实施[B，C] = F（A）的语法（函数f作用在阵列上具有两个或更多的输出阵列） [英] Implementing the [B,C]=f(A) syntax (function f acting on an array with two or more output arrays)
问题描述
我有一个问题这是另外两个问题的扩展我最近已经发布：
I have a question which is an extension of other two questions I have recently posted:
<一个href=\"http://stackoverflow.com/questions/16254797/implementing-b-fa-with-b-and-a-arrays-and-b-already-defined\">Implementing B = F（A），与B和A阵列和B已经
和
<一个href=\"http://stackoverflow.com/questions/16287644/implementing-the-b-fa-syntax-by-move-assignment\">Implementing在B = F（A）由移动分配的语法
假设我有一个数组 A 。我想创建一个函数˚F作用于 A 并返回其他两个数组 B 和 C ，通过启用以下Matlab的语法类似
Suppose I have an array A. I want to create a function f that acts on A and returns two other arrays B and C, by enabling the following Matlab-like syntax
[B,C]=f(A);
在C有可能++？
Is it possible in C++?
解决方案以下列出LEEMESANSWER
#include <tuple>
using std::tie;
std::tuple<TypeOfB,TypeOfC> f(const Matrix<T1>&a,const Matrix<T2>&a) {
// Instruction declaring and defining B_temp and C_temp
return std::make_tuple(B_temp,C_temp); }
int main( int argc, char** argv)
{
// Instruction declaring A, B and C
tie(B,C)=f(A);
// STUFF
return 0;
}
一切工作也在发生变化时，的std ::元组和 make_tuple 到的std ::对和的std :: make_pair </ code>对于这种特殊情况下（只有两个输出）。
Everything works also when changing std::tuple and make_tuple to std::pair and std::make_pair for this particular case (only two outputs).
推荐答案
在一般情况下，如果要返回多个值，你必须做一些小的解决方法，因为C ++不允许这种开箱。
In general, if you want to return multiple values, you have to do some little work-around, since C++ doesn't allow this out of the box.
第一个选项是返回 STD ::对 包含两个值。然后你可以使用 的std ::领带 如果您有C ++ 11可用，这样一个return语句：
The first option is to return a std::pair containing both values. Then you can use std::tie in a return statement if you have C++11 available, like this:
std::tie(B, C) = f(A);
（注：C ++ 11还具有 的std ::元组 超过两个值。）
(Note: C++11 also has std::tuple for more than two values.)
或者你也可以通过引用传递两个目标变量，但随后的函数调用变得像这样（没有工作C ++ 11）：
Or you can pass the two target variables by reference, but then the function call becomes something like this (works without C++11):
f(A, B, C);
为了使函数调用看起来更加详细（有些人不喜欢，你不能告诉˚F修改 B 和 C ），您也可以使用指针而不是引用。然后，函数调用看起来是这样的：
To make the function call look more "verbose" (some people don't like that you can't tell that f changes B and C from looking at this single line of code) you can also use pointers instead of references. Then the function call would look like this:
f(A, &B, &C);
另一种选择是使用一个简单的容器您的多个返回值。如果一个简单的对或元组不给值的特殊意义，这是特别有用的。最好的办法是在code这就要求坚持用这个˚F（不要使用单独的数组 B 和 C ）。使用到code设计的其余部分仅当它适合很好。
Another option is to use a simple "container" for your multiple return values. This is useful in particular if a simple pair or tuple don't give the values a particular meaning. The best option is to use this consistently in the code which calls f (don't use separate arrays B and C). Use this only if it fits nicely into the rest of your code design.
struct TwoArrays {
int B[100];
int C[100];
};
TwoArrays result = f(A);
这篇关于实施[B，C] = F（A）的语法（函数f作用在阵列上具有两个或更多的输出阵列）的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！
