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

[ Boost.MPI ] : Spawning processes, Intercommunicators and send custom data-types ?
by Ashika Umanga Umagiliya 06 Sep '09

06 Sep '09

Greetings all, I use OpenMPI 1.3.3 and I tried to spawn child processes (using MPI_Comm_spawn) and send user-defined-datatypes across processes.The child processes should be ranked from 1-5 while the rank 0 is the manager processes (manager.cpp) . But I get rank 0 for manager process and the child processes also ranked starting from 0 (0-4) . So I cannot send data from manager process to first child processes (which suppose to be rank 1) Please give any suggestion to resolve this. thanks in advance, umanga manager.cpp (manager process which spawns child processes) - rank 0 ------------------------------------------------------------ int main(int argc,char *argv[]) { mpi::environment evn(argc,argv); mpi::communicator world; MPI_Comm everyone; //spawn 5 child processes. MPI_Comm_spawn("./worker", MPI_ARGV_NULL, 5, MPI_INFO_NULL, 0, MPI_COMM_SELF, &everyone, MPI_ERRCODES_IGNORE); intercommunicator intcomm(everyone,comm_duplicate); if(rank==0){ GPSPosition *obj=new GPSPosition(100,200,300); shared_ptr<Position> pos(new Position); pos->setVals(); obj->addP(pos); intcomm.send(0,100,obj); } return 0; } worker.cpp (child process)- should be ranked 1-5 ,but get 0-4 ----------------------------------------------------------------------------------- int main(int argc,char *argv[]) { mpi::environment evn(argc,argv); MPI_Comm parent; MPI_Comm_get_parent(&parent); intercommunicator incomm(parent,comm_duplicate); if(parent==MPI_COMM_NULL){ cout << "Intercommunicator is Null !"<<endl; }else{ int r=incomm.rank(); cout <<"Rank !"<<r<< endl; //get 0-4 if(r==1){ //try receiving data send from manager process } } return 0; }

2 1

Member functions on boost::variant?
by Christopher Jefferson 06 Sep '09

06 Sep '09

Is there a nice way of calling a member function on a boost::variant (where I know said member function is implemented by all the types in the variant)? Chris

2 1

boost::function/boost::bind compilation error
by Ahmed Badran 06 Sep '09

06 Sep '09

Hi, I was trying to use boost function/bind to bind the first argument of a member function that takes two arguments, when I got into a compilation error that I can't quite decipher or understand, below is a sample representation of the problem, and then there's the compilation error, am I doing something incorrectly here? I have tried removing the boost::mem_fn instantiation with no difference. #include <boost/function.hpp> #include <boost/bind.hpp> #include <boost/mem_fn.hpp> #include <iostream> using namespace std; using namespace boost; class my_class { public: void test(bool b) { cout << "got " < test_fn; boost::bind(&my_class::test, &c, _2); /* causes a compilation error */ test_fn = boost::bind<void>(boost::mem_fn(&my_class::test), &c, _2); /* works fine */ test_fn = bind1st(mem_fun(&my_class::test), &c); test_fn(true); test_fn(false); return 0; } /usr/boost/include/boost/bind/arg.hpp: In constructor 'boost::arg::arg(const T&) [with T = boost::arg<2> ()(), int I = 1]': /usr/boost/include/boost/bind.hpp:292: instantiated from 'void boost::_bi::list2<A1, A2>::operator()(boost::_bi::type<void>, F&, A&, int) [with F = boost::_mfi::mf1<void, my_class, bool>, A = boost::_bi::list1<bool&>, A1 = boost::_bi::value<my_class*>, A2 = boost::arg<2>]' /usr/boost/include/boost/bind/bind_template.hpp:32: instantiated from 'typename boost::_bi::result_traits<R, F>::type boost::_bi::bind_t<R, F, L>::operator()(A1&) [with A1 = bool, R = void, F = boost::_mfi::mf1<void, my_class, bool>, L = boost::_bi::list2<boost::_bi::value<my_class*>, boost::arg<2> >]' /usr/boost/include/boost/function/function_template.hpp:152: instantiated from 'static void boost::detail::function::void_function_obj_invoker1<FunctionObj, R, T0>::invoke(boost::detail::function::function_buffer&, T0) [with FunctionObj = boost::_bi::bind_t<void, boost::_mfi::mf1<void, my_class, bool>, boost::_bi::list2<boost::_bi::value<my_class*>, boost::arg<2> > >, R = void, T0 = bool]' /usr/boost/include/boost/function/function_template.hpp:904: instantiated from 'void boost::function1<R, T1>::assign_to(Functor) [with Functor = boost::_bi::bind_t<void, boost::_mfi::mf1<void, my_class, bool>, boost::_bi::list2<boost::_bi::value<my_class*>, boost::arg<2> > >, R = void, T0 = bool]' /usr/boost/include/boost/function/function_template.hpp:785: instantiated from 'typename boost::enable_if_c<boost::type_traits::ice_not<boost::is_integral<Functor>::value>::value, boost::function1<R, T1>&>::type boost::function1<R, T1>::operator=(Functor) [with Functor = boost::_bi::bind_t<void, boost::_mfi::mf1<void, my_class, bool>, boost::_bi::list2<boost::_bi::value<my_class*>, boost::arg<2> > >, R = void, T0 = bool]' test.cpp:28: instantiated from here /usr/boost/include/boost/bind/arg.hpp:37: error: creating array with negative size ('-0x00000000000000001') Regards, Ahmed

