11

I found a mismatch between dates on a posting and the dates in the documents at this page:

http://www.open-std.org/jtc1/sc22/wg14/www/documents

I downloaded the most recent tarball and was reading about a class implementation for C. It was authored by someone from Sun in '95.

My question is are these active proposals that are being considered for the next C standard and is there a better place to read about the next drafting?

user5534993
  • 518
  • 2
  • 17
John Holly
  • 391
  • 1
  • 2
  • 12
  • @MarkRansom I think the likelyhood of this getting into the standard is very low. It would break a ton of currently standard safe code: e.g. all of glib/gtk. – Mgetz Dec 08 '14 at 18:40
  • @Mgetz it seems like a controversial proposal, I am curious though what aspects of the proposal would break existing code? Can you give an example? – Shafik Yaghmour Dec 08 '14 at 18:52
  • 1
    @ShafikYaghmour https://github.com/GNOME/gtk/search?l=c&q=class&utf8=%E2%9C%93 `class` is the big killer in terms of breaking changes. As that would break code that is written in object oriented C using that identifier. Although `this` may have just as many issues. – Mgetz Dec 08 '14 at 18:53
  • 1
    @Mgetz: They can still be called `_Class` and `_This` with `#define` in some `` header, as with `bool` and `static_assert`. – mafso Dec 08 '14 at 19:07
  • @mafso of course, however that creates an intentional disconnect between C and C++, which the author of the proposal explicitly says is not the intent. – Mgetz Dec 08 '14 at 19:13

1 Answers1

9

The documents listed here show the most recent meetings on top. The most recent meeting was in St. Louis and there are the pre-meeting mailing and post-meeting mailing listed on top.

In the pre-meeting documents we find N1875: Adding classes to C which was authored 2014-09-29 which is indeed recent and it would seem this is indeed an active issue.

What may be confusing in the post meeting document lists what look like old proposals from 1995:

N424 12 Jun 95 Jervis, RP: Classes in C, Part 1
N445 22 Aug 95 Jervis, Classes in C, Part 2: Inheritance
N446 22 Aug 95 Jervis, Classes in C, Part 3: Virtual Functions
N447 22 Aug 95 Jervis, Classes in C, Part 4: Constructors and Destructors

They may have been included for reference since the topic has come up again.

Update

Jens pointed out that the discussion of this proposal can be found draft minutes from the post mailing and the conclusion is:

The straw polls above indicate there is no interest by the Committee to persue further development of the proposal presented. However, we would encourage the submitter to develop an implementation that gains user experience with the concepts presented, i.e. Modification of gcc or clang.

So basically there is no interest in pursuing the proposal until there is an implementation that can demonstrate interest from the community.

Shafik Yaghmour
  • 154,301
  • 39
  • 440
  • 740
  • 1
    The discussion of the committee about that proposal can be found in the minutes of the St.Louis meeting: http://www.open-std.org/jtc1/sc22/wg14/www/docs/n1884.pdf – Jens Gustedt Dec 08 '14 at 21:18
  • @JensGustedt thank you, I had looked at that document but somehow missed that discussion at least I know what I am looking for next time. – Shafik Yaghmour Dec 08 '14 at 21:28