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

Boost test support for "Abstract Test Pattern"
by Bruce Trask 23 Jul '08

23 Jul '08

Hi, Does the boost test framework provide facilities out of the box for the Abstract Test Pattern? http://c2.com/cgi/wiki?AbstractTest Thanks in advance. Regards, Bruce

3 3

Boost.Date_Time - How to measure time elapsed?
by Jean-Sebastien Stoezel 23 Jul '08

23 Jul '08

Hello, I've been struggling with how to use the Date_Time library to measure time elapsed (like to measure how long code takes to execute). Could somebody provide me with an example? Resolution should be around msec. I could not find any in the reference that particularly addresses this. Thanks! Jean

2 2

Re: [Boost-users] Boost.Date_Time - How to measure time elapsed?
by Igor R 23 Jul '08

23 Jul '08

// untested!! namespace pt = boost::posix_time; pt::ptime now1 = pt::microsec_clock::universal_time(); ///....do something here... pt::ptime now2 = pt::microsec_clock::universal_time(); pt::time_duration dur = now2 - now1; you can get then dur.total_milliseconds(); > I've been struggling with how to use the Date_Time library to measure > time elapsed (like to measure how long code takes to execute). > > Could somebody provide me with an example? Resolution should be around > msec. I could not find any in the reference that particularly > addresses this. >

1 0

Internal Error VS6 IDE-HELP!
by anil kumar 23 Jul '08

23 Jul '08

Hi All, Please help me out I am new to boost C++. I tried out this code #include "boost\lambda\lambda.hpp" #include <iostream> #include <iterator> #include <algorithm> int main() { using namespace boost::lambda; typedef std::istream_iterator<int> in; std::for_each( in(std::cin), in(), std::cout << (boost::lambda::_1 * 3) << " " ); } I am using Visual studio 6.0 VC++ it shows an error i have included the additional directory and other things as explained in the boost intro tutorial still getting an error. it says c:\program files\boost\boost_1_35_0\boost\lambda\detail\lambda_fwd.hpp(26) : fatal error C1001: INTERNAL COMPILER ERROR (compiler file 'msc1.cpp', line 1794) Please choose the Technical Support command on the Visual C++ How to solve this i hope some one will help why this internal error occurs and what i need to do?. Anil _________________________________________________________________ Missed your favourite programme? Stop surfing TV channels and start planning your weekend TV viewing with our comprehensive TV Listing http://entertainment.in.msn.com/TV/TVListing.aspx

4 3

[boost] [graph] Newbie question: what is the template type of setS and relation with bundled properties?
by Florent Teichteil 23 Jul '08

23 Jul '08

Hi, I read the documentation and the header files, but I can't figure out what is the template type used in vecS, listS or setS. In my application, I need each vertex be unique and ordered so that I think about using setS. I use the bundled vertex property MyVertex to define my vertex class storage. The code perfectly compiles whereas I have not defined a comparison operator for MyVertex objects. It means that setS is not defined as setS<MyVertex>... What is the template type of setS? What is the relation between this template type and the bundled property MyVertex? Thanks a lot, Florent

1 0

Re: [Boost-users] [serialization] Problems with BOOST_IS_ABSTRACT, gcc 3.2.3 and Boost 1.35.0
by Dan Thomas 23 Jul '08

23 Jul '08

> Try using boost/serialization/assume_abstract.hpp Thanks for the quick reply. I tried this change but it made no difference. I still get the same errors. I have dealt with my immediate problems by rolling back to boost 1.33.0 so I will stop pestering you about this :-) Thanks for your help Dan

2 1

please help : Error: Unexpected type name
by Sun, Xiuxin 23 Jul '08

23 Jul '08

> HI all, > when I try to compile the example code only use boost head-file in > SunOS : > > #include <boost/lambda/lambda.hpp> > #include <iostream> > #include <iterator> > #include <algorithm> > > int main() > { > using namespace boost::lambda; > typedef std::istream_iterator<int> in; > > std::for_each( > in(std::cin), in(), std::cout << (_1 * 3) << " " ); > } > > I get many errors like 'Unexpected type name', the first one is : > > CC -g -I../. -c -o main.o main.C > ".././boost/lambda/detail/function_adaptors.hpp", line 203: Error: > Unexpected type name "boost::lambda::Arg1" encountered. > could someone give a hand. > sun

5 9

[fusion] sequence conversion
by Istvan Buki 23 Jul '08

23 Jul '08

Dear Boost experts, I'm trying to copy data between different types of fusion sequences with limited success. The example below shows what I'm trying to achieve. Any good advice would be welcome. namespace fields { struct i ; struct l ; struct d ; } int main() { using namespace boost::fusion ; typedef map < pair< fields::i, int >, pair< fields::l, long >, pair< fields::d, double > > map_type ; typedef vector< int, long, double > vect_t ; vect_t tup1( 1, 2, 3.0 ) ; map_type m = tup1 ; // OK. Works as expected vect_t tup2 = m ; // Error. Conversion not possible // expected result: tup1 == tup2 } Does fusion provide a way to achieve what I'm trying to do? As a side question, I was wondering whether it is possible to synthetize the vect_t type from map_type (with MPL?) A concrete example where this could be useful is an application that manipulates map_type objects and to persist these objects it uses the SOCI database library that supports fusion vectors. Before inserting a row in the database, the map_type object should be converted to a vect_t object and when retrieving a row from the database a vect_t object should converted to a map_type object. Thanks for your help.

7 44

Choice of boost containers
by Ramesh 23 Jul '08

23 Jul '08

Hi, I am a newbie to C++ & boost, I have questions on choosing the right boost container for my requirement. *A base class :* class Player { protected: std::string PlayerId; // real, winamp, Windows media, vlc, divx long capMask // capability mask public: Player(std::string param_Id, long param_Mask) { Id = param_Id; capMask = param_Mask; } }; *Two derived classes:* *Derived Class I:* class Port_Player:public Player { protected: bool IsTouchScreen; // True if touch screen based public: Port_Player(std::string param_Id, long param_Mask, bool TS) :Player(param_Id, param_Mask) { IsTouchScreen = TS; } } *Dervied Class II:* class Non_Port_Player:public Player { protected: long Browser; // Supported by firefox, IE etc public: Non_Port_Player(std::string param_Id, long param_Mask, long pBrowser) :Player(param_Id, param_Mask) { Browser = pBrowser; } } Questions I have: 1) As I would create objects of either Derived Class I or Derived Class II (which would be known to be only at runtime) and I need to add pointers to these objects in a container, I would like to know which container would suit best?? 2) Also, I should be able to perform lookups based on playerId field in the container,playerId would be an auto generated field for every object created (irrespective of Derived Class I or Derived class II), please advice on which container to choose in boost. BTW I had a look at the multi-index container, in which case I would be replicating data from the objects that I create, I am rather looking to see if I can avoid copying data rather than do lookups based on the object addresses in the container. Thanks in advance for any suggestions / advice on this. Regards Ramesh

2 1

problem compiling ublas vector iterator with templates
by Alain Leblanc 22 Jul '08

22 Jul '08

Hi, I'm trying to compile the following piece of code with gcc version 4.2.3 (Ubuntu 4.2.3-2ubuntu7) and I keep getting the error message listed below. A colleague of mine told me that he can compile it on MSVC. If I remove the template and set everything as vector<double> then it compiles without any problem. Anyone has an idea of what I may be doing wrong? Thanks, Alain <... bunch of include statements ...> namespace UBLAS = boost::numeric::ublas; //summation of a vector template <class Tv> Tv vectorSum(const UBLAS::vector< Tv >& A) { Tv sum = 0; for (UBLAS::vector< Tv >::const_iterator it = A.begin(); it!=A.end(); ++it) sum += (*it); return sum; } g++ -I. -I./vector testvector.cc In file included from testvector.cc:43: vector/TVector.h: In function 'Tv vectorSum(const boost::numeric::ublas::vector<Tv, boost::numeric::ublas::unbounded_array<T, std::allocator<_CharT> > >&)': vector/TVector.h:509: error: expected `;' before 'it' vector/TVector.h:509: error: 'it' was not declared in this scope

2 2

1.35.0 && serialization && spirit && thread safety
by Tim St. Clair 22 Jul '08

22 Jul '08

I've seen info in various email-threads regarding thread safety when using xml archives. We had come across some assert conditions in multithreaded code, only to trace down that (BOOST_SPIRIT_THREADSAFE) was not defined during boost compilation. to fix, I modified the Jamfile.v2 for serialization: lib boost_serialization : $(SOURCES).cpp /boost/thread//boost_thread # needed to add this as a dep b/c we need to link threads when we define BOOST_SPIRIT_THREADSAFE : <define>BOOST_SPIRIT_THREADSAFE # needed to add this as a define <toolset>msvc:<cxxflags>/Gy <link>shared:<define>BOOST_SERIALIZATION_DYN_LINK=1 <conditional>@include-spirit ; I've also added BOOST_SPIRIT_THREADSAFE as a <define> in our code as well. I hope this helps others... Cheers, Tim

1 0

[serialization 1_36_0] : Binary Archive portability between 32 and 64 bits platforms...
by Mathieu Peyréga 22 Jul '08

22 Jul '08

Hello, I'm wondering if a binary archive generated on a 32 bits platform would be readable on a 64 bits one... My guess is no... because i am facing an issue i strongly suspect to come from there... Any clues ? I got a basic_string::resize exception evene before trying to call the serialize method... just doing : std::ifstream fichier(filename.c_str(),std::ios::binary); boost::archive::binary_iarchive archive(fichier); the second line causes the exception... Regards, Mathieu -- http://www.incub.net/

3 5

[Serialization][STLport] Linking Error
by Kevin Tang 22 Jul '08

22 Jul '08

Dear All, I am using Boost library version 1.34.1 in VC7.1 sp1. I don't know why I always fail with linking Boost Serialization and STLport library. But I can successfully build with Boost Serialization with VC7.1 build-in STL library. Do I need to modify my STLport library for Boost? I try STLport v5.1.5, v5.0.3, and v4.6.2. both of these version of STLport also fail in Linking process. Thank you. Kevin Tang. Linking... Creating library *Debug/iSentenceRecognizer.lib* and object *Debug/iSentenceRecognizer.exp* *iSentenizer.obj :* *error LNK2019:* /unresolved external symbol "public: __thiscall *boost::archive::text_oarchive_impl*<class boost::archive::text_oarchive>::text_oarchive_impl<class boost::archive::text_oarchive>(...)/ *iSentenizer.obj :* *error LNK2019:* /unresolved external symbol "public: __thiscall *boost::archive::text_iarchive_impl*<class boost::archive::text_iarchive>::text_iarchive_impl<class boost::archive::text_iarchive>//(...)/ *iSentenizer.obj :* *error LNK2019:* /unresolved external symbol "public: __thiscall *boost::archive::basic_text_oprimitive*<class _STL::basic_ostream<char,class _STL::char_traits<char> > >::~basic_text_oprimitive<class _STL::basic_ostream<char,class _STL::char_traits<char> > >(void)" //(...)/ *iSentenizer.obj :* *error LNK2019:* /unresolved external symbol "public: __thiscall *boost::archive::basic_text_iprimitive*<class _STL::basic_istream<char,class _STL::char_traits<char> > >::~basic_text_iprimitive<class _STL::basic_istream<char,class _STL::char_traits<char> > >(void)" //(...)/ *iSentenizer.obj :* *error LNK2019:* /unresolved external symbol "public: void __thiscall *boost::archive::text_oarchive_impl*<class boost::archive::text_oarchive>::save(class _STL::basic_string<char,class _STL::char_traits<char>,class _STL::allocator<char> > const &)"/ /(...)/ *iSentenizer.obj :* *error LNK2019:* /unresolved external symbol "public: void __thiscall *boost::archive::text_iarchive_impl*<class boost::archive::text_iarchive>::load(class _STL::basic_string<char,class _STL::char_traits<char>,class _STL::allocator<char> > &)"/ /(...)/ Y:\2008\iSentenceRecognition\iSentenceRecognition/Debug/iSentenceRecognizer.dll : *fatal error LNK1120: 6 unresolved externals*