3 3

Boost.Variant: Passing visitor as a function argument
by Hawkes Steve-FSH016 06 Sep '09

06 Sep '09

Using the Boost.Variant library, I'm trying to pass a visitor as an argument to a class method (the class instance contains the objects to be visited), but I can't come up with a solution that compiles. For example: typedef std::pair<std::string, double> AttrDouble; typedef std::pair<std::string, int> AttrInt; typedef std::pair<std::string, std::string> AttrString; typedef boost::variant<AttrDouble, AttrInt, AttrString> Attr; typedef std::vector<Attr> AttrList; class Example { public: // Some code omitted for brevity -- void visit(boost::static_visitor<> const& visitor) { BOOST_FOREACH(Attr attr, i_attrList) { boost::apply_visitor(visitor, attr); } } private: AttrList i_attrList; }; The above gives compiler errors like this: boost-1_37/boost/variant/variant.hpp:806: error: no match for call to '(const boost::static_visitor<void>) (std::pair<std::basic_string<char, std::char_traits<char>, std::allocator<char> >, double>&)' boost-1_37/boost/variant/variant.hpp:806: error: return-statement with a value, in function returning 'void' The problem seems to be that static_visitor doesn't actually contain op()--only the user-supplied derived class does. I've read through the Boost.Variant documentation repeatedly, looked at Boost.Variant source, and tried searching the Internet for examples, but I haven't found anything that shows how to do this, leading me to believe it may not be possible. Is there a way to pass a visitor as a function argument, or some alternative approach that allows visiting objects without having direct access to them (such as when they are class private data and the visitor is external to the class)? Thanks, Steve Hawkes

3 4

Bug in serialization::make_nvp?
by stair314 05 Sep '09

05 Sep '09

I'm sorry if this appears twice; I received an e-mail saying it couldn't be sent because I hadn't subscribed to the mailing list. When I run the following code, I get this error: terminate called after throwing an instance of 'boost::archive::xml_archive_exception' what(): uninitialized exception Abort However, if I substitute BOOST_SERIALIZATION_NVP for the make_nvp function, it works fine. Am I doing something wrong, or is this a bug? #include <iostream> #include <fstream> #include <string> #include <boost/config.hpp> #if defined(BOOST_NO_STDC_NAMESPACE) namespace std{ using ::remove; } #endif #include <boost/archive/tmpdir.hpp> #include <boost/archive/xml_iarchive.hpp> #include <boost/archive/xml_oarchive.hpp> #include <boost/serialization/nvp.hpp> class A { public: A() {} virtual ~A() {} int a; private: friend class boost::serialization::access; template<class Archive> void serialize(Archive & ar, const unsigned int version) { //ar & BOOST_SERIALIZATION_NVP(a); ar & boost::serialization::make_nvp("example field",a); }; }; void restore( const char * filename) { A s; // open the archive std::ifstream ifs(filename); assert(ifs.good()); boost::archive::xml_iarchive ia(ifs); // restore the schedule from the archive ia >> BOOST_SERIALIZATION_NVP(s); } int main(int argc, char *argv[]) { std::string filename("/tmp/wtf.xml"); { A s; // make an archive std::ofstream ofs(filename.c_str()); assert(ofs.good()); boost::archive::xml_oarchive oa(ofs); oa << BOOST_SERIALIZATION_NVP(s); } printf("---\n"); restore( filename.c_str()); return 0; } -- View this message in context: http://www.nabble.com/Bug-in-serialization%3A%3Amake_nvp--tp25304627p253046… Sent from the Boost - Users mailing list archive at Nabble.com.

