错误 C2228:“.size"的左边必须有类/结构/联合

2022-01-05 00:00:00 visual-studio-2008 c++

我在调用 vector 的 size() 时遇到这个编译器错误.为什么?

I'm getting this compiler error when calling vector's size(). Why?

#include <vector>
#include <iostream>
#include <fstream>
#include <sstream>
#include <string>
#include <cassert>


using namespace std;

class Vertex {

    float firstValue;
    float secondValue;
    float thirdValue;

    Vertex (float first, float second, float third){
          firstValue=first;
          secondValue=second;
          thirdValue=third;
    }

};


int main()
{
    cout<<"This program loads a 3D .off object. 
Enter the name of the file that describes it "<<endl;
    string inputFileName;
    getline(cin, inputFileName);

    ifstream inputFileStream;

    inputFileStream.open(inputFileName.data());
    assert (inputFileStream.is_open());

    string actualLine;

    for(;;){

        inputFileStream>>actualLine;

        istringstream actualLineStream(actualLine);



        std::vector<float> results( std::istream_iterator<int>(actualLineStream)
                        , std::istream_iterator<int>() );

       int resultsIndex=0;
       int resultsSize=results.size(); //WHY??

       while (resultsIndex<resultsSize){

         cout<<results[resultsIndex]<<endl;
       }


        if (inputFileStream.eof()) break;

    }


    ofstream outputChannel;

    while (true){} // to keep on console view 
    return 0;
}

推荐答案

信不信由你,这一行并没有声明一个名为std::vector的实例code>results,使用开始和结束迭代器调用构造函数:

Believe it or not, this line does not declare an instance of std::vector named results, calling the constructor taking a begin and end iterator:

std::vector<float> results(std::istream_iterator<int>(actualLineStream),
    std::istream_iterator<int>());

这实际上声明了一个名为results的函数,它接受一个名为actualLineStream的参数和另一个未命名的参数,两者都是std类型::istream_iterator.

This actually declares a function called results that takes a parameter named actualLineStream and another unnamed parameter, both of type std::istream_iterator<int>.

一般在 C++ 中,如果某个东西看起来像一个函数,它就会像一个函数一样被解析;C++ 标准要求它.这实际上是为了与 C 向后兼容――但这太违反直觉了,它甚至有自己的名字:"最令人烦恼的解析".一些编译器甚至会在遇到最烦人的解析时发出警告.

Generally in C++, if something looks like a function, it will be parsed like one; the C++ standard requires it. This is really for backward compatibility with C - but this is so counterintuitive that it even has its own name: the "most vexing parse". Some compilers will even issue a warning if it encounters the most vexing parse.

这与这两行在 C++ 中不等价有关:

It is related to the fact that these two lines are not equivalent in C++:

Foo bar;   // Declares an instance of Foo named bar
Foo bar(); // Declares a function named bar that takes no parameters and returns a Foo

要修复它,您可以在其中一个参数周围添加更多括号:

To fix it, you can add more parentheses around one of the arguments:

//                         +--------- Note extra parentheses!! ---------+
//                         |                                            |
//                         V                                            V
std::vector<float> results((std::istream_iterator<int>(actualLineStream)),
    std::istream_iterator<int>());

或者简单地分别声明每个迭代器:

Or simply declare each iterator separately:

std::istream_iterator<int> resultsBegin(actualLineStream);
std::istream_iterator<int> resultsEnd;
std::vector<float> results(resultsBegin, resultsEnd);

相关文章