1 0

Automate call with n template parameters
by Kevin Jenkins 22 Jul '08

22 Jul '08

Is there any way to automate this, to support from 0 to n parameters? template <class P1, class P2, class P3, class P4, class P5> bool DoIt(P1 p1, P2 p2, P3 p3, P4 p4, P5 p5) { DoThat<P1>(p1); DoThat<P2>(p2); DoThat<P3>(p3); DoThat<P4>(p4); DoThat<P5>(p5); }

3 3

[test] Fatal link error LNK1561: entry point must be defined when using boost.test
by Tjernstrom, Staffan 22 Jul '08

22 Jul '08

(Resend with modified subject to better match list standards) Folks, Hopefully this is me having finger trouble, but I can't seem to get around the problem. Boost 1.35 installed using the binary installer, all the vc8 flavours installed. What happens is that I try to compile / link any of the unit test examples in Visual Studio 2005 Pro, for example the code at http://www.boost.org/doc/libs/test/example/unit_test_example_05.cpp I keep getting the LNK1561: entry point must be defined error. I know that it's looking for, and picking up, the libboost_unit_test_framework-vc80-mt-1_35.lib as if I remove the /LIBPATH from the link command options it compains that it cannot find that library. I have my project settings set as follows: Compiler: /O2 /GL /I "C:\Program Files (x86)\boost\boost_1_35_0" /D "_MBCS" /FD /EHa /MD /Fo"Release\\" /Fd"Release\vc80.pdb" /W3 /nologo /c /Wp64 /Zi /TP /errorReport:prompt Linker: /OUT:"D:\staffan\Documents\Visual Studio 2005\Projects\CeT3x\test\AllUnitTests\Release\test_XOLibrary.exe" /NOLOGO /LIBPATH:"C:\Program Files (x86)\boost\boost_1_35_0\lib" /MANIFEST /MANIFESTFILE:"Release\test_XOLibrary.exe.intermediate.manifest" /DEBUG /PDB:"d:\staffan\Documents\Visual Studio 2005\Projects\CeT3x\test\AllUnitTests\release\test_XOLibrary.pdb" /OPT:REF /OPT:ICF /LTCG /MACHINE:X86 /ERRORREPORT:PROMPT kernel32.lib user32.lib gdi32.lib winspool.lib comdlg32.lib advapi32.lib shell32.lib ole32.lib oleaut32.lib uuid.lib odbc32.lib odbccp32.lib Any thoughts anyone? Rgds Staffan Tj. Staffan Tjernström | Sr. Software Engineer - Market Data Systems | www.esignal.com <http://www.esignal.com/> eSignal, Interactive Data Desktop Solutions (NYSE: IDC) 3955 Point Eden Way | Hayward, CA 94545-3720 | 510.723.1723 | FAX: 510.266.6100 DISCLAIMER: This message (including any files transmitted with it) may contain confidential and / or proprietary information, is the property of Interactive Data Corporation and / or its subsidiaries and is directed only to the addressee(s). If you are not the designated recipient or have reason to believe you received this message in error, please delete this message from your system and notify the sender immediately. An unintended recipient's disclosure, copying, distribution or use of this message, or any attachments, is prohibited and may be unlawful.