2 1

Re: [Boost-users] valgrind reports memory still reachable when using shared_lock and shared_mutex
by Rajeev Rao 04 Sep '09

04 Sep '09

Could the problem be with the way I built the boost libraries? I used no options when doing the build. looks like boost::mutex doesn't have the problem (when used with unique_lock). So I guess the problem is in shared_mutex. Rajeev --- On Thu, 9/3/09, Steven Watanabe <watanabesj(a)gmail.com> wrote: From: Steven Watanabe <watanabesj(a)gmail.com> Subject: Re: [Boost-users] valgrind reports memory still reachable when using shared_lock and shared_mutex To: boost-users(a)lists.boost.org Date: Thursday, September 3, 2009, 9:40 AM AMDG Rajeev Rao wrote: > I found similar problems when getting an exclusive lock via unique_lock as well. I wrote this test code to reproduce. I get the still-reachable problem if I use either the read() or the write() call (shared_lock or unique_lock). Multiple calls to these functions did not increase the number of reachable bytes in the valgrind report. I understand my previous use of adopt_lock was undefined (hence i've not used it here). Is my use as described by the code correct now ? If so, is this a bug ? > It is almost certainly exactly the same as with shared_lock. The code that allocates the object is in a lower level component. In Christ, Steven Watanabe _______________________________________________ Boost-users mailing list Boost-users(a)lists.boost.org http://lists.boost.org/mailman/listinfo.cgi/boost-users

2 2

[boost::asio]Question about how to properly make several async requests (reads and writes)
by Matheus Araújo Aguiar 04 Sep '09

04 Sep '09

Dear list, I have learned that one cannot make several async write requests, one after another, without waiting for the completion of each request, otherwise the bytes get merged and the messages will not make sense at the receiving side. I am correct assuming that, right? My doubt is if the same happens to async read requests. Does one have to wait a read request completes before issuing another? Will the bytes get merged too, if one does not wait before making another read request? Assuming things happen that way, one async request waiting for the completion of another earlier request, then one have serial I/O processing and if one wants to have things done in parallel, the simplest way would be having other(s) thread(s) reading and writting in the same I/O object. Is this correct? I hope I have made my doubts clear enough, and if it is not clear please say it and i will try to be more objective. Regards, -- Matheus Araújo Aguiar Computer Scientist matheus.pit(a)gmail.com

3 5

[statechart] Pass event data to state constructor?
by Bill Clark 04 Sep '09

04 Sep '09

I'd like to be able to communicate the data from an event to the state(s) that get constructed as a result of a transition. I don't see an obvious way to do that, since the constructors of a state object either take only a context parameter (this is in an asychronous_state_machine), or also take a transition action plus an event. I can save the event data within the context object and then retrieve it post-construction, but that seems error-prone. Is there a better way to do this? Thanks, Bill

2 8

BGL vertex(std::size_t i, Graph) is in no concept
by Sebastian Weber 04 Sep '09

04 Sep '09

Hi! If I am not wrong, then the vertex(std::size_t i, Graph) function is required in no graph concept at all. However, adjacency_list defines it and csr-graphs use it ... Did I overlook something or shouldn't the function belong to some graph concept? Sebastian

3 5

Correct use of unordered_map
by John Dlugosz 04 Sep '09

04 Sep '09

Consider this loop: const my_map_t::iterator end= my_map.end(); for (my_map_t::iterator it= my_map.begin(); it != end; ++it) { if (is_to_be_removed(it->first)) global_map.erase (it); } The documentation I've seen for unordered_map says that removing doesn't invalidate iterators except those referring to the element being deleted. But that's the case for it itself! Also, what does removing one element do to the ordering of the elements -- is there any guarantee that the iterator will visit exactly the ones that were not visited before? What is the correct, standards-conforming, way to perform this? Thanks, --John TradeStation Group, Inc. is a publicly-traded holding company (NASDAQ GS: TRAD) of three operating subsidiaries, TradeStation Securities, Inc. (Member NYSE, FINRA, SIPC and NFA), TradeStation Technologies, Inc., a trading software and subscription company, and TradeStation Europe Limited, a United Kingdom, FSA-authorized introducing brokerage firm. None of these companies provides trading or investment advice, recommendations or endorsements of any kind. The information transmitted is intended only for the person or entity to which it is addressed and may contain confidential and/or privileged material. Any review, retransmission, dissemination or other use of, or taking of any action in reliance upon, this information by persons or entities other than the intended recipient is prohibited. If you received this in error, please contact the sender and delete the material from any computer.

