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

[1.55.0] Release candidates available
by Marshall Clow 01 Nov '13

01 Nov '13

Release candidate files for 1.55.0 are available at http://boost.cowic.de/rc/ 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. This helps ensure the candidates build OK before we push them out to SourceForge. The files (and associated md5s) are: ba1ae723b8b4d1edeca684bb14b3213b *boost_1_55_0_rc1.tar.gz 66e01be2f328c556835c42be7b6d17d5 *boost_1_55_0_rc1.tar.bz2 925bb6c96082033687e5742b4e77af7b *boost_1_55_0_rc1.zip 830d61075ecf7d5ebf7c0843033393bf *boost_1_55_0_rc1.7z Thanks! -- The release managers

9 16

Connecting signals where one is a member
by Chris Stankevitz 01 Nov '13

01 Nov '13

Hello, "Connect 4" fails to compile in the code below on g++ 4.6.3. Why does "Connect 4" not work while "Connect 2" does work? Connect 4 and 2 are the same thing as far as I can see. Thank you, Chris #include <boost/signals2/signal.hpp> typedef boost::signals2::signal<void( )> TDSignalA; typedef boost::signals2::signal<void(int)> TDSignalB; TDSignalA SignalA1; TDSignalB SignalB1; struct TSItem { TDSignalA SignalA2; TDSignalB SignalB2; }; int main() { TSItem Item; // Connect 1 SignalA1.connect(Item.SignalA2); // Connect 2 SignalA1.connect(TDSignalA::slot_type(&TSItem::SignalA2, &Item)); // Connect 3 SignalB1.connect(Item.SignalB2); // Connect 4 //SignalB1.connect(TDSignalB::slot_type(&TSItem::SignalB2, &Item, _1)); return 0; }

2 2

VS2005 compiler error in swap() definition after including flat_map.hpp
by Фёдор Трушкин 01 Nov '13

01 Nov '13

Dear All, I encountered a strange problem recently. It's perfectly reproducable under VS2005 even with the latest boost 1_55 rc. Here's the minimal reproducable example: // ----- //#include <boost/container/flat_map.hpp> #include <boost/container/detail/config_begin.hpp> #include <boost/intrusive/detail/memory_util.hpp> #include <boost/container/detail/memory_util.hpp> void swap(); // ------ In VS2005 this innocent piece of code yields: // ------ error C2365: 'swap' : redefinition; previous definition was 'formerly unknown identifier' // ------ VS2010 and GCC 4.5.2 have no complaints. Looks like VS2005 compiler is entangled somehow by BOOST_INTRUSIVE_HAS_MEMBER magic. Actually, it's quite a headache that with VS2005 one can't introduce swap() after, say, flat_map is included. Should this incompatibility be regarded as a bug? Best regards, Fedor Trushkin

2 1

[python] Crash right after initialization after ubgrading from 1.48 to 1.54
by Richi Lists 31 Oct '13

31 Oct '13

Now I * downloaded the python 2.75 source * compiled python 2.75 * compiled boost 1.54 against python 2.75 * built our software against these new libs -> The behavior is still the same. ob_type of the first python object I create is 0xffffffff, and trying to access it results in an immediate crash. This only applies to the debug version. Release is fine. Also, the behavior is the same, if I link the boost debug build against python release or debug. Rgds Richard

1 0

[Iostreams] Problem to chain gzip filter with home-made one
by Christophe hugon 31 Oct '13

