- Boost-users - lists.boost.org

Problems with nesting boost::container::small_vector and C++98
by Daniel Trstenjak 02 Jan '17

02 Jan '17

Hi, first the used versions: boost 1.63.0 gcc 5.4.0 20160609 Ubuntu 16.04.1 the example I've problems with is: #include <map> #include <boost/container/small_vector.hpp> using boost::container::small_vector; typedef std::pair<int, small_vector<float, 8> > mypair; typedef std::map<int, mypair> mymap; typedef std::map<int, small_vector<mypair, 8> > mymap2; typedef std::map<int, small_vector<small_vector<float, 8>, 8> > mymap3; int main() { mymap m; mypair& sv = m[0]; mymap2 m2; small_vector<mypair, 8>& sv2 = m2[0]; mymap3 m3; small_vector<small_vector<float, 8>, 8>& sv3 = m3[0]; } I can compile it without errors with 'g++ -std=c++11', but without '-std=c++11' I'm getting an error for the 'operator[]' on mymap2: dan@octa ~> g++ -Iboost_1_63_0/ small_vector_test.cpp In file included from boost_1_63_0/boost/container/small_vector.hpp:27:0, from small_vector_test.cpp:2: boost_1_63_0/boost/container/vector.hpp: In instantiation of ‘void boost::container::vector<T, Allocator>::assign(FwdIt, FwdIt, typename boost::move_detail::disable_if_or<void, boost::move_detail::is_same<typename boost::container::container_detail::version<Allocator>::type, boost::move_detail::integral_constant<unsigned int, 0u> >, boost::move_detail::is_convertible<InIt, typename boost::container::allocator_traits<Allocator>::size_type>, boost::container::container_detail::is_input_iterator<FwdIt> >::type*) [with FwdIt = boost::container::container_detail::vec_iterator<std::pair<int, boost::container::small_vector<float, 8ul> >*, true>; T = std::pair<int, boost::container::small_vector<float, 8ul> >; Allocator = boost::container::small_vector_allocator<boost::container::new_allocator<std::pair<int, boost::container::small_vector<float, 8ul> > > >; typename boost::move_detail::disable_if_or<void, boost::move_detail::is_same<typename boost::container::container_detail::version<Allocator>::type, boost::move_detail::integral_constant<unsigned int, 0u> >, boost::move_detail::is_convertible<InIt, typename boost::container::allocator_traits<Allocator>::size_type>, boost::container::container_detail::is_input_iterator<FwdIt> >::type = void]’: boost_1_63_0/boost/container/small_vector.hpp:563:7: required from ‘boost::container::small_vector<T, N, Allocator>::small_vector(const boost::container::small_vector<T, N, Allocator>&) [with T = std::pair<int, boost::container::small_vector<float, 8ul> >; long unsigned int N = 8ul; Allocator = boost::container::new_allocator<std::pair<int, boost::container::small_vector<float, 8ul> > >]’ /usr/include/c++/5/bits/stl_pair.h:113:31: required from ‘std::pair<_T1, _T2>::pair(const _T1&, const _T2&) [with _T1 = const int; _T2 = boost::container::small_vector<std::pair<int, boost::container::small_vector<float, 8ul> >, 8ul>]’ /usr/include/c++/5/bits/stl_map.h:487:23: required from ‘std::map<_Key, _Tp, _Compare, _Alloc>::mapped_type& std::map<_Key, _Tp, _Compare, _Alloc>::operator[](const key_type&) [with _Key = int; _Tp = boost::container::small_vector<std::pair<int, boost::container::small_vector<float, 8ul> >, 8ul>; _Compare = std::less<int>; _Alloc = std::allocator<std::pair<const int, boost::container::small_vector<std::pair<int, boost::container::small_vector<float, 8ul> >, 8ul> > >; std::map<_Key, _Tp, _Compare, _Alloc>::mapped_type = boost::container::small_vector<std::pair<int, boost::container::small_vector<float, 8ul> >, 8ul>; std::map<_Key, _Tp, _Compare, _Alloc>::key_type = int]’ small_vector_test.cpp:18:39: required from here boost_1_63_0/boost/container/vector.hpp:1261:15: error: binding ‘const std::pair<int, boost::container::small_vector<float, 8ul> >’ to reference of type ‘std::pair<int, boost::container::small_vector<float, 8ul> >&’ discards qualifiers *cur = *first; ^ In file included from /usr/include/c++/5/bits/stl_algobase.h:64:0, from /usr/include/c++/5/bits/stl_tree.h:63, from /usr/include/c++/5/map:60, from small_vector_test.cpp:1: /usr/include/c++/5/bits/stl_pair.h:96:12: note: initializing argument 1 of ‘std::pair<int, boost::container::small_vector<float, 8ul> >& std::pair<int, boost::container::small_vector<float, 8ul> >::operator=(std::pair<int, boost::container::small_vector<float, 8ul> >&)’ struct pair ^ Any ideas? Is 'boost:container::small_vector' supposed to work without C++11 support? Thanks! Greetings, Daniel