5 13

integer to string or char
by pedro chaparro 04 Sep '09

04 Sep '09

hi, i'd like to know what is the equivalent command to itoa in boost? i have to convert a variable from integer to string. how can i do that? thanks -- Ing Pedro Alonso Chaparro Valero Ciudad Politecnica de la Innovación iTEAM - Mobile Communications Group Polytechnic University of Valencia C\ Camino de Vera S/N, Edificio 8G 46022 Valencia, Spain

2 1

Building Boost 1.40.0 with MPI on SUN HPC Developer distribution fails
by Volker Mosert 04 Sep '09

04 Sep '09

Hi, I'm trying to use the Boost MPI library in a virtual Opensolaris environment (specifically the Sun HPC Software, Developer Edition 1.0 for OpenSolaris). System compiler is the sun studio 12u1 compiler. MPI is provided by the Sun HPC ClusterTools which essentially means Open MPI. Trying to build results in following error messages: sun.compile.c++ bin.v2/libs/mpi/build/sun/release/stdlib-sun-stlport/threading-multi/python/py_communicator.o "./boost/mpi/graph_communicator.hpp", line 62: Warning: comm_ptr hides boost::mpi::communicator::comm_ptr. "./boost/mpi/intercommunicator.hpp", line 62: Warning: comm_ptr hides boost::mpi::communicator::comm_ptr. "libs/mpi/src/python/py_communicator.cpp", line 107: Error: Could not find a match for boost::python::class_::def<boost::python::class_<boost::mpi::communicator, boost::python::detail::not_s pecified, boost::python::detail::not_specified, boost::python::detail::not_specified>::Derived>(const char[12] , bool(*)()const) needed in boost::mpi::python::export_communicator(). "libs/mpi/src/python/py_communicator.cpp", line 107: Error: Could not find a match for boost::python::class_::def<boost::python::class_<boost::mpi::communicator, boost::python::detail::not_s pecified, boost::python::detail::not_specified, boost::python::detail::not_specified>::Derived>(const char[12] , bool(*)()const) needed in boost::mpi::python::export_communicator(). "libs/mpi/src/python/py_communicator.cpp", line 107: Error: Could not find a match for boost::python::class_::def<boost::python::class_<boost::mpi::communicator, boost::python::detail::not_s pecified, boost::python::detail::not_specified, boost::python::detail::not_specified>::Derived>(const char[12] , bool(*)()const) needed in boost::mpi::python::export_communicator(). "libs/mpi/src/python/py_communicator.cpp", line 107: Error: Could not find a match for boost::python::class_::def<boost::python::class_<boost::mpi::communicator, boost::python::detail::not_specified, boost::python::detail::not_specified, boost::python::detail::not_specified>::Derived>(const char[12], bool(*)()const) needed in boost::mpi::python::export_communicator(). 4 Error(s) and 2 Warning(s) detected. "CC" -library=stlport4 -xO4 -mt -erroff=%none -KPIC -DBOOST_ALL_NO_LIB=1 -DBOOST_MPI_DYN_LINK=1 -DBOOST_MPI_PYTHON_DYN_LINK=1 -DBOOST_PYTHON_DYN_LINK=1 -DNDEBUG -I"." -I"/opt/SUNWhpc/HPC8.1/sun/include" -I"/opt/SUNWhpc/HPC8.1/sun/include/openmpi" -I"/usr/include/python2.4" -c -o "bin.v2/libs/mpi/build/sun/release/stdlib-sun-stlport/threading-multi/python/py_communicator.o" "libs/mpi/src/python/py_communicator.cpp" sun.compile.c++ bin.v2/libs/mpi/build/sun/release/stdlib-sun-stlport/threading-multi/python/datatypes.o "./boost/mpi/graph_communicator.hpp", line 62: Warning: comm_ptr hides boost::mpi::communicator::comm_ptr. "./boost/mpi/intercommunicator.hpp", line 62: Warning: comm_ptr hides boost::mpi::communicator::comm_ptr. "./boost/mpi/python/serialize.hpp", line 276: Error: Could not find a match for boost::python::detail::direct_serialization_table<boost::mpi::packed_iarchive, boost::mpi::packed_oarchive>::register_type<boost::python::detail::direct_serialization_table<boost::mpi::packed_iarchive, boost::mpi::packed_oarchive>::T>(boost::python::detail::direct_serialization_table<boost::mpi::packed_iarchive, boost::mpi::packed_oarchive>::default_saver<long>, boost::python::detail::direct_serialization_table<boost::mpi::packed_iarchive, boost::mpi::packed_oarchive>::default_loader<long>, const long, _typeobject*) needed in boost::python::detail::direct_serialization_table<boost::mpi::packed_iarchive, boost::mpi::packed_oarchive>::register_type<long>(const long&, _typeobject*). "./boost/mpi/python/serialize.hpp", line 389: Where: While instantiating "boost::python::detail::direct_serialization_table<boost::mpi::packed_iarchive, boost::mpi::packed_oarchive>::register_type<long>(const long&, _typeobject*)". "./boost/mpi/python/serialize.hpp", line 389: Where: Instantiated from boost::python::register_serialized<boost::mpi::packed_iarchive, boost::mpi::packed_oarchive, long>(const long&, _typeobject*). "./boost/mpi/python/serialize.hpp", line 534: Where: Instantiated from boost::mpi::python::register_serialized<long>(const long&, _typeobject*). "libs/mpi/src/python/datatypes.cpp", line 20: Where: Instantiated from non-template code. 1 Error(s) and 2 Warning(s) detected. "CC" -library=stlport4 -xO4 -mt -erroff=%none -KPIC -DBOOST_ALL_NO_LIB=1 -DBOOST_MPI_DYN_LINK=1 -DBOOST_MPI_PYTHON_DYN_LINK=1 -DBOOST_PYTHON_DYN_LINK=1 -DNDEBUG -I"." -I"/opt/SUNWhpc/HPC8.1/sun/include" -I"/opt/SUNWhpc/HPC8.1/sun/include/openmpi" -I"/usr/include/python2.4" -c -o "bin.v2/libs/mpi/build/sun/release/stdlib-sun-stlport/threading-multi/python/datatypes.o" "libs/mpi/src/python/datatypes.cpp" I would be nice if someone helped me resolving this issue. A second problem was the missinterpretation of ">>" in a template argument by the sun compiler. I resolved that by changing the file "boost/archive/detail/register_archive.hpp". (I probably overcompensated with white spaces) 57c57 < get_counter<Archive>::next adjust_counter(get_counter<Archive>::type);\ --- > get_counter< Archive >::next adjust_counter(get_counter< Archive >::type);\ 60,62c60,62 < get_counter<Archive>::type) { \ < ptr_serialization_support<Archive, Serializable> x; \ < instantiate_ptr_serialization(s, get_counter<Archive>::prior()); \ --- > get_counter< Archive >::type) { \ > ptr_serialization_support< Archive, Serializable > x; \ > instantiate_ptr_serialization(s, get_counter< Archive >::prior()); \ 84c84 < BOOST_DEDUCED_TYPENAME _ptr_serialization_support<Archive, Serializable>::type \ --- > BOOST_DEDUCED_TYPENAME _ptr_serialization_support< Archive, Serializable >::type \ ciao, Volker

