- Java 双重比较
- java - 比较器与 Apache BeanComparator
- Objective-C 完成 block 导致额外的方法调用?
- database - RESTful URI 是否应该公开数据库主键?
我正在尝试用 C++ 为类创建一个基本的 ArrayList 实现。该类(class)的要求非常严格,并且需要数据成员和函数。但是,当我运行它时,当它试图删除 deepCopy 中的列表指针时,我得到“无法访问内存位置”。我似乎无法弄清楚为什么。请帮忙。谢谢!
我已经缩小了代码范围。不过,我仍然想为您提供足够的信息以便能够提供帮助。
美国要求完整的代码,所以我把它加回来了。
#ifndef LIST_H
#define LIST_H
/**
* This implentation of list is a pseudo ArrayList. Since we don't guarantee that a class has a hashCode in C++
* this will only work for objects which have an id field.
*
*/
template <class T>
class List{
private:
// If you want the max size of the list (100), then you should probably create a const. Example:
// const int MAX_SIZE;
// T[] list = new T[MAX_SIZE];
// But really, that is not much better. This is just bad, overall. It's like we're implementing a very poor ArrayList.
T * list; // This is our array for our elements.
int numberInList; //
int listSize; // Getter method will be getMaxSize. This seems unnecessary if we're initializing it to 100...
int nextIndex; // this is a number of the index for the next element. Really, this should be inside a private class (who has the pointer to its next).
double loadFactor; // We need to determine when we need to allocate additional slots so we don't run out if they try to add more than they have allocated.
void deepCopy(const List<T> & toBeCopied ); // We need a method to perform a deepCopy. Abstracts the implementation from each method. Private because we don't want it exposed to clients.
bool isAboveThreshold(); // Check if the list is above the loadFactor.
// Publicly available API
public:
List();
List(int size); // Overloaded constructor to initialize the List to a runtime size
List(const List<T> & toBeCopied); // Copy constructor
~List(); // Destructor -- Get rid of the dynamically allocated member.
T next(); // Gets the next element in the list, and increments next
bool add( T & element ); // Adds an element to the list. Also checks to make sure the element hasn't already been added.
bool contains(T & element); // Checks the list to see if the element exists in the list already.
int getSize(); // return the number of elements in the list.
List<T> & List<T>::operator =( const List<T> & toBeCopied );
};
template <class T>
List<T>::List(){
listSize = 100;
list = new T[listSize];
numberInList = 0;
nextIndex = 0; // Initialize the next element to 0.
loadFactor = 0.75;
};
template <class T>
List<T>::List(int size){
list = new T[size];
numberInList = 0;
nextIndex = 0; // Initialize the next element to 0.
listSize = size;
loadFactor = 0.75;
};
template <class T>
List<T>::List(const List<T> & toBeCopied){
deepCopy(toBeCopied);
};
/****
*
* We need to release the allocated heap memory. Non-pointer members will be deallocated when they are out of scope.
*
*/
template <class T>
List<T>::~List(){
delete [] list;
};
/**
* Return the number of elements in the list.
*/
template <class T>
int List<T>::getSize(){
return numberInList;
}
/**
* Return the number of elements in the list.
*/
template <class T>
T List<T>::next(){
return list[nextIndex++];
}
/**
* Check if the element is already in the list
*/
template <class T>
bool List<T>::contains( T & element){
// Now, to check if the item already exists, we could just iterate over the array, but that gives us linear execution time.
// It seems sub-optimal to work in linear time here, when it feels like we shouldn't have to, but honestly, I'm too tired
// to care at this point.
for(int i = 0; i < (numberInList); i++){
if(element != list[i]) // We do this so that if the first item is not equal, we don't even bother checking the second condition.
{
// The element isn't matched. We have to finish iterating, though, before we can add it.
}
else{ // The element matched. Return false.
element = list[i];
return true;
}
}
return false;
}
/**
* The implementation for this is very bad. But, the requirements in the homework documentation are very restrictive.
*
* Ideally, we would have a companion class named Entry which kept the reference to the element via composition.
*
* Obviously, this is a list, so we only want unique entries.
