- Boost-users - lists.boost.org

write_graphviz problem
by Michał Nowotka 09 Dec '07

09 Dec '07

Hello, I'm boost newbie and have some problems with write_graphviz function. I've already asked this question but perhaps with bad subject so i try again. Sorry if you read this second time. In write_graphviz function documentation I can read that PropertyWriter can be used for print both: interior and exterior properties for each vertex of a graph into the graphviz format file. I can find some example of printing graph properties stored in property maps, for example calling: write_graphviz(std::cout, g, make_label_writer(name)); when the graph have 'name' property map. but how to do that when storing properies as bundled properies? I defined following graph: typedef adjacency_list<vecS, vecS, bidirectionalS, Vect_f, Edge_f> Graph; where: struct Vect_f { int factor; int lowlink; int component; int color; bool in_stack; }; struct Edge_f { int num_visited; int color_k; }; and I want to lay out and print my graph with coloured vertexes (I can always change the type of colour property for more suitable if necessary) but I dont't have any idea to do this. Is it possible at all? Cheers, Michał Nowotka

2 2

boost::filesystem: Problem with exists() and wpath (Linux)
by Andreas Henriksson 09 Dec '07

09 Dec '07

Hello Whenever I try to use exists() on a wpath, I get an exception. I'm running Arch Linux, with Boost 1.34.1 and Gcc 4.2.2, if it matters. Here is some code: #include "boost/filesystem.hpp" using namespace boost::filesystem; int main() { wpath test(L"whatever/path"); exists(test); return 0; } Which compiles fine, but gives me this error message on runtime: terminate called after throwing an instance of 'boost::filesystem::basic_filesystem_error<boost::filesystem::basic_path<std::basic_string<wchar_t, std::char_traits<wchar_t>, std::allocator<wchar_t> >, boost::filesystem::wpath_traits> >' what(): boost::filesystem::wpath::to_external conversion error Aborted Does anyone know what I am missing here?

1 0

boost::filesystem and mingw
by Lennart Wagner 09 Dec '07

09 Dec '07

Hi, I am using boost v1.34.1 and the "current" version of mingw. I got a problem (and solution) with the filesystem library: If the file convenience.hpp is included multiple times, mingw/gcc says: In function `ZN5boost10filesystem9extensionERKNS0_10basic_pathISsNS0_11path_traitsEEE':C:/Daten/Programme/Dev/boost_1_34_1/include/boost/filesystem/convenience.hpp:58: multiple definition of `boost::filesystem::extension(boost::filesystem::basic_path const&)' obj\Debug\Source\global.o:C:/Daten/Programme/Dev/boost_1_34_1/include/boost/filesystem/convenience.hpp:58: first defined here I can solve that (at least for me), if line 31 in convenience.hpp # define BOOST_FS_FUNC_STRING std::string is changed to # define BOOST_FS_FUNC_STRING inline std::string Another change I did to avoid an anoying warning is in line 45 boost::throw_exception( basic_filesystem_error<Path>( "boost::filesystem::create_directories", ph, -1 ) ); I changed the third parameter to 0, since it is unsigned. Have a nice day, Lennart

2 2

Re: [Boost-users] [serialization] Thread safety
by Marc Viala 09 Dec '07

09 Dec '07

> De la part de Robert Ramey: > > the SVN branch serialization_next_release contains this code. > > However best wait til monday as I need to upload a couple of > tweaks. > > Robert Ramey > > Kim Barrett wrote: > >> I came > >> with a solution that has no locking which was important > >> to me. But its only applicable in some applications > >> and serialization library is one of them. In anycase > >> I would welcome comments from anyone who has > >> interest in looking at it. > > > > I'm interested. Can you provide some pointers to specific places to > > look? I should have some time for this (and the test case for the > > polymorphic archive performance patch) after next week. > I'm interested me too, but is it possible to merge this SVN branch code of serialization (serialization_next_release) w/ the Boost 1.34.1 distribution? Or is it required to download the Boost SVN current trunk? Best regards and thanks you in advance for your support. Marc Viala

