Title
[tiny] Core closed issues with a previous extension status
Status
open
Section
[intro]
Submitter
Ville Voutilainen

Created on 2014-07-01.00:00:00 last changed 119 months ago

Messages

Date: 2015-04-10.16:59:07

13 extern "C" for Parameters of Function Templates -> request Core to close as NAD, Daveed says aliases with linkage are the solution. 1463 will be opened as an EWG issue to solve this.

203 Type of address-of-member expression -> new EWG issue, 89. The EWG issue is NAD, too late for a breaking change, instruct CWG to also close as NAD.

230 Calls to pure virtual functions -> Instruct Core to close 230 as NAD. Pure virtual functions can be called by non-virtual calls only, and it is by design that virtual calls should never resolve to the definition of a pure virtual function. The behaviour is undefined because implementations aren't required to prevent ending up in the pure virtual function definition. That is also by design.

622 Relational comparisons of arbitrary pointers -> new EWG issue, 91. The issue is NAD -> instruct CWG to do the same.

623 Use of pointers to deallocated storage -> Instruct Core to close 623 as WP. The behavior is implementation-defined rather than undefined in the current working draft, and thus the requirements are no longer draconian. 312, which was resolved as DRWP points to 623, so also close 623 as DRWP.

755 Generalized lambda-captures -> Instruct Core to close 755 as WP. Generalized lambda-captures are in C++14.

822 Additional contexts for template aliases -> new EWG issue, 95. -> issue is NAD, suggest CWG also closes as NAD.

947 Deducing type template arguments from default function arguments -> Open an EWG issue, 97, for 947, immediately close as NAD, explaining that the group doesn't think it's generally implementable. Leave the door open (but not the issue :)) for a restricted implementation.

1259 Deleting a POD via a pointer to base -> new EWG issue, 99. -> issue is NAD, suggest CWG closes as NAD.

1272 Implicit definition of static data member of const literal type -> new EWG issue, 100. -> issue is NAD, suggest CWG closes as NAD.

1461 Narrowing conversions to bit-fields -> Instruct Core to close 1461 as NAD. It would break existing code and EWG doesn't consider the breakage to be worth it. Vandevoorde pointed out valid existing code that initializes bitfields (even with one bit) with -1.

1474 User-defined literals and <inttypes.h> format macros -> Instruct Core to close 1474 as NAD. The fix would be to remove UDLs.

1519 Conflicting default and variadic constructors -> Open an EWG issue for 1519, 107. Mark as NAD, there is a work-around and Concepts Lite will make the work-around much easier.

1564 Template argument deduction from an initializer list -> new EWG issue, 109. -> issue is NAD, suggest CWG closes as NAD as well.

1586 Naming a destructor via decltype -> new EWG issue, 112. -> issue is NAD, suggest CWG closes as NAD as well.

1657 Attributes for namespaces and enumerators -> Open an EWG issue for 1657, 113. Remember to consider inline namespaces and anon namespaces. Rationale for both namespaces and enumerators is [[deprecated]].

1754 Declaration of partial specialization of static data member template -> new EWG issue, 132 (NAD, lack of motivation for the change)

1798 exception-specifications of template arguments -> new EWG issue, 133 (NAD, insufficient motivation for a change, no desire to have exception specifications in the type system, paper requested if there's desire to reopen)

1833 friend declarations naming implicitly-declared member functions

Straw poll, Rapperswil 2014, NAD?

SF 3 - WF 7 - N 7 - WA 2 - SA 0

-> instruct CWG to close as NAD, EWG didn't find sufficient motivation for the extension suggested.

History
Date User Action Args
2014-07-01 00:00:00admincreate