February 2019 - Boost - lists.preview.boost.org

[gsoc2019][astronomy] Program competency test
by rittik verma 27 Feb '19

27 Feb '19

Hello, I was reading through the competency test for boost astronomy and was struck by a question, should each coordinate has its own unit, for example, x - length, y - time, z-force? or just one unit for all coordinates? Thank you

2 1

clang-win clang-cl.exe on Windows 10 - is it time to add clang-win to the Boost test matrix?
by Paul A. Bristow 27 Feb '19

27 Feb '19

Confirming the work of others (pdimov and degski et al), I am now getting good results using clang-win (Clang 8.0.0 rc using GCC provided by Mingw64) on Windows 10. (Clang 8.0.0 should be finally released very soon now). Working the develop branch (fully up-to-date), I can run the Boost.Math examples and tests, and build the a set of most libraries (4 versions of most, for example) 27-Feb-19 14:32 15,872 boost_atomic-clangw8-mt-d-x64-1_70.dll 27-Feb-19 14:32 2,442 boost_atomic-clangw8-mt-d-x64-1_70.lib 27-Feb-19 14:32 9,728 boost_atomic-clangw8-mt-x64-1_70.dll 27-Feb-19 14:32 2,420 boost_atomic-clangw8-mt-x64-1_70.lib (but there are some failures to compile or to link). Examples in Addendum below: My user-config.jam contains using clang-win : # Clang Version. 8.0.0 : # Location of Clang compiler .exe file(s). Must use clang-cl driver, not clang++.exe # Copied path is "C:\LLVM\clang-700\LLVM\bin\clang-cl.exe" "C:/LLVM/clang-800/LLVM/bin/clang-cl.exe" -fuse-ld=lld # Use lld linker https://lld.llvm.org/, not clang++. # : # Options. <cxxflags>-v # Verbose output for testing only. ? --verbose does not work for clang? # <define>BOOST_CONFIG_SUPPRESS_OUTDATED_MESSAGE # <compileflags>-m64 Use address-model=64 on command line. <compatibility>vc14 # MS version compatibility. <cxxflags>-fmsc-version=1915 <cxxflags>-fms-compatibility <cxxflags>-fno-delayed-template-parsing #https://clang.llvm.org/docs/CommandGuide/clang.html#cmdoption-s # options are -flto, -flto=full, -flto=thin, -emit-llvm # Generate output files in LLVM formats, suitable for link time optimization. # The default for -flto is "full", in which the LLVM bitcode is suitable for monolithic Link Time Optimization (LTO), # where the linker merges all such modules into a single combined module for optimization. # With "thin", ThinLTO compilation is invoked instead. <cxxflags>-flto=thin # MS clang-cl compatibility settings: <cxxflags>"/std:c++17" # Must be c++17, not earlier. <cxxflags>"/Zc:forScope" <cxxflags>"/Zc:wchar_t" <cxxflags>"/EHsc" # exception handling # https://docs.microsoft.com/en-us/cpp/build/reference/eh-exception-handling-… # (/EHsc), catches C++ exceptions only and tells the compiler to assume that # functions declared as extern "C" never throw a C++ exception. <cxxflags>"/Zc:inline" <cxxflags>"/Gd" <cxxflags>"/diagnostics:classic" # Clang flags: # https://clang.llvm.org/docs/ClangCommandLineReference.html#compilation-flags # <cxxflags>-std=c++17 - already declared as a MS option. <cxxflags>-Wno-deprecated <cxxflags>-Wno-delete-non-virtual-dtor # <cxxflags>-fmacro-backtrace-limit=0 # <cxxflags>-Wno-invalid-token-paste <cxxflags>-Wno-language-extension-token <cxxflags>-Wno-microsoft-enum-value <cxxflags>-Wno-literal-range <cxxflags>-Wno-unknown-pragmas <cxxflags>-Wno-unused-variable <cxxflags>-Wno-microsoft-enum-value # <cxxflags>-Wno-invalid-constexpr # Need to stop reporting an error in Boost.System? <cxxflags>-Wno-unknown-argument <cxxflags>-Wno-unknown-pragmas <cxxflags>-Wno-macro-redefined <cxxflags>-Wno-unused-variable <cxxflags>-Wno-mismatched-tags <cxxflags>-Wno-deprecated-declarations <define>_CRT_SECURE_NO_DEPRECATE <define>_CRT_SECURE_NO_WARNINGS # Not all these may be needed to get a mostly warning-free build. ... My build command is \modular-boost>b2 -j4 toolset=clang-win-8.0.0 address-model=64 architecture=x86 --libdir=I:/modular-boost/lib --build-type=complete --without=mpi --without=python install 1>../boost_clang-win-8.0.0_64_config_27-Feb-19_1507.log I wonder if this means that it is time to add clang-win-8 configuration to the Boost Test summary at https://www.boost.org/development/tests/develop/developer/summary.html ? Paul PS What I have failed to get working is __float128 which works using GCC, but even after a patch enabling Clang in multiprecision's config and ensuring that quadmath.h is visible to the Clang compiler. It reports "not supported" :-( I have enquired if any progress can be made on making this work. Addendum - problem examples: ======================== LLVM ERROR: Associative COMDAT symbol '??_7codecvt_error_cat@?A0x15DB13AB@@6B@' does not exist. lld-link: error: undefined symbol: __imp_CryptGenRandom >>> referenced by I:\modular-boost\libs\random\src\random_device.cpp:121 >>> lto.tmp:("public: unsigned int __cdecl boost::random::random_device::impl::next(void)" (?next@impl@random_device@random@boost@@QEAAIXZ)) clang-cl: error: linker command failed with exit code 1 (use -v to see invocation) "C:/LLVM/clang-800/LLVM/bin/clang-cl.exe" "-fuse-ld=lld" -m64 /link /incremental:no /DLL /DEBUG /subsystem:console /out:"bin.v2\libs\random\build\clang-win-8.0.0\debug\address-model-64\threading-multi\boost_random-clangw8-mt-d-x64-1_70.dll" /IMPLIB:"bin.v2\libs\random\build\clang-win-8.0.0\debug\address-model-64\threading-multi\boost_random-clangw8-mt-d-x64-1_70.lib" @"bin.v2\libs\random\build\clang-win-8.0.0\debug\address-model-64\threading-multi\boost_random-clangw8-mt-d-x64-1_70.dll.rsp" if %ERRORLEVEL% NEQ 0 EXIT %ERRORLEVEL% and most annoyingly, since Boost.Test is a key library, "I:\modular-boost\lib\boost_unit_test_framework-clangw8-mt-d-x64-1_70.dll" "I:\modular-boost\lib\boost_unit_test_framework-clangw8-mt-d-x64-1_70.lib" but no release variants :-( clang-win.link.dll bin.v2\libs\test\build\clang-win-8.0.0\release\address-model-64\threading-multi\boost_unit_test_framework-clangw8-mt-x64-1_70.dll LLVM ERROR: Associative COMDAT symbol '??_7results_reporter_impl@?A0x22C2280D@results_reporter@unit_test@boost@@6B@' does not exist. clang-cl: error: linker command failed with exit code 1 (use -v to see invocation) "C:/LLVM/clang-800/LLVM/bin/clang-cl.exe" "-fuse-ld=lld" -m64 /link /incremental:no /DLL /subsystem:console /out:"bin.v2\libs\test\build\clang-win-8.0.0\release\address-model-64\threading-multi\boost_unit_test_framework-clangw8-mt-x64-1_70. dll" /IMPLIB:"bin.v2\libs\test\build\clang-win-8.0.0\release\address-model-64\threading-multi\boost_unit_test_framework-clangw8-mt-x64-1_ 70.lib" @"bin.v2\libs\test\build\clang-win-8.0.0\release\address-model-64\threading-multi\boost_unit_test_framework-clangw8-mt-x64-1_70.dll. rsp" if %ERRORLEVEL% NEQ 0 EXIT %ERRORLEVEL% ...failed clang-win.link.dll bin.v2\libs\test\build\clang-win-8.0.0\release\address-model-64\threading-multi\boost_unit_test_framework-clangw8-mt-x64-1_70.dll bin.v2\libs\test\build\clang-win-8.0.0\release\address-model-64\threading-multi\boost_unit_test_framework-clangw8-mt-x64-1_70.lib... --- Paul A. Bristow Prizet Farmhouse Kendal UK LA8 8AB +44 (0) 1539 561830