4 7

Can't call boost::fusion::invoke with >= 7 args
by Kevin Jenkins 21 Jul '08

21 Jul '08

Hi, First time poster. I found Boost over the weekend and have to say it's a great library! It's enabled me to do things in C++ I didn't think were possible before. I'm posting because I'm having a problem with boost::fusion::invoke. I'm using it to call a function with a list of predefined parameters. I found that if I have 6 or fewer parameters, everything compiles fine. If I have 7 or more parameters, it doesn't compile. I verified that it's not the parameters by removing the first 4 and leaving the last 5 (which was OK). Here's the code that I think is relevant: boost::fusion::invoke(func,args); Args is an iterator typename boost::mpl::begin< boost::function_types::parameter_types<Function> >::type func is a function pointer passed from C++ This is on Visual Studio 2005 using boost 1.35. I need at least 9 parameters. Thanks in advance! For reference, the output from the compiler is pasted below ---------------------- 1>c:\boost_1_35_0\boost\fusion\functional\invocation\invoke.hpp(102) : error C2504: 'boost::fusion::detail::invoke_impl<Function,Sequence,N,CBI,RandomAccess>' : base class undefined 1> with 1> [ 1> Function=void (__cdecl *)(A *,A &,C *,C &,D *,D &,NormalizedVector *,NormalizedVector &,RakNet::RPC3 *), 1> Sequence=const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::nil,const boost::fusion::single_view<A *>>,const boost::fusion::single_view<A>>,const boost::fusion::single_view<C *>>,const boost::fusion::single_view<C>>,const boost::fusion::single_view<D *>>,const boost::fusion::single_view<D>>,const boost::fusion::single_view<NormalizedVector *>>,const boost::fusion::single_view<NormalizedVector>>,const boost::fusion::single_view<RakNet::RPC3 *>>, 1> N=9, 1> CBI=false, 1> RandomAccess=false 1> ] 1> c:\boost_1_35_0\boost\fusion\functional\invocation\invoke.hpp(110) : see reference to class template instantiation 'boost::fusion::detail::invoke_nonmember_builtin<F,Sequence,N,RandomAccess>' being compiled 1> with 1> [ 1> F=void (__cdecl *)(A *,A &,C *,C &,D *,D &,NormalizedVector *,NormalizedVector &,RakNet::RPC3 *), 1> Sequence=const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::nil,const boost::fusion::single_view<A *>>,const boost::fusion::single_view<A>>,const boost::fusion::single_view<C *>>,const boost::fusion::single_view<C>>,const boost::fusion::single_view<D *>>,const boost::fusion::single_view<D>>,const boost::fusion::single_view<NormalizedVector *>>,const boost::fusion::single_view<NormalizedVector>>,const boost::fusion::single_view<RakNet::RPC3 *>>, 1> N=9, 1> RandomAccess=false 1> ] 1> c:\boost_1_35_0\boost\fusion\functional\invocation\invoke.hpp(159) : see reference to class template instantiation 'boost::fusion::detail::invoke_impl<Function,Sequence>' being compiled 1> with 1> [ 1> Function=void (__cdecl *)(A *,A &,C *,C &,D *,D &,NormalizedVector *,NormalizedVector &,RakNet::RPC3 *), 1> Sequence=const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::nil,const boost::fusion::single_view<A *>>,const boost::fusion::single_view<A>>,const boost::fusion::single_view<C *>>,const boost::fusion::single_view<C>>,const boost::fusion::single_view<D *>>,const boost::fusion::single_view<D>>,const boost::fusion::single_view<NormalizedVector *>>,const boost::fusion::single_view<NormalizedVector>>,const boost::fusion::single_view<RakNet::RPC3 *>> 1> ] 1> c:\raknet\dependentextensions\rpc3\rpc3_boost.h(261) : see reference to class template instantiation 'boost::fusion::result_of::invoke<Function,Sequence>' being compiled 1> with 1> [ 1> Function=void (__cdecl *)(A *,A &,C *,C &,D *,D &,NormalizedVector *,NormalizedVector &,RakNet::RPC3 *), 1> Sequence=const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::nil,const boost::fusion::single_view<A *>>,const boost::fusion::single_view<A>>,const boost::fusion::single_view<C *>>,const boost::fusion::single_view<C>>,const boost::fusion::single_view<D *>>,const boost::fusion::single_view<D>>,const boost::fusion::single_view<NormalizedVector *>>,const boost::fusion::single_view<NormalizedVector>>,const boost::fusion::single_view<RakNet::RPC3 *>> 1> ] 1> c:\raknet\dependentextensions\rpc3\rpc3_boost.h(247) : see reference to function template instantiation 'RakNet::_RPC3::InvokeResultCodes RakNet::_RPC3::BoostRPCInvoker<Function,From,To>::apply<boost::fusion::joint_view<Sequence1,Sequence2>>(Function,RakNet::_RPC3::InvokeArgs,const Args &)' being compiled 1> with 1> [ 1> Function=void (__cdecl *)(A *,A &,C *,C &,D *,D &,NormalizedVector *,NormalizedVector &,RakNet::RPC3 *), 1> From=next_iter_type, 1> To=boost::mpl::v_iter<boost::mpl::vector9<A *,A &,C *,C &,D *,D &,NormalizedVector *,NormalizedVector &,RakNet::RPC3 *>,9>, 1> Sequence1=const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::nil,const boost::fusion::single_view<A *>>,const boost::fusion::single_view<A>>,const boost::fusion::single_view<C *>>,const boost::fusion::single_view<C>>,const boost::fusion::single_view<D *>>,const boost::fusion::single_view<D>>,const boost::fusion::single_view<NormalizedVector *>>,const boost::fusion::single_view<NormalizedVector>>, 1> Sequence2=const boost::fusion::single_view<RakNet::RPC3 *>, 1> Args=boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::nil,const boost::fusion::single_view<A *>>,const boost::fusion::single_view<A>>,const boost::fusion::single_view<C *>>,const boost::fusion::single_view<C>>,const boost::fusion::single_view<D *>>,const boost::fusion::single_view<D>>,const boost::fusion::single_view<NormalizedVector *>>,const boost::fusion::single_view<NormalizedVector>>,const boost::fusion::single_view<RakNet::RPC3 *>> 1> ] 1> c:\raknet\dependentextensions\rpc3\rpc3_boost.h(247) : see reference to function template instantiation 'RakNet::_RPC3::InvokeResultCodes RakNet::_RPC3::BoostRPCInvoker<Function,From,To>::apply<boost::fusion::joint_view<Sequence1,Sequence2>>(Function,RakNet::_RPC3::InvokeArgs,const Args &)' being compiled 1> with 1> [ 1> Function=void (__cdecl *)(A *,A &,C *,C &,D *,D &,NormalizedVector *,NormalizedVector &,RakNet::RPC3 *), 1> From=next_iter_type, 1> To=boost::mpl::v_iter<boost::mpl::vector9<A *,A &,C *,C &,D *,D &,NormalizedVector *,NormalizedVector &,RakNet::RPC3 *>,9>, 1> Sequence1=const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::nil,const boost::fusion::single_view<A *>>,const boost::fusion::single_view<A>>,const boost::fusion::single_view<C *>>,const boost::fusion::single_view<C>>,const boost::fusion::single_view<D *>>,const boost::fusion::single_view<D>>,const boost::fusion::single_view<NormalizedVector *>>, 1> Sequence2=const boost::fusion::single_view<NormalizedVector>, 1> Args=boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::nil,const boost::fusion::single_view<A *>>,const boost::fusion::single_view<A>>,const boost::fusion::single_view<C *>>,const boost::fusion::single_view<C>>,const boost::fusion::single_view<D *>>,const boost::fusion::single_view<D>>,const boost::fusion::single_view<NormalizedVector *>>,const boost::fusion::single_view<NormalizedVector>> 1> ] 1> c:\raknet\dependentextensions\rpc3\rpc3_boost.h(247) : see reference to function template instantiation 'RakNet::_RPC3::InvokeResultCodes RakNet::_RPC3::BoostRPCInvoker<Function,From,To>::apply<boost::fusion::joint_view<Sequence1,Sequence2>>(Function,RakNet::_RPC3::InvokeArgs,const Args &)' being compiled 1> with 1> [ 1> Function=void (__cdecl *)(A *,A &,C *,C &,D *,D &,NormalizedVector *,NormalizedVector &,RakNet::RPC3 *), 1> From=next_iter_type, 1> To=boost::mpl::v_iter<boost::mpl::vector9<A *,A &,C *,C &,D *,D &,NormalizedVector *,NormalizedVector &,RakNet::RPC3 *>,9>, 1> Sequence1=const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::nil,const boost::fusion::single_view<A *>>,const boost::fusion::single_view<A>>,const boost::fusion::single_view<C *>>,const boost::fusion::single_view<C>>,const boost::fusion::single_view<D *>>,const boost::fusion::single_view<D>>, 1> Sequence2=const boost::fusion::single_view<NormalizedVector *>, 1> Args=boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::nil,const boost::fusion::single_view<A *>>,const boost::fusion::single_view<A>>,const boost::fusion::single_view<C *>>,const boost::fusion::single_view<C>>,const boost::fusion::single_view<D *>>,const boost::fusion::single_view<D>>,const boost::fusion::single_view<NormalizedVector *>> 1> ] 1> c:\raknet\dependentextensions\rpc3\rpc3_boost.h(247) : see reference to function template instantiation 'RakNet::_RPC3::InvokeResultCodes RakNet::_RPC3::BoostRPCInvoker<Function,From,To>::apply<boost::fusion::joint_view<Sequence1,Sequence2>>(Function,RakNet::_RPC3::InvokeArgs,const Args &)' being compiled 1> with 1> [ 1> Function=void (__cdecl *)(A *,A &,C *,C &,D *,D &,NormalizedVector *,NormalizedVector &,RakNet::RPC3 *), 1> From=next_iter_type, 1> To=boost::mpl::v_iter<boost::mpl::vector9<A *,A &,C *,C &,D *,D &,NormalizedVector *,NormalizedVector &,RakNet::RPC3 *>,9>, 1> Sequence1=const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::nil,const boost::fusion::single_view<A *>>,const boost::fusion::single_view<A>>,const boost::fusion::single_view<C *>>,const boost::fusion::single_view<C>>,const boost::fusion::single_view<D *>>, 1> Sequence2=const boost::fusion::single_view<D>, 1> Args=boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::nil,const boost::fusion::single_view<A *>>,const boost::fusion::single_view<A>>,const boost::fusion::single_view<C *>>,const boost::fusion::single_view<C>>,const boost::fusion::single_view<D *>>,const boost::fusion::single_view<D>> 1> ] 1> c:\raknet\dependentextensions\rpc3\rpc3_boost.h(247) : see reference to function template instantiation 'RakNet::_RPC3::InvokeResultCodes RakNet::_RPC3::BoostRPCInvoker<Function,From,To>::apply<boost::fusion::joint_view<Sequence1,Sequence2>>(Function,RakNet::_RPC3::InvokeArgs,const Args &)' being compiled 1> with 1> [ 1> Function=void (__cdecl *)(A *,A &,C *,C &,D *,D &,NormalizedVector *,NormalizedVector &,RakNet::RPC3 *), 1> From=next_iter_type, 1> To=boost::mpl::v_iter<boost::mpl::vector9<A *,A &,C *,C &,D *,D &,NormalizedVector *,NormalizedVector &,RakNet::RPC3 *>,9>, 1> Sequence1=const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::nil,const boost::fusion::single_view<A *>>,const boost::fusion::single_view<A>>,const boost::fusion::single_view<C *>>,const boost::fusion::single_view<C>>, 1> Sequence2=const boost::fusion::single_view<D *>, 1> Args=boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::nil,const boost::fusion::single_view<A *>>,const boost::fusion::single_view<A>>,const boost::fusion::single_view<C *>>,const boost::fusion::single_view<C>>,const boost::fusion::single_view<D *>> 1> ] 1> c:\raknet\dependentextensions\rpc3\rpc3_boost.h(247) : see reference to function template instantiation 'RakNet::_RPC3::InvokeResultCodes RakNet::_RPC3::BoostRPCInvoker<Function,From,To>::apply<boost::fusion::joint_view<Sequence1,Sequence2>>(Function,RakNet::_RPC3::InvokeArgs,const Args &)' being compiled 1> with 1> [ 1> Function=void (__cdecl *)(A *,A &,C *,C &,D *,D &,NormalizedVector *,NormalizedVector &,RakNet::RPC3 *), 1> From=next_iter_type, 1> To=boost::mpl::v_iter<boost::mpl::vector9<A *,A &,C *,C &,D *,D &,NormalizedVector *,NormalizedVector &,RakNet::RPC3 *>,9>, 1> Sequence1=const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::nil,const boost::fusion::single_view<A *>>,const boost::fusion::single_view<A>>,const boost::fusion::single_view<C *>>, 1> Sequence2=const boost::fusion::single_view<C>, 1> Args=boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::nil,const boost::fusion::single_view<A *>>,const boost::fusion::single_view<A>>,const boost::fusion::single_view<C *>>,const boost::fusion::single_view<C>> 1> ] 1> c:\raknet\dependentextensions\rpc3\rpc3_boost.h(247) : see reference to function template instantiation 'RakNet::_RPC3::InvokeResultCodes RakNet::_RPC3::BoostRPCInvoker<Function,From,To>::apply<boost::fusion::joint_view<Sequence1,Sequence2>>(Function,RakNet::_RPC3::InvokeArgs,const Args &)' being compiled 1> with 1> [ 1> Function=void (__cdecl *)(A *,A &,C *,C &,D *,D &,NormalizedVector *,NormalizedVector &,RakNet::RPC3 *), 1> From=next_iter_type, 1> To=boost::mpl::v_iter<boost::mpl::vector9<A *,A &,C *,C &,D *,D &,NormalizedVector *,NormalizedVector &,RakNet::RPC3 *>,9>, 1> Sequence1=const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::nil,const boost::fusion::single_view<A *>>,const boost::fusion::single_view<A>>, 1> Sequence2=const boost::fusion::single_view<C *>, 1> Args=boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::nil,const boost::fusion::single_view<A *>>,const boost::fusion::single_view<A>>,const boost::fusion::single_view<C *>> 1> ] 1> c:\raknet\dependentextensions\rpc3\rpc3_boost.h(247) : see reference to function template instantiation 'RakNet::_RPC3::InvokeResultCodes RakNet::_RPC3::BoostRPCInvoker<Function,From,To>::apply<boost::fusion::joint_view<Sequence1,Sequence2>>(Function,RakNet::_RPC3::InvokeArgs,const Args &)' being compiled 1> with 1> [ 1> Function=void (__cdecl *)(A *,A &,C *,C &,D *,D &,NormalizedVector *,NormalizedVector &,RakNet::RPC3 *), 1> From=next_iter_type, 1> To=boost::mpl::v_iter<boost::mpl::vector9<A *,A &,C *,C &,D *,D &,NormalizedVector *,NormalizedVector &,RakNet::RPC3 *>,9>, 1> Sequence1=const boost::fusion::joint_view<const boost::fusion::nil,const boost::fusion::single_view<A *>>, 1> Sequence2=const boost::fusion::single_view<A>, 1> Args=boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::nil,const boost::fusion::single_view<A *>>,const boost::fusion::single_view<A>> 1> ] 1> c:\raknet\dependentextensions\rpc3\rpc3_boost.h(247) : see reference to function template instantiation 'RakNet::_RPC3::InvokeResultCodes RakNet::_RPC3::BoostRPCInvoker<Function,From,To>::apply<boost::fusion::joint_view<Sequence1,Sequence2>>(Function,RakNet::_RPC3::InvokeArgs,const Args &)' being compiled 1> with 1> [ 1> Function=void (__cdecl *)(A *,A &,C *,C &,D *,D &,NormalizedVector *,NormalizedVector &,RakNet::RPC3 *), 1> From=next_iter_type, 1> To=boost::mpl::v_iter<boost::mpl::vector9<A *,A &,C *,C &,D *,D &,NormalizedVector *,NormalizedVector &,RakNet::RPC3 *>,9>, 1> Sequence1=const boost::fusion::nil, 1> Sequence2=const boost::fusion::single_view<A *>, 1> Args=boost::fusion::joint_view<const boost::fusion::nil,const boost::fusion::single_view<A *>> 1> ] 1> c:\raknet\dependentextensions\rpc3\rpc3_boost.h(336) : see reference to function template instantiation 'RakNet::_RPC3::InvokeResultCodes RakNet::_RPC3::BoostRPCInvoker<Function>::apply<boost::fusion::nil>(Function,RakNet::_RPC3::InvokeArgs,const Args &)' being compiled 1> with 1> [ 1> Function=void (__cdecl *)(A *,A &,C *,C &,D *,D &,NormalizedVector *,NormalizedVector &,RakNet::RPC3 *), 1> Args=boost::fusion::nil 1> ] 1> c:\raknet\dependentextensions\rpc3\rpc3_boost.h(335) : while compiling class template member function 'RakNet::_RPC3::FunctionPointer RakNet::_RPC3::GetBoundPointer_C<Function>::GetBoundPointer(Function)' 1> with 1> [ 1> Function=void (__cdecl *)(A *,A &,C *,C &,D *,D &,NormalizedVector *,NormalizedVector &,RakNet::RPC3 *) 1> ] 1> c:\raknet\dependentextensions\rpc3\rpc3_boost.h(355) : see reference to class template instantiation 'RakNet::_RPC3::GetBoundPointer_C<Function>' being compiled 1> with 1> [ 1> Function=void (__cdecl *)(A *,A &,C *,C &,D *,D &,NormalizedVector *,NormalizedVector &,RakNet::RPC3 *) 1> ] 1> c:\raknet\dependentextensions\rpc3\rpc3.h(100) : see reference to function template instantiation 'RakNet::_RPC3::FunctionPointer RakNet::_RPC3::GetBoundPointer<void(__cdecl *)(A *,A &,C *,C &,D *,D &,NormalizedVector *,NormalizedVector &,RakNet::RPC3 *)>(Function)' being compiled 1> with 1> [ 1> Function=void (__cdecl *)(A *,A &,C *,C &,D *,D &,NormalizedVector *,NormalizedVector &,RakNet::RPC3 *) 1> ] 1> c:\raknet\samples\rpc3\rpc3sample.cpp(127) : see reference to function template instantiation 'bool RakNet::RPC3::RegisterFunction<void(__cdecl *)(A *,A &,C *,C &,D *,D &,NormalizedVector *,NormalizedVector &,RakNet::RPC3 *)>(const char *,Function)' being compiled 1> with 1> [ 1> Function=void (__cdecl *)(A *,A &,C *,C &,D *,D &,NormalizedVector *,NormalizedVector &,RakNet::RPC3 *) 1> ] 1>c:\boost_1_35_0\boost\fusion\functional\invocation\invoke.hpp(159) : error C2039: 'result_type' : is not a member of 'boost::fusion::detail::invoke_impl<Function,Sequence>' 1> with 1> [ 1> Function=void (__cdecl *)(A *,A &,C *,C &,D *,D &,NormalizedVector *,NormalizedVector &,RakNet::RPC3 *), 1> Sequence=const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::joint_view<const boost::fusion::nil,const boost::fusion::single_view<A *>>,const boost::fusion::single_view<A>>,const boost::fusion::single_view<C *>>,const boost::fusion::single_view<C>>,const boost::fusion::single_view<D *>>,const boost::fusion::single_view<D>>,const boost::fusion::single_view<NormalizedVector *>>,const boost::fusion::single_view<NormalizedVector>>,const boost::fusion::single_view<RakNet::RPC3 *>> 1> ] 1>c:\boost_1_35_0\boost\fusion\functional\invocation\invoke.hpp(159) : error C2146: syntax error : missing ';' before identifier 'type' 1>c:\boost_1_35_0\boost\fusion\functional\invocation\invoke.hpp(159) : error C4430: missing type specifier - int assumed. Note: C++ does not support default-int 1>c:\boost_1_35_0\boost\fusion\functional\invocation\invoke.hpp(159) : error C2208: 'boost::type' : no members defined using this type 1>c:\boost_1_35_0\boost\fusion\functional\invocation\invoke.hpp(159) : fatal error C1903: unable to recover from previous error(s); stopping compilation