1 0

[iterator] Does the base of iterator_adaptor need a const version of operator* ?
by Andre Massing 04 Sep '09

04 Sep '09

Hi, I am trying to wrap our kind of iterators (for meshes) with the iterator_adaptor class template, something like class CellBoostIterator : public boost::iterator_adaptor< CellBoostIterator //Derived , CellIterator //Base , Cell //Value , boost::forward_traversal_tag // CategoryOrTraversal > { ... }; Since the base class CellIterator already provides a operator* and a operator-> (both non-const version) I did not redefine any of these within the CellBoostIterator class. If I now want to use with code-snippet typedef Mesh_2::CellBoostIterator MIT; MIT mit; Mesh_2 mesh_2(omega1); for (MIT mit = mesh_2.begin(); mit != mesh_2.end(); ++mit) { std::cout << mit->index() <<std::endl; } I get a compile error /usr/include/boost/iterator/iterator_adaptor.hpp:312: error: passing ‘const dolfin::CellIterator’ as ‘this’ argument of ‘dolfin::Cell& dolfin::CellIterator::operator*()’ discards qualifiers I looked at the corresponding line /usr/include/boost/iterator/iterator_adaptor.hpp:312: typename super_t::reference dereference() const { return *m_iterator; } So it seems to me that I have to provide a const version of operator*, or do I get something wrong? I am a bit confused, since I have not found this kind of requirements in the documentation. Any enlightening, hints or pointers are greatly appreciated! Greetings, Andre

1 0

1.40 and pch - gcc problems?
by Jonathan Brannan 04 Sep '09

04 Sep '09

I just got 1.40 and tried building. I kept having some pieces fail with a complete build, sometimes I had gcc that ran until resources exausted, sometimes the compiler would segfault. Eventaully I tracked it down to the math library. After noticing the many errors from the gch (things like : error _REENTERANT not defined or used diffrent -fpic). I disabled pch for the math library (actually the boost project and the libs/math/test) and I did not have any more issues. This is how I compiled it: ./bootstrap.sh --without-icu --prefix=ID_1_40_0 ./bjam install stage --prefix=/path/to/install -j 16 --buildid=ID_1_40_0 --layout=tagged --build-type=complete --stagedir=/path/to/install The system is sles (suse enterprise) 11. It's an 8 core intel machine. gcc version 4.3.2 [gcc-4_3-branch revision 141291] (SUSE Linux) I did not figure out how to disable this via the command line, so I did this: *** boost_1_40_0-compiled/Jamroot 2009-08-23 12:17:37.000000000 -0400 --- boost_1_40_0-new/Jamroot 2009-09-03 13:25:05.000000000 -0400 *************** *** 177,182 **** --- 177,183 ---- # Disable auto-linking for all targets here, primarily because it caused # troubles with V2. <define>BOOST_ALL_NO_LIB=1 + <toolset>gcc:<pch>off # Used to encode variant in target name. See the 'tag' rule below. <tag>@$(__name__).tag <conditional>@handle-static-runtime *** boost_1_40_0-compiled/libs/math/test/Jamfile.v2 2009-07-23 05:45:33.000000000 -0400 --- boost_1_40_0-new/libs/math/test/Jamfile.v2 2009-09-03 13:24:40.000000000 -0400 *************** *** 23,28 **** --- 23,29 ---- <toolset>intel:<linkflags>-nologo #<toolset>intel-linux:<pch>off <toolset>intel-darwin:<pch>off + <toolset>gcc:<pch>off <toolset>msvc:<warnings>all <toolset>msvc:<asynch-exceptions>on <toolset>msvc:<cxxflags>/wd4996

