Title: Standards Report

Author: Martin Moene

Date: 02 November 2015 09:16:50 +00:00 or Mon, 02 November 2015 09:16:50 +00:00

Summary: Jonathan Wakeley reports on developments in C++.

Body: 

I’m writing this report at Heathrow airport, about to fly out to Kona, Hawaii, for the C++ standards meeting. By the time you read this the C++ committee and the C committee will both have finished their Kona meetings (which run back to back) so expect to see news from those meetings in my next report. Although going to Kona sounds exciting, the C++ committee have well over 100 papers to discuss during the week and I don’t expect to see much outside the meeting rooms. I wonder if the C committee have a more relaxed schedule and can enjoy the nice location, and if I’m on the wrong committee!

Since my last report for CVu there haven’t been any face-to-face meetings, but the pre-meeting mailing [1] for the C++ meeting has lots of papers. The sharp-eyed will notice a change in the naming scheme for papers. From now on N-numbers will only be used for official ISO documents such as meeting notices, agendas and minutes, and working drafts and project editors’ reports. The informal proposals and position papers will get a P-number, with a suffix indicating whether it is a revision of an earlier paper.

The mailing includes a new working draft for ‘Ranges’ [2], which I mentioned in my last report [3]. The lengthy email discussions on the std::experimental::variant design have abated, but there are several papers in the mailing about variant and it will be a hot topic during the Kona meeting. At the end of the Kona meeting the Networking proposal [4] should be in good shape to turn it into a working draft, which is the next step towards publishing a TS. There is a proposal to add the content of the Parallelism TS to C++17, which will no doubt be discussed in Kona too.

Apart from the upcoming meeting, the other big C++ news is the announcement at CppCon of the ‘C++ Core Guidelines’ [5] which are a set of (still evolving) coding guidelines intended to encourage people to use the modern, safe features of the language. Part of that also requires discouraging people from using the dark corners of the language which are still part of the standard but have no place in most code. Accompanying the guidelines will be an open-source library providing a set of useful vocabulary types and utilities to help follow the guidelines, and a static analysis tool to check that code conforms to the guidelines. Keep your eyes peeled for these soon.

References

[1] http://www.open-std.org/jtc1/sc22/wg21/docs/papers/2015/#mailing2015-09

[2] http://www.open-std.org/jtc1/sc22/wg21/docs/papers/2015/p0021r0.pdf

[3] CVu 27-4, September 2015

[4] http://www.open-std.org/jtc1/sc22/wg21/docs/papers/2015/p0112r0.html

[5] https://isocpp.org/blog/2015/09/bjarne-stroustrup-announces-cpp-core-guidelines

Notes: 

More fields may be available via dynamicdata ..