1 0

Running regression tests for msvc-14.2 (preview)
by Tom Kent 27 Feb '19

27 Feb '19

I was hoping to start running regression tests for the upcoming Visual Studio 2019 (toolset 14.2) and got stumped by an issue. I've set a user-config.jam: using msvc : 14.2 ; using python : 2.7 # version : C:\\Python27-32\\python.exe # Interperter/path to dir : C:\\Python27-32\\include # includes : C:\\Python27-32\\libs # libs : <address-model>32 <address-model> # conditions ; using python : 2.7 # version : C:\\Python27-64\\python.exe # Interperter/path to dir : C:\\Python27-64\\include # includes : C:\\Python27-64\\libs # libs : <address-model>64 # conditions ; Then I ran the command, from a vs2019 developer command prompt: D:\tmp>C:\Python27-64\python.exe run.py --runner=teeks99-test02 --toolsets=msvc-14.2 --tag=develop The full build output (minus some repetitive git and mklink commands) can be found here: https://gist.github.com/teeks99/2d72382bf13569855dbdcd2e0850c1cf A couple notable parts... Bootstrap says it is using the vc7 toolset, which isn't installed D:\tmp\boost_bb\src\engine>build.bat could not find "vswhere" ### ### Using 'vc7' toolset. ### but succeeds anyway, probably because cl.exe is available in the path in the vs2019 command prompt. B2.exe is successfully created and seems to work. During... notice: [python-cfg] Configuring python... [snip] ImportError: No module named numpy but that is normal. Finally, at the end...the real issue: ...skipped <p..\..\bin\testing\build\msvc-14.2\release\threading-multi>process_jam_log.obj for lack of <p..\..\bin\standalone\msvc\msvc-14.2>msvc-setup.nup... ...skipped <p..\..\bin\common\build\msvc-14.2\release\threading-multi>process_jam_log.obj for lack of <p..\..\bin\standalone\msvc\msvc-14.2>msvc-setup.nup... common.mkdir D:\tmp\boost_root\bin.v2 if not exist "D:\tmp\boost_root\bin.v2\\" mkdir "D:\tmp\boost_root\bin.v2" common.mkdir D:\tmp\boost_root\bin.v2\libs if not exist "D:\tmp\boost_root\bin.v2\libs\\" mkdir "D:\tmp\boost_root\bin.v2\libs" common.mkdir D:\tmp\boost_root\bin.v2\libs\filesystem if not exist "D:\tmp\boost_root\bin.v2\libs\filesystem\\" mkdir "D:\tmp\boost_root\bin.v2\libs\filesystem" common.mkdir D:\tmp\boost_root\bin.v2\libs\filesystem\build if not exist "D:\tmp\boost_root\bin.v2\libs\filesystem\build\\" mkdir "D:\tmp\boost_root\bin.v2\libs\filesystem\build" common.mkdir D:\tmp\boost_root\bin.v2\libs\filesystem\build\msvc-14.2 if not exist "D:\tmp\boost_root\bin.v2\libs\filesystem\build\msvc-14.2\\" mkdir "D:\tmp\boost_root\bin.v2\libs\filesystem\build\msvc-14.2" common.mkdir D:\tmp\boost_root\bin.v2\libs\filesystem\build\msvc-14.2\release if not exist "D:\tmp\boost_root\bin.v2\libs\filesystem\build\msvc-14.2\release\\" mkdir "D:\tmp\boost_root\bin.v2\libs\filesystem\build\msvc-14.2\release" common.mkdir D:\tmp\boost_root\bin.v2\libs\filesystem\build\msvc-14.2\release\link-static if not exist "D:\tmp\boost_root\bin.v2\libs\filesystem\build\msvc-14.2\release\link-static\\" mkdir "D:\tmp\boost_root\bin.v2\libs\filesystem\build\msvc-14.2\release\link-static" common.mkdir D:\tmp\boost_root\bin.v2\libs\filesystem\build\msvc-14.2\release\link-static\threading-multi if not exist "D:\tmp\boost_root\bin.v2\libs\filesystem\build\msvc-14.2\release\link-static\threading-multi\\" mkdir "D:\tmp\boost_root\bin.v2\libs\filesystem\build\msvc-14.2\release\link-static\threading-multi" ...skipped <pD:\tmp\boost_root\bin.v2\libs\filesystem\build\msvc-14.2\release\link-static\threading-multi>codecvt_error_category.obj for lack of <p..