libc++ Kona Status
This is a temporary page; please check the c++1z status here
This page shows the status of the papers and issues that are expected to be adopted in Kona.
The groups that have contributed papers:
- LWG - Library working group
- CWG - Core Language Working group
- SG1 - Study group #1 (Concurrency working group)
Paper Status
| Paper # | Group | Paper Name | Meeting | Status | First released version |
|---|
Library Working group Issues Status
| Issue # | Issue Name | Meeting | Status |
|---|---|---|---|
| 1169 | num_get not fully compatible with strto* | Kona | |
| 2072 | Unclear wording about capacity of temporary buffers | Kona | Complete |
| 2101 | Some transformation types can produce impossible types | Kona | |
| 2111 | Which unexpected/terminate handler is called from the exception handling runtime? | Kona | Complete |
| 2119 | Missing hash specializations for extended integer types | Kona | |
| 2127 | Move-construction with raw_storage_iterator | Kona | Patch Ready |
| 2133 | Attitude to overloaded comma for iterators | Kona | Complete |
| 2156 | Unordered containers' reserve(n) reserves for n-1 elements | Kona | |
| 2181 | Exceptions from seed sequence operations | Kona | |
| 2218 | Unclear how containers use allocator_traits::construct() | Kona | |
| 2219 | INVOKE-ing a pointer to member with a reference_wrapper as the object expression | Kona | |
| 2244 | Issue on basic_istream::seekg | Kona | Patch Ready |
| 2250 | Follow-up On Library Issue 2207 | Kona | |
| 2259 | Issues in 17.6.5.5 rules for member functions | Kona | Complete |
| 2336 | is_trivially_constructible/is_trivially_assignable traits are always false | Kona | |
| 2353 | std::next is over-constrained | Kona | |
| 2367 | pair and tuple are not correctly implemented for is_constructible with no args | Kona | |
| 2380 | May <cstdlib> provide long ::abs(long) and long long ::abs(long long)? | Kona | Complete |
| 2384 | Allocator's deallocate function needs better specification | Kona | Complete |
| 2385 | function::assign allocator argument doesn't make sense | Kona | |
| 2435 | reference_wrapper::operator()'s Remark should be deleted | Kona | Complete |
| 2447 | Allocators and volatile-qualified value types | Kona | Complete |
| 2462 | std::ios_base::failure is overspecified | Kona | Complete |
| 2466 | allocator_traits::max_size() default behavior is incorrect | Kona | Patch Ready |
| 2469 | Wrong specification of Requires clause of operator[] for map and unordered_map | Kona | |
| 2473 | basic_filebuf's relation to C FILE semantics | Kona | Complete |
| 2476 | scoped_allocator_adaptor is not assignable | Kona | Patch Ready |
| 2477 | Inconsistency of wordings in std::vector::erase() and std::deque::erase() | Kona | Complete |
| 2483 | throw_with_nested() should use is_final | Kona | Complete |
| 2484 | rethrow_if_nested() is doubly unimplementable | Kona | Complete |
| 2485 | get() should be overloaded for const tuple&& | Kona | |
| 2486 | mem_fn() should be required to use perfect forwarding | Kona | |
| 2487 | bind() should be const-overloaded, not cv-overloaded | Kona | Complete |
| 2489 | mem_fn() should be noexcept | Kona | Patch Ready |
| 2492 | Clarify requirements for comp | Kona | Complete |
| 2494 | [fund.ts.v2] ostream_joiner needs noexcept | Kona |
Comments about the issues
- 1169 - We currently have a single function __num_get_float that calls strtold_l, and then casts down to the appropriate floating point type. That will have to change.
- 2072 - We already do this; this is just making the wording better.
- 2101 - Need to write some careful test cases. In particular, need to check function types with/without const/ref qualifiers. Currently we get this wrong. Installing metashell to play with these.
- 2111 - Resolved an ambiguity by calling it out. No code change required.
- 2119 - Hashes for all integral and enumeration types. Research needed
- 2127 - Add a new member to raw_storage_iterator. Looks simple. Patch Available
- 2133 - We do this already; thanks Eric.
- 2156 - check and make sure that we already do this. Write a test.
- 2181 - I suspect that this will not require any code changes, but will need to be read carefully.
- 2218 - Shouldn't require any code changes.
- 2219 - Sizable changes required. INVOKE needs 2 additional overloads and plenty of tests. I'm not going to back port this to C++03. (Eric)
- 2244 - We don't do this; easy fix. Patch Available
- 2250 - Looks like wording cleanup. Need to check more closely, but I think there's no code changes here.
- 2259 - No code changes needed here.
- 2336 - Check later
- 2353 - Simple change, needs a test. (test probably used to exist)
- 2367 - Already done in tuple, pair needs to be done. Tests. (Eric)
- 2380 - No code changes here; we already do this.
- 2384 - Wording cleanup; no code change required
- 2385 - Removing broken signatures. Only question is "how far back"?
- 2435 - Wording cleanup; no code change required
- 2447 - This is a relaxation of a requirement that we satisfy. No changes needed..
- 2462 - Added a simple test. No code change needed Patch Available
- 2466 - Simple change; need a test. Patch Available
- 2469 - This is a follow on to 2464, which we're done. This restates a bunch of operations in terms of newer ops. Probably refactoring to do here, but tests shouldn't change.
- 2473 - There are two changes here; one to filebuf::seekpos and the other to filebuf::sync. We do both of these already.
- 2476 - Just needed tests. Patch Available
- 2477 - Wording cleanup.
- 2483 - We already do this.
- 2484 - We already do this.
- 2485 - Ask Eric to do it.
- 2486 - Lots of code changes, all mechanical. Tests will be sizable.
- 2487 - This is removing a requirement on bind(), which we don't currently satisfy. So, nothing to do.
- 2489 - Looks easy. Just add some NOEXCEPT, and tests. Patch Available
- 2492 - Wording cleanup; no code changes needed.
- 2494 - My implementation of this (not checked in) already has these.
Last Updated: 5-Oct-2015