition: (i > 0) && (i < many_items)
// Postcondition: heap[i] has been swapped with heap[parent_index(i)]
{
    assert( i>0 && i<many_items);
    OneItemInfo temp_parent=heap[parent_index(i)];
    OneItemInfo temp_child=heap[i];
    heap[i]=temp_parent;
    heap[parent_index(i)]=temp_child;
}

1 个答案:

答案 0 :(得分:3)

我认为is_leaf不正确。条件应为(2*i+1 >= many_items)

相关问题