3 3

Random numbers
by Olivier Tournaire 21 Jul '08

21 Jul '08

Hi all, I am currently using boost::random, and I wonder how it is possible to get uniform random numbers on a union of two intervalls. For instance, I would like to generate uniform_real numbers in the intervall [-a;-b]U[b;a]. How can I do such a task ? Best regards, Olivier Tournaire

3 5

[bind][lambda] overriding return value
by David Rodríguez Ibeas 21 Jul '08

21 Jul '08

Hi, Is it possible to override/generate a return value? I want to connect a signal that has a bool return value to a method that has a void return type, generating always a true return. class X { public: void f(); }; boost::signal< bool () > sgn; X x; sgn.connect( boost::bind( &X::f, &x ) ); // error, signal has bool return, method doesn't David

3 2

ImportError: dynamic module does not define init function
by Tony May 21 Jul '08

21 Jul '08

I get the following error when trying to import a pyd file in python. I'm trying to use this with the Delta3D game engine to access VRPN data. ------ Traceback (most recent call last): File "hello.py", line 2, in <module> import PyVrpnData ImportError: dynamic module does not define init function (initPyVrpnData) ------ Here's the python file. ------ import PyVrpnData print "hello" print PyVrpnData.getPos(1) print "done" ------ I'm trying to use boost but not use bjam as it appears Delta3D has done. Here is the source. -------- float getPos(int i); float getQuat(int i); void initPyVrpnData() { using namespace boost::python; def("getPos",getPos); def("getQuat",getQuat); } ------- #include <boost/python.hpp> extern "C" __declspec(dllexport) void initPyVrpnData(); BOOST_PYTHON_MODULE(PyVrpnData) { initPyVrpnData(); } --------

1 0

Problem with type traits
by Robert Dailey 21 Jul '08

21 Jul '08

Hi, I'm trying to create my own struct that is compatible with boost::mpl::or_ as well as boost::enable_if. I find that using boost::mpl::or_ requires that my structure have a 'type' member, and enable_if looks for 'value'. Must I define both? What is the proper way to handle this? Below is my structure: template< typename t_type > struct is_sink { typedef typename boost::mpl::or_< vfx::io::is_device<t_type>, boost::is_same<t_type, vfx::io::Sink> > type; }; The above structure works perfectly fine with boost::mpl::or_, but does not work for boost::enable_if because it is looking for a 'value' member. I feel what it should be looking for is is_sink::type::value. An example of how I use the above structure is below: template< typename t_source, typename t_data > typename boost::enable_if< boost::mpl::and_<*is_source<t_source>*, boost::is_pod<t_data> >, t_source >::type& operator>> ( t_source& source, t_data& data ) { source.Read( reinterpret_cast<boost::uint8_t*>( &data ), sizeof( t_data ) ); return source; }

4 6

"invalid next size" in lu_factorize
by Sylvain Bertrand 21 Jul '08

21 Jul '08

Hi, When running lu_factorize() on a 10x10 matrix, I get the following message and have no clue of the reason why (googling or searching this ML didn't help). *** glibc detected *** ./a.out: free(): invalid next size (normal): 0x0806b8a0 *** ======= Backtrace: ========= /lib/i686/cmov/libc.so.6[0xb7d068f5] /lib/i686/cmov/libc.so.6(cfree+0x90)[0xb7d0a360] /usr/lib/libstdc++.so.6(_ZdlPv+0x21)[0xb7ed4171] ./a.out[0x805d28d] ./a.out[0x8058969] ./a.out[0x8051bcd] ./a.out[0x8057481] ./a.out[0x8057632] ./a.out(__gxx_personality_v0+0x317)[0x804adf7] /lib/i686/cmov/libc.so.6(__libc_start_main+0xe0)[0xb7cb1450] ./a.out(__gxx_personality_v0+0xb1)[0x804ab91] ======= Memory map: ======== Any help is appreciated. Regards, Sylvain

1 1

[lambda] Exploring bind syntax.
by Robert Jones 21 Jul '08

21 Jul '08

Using the code snippet below as an example,... #include <vector> #include <algorithm> #include <functional> #include "boost/lambda/lambda.hpp" #include "boost/lambda/bind.hpp" #include <boost/function.hpp> struct S { int methodOfS( ); }; int main( ) { typedef std :: vector<S *> Vec; using namespace boost :: lambda; boost :: function<int( S * )> fn = bind( & S :: methodOfS, _1 ); Vec v; Vec :: iterator match; match = find_if( v.begin( ), v.end( ), bind( fn, _1 ) != fn( * v.begin( ) ) ); // OK match = find_if( v.begin( ), v.end( ), fn( _1 ) != fn( * v.begin( ) ) ); // Fail } I find the line marked 'OK' compiles, but not the line marked 'Fail'. Isn't the very point of Boost.Bind to enable this kind of syntax? Is there a better way to write it than my 'Fail' line, but which will still compile? Thanks, Rob. -- ACCU - Professionalism in programming - http://www.accu.org

2 1

[serialization] Problems with BOOST_IS_ABSTRACT, gcc 3.2.3 and Boost 1.35.0
by Dan Thomas 21 Jul '08

21 Jul '08

Hi If I use gcc 3.2.3 (Linux) and Boost 1.35.0 to compile the example program at... http://lists.boost.org/boost-users/att-26574/test_serialize.cpp ...I get extensive compile errors which I can summarise as: <path>/boost_1_35_0/boost/serialization/export.hpp:125: `instantiate' is not a member of type `boost::archive::detail::ptr_serialization_support<boost::archive::text_iarchive, base>' These errors do not occur if I use either Boost 1.33.0 or Microsoft Visual Studio. The fact that the test program's base class is abstract seems to be central to the problem. The error is eliminated if I remove the BOOST_IS_ABSTRACT(base) statement from the test program. Doing so naturally causes other problems but these are in turn fixed by making "base" concrete i.e. by removing the =0 qualifier from base::get_id() and providing an implementation. It looks as if code in Boost 1.35.0 expects all classes to provide an instantiate method, even though the method is not in fact present on classes that have been marked as abstract. Is there any chance of a fix or work-around? Many Thanks Dan

2 3

[lambda] Writing pass-through 'standard' algorithms
by Robert Jones 21 Jul '08

21 Jul '08

I encountered this difficulty in the context of passing lambda functors, but on reflection I don't the lambda aspect particularly matters. Consider this templated function template <typename InputIterator, typename Predicate> inline bool all_if( InputIterator first, InputIterator last, Predicate predicate ) { return std :: find_if( first, last, std :: not1( predicate ) ) == last; } Obviously the intent is to enquire if all members of a range satisfy the predicate. Now, as is, this doesn't compile, because it needs ptr_fun() on the predicate, but then a couple of things come up. - ptr_fun( ) requires that it's argument is a pointer-to-function - I'd like my all_if( ) to permit functions or functors, including lambda functors How can this be done? Is it impossible to dress-up a standard algorithm in this way? Thanks, Rob. -- ACCU - Professionalism in programming - http://www.accu.org

2 1

Re: [Boost-users] question on boost-compliant code conventions
by Alan M. Carroll 21 Jul '08

21 Jul '08

1 0

[asio] Copying an asio::stream_buf question
by Graham Reitz 21 Jul '08

21 Jul '08

Is this a reasonable method to copy the contents of one asio::streambuf into another (i.e. in a copy constructor of a class that contains an asio::streambuf)? istreambuf_iterator<char> i( ostream_1.rdbuf() ); istreambuf_iterator<char> eos; // end-of-stream ostreambuf_iterator<char> o( ostream_2.rdbuf() ); std::copy( i, eos, o ); thanks, graham

