Doc. No.: N4309
Date: 2014-11-17
Project: Programming Language C++, Evolution Working Group
Reply To: Michael Price

Return type deduction for explicitly-defaulted and deleted special member functions

I. Introduction

Proposes to allow auto and declspec(auto) as the type-specifiers for the return type of explicitly-defaulted and deleted special member functions. It seems this case was left out of N3638.

II. Motivation

Paragraph, clause 2 of the draft standard (N4141) states:

The placeholder type can appear with a function declarator in the decl-specifier-seq, type-specifier-seq, conversion-function-id, or trailing-return-type, in any context where such a declarator is valid. If the function declarator includes a trailing-return-type (8.3.5), that specifies the declared return type of the function. If the declared return type of the function contains a placeholder type, the return type of the function is deduced from return statements in the body of the function, if any.

This wording allows the usage of auto or declspec(auto) in a function definition, only if there is a trailing return type or a function body that contains at least one return statement that can be used to deduce the return type. This proposal would allow the placeholder type-specifiers to be valid for special member functions that are explicitly-defaulted or deleted.

  struct VersionOne
      VersionOne() = default;
      ~VersionOne() = default;
      VersionOne(const VersionOne&) = default;
      VersionOne(VersionOne&&) = default;
      VersionOne& operator=(const VersionOne&) = default;
      VersionOne& operator=(VersionOne&&) = default;

  // For some reason, the author decided that I wanted non-default
  // copy operations and thought it would be nice to throw in 'auto'.
  struct VersionTwo
      VersionTwo() = default;
      ~VersionTwo() = default;
      VersionTwo(const VersionTwo&) { /* ... */ }
      VersionTwo(VersionTwo&&) = default;
      auto operator=(const VersionTwo&) { /* ... */ return *this; }
      auto operator=(VersionTwo&&) = default;  // Currently an error!

III. Technical Specification

This paper is looking for guidance on whether this feature should be pursued. It provides no wording.