3 2

date_time formatting issues
by ariasgore＠gmx.de 04 Sep '09

04 Sep '09

Hi, I have been trying to format a simple posix date time for string representation. The basic part is plain and simple: ---------------------------------------------- #include <boost/date_time/posix_time/posix_time.hpp> namespace ps = boost::posix_time; ps::ptime t(ps::second_clock::local_time()); std::string foo = ps::to_simple_string(t); ---------------------------------------------- But how can I format this returned string? I think the facets are required but as mentioned here : http://www.boost.org/doc/libs/1_40_0/doc/html/date_time/date_time_io.html#d… I tried something with the boost::date_time::date_input_facet: e.g ---------------------------------------------- ps::date_input_facet("%A"); ps::ptime t(ps::second_clock::local_time()); std::string foo = ps::to_simple_string(t); ---------------------------------------------- by using it before the simple_string call. I am sure this is wrong but I wouldn't like to use streams in addition to that to simply rearange some parts of a date for that reason. thanks sam -- Jetzt kostenlos herunterladen: Internet Explorer 8 und Mozilla Firefox 3 - sicherer, schneller und einfacher! http://portal.gmx.net/de/go/chbrowser

2 1

[Boost-Serialization] Problem with BOOST_CLASS_EXPORT
by Soumen 03 Sep '09

03 Sep '09

Hi, I'm getting conflicting declaration with BOOST_CLASS_EXPORT and I think I've got the reason. But I'm not aware of clean solution to overcome this. Let me explain the problem here. Say in FileAB.h, I've class A, class B defined. I've a FileSExportAB.h where I've BOOST_CLASS_EXPORT(A) at line say 10 and BOOST_CLASS_EXPORT(B) at line 11. Similarly I've class C and class D defined FileCD.h. And FileSExportCD.h has BOOST_CLASS_EXPORT(C) @ line 10 and BOOST_CLASS_EXPORT(D) @ line 11. And while building, I get error like following: ------------------------------------------------------------------------------------------------ FileSExportCD.h:10: error: conflicting declaration ‘const boost::archive::detail::guid_initializer<C>& <unnamed>::boost_serialization_guid_initializer_10’ FileSExportAB.h:10: error: ‘<unnamed>::boost_serialization_guid_initializer_10’ has a previous declaration as ‘const boost::archive::detail::guid_initializer & <unnamed>::boost_serialization_guid_initializer_10’ ------------------------------------------------------------------------------------------------ All A, B, C and D are in same namespace. Putting things at different namespace is overkill on legacy code. Is using BOOST_CLASS_EXPORT_GUID only better solution? Is it not pssible for boost to resolve it automatically with BOOST_CLASS_EXPORT macro? Regards, ~ Soumen -- View this message in context: http://www.nabble.com/-Boost-Serialization--Problem-with-BOOST_CLASS_EXPORT… Sent from the Boost - Users mailing list archive at Nabble.com.

3 3

cycle iterators
by Neal Becker 03 Sep '09

03 Sep '09

I use these regularly. Work for me.

4 13

[iterator_adaptor] Does the base need a const version of operator* ?
by Andre Massing 03 Sep '09

03 Sep '09

Hi, I am trying to wrap our kind of iterators (for meshes) with the iterator_adaptor class template, something like class CellBoostIterator : public boost::iterator_adaptor< CellBoostIterator //Derived , CellIterator //Base , Cell //Value , boost::forward_traversal_tag // CategoryOrTraversal > { ... }; Since the base class CellIterator already provides a operator* and a operator-> (both non-const version) I did not redefine any of these within the CellBoostIterator class. If I now want to use with code-snippet typedef Mesh_2::CellBoostIterator MIT; MIT mit; Mesh_2 mesh_2(omega1); for (MIT mit = mesh_2.begin(); mit != mesh_2.end(); ++mit) { std::cout << mit->index() <<std::endl; } I get a compile error /usr/include/boost/iterator/iterator_adaptor.hpp:312: error: passing ‘const dolfin::CellIterator’ as ‘this’ argument of ‘dolfin::Cell& dolfin::CellIterator::operator*()’ discards qualifiers I looked at the corresponding line /usr/include/boost/iterator/iterator_adaptor.hpp:312: typename super_t::reference dereference() const { return *m_iterator; } So it seems to me that I have to provide a const version of operator*, or do I get something wrong? I am a bit confused, since I have not found this kind of requirements in the documentation. Any enlightening, hints or pointers are greatly appreciated! Greetings, Andre