2 4

How to use Boost libs from projects?
by Andy 01 Jan '17

01 Jan '17

I have project compiled as Debug,x64. And boost 1.63 compiled by: - bootstrap.bat - b2 variant=debug,release link=static runtime-link=static address-model=64 In project I have cpp files with #include "boost/filesystem.hpp" and when compile project I have error: cannot open file 'libboost_filesystem-vc140-mt-gd-1_61.lib' whereas in c:\boost\stage\lib\ I have libboost_filesystem-vc140-mt-sgd-1_63.lib but not libboost_filesystem-vc140-mt-gd-1_61.lib and in VS2015 project I have no Boost lib. (sorry for my English)

4 13

How to compile boost-thread for Universal Windows?
by Petr Hudeček 31 Dec '16

31 Dec '16

Hello, I'm trying to use the boost-thread library in a Windows Runtime (UWP) project. The tutorial at https://blogs.msdn.microsoft.com/vcblog/2014/07/18/using-boost-libraries-in… doesn't seem to work anymore. If I compile boost sources, either from the primary distribution, or from https://github.com/wpbest/Boost, I can compile it as a static library but when I link to it from the project, I receive linker errors that four identifiers could not be found. These identifiers are API functions such as CreateWaitableTimer. Is there a way to use boost-thread from a UWP project despite this? Perhaps setting some preprocessor symbols? Thank you, Petr Hudecek

4 5

Echo AsyncReadStream/AsyncWriteStream example
by Adam Zegelin 30 Dec '16

30 Dec '16

Hi. Is there a boost::asio example that shows an implementation of AsyncReadStream/AsyncWriteStream that copies write->read? i.e. echoes what was sent I’m trying to emulate serial device that takes input commands separated by \r, executes them, and then outputs the result. i.e., I'm trying to create a fake asio::serial_port that can read input, process it and then provide output. The device I'm trying to emulate echoes the sent characters. Commands are executed once '\r' is received. Ideally the fake serial port also echos the received data immediately. I've got the read part working by implementing async_write_some(…), but I'm having trouble implementing the async_read_some(…) method so that it can be called at anytime, especially when io_service.run() is run from multiple threads. Thanks, Adam

1 0

Boost 1.63.0 is released!
by Marshall Clow 28 Dec '16

28 Dec '16

Release 1.63.0 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 numerous enhancements and bug fixes for existing libraries. For details, including download links, see http://www.boost.org/users/news/version_1.63.0 You can also download directly from SourceForge: https://sourceforge.net/projects/boost/files/boost/1.63.0/ 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 Vladimir Prus Rene Rivera Marshall Clow Eric Niebler Daniel James Beman Dawes

2 1

[msm] Transitions not atomic with orthogonal regions
by Stian Zeljko Vrba 26 Dec '16

26 Dec '16

I have spent almost an hour debugging the following issue. I have a "main" orthogonal region with a couple of states and an "error" region with RUNNING and PAUSED states, like this: struct PAUSED : public msm::front::interrupt_state<new_job> {}; struct RUNNING : public msm::front::state<> {}; There are transitions PAUSED --new_job--> RUNNING and RUNNING --failed--> PAUSED. Then, in one of the main states, I have struct MAIN_STATE1 : public msm::front::state<> { template<typename Event, typename FSM> void operator()(const Event& e, FSM& fsm) { if (error_condition) fsm.process_event(failed{}); } } and in the FSM I have using initial_state = mpl::vector<MAIN_STATE0, PAUSED>; Now, when new_job is processed and MAIN_STATE1::on_entry called, the machine is in (MAIN_STATE1, PAUSED). In this state, process_event(failed{}) is ignored and not queued because PAUSED is interrupt state. Then, the next orthogonal region is processed and the machine goes into (MAIN_STATE1, RUNNING), which destroys the rest of the logic around which the FSM is designed... Adding using active_state_switch_policy = msm::active_state_switch_after_exit; to the machine did not help in any way. Reordering the states in initial_state fixed the problem, but this is a rather unsatisfactory fix because it doesn't solve the underlying problem of transition atomicity. This is counter-intuitive, being on the border between a bug and lack of clear documentation. Specifically: 1. State switch should be atomic. By using active state switch after exit, I had *expected* that first all exit handlers of the current state are called, the state is atomically switched to the new state, and then all entry handlers are called. What I have *observed* is that on_entry handler of a state in one orthogonal region is called before the other orthogonal region has switched *its* state. Thus, with orthogonal regions, the machine can be in an "impossible" intermediate combination of states. Because not all side-effects have run, this has consequences for all actions/guards acting on state or FSM data, which is illustrated by the following: 2. process_event eagerly evaluates is_event_handling_blocked_helper for each orthogonal region in order. The event should be queued unconditionally, and only after the transition has completed, the machine should dequeue events and check whether processing is blocked etc. This, I think, would make at least process_event to appear atomic wrt state switch. Which leads me to conclude that MSM is seductively declarative on the surface, but that it's rather tricky operationally.

