April 2020 - Boost-users - lists.preview.boost.org

Boost.Numeric.Interval: Please specify rounding control mechanism
by Bastien Commelongue 01 Dec '20

01 Dec '20

Hello, I am building an app for IOS (ARM) and I had this issue while compiling: *Boost.Numeric.Interval: Please specify rounding control mechanism.* C99 version should be used in this case but __USE_ISOC99 is not defined. Modifying the condition with this code (using __cplusplus) fixed the issue: *boost/numeric/interval/hw_rounding.hpp(36)* *#if defined(BOOST_NUMERIC_INTERVAL_NO_HARDWARE) && (defined(__USE_ISOC99) || defined(__MSL__) || (defined __cplusplus && __cplusplus >= 201103L))# include <boost/numeric/interval/detail/c99_rounding_control.hpp>#endif* Let me know if I missed something or if you plan to integrate this modification in a next version. Cheers, Bastien Commelongue

2 1

[Boost.Test] Three-way comparison operator
by Leo Carreon 17 May '20

17 May '20

Hi Boosters, Does anyone know if Boost.Test has been updated to provide support for the three-way comparison operator? I’m asking because I’ve encountered a situation where my user-defined class provides the three-way comparison operator but when I perform comparisons using Boost.Test it complains that my class has no ‘operator==’ which is true because it provides ‘operator<=>’. Please note that I’m using Boost 1.73.0 and g++ 10.0.1. Is there any plan to add the three-way comparison operator support? Kind regards, Leo

2 1

Three-way comparison operator
by Leo Carreon 30 Apr '20

30 Apr '20

Hi Boosters, Does anyone know if Boost.Test has been updated to provide support for the three-way comparison operator? I’m asking because I’ve encountered a situation where my user-defined class provides the three-way comparison operator but when I perform comparisons using Boost.Test it complains that my class has no ‘operator==’ which is true because it provides ‘operator<=>’. Please note that I’m using Boost 1.73.0 and g++ 10.0.1. Is there any plan to add the three-way comparison operator support? Kind regards, Leo

4 4

Windows boost::filesystem canonical on NTFS network drives mounted in a directory
by gga 29 Apr '20

29 Apr '20

When on Windows, if you use boost::filesystem::canonical() to resolve a NTFS network drive mounted in a directory, the result is what I see as a useless result. For example: C:\ntfs_mount\other\dirs\file boost::filesystem returns: C:\?VolumeXXX-XXX-XXX-XXX\ other\dirs\file This volume notation is useless as it cannot be passed to any file function. Also, the path is divided by a space to other\dirs. It would be better if canonical does not mess with the network drive at all and returns the directory as is, like boost::filesystem::absolute() does. -- Gonzalo Garramuño

1 0

[release] Boost 1.73.0
by Marshall Clow 28 Apr '20

28 Apr '20

Release 1.73.0 of the Boost C++ Libraries is now available. These open-source libraries work well with the C++ Standard Library, and are usable across a broad spectrum of applications. The Boost license encourages both commercial and non-commercial use. This release contains 2 new libraries and numerous enhancements and bug fixes for existing libraries. For details, including download links, see < http://www.boost.org/users/news/version_1.73.0> You can also download directly from: < https://dl.bintray.com/boostorg/release/1.73.0/> To install this release on your system, see < http://www.boost.org/doc/libs/release/more/getting_started/index.html> Thanks to everyone who participated in this release. Thanks, -- The Boost Release Team Marshall Clow, Micheal Caisse and Glen Fernandes

1 0

[release] Boost 1.73.0 Release Candidate 1
by Marshall Clow 26 Apr '20

26 Apr '20

The first release candidates for the 1.73.0 release are now available at: <https://dl.bintray.com/boostorg/release/1.73.0/source/> The SHA256 checksums are as follows: d2e7501bb04fe7abc09aa93f013ae997604286a882da1bd36ddd96ea1163ea71 boost_1_73_0_rc1.7z 4eb3b8d442b426dc35346235c8733b5ae35ba431690e38c6a8263dce9fcbb402 boost_1_73_0_rc1.tar.bz2 9995e192e68528793755692917f9eb6422f3052a53c5e13ba278a228af6c7acf boost_1_73_0_rc1.tar.gz 0909a79524f857ef54570ceef8f397cc0629202532cc997785479c7c08bbc2a4 boost_1_73_0_rc1.zip As always, the release managers would appreciate it if you download the candidate of your choice and give building it a try. Please report both success and failure, and anything else that is noteworthy. -- The Release managers

