P1642R1 Freestanding Library: Easy [utilities], [ranges], and [iterators] (Ben Craig)

P1642R2 Freestanding Library: Easy [utilities], [ranges], and [iterators] (Ben Craig)

Prague 2020-02 LEWGI Minutes

P1642R2 Freestanding [utilities], [ranges], and [iterators]: Design Review

Chair: Bryce Adelstein Lelbach

Champion: Ben Craig

Minute Taker: Peter Bindels

Start Review: 2020-02-10 11:06

Major design points:

• Splitting std::swap overload set: some std::swap overloads will be freestanding, some will not.
• Omitting feature test macros.

Should std::function be in freestanding?

End: 11:31

CONSENSUS: LEWGI sends P1642R2 (Freestanding Status Quo) to LEWG.

P1642R3: Freestanding [utilities], [ranges], and [iterators]

2020-06-09 Library Evolution Telecon Minutes

Chair: Bryce Adelstein Lelbach

Champion: Ben Craig

Minute Taker: Mark Hoemmen

Start Review: 2020-06-09 10:06 Pacific

POLL: Add __cpp_lib_freestanding_* feature test macros (either per header or per feature).

Attendance: 26

# of Authors: 1

Author Position: WF

That has strong consensus.

POLL: Add __cpp_lib_freestanding_* per-header feature test macros (proposed by P1642R3).

Attendance: 26

# of Authors: 1

Author Position: SF

That does not have consensus.

POLL: The freestanding versions of <tuple>, <ratio>, and <utility> proposed by P1642R3, excluding feature test macros, are ready to forward to LWG.

Attendance: 26

# of Authors: 1

Author Position: SF

That has strong consensus.

POLL: The freestanding versions of <ranges>, <memory>, <iterator>, and <functional> proposed by P1642R3, excluding feature test macros, are ready to forward to Library.

Attendance: 26

# of Authors: 1

Author Position: SF

That has strong consensus.

End: 11:40

SUMMARY:

We reviewed P1642R3, which proposes adding library features to freestanding that
are relatively straightforward to include. The proposed additions to
freestanding - <tuple>, <utility> and <ratio>, and a subsetted versions of
<ranges>, <memory>, <iterator>, and <functional> - were largely
non-controversial and we had strong consensus that these parts were ready to move
forward to Library.

We discussed how feature test macros should work in freestanding at great length.
We agreed that we wanted freestanding-specific feature test macros, although it
is an open question whether said feature test macros should also be defined in
hosted. Additional investigation on freestanding feature test macros is needed.

P1642R3 depends on P1641R3, which adds some changes how the freestanding library
is defined and adds some front matter; we reviewed P1641R3 last week.
We've asked the author to take the freestanding feature test macro content
out of P1641R3 and P1642R3 and put it in a new paper, and then combine P1641R3
and P1642R3 into a single paper, have a wordsmith (Walter Brown) review the
more complicated wording from P1641R3, and then bring the combined paper back
to Library Evolution for a final review before forwarding it to Library.

CONSENSUS:

• Split the feature test macro parts of P1641 and P1642 out into a third
  paper. Explore whether __cpp_lib_freestanding_* feature test macros
  should be defined in both hosted and freestanding. Show examples of usage.
• Combine P1641 and P1642 minus feature test macros into one paper; return to
  Library Evolution with said paper. We will review the non P1642 parts with
  the goal of advancing them to Library.
• We need to make some policy decisions here at some point, such as how to
  handle partially supported standard library classes in freestanding, how to
  handle partial headers, the rubric for what should be included in
  freestanding. P0829, P1641, and P1642 cover some of this. Consider writing
  a policy paper and proposing a standing document.

P1642R4 Freestanding Library: Easy [utilities], [ranges], and [iterators] (Ben Craig)

P1642R4: Freestanding [utilities], [ranges], and [iterators]

2020-11-30 Library Evolution Telecon Minutes

Chair: Nevin Liber

Champion: Ben Craig

Minute Taker: Inbal Levi

POLL: Send P1642R4 (Freestanding [utilities], [ranges], and [iterators]) after rebasing against the current WD and add _Exit to electronic ballot to be forwarded to LWG for C++IS.

Attendance: 19

Author Position: SF

Conclusion: Consensus in favor.

POLL: Which priority bucket should P1642 (Freestanding [utilities], [ranges], and [iterators]) be placed in?

Bucket 2: 10

Bucket 3: 1

Attendance: 19

Author Position: Bucket 2

Conclusion: Consensus for Bucket 2.

Summary

TODO (Nevin)

Outcome

TODO (Nevin)

P1642R6 Freestanding Library: Easy [utilities], [ranges], and [iterators] (Ben Craig)

Related polling history in #428

P1642R7 Freestanding Library: Easy [utilities], [ranges], and [iterators] (Ben Craig)

The 2021 Winter Library Evolution poll did not specify ship vehicle due to a typo, but I believe that C++23 Library Evolution's understanding and desire.

P1642R8 Freestanding Library: Easy [utilities], [ranges], and [iterators] (Ben Craig)

P1642R9 Freestanding Library: Easy [utilities], [ranges], and [iterators] (Ben Craig)

P1642R10 Freestanding Library: Easy [utilities], [ranges], and [iterators] (Ben Craig)

CWG telecon 2022-07-01: Approved after minor adjustments. LWG to move in plenary.

Adding notes from earlier LWG review of R9 on 2022-06-17

https://wiki.edg.com/bin/view/Wg21telecons2022/P1642-20220617

poll: put P1642r10 (to be checked by JG and TK and pending Core approval) into C++23

poll: put P1642r10 (to be checked by JG and TK and pending Core approval) into C++23

Core requested another change, so the poll will be for P1642R11

P1642R11 Freestanding Library: Easy [utilities], [ranges], and [iterators] (Ben Craig)

I'm sorry for bothering here... Has <numbers> ever been discussed?

I think anything involving floating-point types was out of scope of this paper and will be addressed later.