31 Oct '13

Hello everybody, I meeting a execution issue while using the gzip filter of the Boost.Iostreams, combined with a buffered home-made one. I would like to know what is wrong in my way to do it. Firstly, my filter is a general DualFilter. It uses an buffer (of 16 bytes) before the transformation and the put/return. The categories of my filter is closable and dual_use. You can find code details in the end of this mail. My problem is ([a/b] means for the separates case with the letter a and b): If I take the 6 possibilities for write and read: 1) io::filtering_[i/o]stream stm; 2) stm.push(io::gzip_[de/]compressor()); 3) stm.push(my_[Unt/T]ransFilter); at the moment to use the stm.[get/put] to read/write 1)-2) works perfectly 1)-3) works perfectly 1)-3)-2) works perfectly 1)-2)-3) Stay stuck (obviously it is the one I really need) In fact in the last case, the gzip filter give this infinite loop to the put of the TransFilter : 31 -117 8 0 0 0 0 0 0 31 -117 8 0 0... A similar case occur with the bzip filter (even if it is a bit longer loop). What do I do wrong ? Somebody can help me ? Thanks in advance Christophe PS: here a simplified implementation of my filter template<typename Source> int get(Source& src) { //at the very beginning or if the buffer are fully returned if (fcurrent_pos == fblkLength) { fcurrent_pos =0; FillThem (src); //fill&transform the buffers with the right number of src.get() fc=CheckNSolveBufferIsEnd (src); //eventually prepare read the suffix of the file } //The effective return if (fcurrent_pos < fUntrans.size ()) return (int)fUntrans[fcurrent_pos++]; else return fc; } template<typename Sink> bool put(Sink& dest, int c) { if (c==EOF || c ==boost::iostreams::WOULD_BLOCK) return false; if (fUntrans.size () < fblkLength - 1) { fUntrans.AddData((char*)(&c),1); return true; } fUntrans.AddData((char*)(&c),1); fenc->TransformThis (fUntrans, fTrans, fKey); for (int i=0; i<fblkLength; i++) { if (!boost::iostreams::put(dest, (int)fTrans[i])) return false; } return true; } template<typename Sink> bool close (Sink &dest, BOOST_IOS::openmode m) { if (fFlushed) return true; PrepareSuffix (); for (int i=0; i< fblkLength*2; //suffix length i++) { if (!boost::iostreams::put(dest, (int)fTrans[i])) return false; } return true; }

1 0

BOOST ASIO: Having successful READ from the SERVER ..??
by Rahul Mathur 31 Oct '13

31 Oct '13

All, Initially after having done successful CONNECT with the server, sending successful LOGON message and finally KEEP ALIVE status, I tried to read from the server as - -- boost::asio::streambuf data; std::size_t length = boost::asio::read(socket, data, boost::asio::transfer_at_least(socket.available())); -- but was able to partially receive response from remote server which I don't think was completely successful, so to have COMPLETE RESPONSE from the server, tried with below modification - -- boost::system::error_code ec; boost::asio::streambuf data; std::size_t length = boost::asio::read(socket, data, boost::asio::transfer_all(), ec); --- but this modification doesn't work. Any clue what should be the proper way? Thanks!!

2 1

SIGSEGV when removing Ethernet/Serial Port cables
by Ulf Samuelsson 30 Oct '13

30 Oct '13

Have a boost enable application which communicated over USB serial ports and UDP If the cables are removed, the library fires off SIGSEGV signals. I am trying to catch exceptions but neither } catch (boost::system::system_error &e) { } catch (std::exception &e) { seems to work, and a SIGSEGV signal is sent when the port->get_io_service().reset(); statement is executed. How can I handle the problem, that the USB serial cable is removed? I know I can test for the existence of /dev/ttyUSBx in the file system or /sys/class/tty/ttyUSBx" I know I can write a SIGSEGV handler, but would like to know if there is any exception I can catch instead. Have the same problem when the Ethernet Cable is removed. How handle this? // Serial Port Read. // Reads a character or times out bool SerialPort::read_char(unsigned char& val) { val = c = '\0'; try { // After a timeout & cancel it seems we need // to do a reset for subsequent reads to work. port->get_io_service().reset(); // Asynchronously read 1 character. async_read(*port, buffer(&c, 1), boost::bind(&SerialPort::read_complete, this, placeholders::error, placeholders::bytes_transferred)); // Setup a deadline time to implement our timeout. timer->expires_from_now(boost::posix_time::milliseconds(1)); timer->async_wait(boost::bind(&SerialPort::time_out, this, placeholders::error)); // This will block until a character is read // or until the it is cancelled. port->get_io_service().run(); if (!read_error) { val = c; } else { val = 0; // Timeout } } catch (boost::system::system_error &e) { val = 2; DebugPort.WriteLine("Serial Port System Error", RED); return false; } catch (std::exception &e) { val = 1; DebugPort.WriteLine("Serial Port General Error", RED); return false; } return !read_error; } // Network port open bool UdpClient::Open(void) { udp::resolver resolver(io_service); udp::resolver::query query(udp::v4(), ipAddress, Port); try { Endpoint = *resolver.resolve(query); } catch (boost::system::system_error &e) { // We might not have a network connection return false; } if (!CreateSocket()) { return false; } if (!OpenSocket()) { return false; } init = true; return true; } Best Regards Ulf Samuelsson