*
* if successful, we return a true. Else, we're returning false.
*
*/
template<class T>
bool List<T>::add( T & element ){
// If we've exceeded the loadFactor, we want to expand our array before we add.
if( isAboveThreshold() ){
int newSize = listSize*2;
List<T> tempPtr = List<T>(newSize);
for(int i = 0; i < numberInList; i++){
tempPtr.add(list[i]);
}
deepCopy(tempPtr);
}
if(!contains( element )){
list[numberInList] = element; // if there are 4 in the list, the next empty index is 4, so this works. We get our element, then post-increment.
numberInList++;
return true;
}
return false;
}
/**
* Deep copy mechanism
*/
template<class T>
void List<T>::deepCopy(const List<T> & toBeCopied){
// Take care of shallow copying first.
numberInList = toBeCopied.numberInList;
listSize = toBeCopied.listSize;
nextIndex = 0; // We're getting a new list, so our iterator should start over.
// Now, to initialize the new list
T *tempList = new T[listSize];
for(int i = 0; i < toBeCopied.numberInList; i++){
// We can do this because we're in the List class. We have access to private members.
tempList[i] = toBeCopied.list[i];
}
delete [] list;
list = tempList;
}
/**
* boolean for if we've exceeded the loadFactor threshold.
*/
template<class T>
bool List<T>::isAboveThreshold(){
if(numberInList != 0){
double division = (double)numberInList/listSize;
return (division >= loadFactor)? true : false;
}else
return false;
}
/***
* Overloaded assignment operator
*/
template <class T>
List<T> & List<T>::operator =( const List<T> & assigner ){
if(*this == &assigner)
return *this;
delete[] list;
deepCopy(assigner);
return *this;
}
#endif
#include "List.h"
#include <string>
#include<iomanip>
#include <iostream>
#include <fstream>
#include <stdlib.h>
using namespace std;
struct Customer{
int id;
string name;
string city;
string address;
float amount;
Customer(){id=0; city="Default"; name="N/A"; address="N/A", amount = 0;}
bool operator==( const Customer & assign){
if(assign.id == id)
return true;
else
return false;
}
bool operator!=( const Customer & assign){
if(assign.id != id)
return true;
else
return false;
}
};
List<Customer> readCustomers();
void printCustomers();
int main(){
cout.setf(std::ios::fixed);
printCustomers();
return 0;
}
// Definitions
List<Customer> readCustomers(){
List<Customer> WebsterCommunications(50);
ifstream custFile;
custFile.open("Customers.csv");
// This could be abstracted out into another method, where we took the struct, a struct name, and the inFile, and spit back a
// customer from the file. But, for now, we'll just settle with the code duplication.
if(!custFile){
cout << "There was a problem reading the Customer File..." << endl;
exit(99);
}
while(!custFile.eof()){
Customer tempCust;
custFile>>tempCust.id;
if(tempCust.id == 0)
break;
custFile.ignore();
getline(custFile, tempCust.name, ',');
getline(custFile, tempCust.address, ',');
getline(custFile, tempCust.city, ',');
custFile>>tempCust.amount;
custFile.ignore();
WebsterCommunications.add(tempCust);
}