1 0

[format] error when used with DEBUG_NEW in VC7.1 MFC
by hongleij 21 Jul '08

21 Jul '08

when i comment these codes ,it will be OK. #ifdef _DEBUG #define new DEBUG_NEW #undef THIS_FILE static char THIS_FILE[] = __FILE__; #endif ////////////////////souce code #include <iostream> #include <tchar.h> #define WIN32_LEAN_AND_MEAN // #pragma warning(disable:4200) // nonstandard extension used : zero-sized array in struct/union #pragma warning(disable:4244) // 'conversion' conversion from 'type1' to 'type2', possible loss of data #pragma warning(disable:4800) // 'type' : forcing value to bool 'true' or 'false' (performance warning) // Windows Header Files: #ifndef WINVER #define WINVER 0x0400 // 0x0400 == Windows 98 and Windows NT 4.0 (because of '_WIN32_WINDOWS=0x0410') #endif #ifndef _WIN32_WINNT #define _WIN32_WINNT 0x0400 // 0x0400 == Windows NT 4.0 #endif #ifndef _WIN32_WINDOWS #define _WIN32_WINDOWS 0x0410 // 0x0410 == Windows 98 #endif #ifndef _WIN32_IE #define _WIN32_IE 0x0400 // 0x0400 == Internet Explorer 4.0 -> Comctl32.dll v4.71 #endif #define _ATL_CSTRING_EXPLICIT_CONSTRUCTORS #define _AFX_ALL_WARNINGS #include <afxwin.h> #include <afxext.h> #include <afxdisp.h> //#include <afxdtctl.h> #ifndef _AFX_NO_AFXCMN_SUPPORT #include <afxcmn.h> #endif // _AFX_NO_AFXCMN_SUPPORT #include <winsock2.h> #define _WINSOCKAPI_ #include <afxsock.h> #include <afxdhtml.h> #include <afxmt.h> #ifdef _DEBUG #define new DEBUG_NEW #undef THIS_FILE static char THIS_FILE[] = __FILE__; #endif #include <boost/format.hpp> void ipAddressToString(const unsigned int ip, std::string &str) { //char temp[64]; //_snprintf(temp,sizeof(temp),"%ld.%ld.%ld.%ld", // ((ip >> 24) & 0xFF), // ((ip >> 16) & 0xFF), // ((ip >> 8) & 0xFF), // ((ip ) & 0xFF) ); //str = temp; str = boost::str( boost::format("%ld.%ld.%ld.%ld") % ((ip >> 24) & 0xFF) % ((ip >> 16) & 0xFF) % ((ip >> 8) & 0xFF) % ((ip ) & 0xFF) ); } class A {}; int _tmain(int argc, _TCHAR* argv[]) { std::string str; ipAddressToString(0xf1c2d3e4,str); int *p = new int(32); // A *q = new(p)(A); return 0; } ////////////////////////errors e:\boost_1_35_0\boost\optional\optional.hpp(315) : error C2061: syntax error : identifier 'm_storage' e:\boost_1_35_0\boost\optional\optional.hpp(314) : while compiling class-template member function 'void boost::optional_detail::optional_base<T>::construct(boost::optional_detail::optional_base<T>::argument_type)' with [ T=boost::io::detail::locale_t ] e:\boost_1_35_0\boost\optional\optional.hpp(455) : see reference to class template instantiation 'boost::optional_detail::optional_base<T>' being compiled with [ T=boost::io::detail::locale_t ] e:\boost_1_35_0\boost\format\internals.hpp(56) : see reference to class template instantiation 'boost::optional<T>' being compiled with [ T=boost::io::detail::locale_t ] e:\boost_1_35_0\boost\format\internals.hpp(57) : see reference to class template instantiation 'boost::io::detail::stream_format_state<Ch,Tr>' being compiled Build log was saved at "file://c:\Documents and Settings\jiang\桌面\Voip\kademlia\Debug\BuildLog.htm" Kad - 1 error(s), 1 warning(s)

1 0

Spirit: skip parser and rules
by Alessandro Re 21 Jul '08

21 Jul '08

Hello, I've this code: // w and space_signaler are a function objects which print out some info. bool read_str(const char *in, StrList &w) { rule<> identifier = +alnum_p; rule<> statement = identifier[w] >> ch_p(';')[w]; return parse( in, (statement)[w], space_p[space_signaler] // Called for initial spaces only ).full; } Well, I don't understand why if i use (statement)[w] where signaled, the space_signaler is never called for non-initial spaces, while if i write directly the rule, it does correctly: return parse( in, (identifier[w] >> ch_p(';')[w])[w], space_p[space_signaler] // Called for every space ).full; In the latter case, it parses (skipping spaces) inputs like: "hello;" "hello ;" " hello ; " while in the former it parses correctly inputs like " hello;" "hello;" disallowing spaces between the word and the semicolon. Can someone explain me why does this happens? Thanks very much

2 2

question on boost-compliant code conventions
by Ákos Maróy 21 Jul '08

21 Jul '08

Hi, This is probably a lame question, but I have some open issues I don't know how are handled in contemporary C++ style, for example with using boost coding convetions. I've been using the CamelCase naming convention from Java recently, where the issues below are non-issues actually. for example: - how do I create an instance of the same name as its type? in CamelCase: class Foo; Foo foo; in C++: class foo; foo foo; // won't compile what's the solution here? add some prefix to the instance's name? - how do I declare accessor functions to private members? in Java / CamelCase: class Foo { private int bar; public int getBar() { return bar; } }; in C++: class foo { private: int bar; public: int bar() const { return bar; } // won't compile }; should I put some prefix to the private member's name? Sorry for the lame questions :) But I just want to follow the guideline of 'in Rome, do as the Romans do' :) Akos

3 2

[spirit] reading vector of string
by Christian Henning 21 Jul '08

21 Jul '08

Hi there, can I use push_back_a with std::vector<std::string>? I want something like this: std::vector< std::string > lines; rule<> l_entry = blanks_p >> *print_p[push_back_a( lines )] >> blanks_p >> eol_p; Regards, Christian

2 1

Fusion transform question
by Joel FALCOU 21 Jul '08

21 Jul '08

I started porting over some old codes that used hand-made MPL<->concrete object code to Fusion. I'm currently trying to have the following to work with no results so far : struct make_arg { template<class Type> struct result { typedef typename boost::add_reference<typename boost::add_const<Type>::type >::type type; }; }; template<class Args> struct tree { typedef Args children; typedef typename bf::result_of::transform<Args,make_arg>::type contents; template<class T0,class T1> tree( const T0& t0 , const T1& t1 ) : mArgs( /* what to put here ??? */ ) {} template<size_t I> typename bf::result_of::at_c<contents,I>::type get() const { return bf::at_c<I>(mArgs); } contents mArgs; }; int main() { int k=3; float u=7.2; tree< bf::vector<int, float> > tree(k,u); return 0; } The question - as stated - in the code is how to simply build the mArgs member from the t0,...,tn arguments ? Thanks in advance. -- Joel FALCOU Research Engineer @ Institut d'Electronique Fondamentale Université PARIS SUD XI France

5 12

[Boost.thread] Interruption points ?
by Ruediger Berlich 21 Jul '08

21 Jul '08

Hi there, just a short question: http://www.boost.org/doc/libs/1_35_0/doc/html/thread/thread_management.html… lists the following interruption points: * boost::thread::join() * boost::thread::timed_join() * boost::condition_variable::wait() * boost::condition_variable::timed_wait() * boost::condition_variable_any::wait() * boost::condition_variable_any::timed_wait() * boost::thread::sleep() * boost::this_thread::sleep() * boost::this_thread::interruption_point() Is (and if so: why is) a lock on a mutex no interruption point ? Thanks and Best Regards, Ruediger

2 1

Re: [Boost-users] [fusion] sequence conversion
by Zeljko Vrba 21 Jul '08

21 Jul '08

On Sun, Jul 20, 2008 at 07:34:29PM -0400, Daryle Walker wrote: > > I'm not really into the whole MPL scene myself, but I'll give you > examples where I used it. > Thanks for the long explanation.

1 0

[dataflow] new frameworks and GUI editors + visually edit a BGL graph
by Stjepan Rajko 20 Jul '08

20 Jul '08

Hello, I uploaded some youtube videos of a new visual editor for the Dataflow library. There are three examples, each using a different underlying framework (the visual editor code is generic): visually editing a Dataflow.Signals network (data flows instantly): * http://www.youtube.com/watch?v=7a4HJ4zVPXc visually editing a Dataflow.Managed network (data flows on request): * http://www.youtube.com/watch?v=KwrnSjhJyf4 visually editing a BGL graph (this simple example lets you construct a directed graph and query the number of strongly connected components): * http://www.youtube.com/watch?v=gVLetAAnPrc The latter two examples use proof-of-concept implementations of alternate frameworks built on top of the generic framework that is part of the Dataflow library: Dataflow.Managed is unlike Dataflow.Signals in that doesn't couple data transfer with component invocation. In other words, if producer A is connected to consumer B, when A changes its value B will not be notified immediately. Instead, after any number of values in the network has changed, the network as a whole can be told to update (the network will then notify all components whose inputs have been modified, in the order of a topological sort of the components). Alternatively, components can be invoked individually, which gives them a chance to check their inputs / modify their outputs. The BGL support layer for the Dataflow library is an example of how something that is not strictly a dataflow thing (e.g., a graph) can still satisfy Dataflow concepts (there are Components (vertices) that can be connected), and be used with the framework (e.g., you get a GUI editor for free). If you're interested in playing with this: The new GUI uses GLV (http://mat.ucsb.edu/glv/) which is included with the dataflow library in the sandbox and integrated in the bjam build of the examples. As long as you have glut/freeglut available, it should build using gcc/darwin, MSVC8+, gcc/minGW, and if you are better than me at knowing which X libraries to link with and know how to manipulate a Jamfile, gcc/linux (any GCC version 4.x should work). If you run into any problems, please let me know. The latest code is currently available from the sandbox only, instructions/docs at: http://www.dancinghacker.com/code/dataflow/ Any feedback would be appreciated. Kind regards, Stjepan

1 0

regular expressions and binary files
by Brad Tilley 20 Jul '08

20 Jul '08

I'm using boost regular expressions with good results on text files (html, xml, txt, etc), however, I'm having issues on binary file formats (i.e. MS Office files, etc). I've goggled around, but not found any hints. Just wondering if someone had experience doing REs on binary files. I've tried different open methods ios::binary, etc... still no luck. Thanks

2 1

[lambda] Using lambda expressions with STL function adaptors
by Robert Jones 20 Jul '08

20 Jul '08

This little piece of code, #include <algorithm> #include <functional> template <typename InputIterator, typename Predicate> inline bool if_all( InputIterator first, InputIterator last, Predicate predicate ) { return std :: find_if( first, last, std :: not1( predicate ) ) == last; } #include <vector> #include "boost/lambda/lambda.hpp" int main( ) { using namespace boost :: lambda; std :: vector<int> v; if_all( v.begin( ), v.end( ), _1 == 3 ); return 0; } Gives a compiler error, which I think tells me that my lambda expression, _1 == 3 does not provide the argument_type typedef. I thought lambda expressions were careful to to do this. What am I doing wrong? Thanks, Rob. -- ACCU - Professionalism in programming - http://www.accu.org

2 4

[filesystem] linker error with trunk
by Christian Henning 20 Jul '08

20 Jul '08

Hi there, when including boost\filesystem.hpp I get the following linker error with VC8. error LNK2019: unresolved external symbol "class boost::system::error_category const & __cdecl boost::system::get_generic_category(void)" (?get_generic_category@system@boost@@YAABVerror_category@12@XZ) referenced in function "void __cdecl boost::system::`dynamic initializer for 'generic_category''(void)" (??__Egeneric_category@system@boost@@YAXXZ) Since I'm using autolinking I don't think it's due to a missing library. Anyone? Thanks, Christian

1 0

checking that two types are equal
by e r 20 Jul '08

20 Jul '08

Hi all, what is the command to check that two types, say Type0 and Type1 are equal? I'm looking for something like BOOST_STATIC_ASSERT(...) Thanks.