\..\bin\standalone\msvc\msvc-14.2>msvc-setup.nup... ...skipped <pD:\tmp\boost_root\bin.v2\libs\filesystem\build\msvc-14.2\release\link-static\threading-multi>operations.obj for lack of <p..\..\bin\standalone\msvc\msvc-14.2>msvc-setup.nup... ...skipped <pD:\tmp\boost_root\bin.v2\libs\filesystem\build\msvc-14.2\release\link-static\threading-multi>path.obj for lack of <p..\..\bin\standalone\msvc\msvc-14.2>msvc-setup.nup... ...skipped <pD:\tmp\boost_root\bin.v2\libs\filesystem\build\msvc-14.2\release\link-static\threading-multi>path_traits.obj for lack of <p..\..\bin\standalone\msvc\msvc-14.2>msvc-setup.nup... ...skipped <pD:\tmp\boost_root\bin.v2\libs\filesystem\build\msvc-14.2\release\link-static\threading-multi>portability.obj for lack of <p..\..\bin\standalone\msvc\msvc-14.2>msvc-setup.nup... ...skipped <pD:\tmp\boost_root\bin.v2\libs\filesystem\build\msvc-14.2\release\link-static\threading-multi>unique_path.obj for lack of <p..\..\bin\standalone\msvc\msvc-14.2>msvc-setup.nup... ...skipped <pD:\tmp\boost_root\bin.v2\libs\filesystem\build\msvc-14.2\release\link-static\threading-multi>utf8_codecvt_facet.obj for lack of <p..\..\bin\standalone\msvc\msvc-14.2>msvc-setup.nup... ...skipped <pD:\tmp\boost_root\bin.v2\libs\filesystem\build\msvc-14.2\release\link-static\threading-multi>windows_file_codecvt.obj for lack of <p..\..\bin\standalone\msvc\msvc-14.2>msvc-setup.nup... ...skipped <pD:\tmp\boost_root\bin.v2\libs\filesystem\build\msvc-14.2\release\link-static\threading-multi>libboost_filesystem-vc142-mt-1_70.lib for lack of <pD:\tmp\boost_root\bin.v2\libs\filesystem\build\msvc-14.2\release\link-static\threading-multi>codecvt_error_category.obj... ...skipped <p..\..\bin\testing\build\msvc-14.2\release\threading-multi>process_jam_log.exe for lack of <p..\..\bin\testing\build\msvc-14.2\release\threading-multi>process_jam_log.obj... ...skipped <pD:\tmp\boost_regression\stage\bin>process_jam_log.exe for lack of <p..\..\bin\testing\build\msvc-14.2\release\threading-multi>process_jam_log.exe... ...skipped 14 targets... ...updated 599 targets... # Searching for "process_jam_log.exe" in "D:\tmp\boost_regression\stage\bin"... Traceback (most recent call last): File "run.py", line 71, in <module> runner(root) File "D:\tmp\boost_regression_src\regression.py", line 286, in __init__ self.main() File "D:\tmp\boost_regression_src\regression.py", line 623, in main getattr(self,action_m)() File "D:\tmp\boost_regression_src\regression.py", line 580, in command_regression self.command_setup() File "D:\tmp\boost_regression_src\regression.py", line 351, in command_setup self.build_if_needed(self.process_jam_log,self.pjl_toolset) File "D:\tmp\boost_regression_src\regression.py", line 714, in build_if_needed tool[ 'build_path' ] = self.tool_path( tool ) File "D:\tmp\boost_regression_src\regression.py", line 739, in tool_path , '\n'.join( [ name_or_spec[ 'path' ], build_dir ] ) Exception: Cannot find "process_jam_log.exe" in any of the following locations: D:\tmp\boost_regression\stage\bin\process_jam_log.exe D:\tmp\boost_regression\stage\bin What is the msvc-setup.nup and where does it come from? Thanks, Tom

