[Insight-developers] Postfix versus Prefix operators

[David Cole]

On Fri, May 27, 2011 at 9:24 AM, Brad King <brad.king at kitware.com> wrote:
> On 05/27/2011 09:00 AM, Bill Lorensen wrote:
>> I noticed a comment you made in a review asking why cppcheck prefers
>> prefix operators over postfix operators. I did a little searching and
>> found this discussion regarding  the advantage of prefix over postfix:
>> http://www.thunderguy.com/semicolon/2002/08/13/prefer-prefix-operators-over-postfix/
>>
>> Perhaps other c++ experts may comment.
>
> Everything on that page is accurate.  However, it is possible to write a
> postfix operator for one's own class that does not need to save the value
> in a temporary.  See below for an example.  In C++0x the tag_postincr
> constructor will be able to call the real copy constructor to avoid the
> duplication.
>
> Most of the performance discussion is moot given modern compiler optimizers.
> The code below depends on the "Return Value Optimization" (RVO) to construct
> the return value of the postfix increment operator directly in its final
> destination rather than in an unnamed temporary which is then copied to the
> final destination.  Compilers also implement a "Named RVO" (NRVO) which will
> identify a local variable that is used as the only return value and allocate
> it directly in the final destination (when the code is inlined).  Therefore
> the two approaches will have no difference in practice.
>
> The other arguments on the page are not performance related and should be
> considered important.  I always use prefix unless I really need postfix.
>
> -Brad
>
>
> #include <iostream>
> class A
> {
>  struct tag_postincr {};
>  A(A& a, tag_postincr): Value(a.Value) // Duplicate copy constructor
>    { ++a; } // Increment copied value afterward
>  int Value;
> public:
>  A(): Value() {}
>  A(A const& a): Value(a.Value) {}
>  A& operator++() { ++this->Value; return *this; }
>  A operator++(int) { return A(*this, tag_postincr()); }
>  int GetValue() const { return this->Value; }
> };
>
> int main()
> {
>  A a;
>  A ar1 = ++a;
>  std::cout << "++a return = " << ar1.GetValue() << std::endl;
>  std::cout << "++a after  = " << a.GetValue() << std::endl;
>  A ar2 = a++;
>  std::cout << "a++ return = " << ar2.GetValue() << std::endl;
>  std::cout << "a++ after  = " << a.GetValue() << std::endl;
>  return 0;
> }
> _______________________________________________
> Powered by www.kitware.com
>
> Visit other Kitware open-source projects at
> http://www.kitware.com/opensource/opensource.html
>
> Kitware offers ITK Training Courses, for more information visit:
> http://kitware.com/products/protraining.html
>
> Please keep messages on-topic and check the ITK FAQ at:
> http://www.itk.org/Wiki/ITK_FAQ
>
> Follow this link to subscribe/unsubscribe:
> http://www.itk.org/mailman/listinfo/insight-developers
>

See also: "Item 6: Distinguish between prefix and postfix forms of
increment and decrement operators." from Scott Meyers "More Effective
C++" (1996)

  http://www.aristeia.com/books.html

  http://www.amazon.com/More-Effective-Improve-Programs-Designs/dp/020163371X