1 0

BGL new graph type implementation trouble
by Sebastian Weber 03 Sep '09

03 Sep '09

Hi! I am currently trying to implement a new graph type which I want to use with the BGL. The graph implements the Incidence Graph concept and should work out of the box with DFS algorithm of the BGL. However, I had to find out that the edge_descriptor must obey an EqualityComparison concept which keeps DFS from compiling. My current edge_descriptor is just a typedef to std::pair<vertex_descriptor, vertex_descriptor>. The vertex_descriptor itself is a custom object I wrote myself. Another problem turns out to appear with the copy-graph algorithm. This algorithm tries to copy vertex_all-property maps and edge_all property maps which I simply do not have defined in my custom graph as I only support the vertex_index mechanism. So what to do here? How can I tell copy-graph that these maps are not existent? Any help would be great here as I have the impression to hit limits of the documentation (the Equality-thing is not mentioned in the docs, at least I did not find it...). Greetings, Sebastian

3 10

compile boost thread lib for iphone failed! the darwin g++ cant find head files for STL (string, cstddef)..how to solve this problem?
by terryc tang 03 Sep '09

03 Sep '09

i do the work as following: 0.install bjam 1. down boost_1_40_0 2. untar to disk root directory 3. ./bootstrap.sh ---with-library=thread 4. configure user-config.jam as following: using darwin : 4.2.1~iphone : /Developer/Platforms/iPhoneOS.platform/Developer/usr/bin/g++ : <striper> : <architecture>arm <target-os>iphone <macosx-version>iphone-2.1 ; using darwin : 4.2.1~iphonesim : /Developer/Platforms/iPhoneSimulator.platform/Developer/usr/bin/g++ : <striper> : <architecture>x86 <target-os>iphone <macosx-version>iphonesim-2.1 ; 5. command at console: bjam --prefix=/usrLibs/boostiPhoneDev toolset=darwin architecture=arm target-os=iphone macosx-version=iphone-2.1 define=_LITTLE_ENDIAN link=static install ... ... .... .. Any setting missing ? how to deal with it? In file included from ./boost/config.hpp:40, from ./boost/thread/detail/config.hpp:10, from libs/thread/src/pthread/thread.cpp:8: ./boost/config/select_stdlib_config.hpp:17:19: error: cstddef: No such file or directory In file included from ./boost/config/select_stdlib_config.hpp:33, from ./boost/config.hpp:40, from ./boost/thread/detail/config.hpp:10, from libs/thread/src/pthread/thread.cpp:8: ./boost/config/no_tr1/utility.hpp:21:21: error: utility: No such file or directory In file included from ./boost/thread/pthread/thread_data.hpp:9, from ./boost/thread/thread.hpp:17, from libs/thread/src/pthread/thread.cpp:10: ./boost/thread/exceptions.hpp:19:18: error: string: No such file or directory ./boost/thread/exceptions.hpp:20:21: error: stdexcept: No such file or directory In file included from ./boost/smart_ptr/shared_ptr.hpp:27, from ./boost/shared_ptr.hpp:17, from ./boost/thread/pthread/thread_data.hpp:10, from ./boost/thread/thread.hpp:17,

2 3

[proto] char vs. int behaviour when matching
by Maurizio Vitale 03 Sep '09

03 Sep '09

Proto behaves in a different way when matching terminal<char> and terminal<int>. The only difference I'm aware is that it is implementation defined whether 'char' means 'signed char' or 'unsigned char' but it has to mean one of the two. I was surprised to hit the following: struct char_grammar : proto::or_< proto::terminal<signed char>, proto::terminal<unsigned char> > {}; struct extended_char_grammar : proto::or_< proto::terminal<char>, proto::terminal<signed char>, proto::terminal<unsigned char> > {}; struct int_grammar : proto::or_< proto::terminal<signed int>, proto::terminal<unsigned int> > {}; int main() { // WHY _NOT? BOOST_MPL_ASSERT_NOT ((proto::matches<proto::terminal<char>, char_grammar>)); BOOST_MPL_ASSERT ((proto::matches<proto::terminal<char>, extended_char_grammar>)); // GIVEN THAT: BOOST_MPL_ASSERT ((proto::matches<proto::terminal<int>, int_grammar>)); return 0; } Is this behaviour intentional? Regards, Maurizio