2 2

Looking for perl interface to boost::serialization...
by Sangeeta Singh 20 Jul '08

20 Jul '08

Hi, I was wondering if a perl interface to the boost serialization library is available some where like the boost::graph cpan perl module. Thanks, Sangeeta

2 1

implicit astar_search weirdness, bug in mutable_queue?
by Christopher Flesher 20 Jul '08

20 Jul '08

Hello, I am trying to use this boost::graph implicit a-star search using boost 1.35.0 but have been having some difficulty. There seems to be a bug in the mutable_queue class where it will try to write to an index that is not valid: index_array[ tmp ] = c.size()-1; It seems someone else had posted an e-mail about this problem about four months ago. I checked CVS for any recent updates but this code hasn't been changed in the past 3 years. Anyways I got around this problem temporarily by adding some code to push indices onto the index_array until the code doesn't crash... not sure what the side effects of this would be though. Anyways after getting that working I find that my cost heuristic is never getting called! My implicit search basically goes off in the wrong direction to infinity. Anyways I've attached my test case. I've tried 2-3 different code samples that try to do this and none of them have worked so far! Chris

1 0

[multi_array] more on assignability
by Dr Johann A. Briffa 20 Jul '08

20 Jul '08

I realize this topic has been discussed before (I've seen at least a thread in 2006 and another in 2007), but I believe there is more to add. At issue is the decision that multi_array objects cannot be assigned [with the operator=() that is] unless the target is already the same size as the source. Further, there is no documented way to easily resize an existing array to the same size and range as some other array. Let me start with the first issue: others have already said before that making multi_array assignable would be a good thing. Ron disagrees on this point, and from what I gather his principal reason is that multi_array should have the same semantics as multi_array_ref etc. [please do correct me if I'm wrong here, I've only had a day to sort this out]. I am unsure why multi_array_ref etc cannot also have such an interface, but then I haven't used that class enough so my knowledge is very limited there. As far as my needs are concerned, a multi_array that cannot be assigned is next to useless. Allow me to elaborate: I am using multi_array as a replacement to my own (old) 2/3D containers, as recently I've had need for higher dimensions. I have classes containing array members, where the size of the arrays is dynamic through the object lifetime. So my class default constructor uses the array default constructor for its member objects, creating empty arrays. Eventually as my classes get used, the array members are updated accordingly, and get to have some non-zero size. Now the problem is that my classes need to be assignable, and the only way that can be achieved is if the array members are assignable; the alternative is to have to write a specific assignment operator for any of my classes that use arrays, resizing each member of the target before copying. Clearly that would involve a lot of extra effort. The second (admittedly minor) problem is that the only way to resize the target is unclean and the semantics are bound to the dimensionality: target.resize(boost::extents[source.shape()[0]][source.shape()[1]]); etc. I would rather have something like: target.resize(source.extents()); Ron had mentioned in 2007 that the lack of such an interface was an oversight, but there is still no solution. I am bringing up these issues again because: 1) I am not convinced there is any valid reason not to have multi_array assignable. I do not buy the performance issue, as this would only kick in if (a) assignment is within a performance-sensitive path and (b) the target is not already of the same size. Further, at the very least there could be a derived class (somewhat like what I document below) that allows such assignment. 2) It would be nice to have the clean resize interface as part of the multi_array class. As an interim solution for myself, I derived a new class from multi_array that allows assignment and does the resize automagically. I am copying the code below for public benefit. Notes: i) anyone is free to (re)use this code under the same license as multi_array; in particular, Ron, please do feel free to adapt and reuse within a future multi_array as you see fit. ii) while the interface provided is sufficient for the way I use the class, it certainly needs to be adapted to be completely generic. Johann #ifndef __multi_array_h #define __multi_array_h #include "config.h" #include <boost/multi_array.hpp> namespace boost { /*! \brief Assignable version of Boost MultiArray. \author Johann Briffa \par Version Control: - $Revision: 1488 $ - $Date: 2008-07-11 10:35:16 +0100 (Fri, 11 Jul 2008) $ - $Author: jabriffa $ */ template<typename T, std::size_t NumDims> class assignable_multi_array : public multi_array<T,NumDims> { public: /*! \name Constructors / Destructors */ explicit assignable_multi_array() : multi_array<T,NumDims>() {}; explicit assignable_multi_array(const detail::multi_array::extent_gen<NumDims>& ranges) : multi_array<T,NumDims>(ranges) {}; explicit assignable_multi_array(const assignable_multi_array& x) : multi_array<T,NumDims>(dynamic_cast< const multi_array<T,NumDims>& >(x)) {}; // @} /*! \name Assignment */ assignable_multi_array& operator=(const assignable_multi_array& x) { if(!std::equal(this->shape(), this->shape()+this->num_dimensions(), x.shape())) { resize(x.extents()); for(std::size_t i=0; i<NumDims; i++) libbase::trace << "Output Extent " << i << " = " << this->shape()[i] << "\n"; } dynamic_cast< multi_array<T,NumDims>& >(*this) = dynamic_cast< const multi_array<T,NumDims>& >(x); return *this; } // @} /*! \name Informative functions */ /*! \brief Get array extents description Returns an object describing the array extents, in a format suitable for use with resize(). */ detail::multi_array::extent_gen<NumDims> extents() const { typedef typename multi_array<T,NumDims>::extent_range extent_range; detail::multi_array::extent_gen<NumDims> extents_list; for(std::size_t i=0; i<NumDims; i++) { extents_list.ranges_[i] = extent_range(this->index_bases()[i], this->index_bases()[i]+this->shape()[i]); libbase::trace << "Input Extent " << i << " = " << this->shape()[i] << "\n"; } return extents_list; } // @} }; }; // end namespace #endif

4 9

Testing a boost::signal
by Simon Pickles 19 Jul '08

19 Jul '08

Hi, What is the correct syntax to test whether a signal has any slots connected? boost::signal<void(void)> sig; if (sig) DoSomething(); This doesn't compile. Thanks Simon

3 2

Specifying starting vertex in depth_first_search
by Hiren Patel 19 Jul '08

19 Jul '08

I'm trying to use depth_first_search() and I want to specify the start node. However, I can't seem to specify the color map correctly. I'm encountering the same problem as described in one of the archive posts at ( http://groups.google.ca/group/boost-list/browse_thread/thread/679106e227e0e… ). Any small examples available? Regards, ~ Hiren

4 4

[serialization 1_36_0] : Binary Archive portability between 32 and 64 bits platforms...
by Mathieu Peyréga 18 Jul '08

18 Jul '08

Now i'm even more sure the issue come from the 32 bits 64 bit point... See attached the HEX view of the same object kind serialized on a binary archive under a 32 bits and a 64 bits platform... (resp Vista and Ubuntu 8.04) Anything that i can configure to avoid that behaviour ? Any work arounds ? regards, Mathieu -- http://www.incub.net/

2 2

[boost-users][lambda] calling a functor in-place
by Igor R 18 Jul '08

18 Jul '08

Hi, std::vector<boost::function<void(void)> > functors; using namespace boost::lambda; std::for_each (functors.begin(), functor.end(), _1); The above compiles, but...does nothing. How can I enforce calling of operator () ? Thank you in advance!

6 9

Re: [Boost-users] run_test parameter not working (Boost Unit Test Framework)
by Marc Oliver Castañeda 18 Jul '08

18 Jul '08

Hi, David is right, the issue can be reproduced with the following simple program: #define BOOST_TEST_MODULE example #include <boost/test/included/unit_test.hpp> BOOST_AUTO_TEST_CASE( testA ) { } BOOST_AUTO_TEST_CASE( testB ) { } BOOST_AUTO_TEST_SUITE( s1 ) BOOST_AUTO_TEST_CASE( test1 ) { } BOOST_AUTO_TEST_CASE( lest2 ) { } BOOST_AUTO_TEST_SUITE_END() BOOST_AUTO_TEST_SUITE( s2 ) BOOST_AUTO_TEST_CASE( test1 ) { } BOOST_AUTO_TEST_CASE( test11 ) { } BOOST_AUTO_TEST_SUITE( in ) BOOST_AUTO_TEST_CASE( test ) { } BOOST_AUTO_TEST_SUITE_END() BOOST_AUTO_TEST_SUITE_END() Compile and link this with the boost unit framework shared library. Run the executable like the following: >example --log_level=test_suite --run_test=testA Output should be.. Running 1 test case... Entering test suite "example" Entering test case "testA" Test case testA doesn't include any assertions Leaving test case "testA" Leaving test suite "example" *** No errors detected However, output is always like the following: Running 7 test cases... Entering test suite "example" Entering test case "testA" Test case testA doesn't include any assertions Leaving test case "testA" Entering test case "testB" Test case testB doesn't include any assertions Leaving test case "testB" Entering test suite "s1" Entering test case "test1" Test case test1 doesn't include any assertions Leaving test case "test1" Entering test case "lest2" Test case lest2 doesn't include any assertions Leaving test case "lest2" Leaving test suite "s1" Entering test suite "s2" Entering test case "test1" Test case test1 doesn't include any assertions Leaving test case "test1" Entering test case "test11" Test case test11 doesn't include any assertions Leaving test case "test11" Entering test suite "in" Entering test case "test" Test case test doesn't include any assertions Leaving test case "test" Leaving test suite "in" Leaving test suite "s2" Leaving test suite "example" *** No errors detected Thanks Marc ----- Original Message ---- From: Gennadiy Rozental <rogeeff(a)gmail.com> To: boost-users(a)lists.boost.org Sent: Friday, July 18, 2008 1:53:00 AM Subject: Re: [Boost-users] run_test parameter not working (Boost Unit Test Framework) David Abrahams <dave <at> boostpro.com> writes: > > > on Thu Jul 17 2008, Marc Oliver Castañeda <floranteatlaura-AT-yahoo.com> wrote: > > > Hi, > > > > I've downloaded the 1.35.0 source and built my own boost libraries in Redhat 5.1 with > > the intention of using the boost unit test framework. However, I noticed that the run_test > > parameter is not working. > > parameter to what? e.g., a function? a command-line tool? > he mean command line parameter that is used to specify which test units to run based on their name _______________________________________________ Boost-users mailing list Boost-users(a)lists.boost.org http://lists.boost.org/mailman/listinfo.cgi/boost-users

1 0

Boost.Test: time to execute a unit test?
by Fadi Beyrouti 17 Jul '08

17 Jul '08

Hi, The time it takes to execute a unit tests is rarely showing up in the log. It shows up for some test cases but not others. Furthermore, the individual test cases for which the time does appear seem to change from run to run. Here is an example: For the first run of one of our tests we get (an excerpt only): ... Entering test case "QuaternionTest" Leaving test case "QuaternionTest" Entering test case "QuaternionOperatorsTest" Leaving test case "QuaternionOperatorsTest" Entering test case "QuaternionMathTest" Leaving test case "QuaternionMathTest" Entering test case "QuaternionTransformsTest" Leaving test case "QuaternionTransformsTest"; testing time: 16ms Entering test case "QuaternionLerpTest" Leaving test case "QuaternionLerpTest" Entering test case "QuaternionConjugateTest" Leaving test case "QuaternionConjugateTest" ... However, another run will yield a different output. ... Entering test case "QuaternionTest" Leaving test case "QuaternionTest" Entering test case "QuaternionOperatorsTest" Leaving test case "QuaternionOperatorsTest"; testing time: 16ms Entering test case "QuaternionMathTest" Leaving test case "QuaternionMathTest" Entering test case "QuaternionTransformsTest" Leaving test case "QuaternionTransformsTest" Entering test case "QuaternionLerpTest" Leaving test case "QuaternionLerpTest" Entering test case "QuaternionConjugateTest" Leaving test case "QuaternionConjugateTest" ... It seems to me that the testing time is not being outputted if it is less than 15ms. The unit tests are being run on Windows XP professional. Any help would be appreciated. Thanks. -- Fadi,