custFile.close();
return WebsterCommunications;
}
void printCustomers(){
List<Customer> customers = readCustomers();
double addCalc = 0.0;
cout << string( 100, '\n' );
for(int i = 0; i < customers.getSize(); i++){
Customer cust;
cust = customers.next();
cout << "id: " << cust.id << " name: " << cust.name << " address: " << cust.address << " balance: " << cust.amount << endl;
addCalc += cust.amount;
}
cout.precision(2);
cout << "average: " << (addCalc / customers.getSize()) << endl;
int isActive = 1;
cout << "Please enter a customer's id to find them (0 to exit):" << endl;
while(isActive){
cin >> isActive;
if(!isActive){
return;
}
Customer tempCust;
tempCust.id = isActive;
if(customers.contains(tempCust)){
cout << "id: " << tempCust.id << " name: " << tempCust.name << " address: " << tempCust.address << " balance: " << tempCust.amount << endl;
}
else{
cout << "That customer is not found" << endl;
}
cout << "Please enter a customer's id to find them (0 to exit):" << endl;
}
}
最佳答案
在 operator=
中,您尝试删除当前附加到对象的内部列表,但实际上您所做的是删除要复制的 list
- 在复制构造函数中,list
命名参数而不是类成员。
通过为类成员和参数赋予不同的名称来解决此问题。 C++ 中有明确定义的名称隐藏规则,但依赖它们几乎总是会导致混淆。
而且你还在重复删除 list
!一旦进入你的 deepCopy
并且一旦进入你的 operator=
!您需要重新考虑您的设计以及应该清理哪些资源。一个快速修复方法是在 deepCopy
中不删除,但我还没有想过这会如何泄漏。我会留给您寻找更好的解决方案。
关于C++编程练习--深拷贝,我们在Stack Overflow上找到一个类似的问题: https://stackoverflow.com/questions/14674844/
拿这个: var lists:{ item1:{} ,item2:{} ,item3:{} ,item4:{} } 由于它基本上是空的,我想要一个函数(可能但不一定是 _lodash
我想更改 ng bootstrap 分页组件的样式并在 Angular 6 应用程序中使用 /deep/ 链接。以下代码工作正常,但控制台显示警告,指出该代码已被弃用。 那么,我应该如何更改它以消除警
使用 webcomponents,您可以使用 /deep/ 定位 shadowdom 的内部元素,在我尝试使用事件委托(delegate)之前它工作正常。 一个常规的点击功能将起作用: $('html
在 Swift 中,我试图实现一个单词 Trie,使用文字表示作为一系列嵌套的 NSObject。这是 Trie。 let GEENITRIE:NSObject = [ "i":[
运行 npm update 更新 package.json 中列出的项目;但是,这些项目的依赖关系仍然过时。 明显的解决方法是再次运行 npm update。有时我需要运行它 3 次以上才能使 npm
我创建了 2 级嵌套 linq 查询: var data = (from p in Departments join e in Employees on p.Id equals
首先是代码 #include typedef wchar_t* BSTR; wchar_t hello[] = L"Hello"; class _bstr_t { public: opera
我要编写一个 lisp 程序来生成十六进制数的实际值。我已经编写了一个函数,但似乎出现了 stackoverflow(深度)错误。我想知道是否有人可以指出我的错误或指导我朝着正确的方向前进。 如果没有
我想将每种类型都转换为boolean 或object type CastDeep = { [P in keyof T]: K extends K[] ? K[] : T[P] ex
我刚刚发现自己在写这个: fn init_timestamps(dir: &PathBuf, file_timestamps: &'static HashMap) { match fs::re
我有一个现有的 pybind11::array_t,需要进行复制构造。 pybind11 中是否有一个函数允许我对 array_t 进行深度复制? 我知道我可以创建一个新的 array_t,适当调整大
引用http://www.devx.com/tips/Tip/13291 ,它说我可以使用 memcpy 来复制由 sizeof() 确定的大小,但是,数组不是指向指针吗? 除了遍历数组,如何使用 m
我有多个成员(member),每个成员(member)都有一条记录,其中包含几个备注字段: Member ID Entry A Entry B 1 [memo t
$watchCollection 是否能够忽略对以 $ 开头的属性的更改?使用深度 $watch 时已经存在此行为,因为它依赖于 angular.equals 进行比较。 理想情况下,$watchCo
我有一个带有复杂键的 map - 例如,二维数组: m := make(map[[2][3]int]int) 当我在映射中插入一个新键时,Go 是否会对该键进行深度复制? a := [2][3]int
我需要查询如下所述的三个表。我了解简单的 JOIN,但是这个有点超出我的水平。 courses 有两列 id (PK) 和 courseTitle(示例 id 1,courseTitle 物理) cl
我有一个对象,它是 Realm 的 Object 子类,并且符合 Codable 以便在与 API 对话时与 JSON 相互转换。 如何利用 Codable 协议(protocol)进行深度复制(包括
我是一名优秀的程序员,十分优秀!