5 5

[Unordered] [Multi-Index] Set lookup from something else than key_type
by Dominique Devienne 02 Sep '09

02 Sep '09

We currently have a unordered_map< Uid, shared_ptr<Identity> >, where Identity already contains the Uid. So I thought I could avoid the duplicated Uid by using an unordered_set< shared_ptr<Identity>, MyHash, MyPred > to hash/compare by the internal Uid, but unlike a Multi-Index hashed_index [1], unordered doesn't seem to provide: "lookup operations accepting search keys different from the key_type of the index, which is a specially useful facility when key_type object are expensive to create." Any reason why unordered cannot allow similar "compatible key" lookups? What kind of overhead is there (memory use, runtime), if any, in using Multi-Index as opposed to unordered_set in this scenario? Using Multi-Index yields a more complex type and slower compiles it seems due to its meta-programming uses I guess. Thanks for any insight. --DD [1] http://www.boost.org/doc/libs/1_40_0/libs/multi_index/doc/tutorial/basics.h…

5 14

Boost::bind: compiling error from sample code in book
by Boost lzw 02 Sep '09

02 Sep '09

Hi there, I am testing the following sample code on page 264 of the book "Beyond the C++ Standard Library" and got an error msg: "result_type' : is not a member of '`global namespace'' from MSVC 2008 Pro compiler. Can anyone point out what is wrong with it? Thanks. ------------ #include<iostream> #include<string> #include<map> #include<vector> #include<algorithm> #include "boost/bind.hpp" class print_size { typedef std::map<std::string, std::vector<int> > map_type; public: typedef void result_type; result_type operator()(std::ostream& os, const map_type::value_type& x) const { os << x.second.size() << '\n'; } }; int main() { typedef std::map<std::string, std::vector<int> > map_type; map_type m; m["Strange?"].push_back(1); m["Strange?"].push_back(2); m["Wierd?"].push_back(3); m["Wierd?"].push_back(4); m["Wierd?"].push_back(5); std::for_each(m.begin(), m.end(), boost::bind(&print_size(), boost::ref(std::cout), _1)); } ------------

2 2

[GIL] GIL and FreeImage interaction sample code suggestions ?
by Nicolas Lelong 02 Sep '09

02 Sep '09

Hi, FWIW I'm considering boost GIL to implement the few image algorithms I need, and I need proper image I/O on top of that. As I'm using FreeImage in my current codebase, I thought about exposing FreeImage image memory layout to GIL. You can find the, quite simplistic, source code here : http://code.google.com/p/rotoglup-scratchpad/source/browse/trunk/rtgu/image… http://code.google.com/p/rotoglup-scratchpad/source/browse/trunk/rtgu/image… http://code.google.com/p/rotoglup-scratchpad/source/browse/trunk/rtgu/image… I created a 'borrowed_image' class that mimics the behaviour of gil::image : it is not allocating memory itself, it takes an existing memory buffer and a 'deleter' (under the form of a boost::shared_ptr<void>) as input. This allows to use directly the memory allocated by FreeImage, avoiding extra copy, in case of file input. Feel free to use the code as a starting point, if you have similar needs, and I'd be glad to have comments about the concept of 'borrowed_image' class, and about its interface, as I feel I'm missing something. Cheers, Nicolas.

2 3

Re: [Boost-users] [mpl, units] rational and static_rational
by Matthias Schabel 02 Sep '09

02 Sep '09

> I am looking for a template library that can handle compile time > rationals, ideally with a similar interface to mpl::int_ and that > plays well with boost::rational (the dynamic type). > > First I found boost/mpl/math/rational_c.hpp but soon I realized that > it doesn't work with basic operations like mpl::plus, it doesn't even > simplify reducible fractions (i.e. 2/4 -> 1/2), in fact it looks like > it is just the storage of two (static) integers. Is this something > that will be developed in the future or am I using it in the wrong > way? > > Then I found boost::units::static_rational which is more promising, > because it simplifies fractions and works with mpl::plus, but it is > not integrated with boost::rational nor has an interface similar to > that of mpl metatypes. > > Is there a way of extending mpl::math::rational_c with the existing > libraries in boost, or a simple way to make > boost::units::static_rational have an interface similar to mpl and > also have a value type that returns it runtime boost::rational value? I certainly would have no objection to repackaging the static_rational type used in Boost.Units as an MPL type that plays nicely with Boost.Rational. My advice would be to put together a functional implementation and propose it for a mini-review. We could easily typedef units::static_rational -> mpl::rational_ in the units library. Matthias

2 1