2 1

run_test parameter not working (Boost Unit Test Framework)
by Marc Oliver Castañeda 17 Jul '08

17 Jul '08

Hi, I've downloaded the 1.35.0 source and built my own boost libraries in Redhat 5.1 with the intention of using the boost unit test framework. However, I noticed that the run_test parameter is not working. Has anyone experienced this and came up with a solution to resolve this issue? The build in windows is working fine. I've verified using gdb that the parameter value is being passed correctly to boost layer. any help will be appreciated. thanks Marc

3 3

Undefined symbols "boost::system::get_posix_category()" and "boost::system::get_system_category()"
by Frank van Meurs 17 Jul '08

17 Jul '08

Hello, I'm trying to use the filesystem libraries provided by Boost (1.35) through "boost/filesystem.hpp". I've installed Boost via macports and checked whether or not the installation is up to date, which it is. Moreover, I can use (some) other Boost libraries just fine, but not the abovementioned one. I'm using the following code to access several libraries: ----SNIP---- #include <boost/regex.hpp> #include <boost/signals.hpp> #include <boost/system/error_code.hpp> #include <boost/filesystem.hpp> ---/SNIP---- Including any of the first three libraries works just fine and results in a successful build. However, adding the fourth results in: ----SNIP---- Undefined symbols: "boost::system::get_posix_category()", referenced from: __static_initialization_and_destruction_0(int, int)in test.o __static_initialization_and_destruction_0(int, int)in test.o "boost::system::get_system_category()", referenced from: __static_initialization_and_destruction_0(int, int)in test.o __static_initialization_and_destruction_0(int, int)in test.o ---/SNIP--- Can anybody point me to a solution for this issue? I really prefer using the existing functionality from Boost over wirting them myself... Thanks in advance. Kind regards, Frank -------------------- Frank van Meurs Dept. of Information and Computer Sciences, Utrecht University Padualaan 14, 3584 CH Utrecht, The Netherlands P: +31 (0)30 - 253 9128 F: +31 (0)30 - 253 4619 W: http://www.cs.uu.nl/staff/fvm.html

3 2

how can I return a 'null' smart pointer?
by Meryl Silverburgh 17 Jul '08

17 Jul '08

Hi, Can you please tell me how can I return a 'null' smart pointer? typedef shared_ptr<MyA> MyAPtr; typedef vector<MyAPtr> MyAVector; For example,I have this code, which return NULL in an error situation: MyAPtr myfunction(MyAVector& v, int size) { if (index < v.size()) { return v[index]; } else { // how can I return NULL here? } } How can I achieve that when I use boost smart pointer? Thank you for any tip.

10 15

Boost.Filesystem path string() confusion
by PETER BARKER 17 Jul '08

17 Jul '08

Hello, I want to use the Boost.Filesystem library in my Win32 application. I've stumbled upon a problem which may very well be my misunderstanding of how to use the library but reading the documentation doesn't clear up my confusion. The user choses a path from a native UI and a bfs::path is constructed with the native name checker. This path needs to be persisted so I call the string() method which according to the documentation "Returns: The contents of m_name, formatted according to the rules of the generic path string grammar." I figured this string would be acceptable to the bfs::path constructor without specifying native because it's in the corrct grammar but I get an exception thrown. Here's a stripped down version of my code: #include <iostream> #include <iterator> #include <boost/filesystem/operations.hpp> namespace bfs = boost::filesystem; int main() { try { // Path constructed from a native UI so constructed with native bfs::path myPath("C:/autoexec.bat",bfs::native); // Persist to grammar conformant string std::string persisted = myPath.string(); // Recreate from persisted string - *THROWS EXCEPTION* bfs::path otherPath(persisted); } catch(const std::exception& e) { std::cout << e.what() << std::endl; } return 0; } The exception string is "boost::filesystem::path: invalid name "C:" in path: "C:/autoexec.bat"" Can someone explain how I should be doing this please? Thanks, Pete

2 3

How to match '}' in regex
by 王文华 17 Jul '08

17 Jul '08

Hello everyone, when i use regex("}"), there is always an error. How can I match it? Regards NextRainbow.

2 1

[boost-users][bind, signal] how to
by Igor R 17 Jul '08

17 Jul '08

Hello, I've got the following design: several functors ("Tasks"), when executed, "delegate" to a shared Processor object that performs some coordination between the tasks. The Processor runs the tasks at some stage, and wants to be notified by the task when it's over. However, I wouldn't like them to "know" each other. I even wouldn't like the Processor making assumptions about exact Task's "connection point" signature/name. For this purpose I want the Task to pass its "connection point" as a generic functor - so that the Processor would execute it to subscribe. However, I haven't managed to bind signal::connect. I'd appreciate any idea about what's wrong in the following schematic code: (In the real code the life-time of the objects is managed by means of shared/weak pointers, which are omitted here for the sake of simplicity!) //task.hpp class Task { public: typedef signal<void (void)> CompletionEvent; void operator ()() { // forward the execution to the processor; (*processor)(bind(&Task::run, this), bind(&CompetionEvent::connect, &done_, _1)); // this bind dosen't compile! } private: void run(); Processor *processor_; signal<void(void)> done_; }; // processor.hpp class Processor { public: void operator ()(boost::function<void(void)> task, boost::function<void(boost::function<void(void)>)> completionNotifier) { // this way I'd like to call signal::connect(bind(&Processor::taskDone, this)); completionNotifier(bind(&Processor::taskDone, this)); } private: void taskDone(); }; Thank you!

4 4

performance weirdness with AVL set
by Lothar Werzinger 17 Jul '08

17 Jul '08

Hi, If I compile the attached test with and without compile optimizations with G++ 4.3.1 the non optimized version (using optimize_size<false>) is more than 10 times faster than the optimized one. I did use oprofile on the application where I use the avl set and the time is spent in the 'find' call. Can you please try to shed some light on this, I am completely stumped. Any help is appreciated! here is how I built the attached test: $ /opt2/linux/ix86/bin/g++-4.3.1 -m64 -O3 -g -I/opt2/linux/x86_64/include/boost-1_35 -DUSE_OPTIMIZATION=0 -o ./avltest-no ./avltest.cpp ./avltest.cpp:21:2: warning: #warning without optimization $ /opt2/linux/ix86/bin/g++-4.3.1 -m64 -O0 -g -I/opt2/linux/x86_64/include/boost-1_35 -DUSE_OPTIMIZATION=0 -o ./avltest-nn ./avltest.cpp ./avltest.cpp:21:2: warning: #warning without optimization $ /opt2/linux/ix86/bin/g++-4.3.1 -m64 -O3 -g -I/opt2/linux/x86_64/include/boost-1_35 -DUSE_OPTIMIZATION=1 -o ./avltest-oo ./avltest.cpp ./avltest.cpp:21:2: warning: #warning with optimization $ /opt2/linux/ix86/bin/g++-4.3.1 -m64 -O0 -g -I/opt2/linux/x86_64/include/boost-1_35 -DUSE_OPTIMIZATION=1 -o ./avltest-on ./avltest.cpp ./avltest.cpp:21:2: warning: #warning with optimization $ time ./avltest-nn sizeof(BalanceNode)=32 sizeof(BalanceSet)=40 creating file /tmp/avltest.img done region.get_address()=0x7fd715fc2000 region.get_size()=1600056 p_set=0x7fd715fc2000 p_values=0x7fd715fc2030 created/inserted 50000 entries region.get_address()=0x7fd715fc2000 region.get_size()=1600056 p_set=0x7fd715fc2000 p_values=0x7fd715fc2030 checked 50000 entries real 0m1.031s user 0m0.956s sys 0m0.008s $ time ./avltest-no sizeof(BalanceNode)=32 sizeof(BalanceSet)=40 creating file /tmp/avltest.img done region.get_address()=0x7f70b6db6000 region.get_size()=1600056 p_set=0x7f70b6db6000 p_values=0x7f70b6db6030 created/inserted 50000 entries region.get_address()=0x7f70b6db6000 region.get_size()=1600056 p_set=0x7f70b6db6000 p_values=0x7f70b6db6030 checked 50000 entries real 0m20.803s user 0m20.233s sys 0m0.028s $ time ./avltest-on sizeof(BalanceNode)=32 sizeof(BalanceSet)=32 creating file /tmp/avltest.img done region.get_address()=0x7f8f9750d000 region.get_size()=1600048 p_set=0x7f8f9750d000 p_values=0x7f8f9750d020 created/inserted 50000 entries region.get_address()=0x7f8f9750d000 region.get_size()=1600048 p_set=0x7f8f9750d000 p_values=0x7f8f9750d020 checked 50000 entries real 0m1.890s user 0m1.844s sys 0m0.012s $ time ./avltest-oo sizeof(BalanceNode)=32 sizeof(BalanceSet)=32 creating file /tmp/avltest.img done region.get_address()=0x7fb9e3b2d000 region.get_size()=1600048 p_set=0x7fb9e3b2d000 p_values=0x7fb9e3b2d020 created/inserted 50000 entries region.get_address()=0x7fb9e3b2d000 region.get_size()=1600048 p_set=0x7fb9e3b2d000 p_values=0x7fb9e3b2d020 checked 50000 entries real 0m0.248s user 0m0.144s sys 0m0.016s Lothar -- Lothar Werzinger Dipl.-Ing. Univ. framework & platform architect Tradescape Inc. - Enabling Efficient Digital Marketplaces 1754 Technology Drive, Suite 128 San Jose, CA 95110 email: lothar(a)tradescape.biz web: http://www.tradescape.biz

2 3

Question on the Boost Preprocessor library
by Wong, Shin Guey 17 Jul '08

17 Jul '08

Hi all, This is my first email to this list; please point me out if I done anything wrong in this mail like the title format or etc... I am a newbie to the Boost library. I am really interesting on the usage of the Boost preprocessor library. I would like to archive 1 of the thing but I am not sure whether it is possible with Boost preprocessor library: Converting the binary number to hexadecimal or decimal value Eg: BIN2HEX(011) -> 0x3 or BIN2DEC(011) -> 3 BIN2HEX(1111) -> 0xf or BIN2DEC(1111) -> 15 I had read the C-Template-Metaprogramming-Concepts-Tools-and-Techniques-from-Boost-and- Beyond-C-in-Depth-Series which tell me how to archive that using template. But what I really want is to done it with the preprocessor without using any template which will not affect the run time but only the compile time. I know that I can use the loop and the arithmetic function from the Boost library but how do I get the binary number characters count, eg. COUNT(011) -> 3 or COUNT(0111) -> 4. So that I can use this count with loop and the arithmetic function to perform the conversion from binary to hexadecimal or decimal. Thanks in advance. Regards, Shin Guey