3 4

Building the Release Notes locally
by Vinnie Falco 27 Feb '19

27 Feb '19

Does anyone know how to build the release notes locally? I would like to know if they render correctly without going through an 8-hour Travis turnaround. The rumor mill says @Daniel James knows how to do it... Anyone?? Thanks

2 2

[Boost.astronomy]
by Viraj Sonaje 24 Feb '19

24 Feb '19

This is for the respective authority I am a 4th year engineering student. I want to contribute to Boost.astronomy library for GSOC 2019. I am familiar with the co-ordinate systems but I am new to boost.

2 1

Graph Algorithms
by Siddhartha Sen 24 Feb '19

24 Feb '19

Is there any interest in a library containing graph algorithms? Also, I am unable to find the list of maintainers of the respective libraries. Any information on that would be appreciated.

7 7

add serialization to circular_buffer
by Yuval Lifshitz 24 Feb '19

24 Feb '19

Dear Community,I have a PR [1] for supporting serialization of the circular_buffer. Main drive for this change, is my effort for adding serialization to boost accumulators [2] (some accumulators hold data internally in a circular buffer).The maintainers list [3] is not up to date regarding circular buffer, so, I'm not sure who the maintainer is (or if there is one). Would appreciate your help, so I can drive that to completion. Yuval Lifshitz [1] https://github.com/boostorg/circular_buffer/pull/25[2] https://github.com/boostorg/accumulators/pull/18 [3] https://github.com/boostorg/boost/blob/master/libs/maintainers.txt

1 0

BGL
by Siddhartha Sen 22 Feb '19

22 Feb '19

I looked up https://github.com/boostorg/boost/wiki/Google-Summer-of-Code:-2019 . No project ideas have been added for BGL as of yet, it seems. I hope there is something in BGL in this years GSOC, as I am really passionate about graphs and graph algorithms. Also, will you be mentoring this year? If not, do you have any information on the potential mentors for this BGL project? Thanks.

2 1

Re: [boost] Graph Algorithms
by Siddhartha Sen 22 Feb '19

22 Feb '19

Thanks for the feedback! But what if I focus on C++14 then? Is that a good idea? On Fri, 22 Feb, 2019, 6:14 PM Vinnie Falco, <vinnie.falco(a)gmail.com> wrote: > On Fri, Feb 22, 2019 at 12:49 AM Siddhartha Sen > <siddhartha99.sen(a)gmail.com> wrote: > > From what I've read in the mailing lists, I think the demand has > shifted > > from C++11 to more modern versions, like C++14 and C++17. Anyways, > > thanks for the tip. Will you be mentoring this year? > > I will not be mentoring directly although I do offer advice on the mailing > list. > > People say they want C++14 and C++17 but I don't think its a good idea > to require those language versions in a library, because most people > are still on C++11. Have a look at the latest developer survey: > > <https://www.jetbrains.com/research/devecosystem-2018/cpp/> > > If a library requires C++17 then most C++ users will not be able to > use it. And this is bad for the library, because fewer users means > less testing, less feedback, less popularity, and in general this > makes the library not as good. > > In terms of the API (interface) of a library, C++17 does not offer > much more than C++11 except for things like, avalability of > string_view, variant. But Boost has these things already (e.g. > boost::string_view, boost::variant) and they work in C++11. So there > is really no benefit to C++17 other than some improvements in the way > you can write the implementation. But that doesn't affect users > directly. > > In summary, by requiring C++17 most users will be unable to use the > code, and there will be very little benefit to C++17 users. > > Regards >

3 7

BGL
by Siddhartha Sen 22 Feb '19

22 Feb '19

I apologise. Actually, there was a suggestion in the boost mailing lists that there is a demand for this. Anyways, I am familiar with C++14 and a C++17 as well(a bit). Will you be mentoring this year? I have already read the documentation of LEMON and started using it as well. Please let me know what should I do next, or if there are any particular issues you would like me to work on. Thanks. Regards. Siddhartha.

2 1