2 1

Looking for official .m4 script
by Yaroslav Panych 30 Oct '13

30 Oct '13

HI Looking for official .m4 script to include it into distribution package in order to use its benefits while configuring/build software. Yes, I know about Tsuna's boost.m4 library but I never heard it was approved by Boost officials. Regards

2 1

[graph] ambiguity moving from 1.48 to 1.54
by Richi Lists 30 Oct '13

30 Oct '13

Hi, I'm in the process of upgrading from boost 1.48 to 1.54. We're using MSVC10. Most stuff that comes up is easy to fix, but this one gives me headaches: ...boost\boost_1_54_0\boost/pending/property.hpp(108): error C2752: 'boost::lookup_one_property_internal<PList,PropName>' : more than one partial specialization matches the template argument list with [ PList=boost::property<boost::edge_bundle_t,EdgeData>, PropName=boost::edge_bundle_t ] ...boost\boost_1_54_0\boost/pending/property.hpp(114): could be 'boost::lookup_one_property_internal<boost::property<Tag,T,Base>,Tag>' ...boost\boost_1_54_0\boost/pending/property.hpp(108): or 'boost::lookup_one_property_internal<boost::property<Tag,T,Base>,boost::edge_bundle_t>' ...boost\boost_1_54_0\boost/pending/property.hpp(108): or 'boost::lookup_one_property_internal<T,boost::edge_bundle_t>' ...boost\boost_1_54_0\boost/pending/property.hpp(158) : see reference to class template instantiation 'boost::lookup_one_property_internal<PList,PropName>' being compiled with [ PList=boost::property<boost::edge_index_t,int,boost::property<boost::edge_bundle_t,EdgeData>>, PropName=boost::edge_bundle_t ] ...boost\boost_1_54_0\boost/graph/detail/adjacency_list.hpp(1505) : see reference to class template instantiation 'boost::lookup_one_property<T,Tag>' being compiled with [ T=boost::property<boost::edge_index_t,int,boost::property<boost::edge_bundle_t,EdgeData>>, Tag=boost::edge_bundle_t ] ...boost\boost_1_54_0\boost/graph/detail/adjacency_list.hpp(2028) : see reference to class template instantiation 'boost::adj_list_helper<Config,Base>' being compiled with [ Config=boost::detail::adj_list_gen<boost::adjacency_list<boost::vecS,boost::vecS,boost::undirectedS,VertexProps,EdgeProps>,boost::vecS,boost::vecS,boost::undirectedS,VertexProps,EdgeProps,boost::no_property,boost::listS>::config, Base=boost::undirected_graph_helper<boost::detail::adj_list_gen<boost::adjacency_list<boost::vecS,boost::vecS,boost::undirectedS,VertexProps,EdgeProps>,boost::vecS,boost::vecS,boost::undirectedS,VertexProps,EdgeProps,boost::no_property,boost::listS>::config> ] ...boost\boost_1_54_0\boost/graph/adjacency_list.hpp(267) : see reference to class template instantiation 'boost::vec_adj_list_impl<Graph,Config,Base>' being compiled with [ Graph=boost::adjacency_list<boost::vecS,boost::vecS,boost::undirectedS,VertexProps,EdgeProps>, Config=boost::detail::adj_list_gen<boost::adjacency_list<boost::vecS,boost::vecS,boost::undirectedS,VertexProps,EdgeProps>,boost::vecS,boost::vecS,boost::undirectedS,VertexProps,EdgeProps,boost::no_property,boost::listS>::config, Base=boost::undirected_graph_helper<boost::detail::adj_list_gen<boost::adjacency_list<boost::vecS,boost::vecS,boost::undirectedS,VertexProps,EdgeProps>,boost::vecS,boost::vecS,boost::undirectedS,VertexProps,EdgeProps,boost::no_property,boost::listS>::config> ] ..