3 7

Using coroutines in ASIO
by Pedro Pinto 25 Apr '20

25 Apr '20

Hi there, I am trying to compile the asio co-routine example but as far as I can tell boost is looking for an <experimental/coroutine> header that does not exist on my system. I am using boost 1.71 and gcc on ubuntu 20.04. I tried both stock gcc (v9 on ubuntu) which does not have a coroutine header and gcc 10 which does have it but not under the experimental directory. Here is the output with gcc 10: /usr/lib/ccache/g++ -I../../src -g -Wall -pedantic -Wextra -Werror -DASIO_HAS_CO_AWAIT -DBOOST_ASIO_HAS_CO_AWAIT -fcoroutines --std=c++2a -fdiagnostics-color -MD -MT CMakeFiles/mcbridge.dir /src/main.cpp.o -MF CMakeFiles/mcbridge.dir/src/main.cpp.o.d -o CMakeFiles/mcbridge.dir/src/main.cpp.o -c ../../src/main.cpp In file included from /usr/include/boost/asio/co_spawn.hpp:22, from ../../src/main.cpp:4: /usr/include/boost/asio/awaitable.hpp:22:10: fatal error: experimental/coroutine: No such file or directory 22 | #include <experimental/coroutine> Any ideas on how to compile this example?

2 1

10 slots for the Google Summer of Code 2020
by David Bellot 22 Apr '20

22 Apr '20

Hi Boost community, this year Google allocated 10 slots for the Google Summer of Code. It's again a great achievement thanks to the hard work of our team of dedicated mentors. We are now selecting the best students among all the proposals we receive. We received 79 proposals this year. 10 of them will be selected. This year, a lot of proposals if not all are targeting a finalization of the work which has been done in the previous GSoCs and we should see a lot of code being pushed in production. It's a very exciting time for all of us. The mentors are now hard at work to choose the students. We'll keep you informed when it's official. Best regards, David

1 0

Re: [Boost-users] [random_device] App crash on boost::ramdom::random_device
by 박형근(HYUNGGUN PARK) 22 Apr '20

22 Apr '20

> On Fri, Apr 17, 2020 at 10:29 AM 박형근(HYUNGGUN PARK) via Boost-users > <boost-users(a)lists.boost.org> wrote: > > > > My app crash on customer's site. It crashes everytime the app is > starting. > > It happens on Chinese Windows7 (64bit) OS but it runs well on my PC > > with > Korean Windows7 OS and English Windows7 OS. > > My app is built on VisualStudio 2010 (x86 build, C++) using > > websocketpp > library and boost 1.55. > > The call-stack from the dump file shows my app(TMSClient100) called > websocketpp::endpoint, then it called boost::random::random_device > then ThrowException. Here's the call-stack info: > > > > 14 001cf6ac 7113872d e06d7363 00000001 00000003 > > KERNELBASE!RaiseException+0x58 > > 15 001cf6e4 713b7c5a 001cf704 713d9350 b621b77b > > msvcr100!_CxxThrowException+0x48 > > [f:\dd\vctools\crt_bld\self_x86\crt\prebuild\eh\throw.cpp @ 157] > > 16 001cf814 713b7da3 001cf840 b621b9a7 002a19bc > > TMSClient100!boost::random::random_device::impl::error+0x15a > > 17 001cf8b8 713b7f7b 001cf8dc b621b94b 00000000 > > TMSClient100!boost::random::random_device::impl::impl+0x143 > > Check the exception message. Is it failing because it: > 1. Could not find a provider name > 2. Could not acquire CSP context > > If #1, this might be solved by upgrading from Boost 1.55 to a newer > version. This fix in particular: > https://github.com/boostorg/random/commit/dc2fdeac443fb56c87a03a0887c9 > e7c8 > e1d3296d > went into Boost 1.56. > > Glen Glen, Thank you for your reply. Thanks to your suggestion, upgrade from Boost 1.55 to 1.56 solved this problem. Thank you very much. Harry

1 0

[Boost] [test] strange behaviour when comparing floating point numbers
by Matthijs Möhlmann 21 Apr '20

21 Apr '20

Hi, I am trying to compare two floating point numbers but am encountering somewhat strange behaviour. The minimal compilable example: #define BOOST_TEST_MODULE tests #define BOOST_TEST_DYN_LINK #include <boost/test/unit_test.hpp> BOOST_AUTO_TEST_SUITE(testsuite) BOOST_AUTO_TEST_CASE(test1) { namespace tt = boost::test_tools; BOOST_TEST(0.1 == 0.2, tt::tolerance(0.01)); // will fail BOOST_TEST(0.1f == 0.2f, tt::tolerance(0.02)); // will fail too, but different error message... } BOOST_AUTO_TEST_SUITE_END() When compiling with: clang++ (clang version 7.0.1-8 (tags/RELEASE_701/final)) flags: -o main -g -std=c++17 -stdlib=libc++ -lboost_unit_test_framework main.cpp boost: 1.72.0 I'll get the following output: Running 1 test case... main.cpp(10): error: in "testsuite/test1": check 0.1 == 0.2 has failed [0.10000000000000001 != 0.20000000000000001]. Relative difference exceeds tolerance [1 > 0.01] main.cpp(11): error: in "testsuite/test1": check 0.1f == 0.2f has failed [0.100000001 != 0.200000003] *** 2 failures are detected in the test module "tests" Why am I getting two different messages? I expected those to be the same... Regards, Matthijs

2 2