I've seen it in coding standards for C++ that were a hang-over from C, as if you don't have RAII or other automatic memory management then you have to clean up for each return, which either means cut-and-paste of the clean-up or a goto (logically the same as 'finally' in managed languages), both of which are considered bad form. If your practices are to use smart pointers and collections in C++ or another automatic memory system, then there isn't a strong reason for it, and it become all about readability, and more of a judgement call.

我更喜欢一个return语句。一个尚未指出的原因是，一些重构工具在单点退出时工作得更好，例如Eclipse JDT提取/内联方法。

我通常支持多个return语句。它们是最容易阅读的。

在某些情况下，这并不好。有时从函数返回可能非常复杂。我记得有一种情况，所有函数都必须链接到多个不同的库。一个库期望返回值是错误/状态代码，而其他库则不期望。使用一个return语句可以节省时间。

我很惊讶没人提到goto。Goto并不是每个人都想让你相信的编程的祸害。如果必须在每个函数中只有一个return语句，请将它放在末尾，并根据需要使用gotos跳转到该return语句。绝对避免标记和箭头编程，它们既丑陋又运行缓慢。

函数中return语句越多，该方法的复杂性就越高。如果您发现自己想知道是否有太多的return语句，那么您可能需要问问自己，该函数中是否有太多的代码行。

但是，不是，一个/多个返回语句并没有错。在某些语言中，这是一种比其他语言(C)更好的实践(c++)。

如果管理得当，多次退出是件好事

第一步是明确退出的原因。我的通常是这样的: 1. 不需要执行函数 2. 发现错误 3.提前完成 4. 正常完成 我想你可以把“1”归为一类。不需要执行函数“into”3。提前完成”(如果你愿意的话，可以说是非常提前完成)。

第二步是让函数外部的人知道退出的原因。伪代码看起来像这样:

function foo (input, output, exit_status)

  exit_status == UNDEFINED
  if (check_the_need_to_execute == false) then
    exit_status = NO_NEED_TO_EXECUTE  // reason #1 
    exit

  useful_work

  if (error_is_found == true) then
    exit_status = ERROR               // reason #2
    exit
  if (need_to_go_further == false) then
    exit_status = EARLY_COMPLETION    // reason #3
    exit

  more_work

  if (error_is_found == true) then
    exit_status = ERROR
  else
    exit_status = NORMAL_COMPLETION   // reason #4

end function

显然，如果将上图中的一大块工作移到一个单独的函数中是有益的，那么您应该这样做。

如果您愿意，您可以使用更具体的退出状态，例如，使用几个错误代码和早期完成代码来精确定位退出的原因(甚至位置)。

即使您强制这个函数只有一个出口，我认为您仍然需要指定出口状态。调用者需要知道是否可以使用输出，这有助于维护。

I always avoid multiple return statements. Even in small functions. Small functions can become larger, and tracking the multiple return paths makes it harder (to my small mind) to keep track of what is going on. A single return also makes debugging easier. I've seen people post that the only alternative to multiple return statements is a messy arrow of nested IF statements 10 levels deep. While I certain agree that such coding does occur, it isn't the only option. I wouldn't make the choice between a multiple return statements and a nest of IFs, I'd refactor it so you'd eliminate both. And that is how I code. The following code eliminates both issues and, in my mind, is very easy to read:

public string GetResult()
{
    string rv = null;
    bool okay = false;

    okay = PerformTest(1);

    if (okay)
    {
        okay = PerformTest(2);
    }

    if (okay)
    {
        okay = PerformTest(3);
    }

    if (okay)
    {
        okay = PerformTest(4);
    };

    if (okay)
    {
        okay = PerformTest(5);
    }

    if (okay)
    {
        rv = "All Tests Passed";
    }

    return rv;
}