commit ee18548b1eb68a592d0980611b5f96216ee71a2d (tag: boost-1.82.0.beta1)
Author: Automated Commit <automated(a)calamity.org.uk>
Date: Thu Mar 9 00:00:15 2023 +0000
Update multiprecision from develop
Looks like the boost-1.82.0.beta1 tag points to develop rather than master.
El 27/03/2023 a las 19:39, Peter Dimov via Boost escribió:
> Joaquin M López Muñoz wrote:
>> El 16/03/2023 a las 16:38, Marshall Clow via Boost escribió:
>>> The master branch is now open for bug fixes and documentation changes.
>>> Other changes by release manager permission, as described in
>>> https://github.com/boostorg/wiki/wiki/Releases%3A-Beta-Merge-Policy
>> Hi, I'd like to commit the following changes:
>>
>> * multi_index: removal of foreach and serialization from dependencies listed
>> in CMakeLists.txt:
>> https://github.com/boostorg/multi_index/commit/b6d5c88629d3ca3b0822e
>> 87629c28ebf151c6a9e
>> https://github.com/boostorg/multi_index/commit/0e55b4878952375c8f8c0
>> 0b6dc6964423644970e
>>
>> * flyweight: removal of serialization from dependencies listed in
>> CMakeLists.txt:
>> https://github.com/boostorg/flyweight/commit/b37873c46902f06d6c131a5
>> 25493ff6d29180efa
> As a general reminder to everyone, if you make changes that add or remove
> dependencies, and your CMakeLists.txt file is generated by `boostdep --cmake`,
> and contains no further changes, you should regenerate it (by replacing it with
> the output of `boostdep --cmake my_library`.)
Thanks for the tip, I've just run boostdep and results coincide with
what there's already on the CML
files. Marshall, OK to merge to master?
Joaquín M López Muñoz
On Wed, Mar 22, 2023 at 2:41 AM Andrey Semashev via Boost <
boost(a)lists.boost.org> wrote:
> https://github.com/Lastique/scope
Hi. Maybe you should tag your library on GitHub, so it can be easily found,
among the several which "compete" with it. Like those listed at the link
below:
https://github.com/topics/scope-guard?l=c%2B%2B
One in particular (https://github.com/offa/scope-guard) refers to the WG21
paper.
PS: Actually, I wonder whether Boost Libraries in general use GH tagging
much.
Joaquin M López Muñoz wrote:
> El 16/03/2023 a las 16:38, Marshall Clow via Boost escribió:
> > The master branch is now open for bug fixes and documentation changes.
> > Other changes by release manager permission, as described in
> > https://github.com/boostorg/wiki/wiki/Releases%3A-Beta-Merge-Policy
>
> Hi, I'd like to commit the following changes:
>
> * multi_index: removal of foreach and serialization from dependencies listed
> in CMakeLists.txt:
> https://github.com/boostorg/multi_index/commit/b6d5c88629d3ca3b0822e
> 87629c28ebf151c6a9e
> https://github.com/boostorg/multi_index/commit/0e55b4878952375c8f8c0
> 0b6dc6964423644970e
>
> * flyweight: removal of serialization from dependencies listed in
> CMakeLists.txt:
> https://github.com/boostorg/flyweight/commit/b37873c46902f06d6c131a5
> 25493ff6d29180efa
As a general reminder to everyone, if you make changes that add or remove
dependencies, and your CMakeLists.txt file is generated by `boostdep --cmake`,
and contains no further changes, you should regenerate it (by replacing it with
the output of `boostdep --cmake my_library`.)
On 3/15/23 18:38, Marshall Clow via Boost wrote:
> Boost release 1.82.0 beta 1 is now available at:
> https://boostorg.jfrog.io/artifactory/main/beta/1.82.0.beta1/source/
>
> The SHA256 checksums are:
> efe17bea3f0b2c11f34828f37d5777f30eb6fef57af355c6a6664b08a4339999 boost_1_82_0_b1.7z
> 0af2a60f76abb4f411b1bd9c1a553f64454d9414990e8cc12f02f3b431e2ac45 boost_1_82_0_b1.tar.gz
> 495b2ae2f6970ea1e635f80222d9f6978148496cdd163e87c37cdd6de3b5ccfc boost_1_82_0_b1.zip
> edf4cb9e8e5a14412b91311438478ef6ce12afe2a54714eb2f0632a454c3d8d5 boost_1_82_0_b1.tar.bz2
>
> Release notes (still need updates):
> https://www.boost.org/users/history/version_1_82_0.html
>
> Please try the beta and report any problems you encounter.
There is a bug in Boost.MultiIndex CMakeLists.txt, although it may not
affect the monolithic release tarball.
The CMakeLists.txt lists a dependency on Boost::foreach target while no
Boost.MultiIndex headers require it. This causes CI failures in
Boost.Log[1] because boostdep does not find the dependency from headers
and doesn't checkout Boost.ForEach git repository, which makes CMake to
fail to resolve the dependency later.
[1]: https://github.com/boostorg/log/actions/runs/4531415733/jobs/7981536579
On 3/15/23 18:38, Marshall Clow via Boost wrote:
> Boost release 1.82.0 beta 1 is now available at:
> https://boostorg.jfrog.io/artifactory/main/beta/1.82.0.beta1/source/
>
> The SHA256 checksums are:
> efe17bea3f0b2c11f34828f37d5777f30eb6fef57af355c6a6664b08a4339999 boost_1_82_0_b1.7z
> 0af2a60f76abb4f411b1bd9c1a553f64454d9414990e8cc12f02f3b431e2ac45 boost_1_82_0_b1.tar.gz
> 495b2ae2f6970ea1e635f80222d9f6978148496cdd163e87c37cdd6de3b5ccfc boost_1_82_0_b1.zip
> edf4cb9e8e5a14412b91311438478ef6ce12afe2a54714eb2f0632a454c3d8d5 boost_1_82_0_b1.tar.bz2
>
> Release notes (still need updates):
> https://www.boost.org/users/history/version_1_82_0.html
>
> Please try the beta and report any problems you encounter.
There is a bug in Boost.ASIO that prevents Boost.Log from compiling with
gcc 4.4 and clang 3.5. The bug is in BOOST_ASIO_ASSUME definition:
https://github.com/boostorg/asio/blob/boost-1.82.0.beta1/include/boost/asio…
1. gcc 4.4 does not support __builtin_unreachable, which makes the
__GNUC__ branch definition to fail.
2. clang 3.5 doesn't support __builtin_assume and uses the last fallback
option. There, BOOST_ASIO_ASSUME is defined without arguments, meaning
it expands to incorrect code like `(void)0(base != 0)`.
GitHub is dead for me ATM, so I can't create a PR.