2 1

[serialization, boost::function] stateful function objects
by Ruediger Berlich 09 Dec '07

09 Dec '07

Hi there, I have a class that is being serialized in regular intervals. It contains a vector<shared_ptr<somefunctionObject> > . As the somefunctionObject objects can maintain state, I want to be able to serialize them along with the surrounding class. This works fine with the current design. I would find it much more pleasing to replace the above vector e.g. with a vector<boost::function<void(someType)> >. From the program logic this would have the same effect, but would be more elegant. I can not see, though, how serialization would work in this case. As far as I know it is not possible to retrieve the object back from boost::function. And it is also not possible to specify more than one interface to boost::function, e.g. in order to support two operator() with different signatures like this: "boost::function<void(someFunctionObkject), someFunctionObject(void)> f;" If that was possible, I could just give the function object a "clone" function and write my own (de-)serialization code. Any ideas ? Thanks and Best Regards, Ruediger Berlich

5 9

trouble compiling with -m32 or address-model=32
by John Salmon 09 Dec '07

09 Dec '07

I'm trying to build boost_1_34_0 on an x86_64 linux box with a 32-bit address model. After letting configure create a bjam for me, I have tried many variations on the following: tools/jam/src/bin.linux/bjam address-model=64 -d2 release And also: tools/jam/src/bin.linux/bjam cxxflags=-m32 linkflags=-m32 cflags=-m32 -d2 release The -d2 appears to show me what the compiler is doing. I see no sign of -m32 command-line options when the compiler is invoked. I also checked the .so file that was created and verifited that it's dynamically linked to libraries in /lib64. What am I doing wrong?? Here's some of the -d2 output with address-model: salmonj(a)drdws032.nyc$ tools/jam/src/bin.linux/bjam address-model=32 -d2 release Building Boost.Regex with the optional Unicode/ICU support disabled. Please refer to the Boost.Regex documentation for more information (don't panic: this is a strictly optional feature). warning: No toolsets are configured. warning: Configuring default toolset "gcc". warning: If the default is wrong, you may not be able to build C++ programs. warning: Use the "--toolset=xxxxx" option to override our guess. warning: For more configuration options, please consult warning: http://boost.org/boost-build2/doc/html/bbv2/advanced/configuration.html ...patience... ...patience... ...found 4006 targets... ...updating 756 targets... gcc.compile.c++ bin.v2/libs/python/build/gcc-3.4.4/release/address-model-32/threading-multi/type_id.o "g++" -ftemplate-depth-128 -O3 -finline-functions -Wno-inline -Wall -fPIC -pthread -DBOOST_ALL_NO_LIB=1 -DBOOST_PYTHON_SOURCE -DNDEBUG -I"." -I"/usr/include/python2.3" -c -o "bin.v2/libs/python/build/gcc-3.4.4/release/address-model-32/threading-multi/type_id.o" "libs/python/src/converter/type_id.cpp" gcc.compile.c++ bin.v2/libs/python/build/gcc-3.4.4/release/address-model-32/threading-multi/enum.o "g++" -ftemplate-depth-128 -O3 -finline-functions -Wno-inline -Wall -fPIC -pthread -DBOOST_ALL_NO_LIB=1 -DBOOST_PYTHON_SOURCE -DNDEBUG -I"." -I"/usr/include/python2.3" -c -o "bin.v2/libs/python/build/gcc-3.4.4/release/address-model-32/threading-multi/enum.o" "libs/python/src/object/enum.cpp" And here's some more with cxxflags/cflags/linkflags: salmonj(a)drdws032.nyc$ tools/jam/src/bin.linux/bjam cflags=32 cxxflags=-m32 linkflags=-m32 -d2 release Building Boost.Regex with the optional Unicode/ICU support disabled. Please refer to the Boost.Regex documentation for more information (don't panic: this is a strictly optional feature). warning: No toolsets are configured. warning: Configuring default toolset "gcc". warning: If the default is wrong, you may not be able to build C++ programs. warning: Use the "--toolset=xxxxx" option to override our guess. warning: For more configuration options, please consult warning: http://boost.org/boost-build2/doc/html/bbv2/advanced/configuration.html ...patience... ...patience... ...found 3994 targets... ...updating 747 targets... MkDir1 bin.v2/libs/python/build/gcc-3.4.4/release/threading-multi mkdir "bin.v2/libs/python/build/gcc-3.4.4/release/threading-multi" gcc.compile.c++ bin.v2/libs/python/build/gcc-3.4.4/release/threading-multi/numeric.o "g++" -ftemplate-depth-128 -O3 -finline-functions -Wno-inline -Wall -fPIC -pthread -DBOOST_ALL_NO_LIB=1 -DBOOST_PYTHON_SOURCE -DNDEBUG -I"." -I"/usr/include/python2.3" -c -o "bin.v2/libs/python/build/gcc-3.4.4/release/threading-multi/numeric.o" "libs/python/src/numeric.cpp" gcc.compile.c++ bin.v2/libs/python/build/gcc-3.4.4/release/threading-multi/list.o "g++" -ftemplate-depth-128 -O3 -finline-functions -Wno-inline -Wall -fPIC -pthread -DBOOST_ALL_NO_LIB=1 -DBOOST_PYTHON_SOURCE -DNDEBUG -I"." -I"/usr/include/python2.3" -c -o "bin.v2/libs/python/build/gcc-3.4.4/release/threading-multi/list.o" "libs/python/src/list.cpp" gcc.compile.c++ bin.v2/libs/python/build/gcc-3.4.4/release/threading-multi/long.o "g++" -ftemplate-depth-128 -O3 -finline-functions -Wno-inline -Wall -fPIC -pthread -DBOOST_ALL_NO_LIB=1 -DBOOST_PYTHON_SOURCE -DNDEBUG -I"." -I"/usr/include/python2.3" -c -o "bin.v2/libs/python/build/gcc-3.4.4/release/threading-multi/long.o" "libs/python/src/long.cpp"