1 0

Boost 1.63.0 Release Candidate 2
by Marshall Clow 25 Dec '16

25 Dec '16

The second release candidate for the 1.63.0 release is now available at: https://sourceforge.net/projects/boost/files/boost/1.63.0/ The only changes are in Boost.Config, Boost.Context, and Boost.Fiber. The SHA256 checksums are as follows: 25db3956a8d58187ac7a0702cc917e9bab47ff90baafc35e4e789dca1ce5f423 boost_1_63_0_rc2.7z beae2529f759f6b3bf3f4969a19c2e9d6f0c503edcb2de4a61d1428519fcb3b0 boost_1_63_0_rc2.tar.bz2 fe34a4e119798e10b8cc9e565b3b0284e9fd3977ec8a1b19586ad1dec397088b boost_1_63_0_rc2.tar.gz 97ce6635df1f44653a597343cd5757bb8b6b992beb3720f5fc761e3644bcbe7b boost_1_63_0_rc2.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

2 2

cross compile application include boost-lib
by Thomas Schmiedl 24 Dec '16

24 Dec '16

Hello, I try to use this application https://github.com/repetier/Repetier-Server on my mips-based router. Sorry, I'm not a developer. Please could someone explain the cross-compiling steps for Linux (I use Ubuntu 14 x86 as build system). gcc/g++ toolchain for mips is from this project http://freetz.org/wiki/help/howtos/common/newbie.en. Do I have also to cross-compile boost for mips or only install the Ubuntu-source-package (sudo apt-get install libboost-all-dev). Thanks for your help and best regards, Thomas

2 2

[boost] [build] [context] Mac build options for Boost.Context
by Nat Goodspeed 23 Dec '16

23 Dec '16

Building Boost 1.57 on OS X 10.11, but building it *for* OS X down to 10.9, we're passing bjam a cxxflags command-line argument including -mmacosx-version-min=10.9. We see that switch (along with our other cxxflags switches) in most "g++" lines in the build output, since we also pass -d2 -d+4. But for the assembly source files in Boost.Context, the lines do not contain our requested -mmacosx-version-min switch. In fact they don't seem to contain any of our cxxflags switches. As a result, the assembly object files are built for the OS X version on the build host, 10.11. At executable link time, this produces messages: ld: warning: object file (packages/lib/Release/libboost_context-mt.a(jump_x86_64_sysv_macho_gas.o)) was built for newer OSX version (10.11) than being linked (10.9) ld: warning: object file (packages/lib/Release/libboost_context-mt.a(make_x86_64_sysv_macho_gas.o)) was built for newer OSX version (10.11) than being linked (10.9) (These are the only two .o files producing such messages.) I'm afraid that the resulting executable won't run on OS X 10.9 or 10.10, and that segment of the population is still interesting to us. How should we propagate the cxxflags=-mmacosx-version-min=10.9 feature to the assembler command for Boost.Context?

2 2

Units: Are scales for bytes really working?
by Akim Demaille 21 Dec '16

21 Dec '16

Hi, I’m trying to run the example given in http://www.boost.org/doc/libs/1_62_0/doc/html/boost_units/Examples.html#boo…, and the result is puzzling. $ cat foo2.cc #include <iostream> #include <boost/units/systems/information/byte.hpp> #include <boost/units/io.hpp> using namespace std; using boost::units::binary_prefix; using boost::units::engineering_prefix; using boost::units::no_prefix; using boost::units::symbol_format; using boost::units::quantity; using boost::units::information::byte_base_unit; int main() { cout << BOOST_VERSION << endl; // Don't forget that the units name or symbol format specification is persistent. cout << symbol_format; // Resets the format to the default symbol format. quantity<byte_base_unit::unit_type> b = 2048. * byte_base_unit::unit_type(); cout << b << endl; cout << engineering_prefix << b << endl; // 2.048 kb cout << symbol_format << binary_prefix << b << endl; // "2 Kib" } $ g++-6 foo2.cc -o foo2 $ ./foo2 106200 2048 B 16.384 kb 16 Kib What am I doing wrong? Note that there are a couple of issues in the example as given: - there should not be any endl after cout << symbol_format; - using ‘\n’ instead of endl, too many people seem to believe that std::endl is the normal way to end the lines - it would be nice to provide all the needed includes

1 0