Meta LWG issue: 2024-03 meeting

[frederick-vs-ja]

(Previous meta-issue: #4158)

At the March 2024 meeting, the following LWG issues were resolved in the C++ Working Paper.

❔ Not yet analyzed

• Remaining issues:
  • All done!

❌ Not applicable

If an issue requires no action from implementers, we mark it as N/A. Categories:

• Pure wording clarifications with nothing to implement (these can be changes to non-normative text like examples and informative notes, or wording cleanups to normative text that don't impact observable behavior)
  • LWG-4011 "Effects: Equivalent to return" in [span.elem]
  • LWG-4030 Clarify whether arithmetic expressions in [numeric.sat.func] are mathematical or C++
  • LWG-4036 __alignof_is_defined is only implicitly specified in C++ and not yet deprecated
    • Like LWG-3827, there's nothing for us to do here. Should we ping UCRT to implement macro deprecation? See also DevCom-10559200.
  • LWG-4043 "ASCII" is not a registered character encoding
• Something that increases the restrictions placed on users, but implementers aren't expected to enforce those restrictions
  • LWG-3975 Specializations of basic_format_context should not be permitted
  • LWG-4023 Preconditions of std::basic_streambuf::setg/setp
• Fixes for obviously broken wording, where implementers would have done the right thing anyways

😸 Already implemented

Sometimes we cite LWG issues in product code comments as we're implementing their proposed resolutions. When the resolutions are officially accepted, we should remove the citations (as the default assumption is that we're implementing what the Standard says). If something is especially subtle, we can convert the citation to mention the relevant Standard section. Sometimes we should add test coverage - e.g. when the Standard begins requiring something that we were already doing, but weren't explicitly testing for.

• Already implemented, comments need to be removed and messages need to cite the Standard
• Implemented without comments
  • LWG-4025 Move assignment operator of std::expected<cv void, E> should not be conditionally deleted
  • LWG-4031 bad_expected_access<void> member functions should be noexcept

🩹 Patches an unimplemented feature

We should record this LWG issue in the GitHub issue tracking the feature. That way, we'll remember to verify it, but it doesn't represent net new work.

• LWG-4038 std::text_encoding::aliases_view should have constexpr iterators
  • Recorded in P1885R12 <text_encoding>: Naming Text Encodings To Demystify Them #3791
• LWG-4045 tuple can create dangling references from tuple-like
  • Recorded in P2255R2 Type Traits To Detect References Binding To Temporaries #2534

🐞 Not yet implemented

• Not yet implemented
• Filed a GitHub issue labeled LWG
  • LWG-3767 codecvt<charN_t, char8_t, mbstate_t> incorrectly added to locale #4498
  • LWG-3919 enumerate_view may invoke UB for sized common non-forward underlying ranges #4499
  • LWG-3950 std::basic_string_view comparison operators are overspecified #4500
  • LWG-3984 ranges::to's recursion branch may be ill-formed #4501
  • LWG-4012 common_view::begin/end are missing the simple-view check #4502
  • LWG-4013 lazy_split_view::outer-iterator::value_type should not provide default constructor #4503
  • LWG-4016 container-insertable checks do not match what container-inserter does #4504
  • LWG-4035 single_view should provide empty #4505
  • LWG-4037 Static data members of ctype_base are not yet required to be usable in constant expressions #4506
  • LWG-4053 Unary call to std::views::repeat does not decay the argument #4507
  • LWG-4054 Repeating a repeat_view should repeat the view #4508
• PR out for review

[frederick-vs-ja]

Precondition checking for LWG-4023 and test coverages for LWG-4025/LWG-4031 may be wanted.

[StephanTLavavej]

Precondition checking for LWG-4023 and test coverages for LWG-4025/LWG-4031 may be wanted.

After looking at these issues, while I wouldn't object to PRs addressing them, I don't think it's so urgent that we need tracking issues.

• LWG-4023 involves protected member functions of iostreams. These aren't directly called, and people implementing new iostreams need to be experts.
• LWG-4025 would involve writing test coverage to observe the difference between "defined as deleted" and "constraints" which seems like a ton of effort for little return, given that we automatically got this right to begin with.
• LWG-4031 would involve writing test coverage to observe the noexceptness of protected member functions, which is doable but requires extra effort. We just =defaulted them, getting this automatically correct, and I don't think we need to bother testing this.

(In general, test coverage is most valuable for things that were tricky to achieve, or that might regress due to compiler or library changes.)

I've filed tracking issues for everything else, so we can close this meta-issue. Thanks again for doing the massive work needed to understand our current status, and so rapidly too! 😻 ✅ 🎉