1 0

about some boost-config helper script
by François Mauger 08 Dec '07

08 Dec '07

Hi boost users, I'm new in the boost (amazing and powerfull) universe and had no time yet to explore archives about this topic. I have just installed the current version of boost libs on my laptop (no problem with it) but I have not found a dedicated helper config script (i.e. boost-config [--cflags|--libs...]) like the ones coming with the well-known packages used in high-energy physics (GSL, ROOT, CLHEP). It will help a lot if such a script was provided by the build process, at least for Unix-like systems. It is very useful within makefiles or whatever "manual" build scripts. Maybe someone can tell why this feature is not in boost (probably there is a good reason I'm not aware of) or if there are plans to provide it. Thank you very much for your kind comment. best regards frc -- Pr. Francois Mauger Laboratoire de Physique Corpusculaire de Caen et Universite de Caen ENSICAEN - 6, Boulevard du Marechal Juin, 14050 CAEN Cedex, FRANCE e-mail: mauger(a)lpccaen.in2p3.fr tel.: (0/+33) 2 31 45 25 12 fax: (0/+33) 2 31 45 25 49

1 0

boost.proto optimization on MSVC (2005/2008)
by AJD 08 Dec '07

08 Dec '07

Hi, I've been trying out boost.proto lately and I've noticed that for my expressions it tends to force my operands to be created on the stack. For example, instead of optimizing out an object completely when everything about it is known at compile-time, the compiler (MSVC 2005 or 2008) is creating that object on the stack anyway.. even though it's not necessary. Even worse, it actually folds the simpler expressions at compile-time and never even touches the data that it just put on the stack. The problem is compounded with larger expressions involving many operands. It even seems to build the larger expression objects on the stack unnecessarily (and this looks like it starts to break constant folding). Here's some of the generated assembly: int main() { 00401800 sub esp,18h vector2< float[3] > v1 = { 0, 1, 2 }; 00401803 xorps xmm0,xmm0 vector2< float[3] > v2 = { 3, 4, 5 }; vector2< float[3] > v3 = { 6, 7, 8 }; // Add two vectors lazily and get the 2nd element. #if 1 // this sticks temporary junk all over the stack, even with 2 terms std::cout << (v1 + v2)[0]; 00401806 fld dword ptr [__real@40400000 (402128h)] 0040180C movss dword ptr [esp],xmm0 00401811 movss xmm0,dword ptr [__real@3f800000 (402124h)] 00401819 movss dword ptr [esp+4],xmm0 0040181F movss xmm0,dword ptr [__real@40000000 (402120h)] 00401827 movss dword ptr [esp+8],xmm0 0040182D movss xmm0,dword ptr [__real@40400000 (402128h)] 00401835 movss dword ptr [esp+0Ch],xmm0 0040183B movss xmm0,dword ptr [__real@40800000 (40211Ch)] 00401843 push ecx 00401844 mov ecx,dword ptr [__imp_std::cout (402044h)] 0040184A fstp dword ptr [esp] 0040184D movss dword ptr [esp+14h],xmm0 00401853 movss xmm0,dword ptr [__real@40a00000 (402118h)] 0040185B movss dword ptr [esp+18h],xmm0 00401861 call dword ptr [__imp_std::basic_ostream<char,std::char_traits<char> >::operator<< (402040h)] std::cin.get(); 00401867 mov ecx,dword ptr [__imp_std::cin (40203Ch)] 0040186D call dword ptr [__imp_std::basic_istream<char,std::char_traits<char> >::get (402038h)] return 0; 00401873 xor eax,eax } 00401875 add esp,18h 00401878 ret I've come up with a couple workarounds but these only work for specific cases: // (1) This optimizes out nicely, but only works for adding 2 objects. typedef proto::terminal< vector2< float[3] > >::type expr_type; std::cout << (vector_expr< expr_type >( expr_type::make(v1) ) + vector_expr< expr_type >( expr_type::make(v2) ))[0]; // (2) This never optimizes out std::cout << (v1 + v2)[0]; I'd noticed that the reference at traits.hpp line 144 seemed to play a major role (it was the only significant difference between the expression