\graph_problem.cpp(52) : see reference to class template instantiation 'boost::adjacency_list<OutEdgeListS,VertexListS,DirectedS,VertexProperty,EdgeProperty>' being compiled with [ OutEdgeListS=boost::vecS, VertexListS=boost::vecS, DirectedS=boost::undirectedS, VertexProperty=VertexProps, EdgeProperty=EdgeProps ] ...boost\boost_1_54_0\boost/graph/adjacency_list.hpp(374): error C2182: '[]' : illegal use of type 'void' The code to reproduce is here: // boost #include <boost/graph/adjacency_list.hpp> #include <boost/graph/subgraph.hpp> #include <boost/graph/graph_traits.hpp> class EDGE {}; ////////////////////////////////////////////////////////////////////////// /// graph's vertex data struct VertexData { /// c-tor VertexData() { pos[0] = 0.0; pos[1] = 0.0; pos[2] = 0.0; } VertexData(const double* _pos) { pos[0] = _pos[0]; pos[1] = _pos[1]; pos[2] = _pos[2]; } /// data double pos[3]; }; ////////////////////////////////////////////////////////////////////////// /// graph's edge data struct EdgeData { /// c-tor EdgeData() : pEdge(NULL) {} EdgeData(EDGE* _pEdge) : pEdge(_pEdge) {} /// comparison for equality bool operator==(const EdgeData& dst) const { return pEdge == dst.pEdge; } /// data EDGE* pEdge; }; ////////////////////////////////////////////////////////////////////////// /// types typedef VertexData VertexProps; ///< custom vertex properties typedef boost::property<boost::edge_index_t, int, boost::property<boost::edge_bundle_t, EdgeData> > EdgeProps; ///< custom edge properties typedef boost::adjacency_list<boost::vecS, boost::vecS, boost::undirectedS, VertexProps, EdgeProps> Type; ///< custom graph type ////////////////////////////////////////////////////////////////////////// /// unary predicate for testing if edge has duplicates in graph struct CheckDuplicateEdge { CheckDuplicateEdge(const Type& _g) : g(_g) {} // check if this edge has duplicates bool operator()(const Type::edge_descriptor& e) const { // do something return false; } private: const Type& g; }; ////////////////////////////////////////////////////////////////////////// void RemoveDuplicateEdges(Type& g) { remove_edge_if(CheckDuplicateEdge(g), g); } ////////////////////////////////////////////////////////////////////////// void main() { std::cout << "hello graph"; } ////////////////////////////////////////////////////////////////////////// I have tried to find a solution, and have searched the internet for hours. So, hopefully anybody here saw something similar, and solved it already. Rgds Richard

2 3

[python] Crash right after initialization after ubgrading from 1.48 to 1.54
by Richi Lists 30 Oct '13

30 Oct '13

I upgraded from boost 1.48 to .154 and I thought I was done after all the compiler errors and unit tests were ironed out. But now when I start our application, it terminates immediately. This only happens with the debug build. The release build is fine. I verified that it properly links to boost-python-vc100-mt-gyd-1_54.lib and python27_d.lib The python version (2.73) is still the same that we used with boost 1.48. I tracked it down to here: Py_Initialize(); m_mainModule = boost::python::import(__main__); <-- crash here With the debugger I found out that the string object is invalid. More specifically, ob_type is 0xffffffff. So I added this exception which always throws now: Py_Initialize(); const boost::pathon::str someRandomString("someRandomString"); if(reinterpret_cast<void*>(0xffffffff) == someRandomString.ptr()->ob_type) throw std::runtime_error("Python problem"); What could be going on here? Rgds Richard

1 0