trees generated by the above two snippets.) // (3) This makes (2) optimize just like (1), but doesn't help 3+ // HACK TO REMOVE REF AT boost\xpressive\proto\traits.hpp LINE 144 namespace boost { namespace proto { namespace result_of { template<typename T, typename EnableIf> struct as_arg<T, VectorDomain, EnableIf> { typedef expr<proto::tag::terminal, args0<T /*&*/> > expr_type; typedef typename VectorDomain::template apply<expr_type>::type type; template<typename T2> static type call(T2 &t) { return VectorDomain::make(expr_type::make(t)); } }; }}} Finally, here is the entire source I'm using: #include <cstddef> #include <iostream> #include <boost/xpressive/proto/proto.hpp> #include <boost/xpressive/proto/context.hpp> using namespace boost; template< typename T> struct vector2; template< typename T > struct vector2<T[3]> { T v[3]; typedef T value_type; T const & operator [](std::size_t i) const { return this->v[i]; } T & operator [](std::size_t i) { return this->v[i]; } }; // This grammar describes which lazy vector expressions // are allowed; namely, vector terminals and addition // and subtraction of lazy vector expressions. struct VectorGrammar : proto::or_ < proto::terminal< vector2< proto::_ > > , proto::plus< VectorGrammar, VectorGrammar > > { }; // Expressions in the lazy vector domain must conform // to the lazy vector grammar struct VectorDomain; // Here is an evaluation context that indexes into a lazy vector // expression, and combines the result. struct subscript_context { private: std::size_t i; public: subscript_context(std::size_t i) : i(i) { } // Use default_eval for all the operations ... template< typename Expr, typename Tag = typename Expr::proto_tag > struct eval : proto::default_eval< Expr, subscript_context const > { }; // ... except for terminals, which we index with our subscript template< typename Expr > struct eval< Expr, proto::tag::terminal > { typedef typename proto::result_of::arg< Expr >::type::value_type result_type; result_type operator ()(Expr const & expr, subscript_context const & ctx) const { return proto::arg(expr)[ctx.i]; } }; }; // Here is the domain-specific expression wrapper, which overrides // operator [] to evaluate the expression using the subscript_context. template< typename Expr > struct vector_expr : proto::extends<Expr, vector_expr< Expr >, VectorDomain > { typedef proto::extends< Expr, vector_expr< Expr >, VectorDomain > base_type; vector_expr(Expr const & expr) : base_type(expr) { } // Use the subscript_context to implement subscripting // of a lazy vector expression tree. typename proto::result_of::eval< Expr, subscript_context const >::type operator [](std::size_t i) const { subscript_context const ctx(i); return proto::eval(*this, ctx); } }; // Tell proto that in the VectorDomain, all // expressions should be wrapped in vector_expr<> struct VectorDomain : proto::domain< proto::generator< vector_expr >, VectorGrammar > { }; template< typename T > struct is_vector : mpl::false_ { }; template< typename T > struct is_vector< vector2< T > > : mpl::true_ { }; BOOST_PROTO_DEFINE_BINARY_OPERATOR(+, boost::proto::tag::plus, is_vector, VectorDomain); int main() { vector2< float[3] > v1 = { 0, 1, 2 }; vector2< float[3] > v2 = { 3, 4, 5 }; vector2< float[3] > v3 = { 6, 7, 8 }; // Add two vectors lazily and get the 2nd element. #if 1 // this sticks write-only temporary junk all over the stack, even with 2 terms std::cout << (v1 + v2)[0];// + v3 + v3 + v3 + v3 + v3)[0]; #else // this optimizes out completely for 2 terms... expanding to 3 terms does not typedef proto::terminal< vector2< float[3] > >::type expr_type; std::cout << (vector_expr< expr_type >( expr_type::make(v2) ) + vector_expr< expr_type >( expr_type::make(v3) ))[0]; #endif std::cin.get(); return 0; } Thanks in advance. // AJD < n.outpost(a)gmail.com >

3 7

[serialization] Thread safety
by Ruediger Berlich 08 Dec '07

08 Dec '07

Hi there, I remember that, a few months ago, there was a discussion that Boost.Serialization was currently only thread-safe when using g++ . I cannot find any specific information on this topic, however. I have distinct objects, each of which is serialized in its own (boost) thread and then sent over a network connection and deserialized at the other end. Seems to work fine, without synchronization between different serialization calls. I am using Boost 1.34.1 with Asio 0.3.8, together with threadpool 0.2.4. This is on an OpenSuSE 10.3 box with g++ 4.2.1 . Will I run into trouble when I port this to other compilers + platforms ? Thanks and Best Regards, Ruediger

4 9

BUG ?? Date_time
by paul 08 Dec '07

08 Dec '07

/* There seems to be something wrong with the month() member function of boost::gregorian::date if you valgrind the executable of (*this) source file, you will see that some blocks are "possibly lost" and some are also "still reachable". Yet, were you to comment out the line making use of month(), you would see that valgrind sees no problem with the code. what seems to be the problem ? uname -a = Linux Stanford 2.6.22-14-generic #1 SMP Sun Oct 14 23:05:12 GMT 2007 i686 GNU/Linux */ #include <boost/date_time/gregorian/gregorian.hpp> using namespace boost::gregorian ; #include <iostream> using namespace std ; int main() { date today( day_clock::local_day() ) ; cout << "year = " << today.year() << endl ; cout << "month = " << today.month() << endl ; cout << "day = " << today.day() << endl ; return EXIT_SUCCESS; }

1 0