8 9

[function,bind] Templated function<> function
by Michiel Helvensteijn 16 Jul '08

16 Jul '08

Ok, the subject-line might be a bit confusing. :-) I have the following template function: template<class T> bool is(const string& str); Its implementation is not important. It checks if a string can be converted to type T by std::stringstream. I also have the following function: template <class T> function<bool (T)> neg(function<bool (T)> func) { return !bind(func, _1); } Which I use in the following way: BOOST_CHECK_PREDICATE( neg(is<int>), ("text") ); But the compiler (GCC 4.1.2) complains: error: no matching function for call to 'neg(<unresolved overloaded function type>)' What's the problem. And, also, what's the solution? :-) Thanks in advance! -- Michiel Helvensteijn

2 6

[serialization] Visual Studio 2008 + Boost 1.35 : runtime casting not working
by Steve Sanders 16 Jul '08

16 Jul '08

Hi, Below is a basic sample adapted from the documentation at: http://tinyurl.com/55fr4q For me, with Visual Studio 2008, the deserialization of the base pointer results in an instance of the base class, not the serialized derived class. Also, if I add a virtual function to the base class, deserialization results in an exception (it attempts to instantiate the base class anyway). I added the change cited here and rebuilt the serialization library, which did not help: http://tinyurl.com/5j4e8x Any suggestions would be appreciated. Thanks, Steve Sanders #include "stdafx.h" #include "stdafx.h" #include <sstream> // include headers that implement a archive in simple text format #include <boost/archive/text_oarchive.hpp> #include <boost/archive/text_iarchive.hpp> #include <boost/serialization/export.hpp> class base { public: base(){ m_val = 1; } void PrintVal( void ) { printf(" Value: %d\n", m_val ); } protected: // virtual void foo(void) = 0; friend class boost::serialization::access; //... // only required when using method 1 below // no real serialization required - specify a vestigial one template<class Archive> void serialize(Archive & ar, const unsigned int file_version){} int m_val; }; class derived : public base { public: derived(){ m_val = 2; } private: void foo(void){} friend class boost::serialization::access; template<class Archive> void serialize(Archive & ar, const unsigned int file_version){ // method 1 : invoke base class serialization boost::serialization::base_object<base>(*this); // method 2 : explicitly register base/derived relationship boost::serialization::void_cast_register<derived, base>( static_cast<derived *>(NULL), static_cast<base *>(NULL) ); } }; BOOST_CLASS_EXPORT_GUID(derived, "derived") void test4(){ //... std::stringstream ss; boost::archive::text_oarchive oa(ss); derived *d = new derived(); base *b = d; b->PrintVal(); oa & d; boost::archive::text_iarchive ar(ss); b = NULL; ar & b; b->PrintVal(); } int main() { test4(); return 0; }

3 10

Can I use boost smart pointer to a pointer a vector of smart pointer to my class
by Meryl Silverburgh 16 Jul '08

16 Jul '08

Hi, In my code, I have this: A pointer to a vector of pointer to A: vector<A*>* Can I switch it to: shared_ptr< vector< shared_ptr <A> > > ? a shared pointer to a vector of shared pointer of A? And when the reference count reaches to 0, will it remove everything? the memory of the vector and all the A objects pointed by the vector?

2 1

Switch from a regular C++ pointer to boost shared_ptr
by Meryl Silverburgh 16 Jul '08

16 Jul '08

Hi, I have a class like this: class A { private: B* b; public: A(); B& getB(); } A::A() { b = NULL; } B& A::getB() { if (b == NULL) { b = new B(); } return *b; } when I switch to use boost shared_ptr, i.e. switch from 'B* b' to 'shared_ptr<B> b' in A's class definition. I don't need to do 'b= NULL' in A's constructor, right? But why this line won't compile "b = new B();" ? I get this error: A.cpp:297: error: no match for 'operator=' in '((const A*)this)->A::xBlockMap = (((B*)operator new(128u)), (<anonymous>->B::B(((const A&)((const A*)this)), ((int)strict)), <anonymous>))' /usr/local/include/boost-1_35/boost/shared_ptr.hpp:194: note: candidates are: boost::shared_ptr<T>& boost::shared_ptr<T>::operator=(const boost::shared_ptr<T>&) [with T = B] Thank you for any help.

3 4

Visual Studio 2008: Link Errors in Debug Mode (__imp____wassert not found).
by Vikram Saxena 16 Jul '08

16 Jul '08

I am trying to use the circular_buffer template from Boost (version 1.35.0) with VS 2008 (msvc-9.0). I initially downloaded the compiled libraries from boostconsulting. However when I linked against them, I got a link error that (__imp___wassert not found )whenever I used an iterator. A search on the web revealed this problem seems to occur when the version of MSVC used to compile is different from the one used to link. I then decided to do a local compile using the basic bjam command with the toolset pointing to msvc-9.0. However I still get the same error; I have confirmed that VS is picking up boost libraries from the locally compiled location. The error does not happen in the Release mode. For the time being I am working around the problem by disabling assertion via a compile time flag which forces all boost asserts to (void)0. I will appreciate any thoughts on how to overcome this problem without disabling asserts. -Vikram

1 0

[general] boost 1.35.0 library naming and bjam arguments priorities
by damny＠web.de 16 Jul '08

16 Jul '08

hi, for building boost 1.35.0 with bjam 3.1.16 and mingw gcc 3.4.5 on winxp sp2 i created a *.BAT-file like follows: @echo off SETLOCAL SET MINGW=C:\Development_Compiler\MinGW SET PYTHN=C:\Development_Compiler\Python25 SET BUILD=C:\Development_Libs\Boost_v1.35.0 SET path=C:\WINDOWS;C:\WINDOWS\system32;%MINGW%\Bin cls bjam.exe -a "-sTOOLS=mingw" "-sGCC_ROOT_DIRECTORY=%MINGW%" "-sPYTHON_ROOT=%PYTHN%" "-sPYTHON_VERSION=2.5" --link=static --runtime-link=static --threading=single --optimization=on --toolset=gcc --build-dir=%BUILD% --build-type=complete install ENDLOCAL this works well. but unlike boost 1.34.1 and bjam 3.1.15 [*] i get a lot of *.LIBs instead of *.A-files. is there a new/different library naming than this described in http://www.boost.org/doc/libs/1_35_0/more/getting_started/windows.html#libr… ? in spite of "--threading=single" i also get the "...-mt.lib"-files. why? covers "--build-type=complete" the single threading argument? thank you all daniel [*] i didn't review this, but the last build some months ago was okay :o) _______________________________________________________________________ EINE FÜR ALLE: die kostenlose WEB.DE-Plattform für Freunde und Deine Homepage mit eigenem Namen. Jetzt starten! http://unddu.de/?kid=kid@mf2

7 28

Copy-on-write implementation
by Michiel Helvensteijn 16 Jul '08

16 Jul '08

I am looking for a simple way to implement copy-on-write classes. Several `copies' of an instantiation of such a class can coexist and share the same data in memory until someone changes their `copy' (calls a non-const member function). At that time, a deep copy is made and the change is only made in that one location. The deep copy is not needed, of course, if the instantiation has only one handle. All of this should be transparent to the user of the class. I think implementing such a class would be easier with a special class to inherit from and/or a useful set of macro's designed for this purpose. This seems like something that would (or should) exist in boost. Does boost provide something like this? If not, I will write it myself. But I'm still not sure how to approach it. I imagine shared_ptr would play a role, though. Any ideas? -- Michiel Helvensteijn

3 2

Re: [Boost-users] Boost serialization: is there a way to perform selective read
by ari vol 16 Jul '08

16 Jul '08

The problem with this solution is that it is time consuming, to open file for a key, i would like to use the boost serializeation to improve time preformance. What is realy required is some mechanism, to identify an instance according to its key and only than read its data. Thus reading only the required instnace (accordig to its key) and not also the previous data at the file, someting like seek. I hoped that "XML boost" can supply someting like that, but could not find how. Thanks, -Ariella ----- Original Message ---- From: gchen <chengang31(a)gmail.com> To: boost-users(a)lists.boost.org Sent: Wednesday, July 16, 2008 9:43:56 AM Subject: Re: [Boost-users] Boost serialization: is there a way to perform selective read ari vol wrote: > Hey, > > I would like to know whether there is a way to preform selective read > using Boost serialization > <http://www.boost.org/doc/libs/1_35_0/libs/serialization/doc/index.html>, > thus reading serialized instance with its id/key. > > The implementation that i know is saving instances to a file and then > loading them all according to the saving order. Is there a way to read > only one selected instance from the file. > How about serializing objects separately? I mean instead of saving a object and its children objects into a big file, simply saving them into different files. _______________________________________________ Boost-users mailing list Boost-users(a)lists.boost.org http://lists.boost.org/mailman/listinfo.cgi/boost-users

4 3

boost.program_options no throw
by Aljaz 16 Jul '08

16 Jul '08

Hello I've been wondering if there is no-throw version of boost::program_options::store , command_line_parser and basic_line_parser ? Many thanks in advance!

5 4

[math/distributions/poisson] extract Poisson-distributed random variate?
by Michael Fuller 16 Jul '08

16 Jul '08

I've searched the Boost documentation and mail archive, but I can't find an example for extracting a random variate from a Poisson distribution. What I want to do is similar to extracting a variate from a uniform distribution, such as the C++ rand() function, which returns a single random variate on the range [0, RAND_MAX]. Is there a way to do something similar using the Poisson distribution with a given mean? For example, something like this: double n, lambda(0.30); //lambda is the mean of the Poisson n = poisson_rand(lambda); Thanks, Mike

2 4

boost interprocess
by Marc Magrans de Abril 16 Jul '08

16 Jul '08

Dear colleagues, I have been testing the boost interprocess library for two months in my spare time... everything looked ok until now. Perhaps I am doing something wrong, so be patient. As you can see in the code, I am trying to use the offset_ptr within a class named OffsetPtrClass. However, if I try to allocate the int inside the OffsetPtrClass cnstructor I'll get a segmentation fault in the getter (either the MASTER or the SLAVE forked processes). I a running on Linux kernel 2.6.16 and g++ 3.4.xx. What I am doing wrong? Thanks for your help, marc #include <iostream> #include <unistd.h> #include <boost/interprocess/managed_shared_memory.hpp> #include <boost/interprocess/offset_ptr.hpp> namespace ip=boost::interprocess; class OffsetPtrClass { public: OffsetPtrClass():myint_(0) { //If we allocate the int here we get a segmentation fault!! // //ip::managed_shared_memory segment(ip::open_only,"MySharedSegment"); //myint_ = static_cast<int*>(segment.allocate(sizeof(int))); } ~OffsetPtrClass() { } void setMyint(const int& myint) { *myint_ = myint; } int getMyint() { int myint = *myint_; return myint; } ip::offset_ptr<int> myint_; }; int main( int argc, char* argv[] ) { std::string shName("MySharedSegment"); pid_t pid = fork(); if (pid == 0) { try { std::cout << "CHILD: getpid() = " << getpid() << std::endl; std::cout << "CHILD: press return to continue..." <<std::endl; getchar(); ip::managed_shared_memory segment(ip::open_only, shName.c_str()); std::cout << "CHILD: Retrieving OffsetPtrClass 'optr' instance: " << std::endl; std::pair<OffsetPtrClass*, std::size_t> res; res = segment.find<OffsetPtrClass> ("optr"); assert(res.second == 1); OffsetPtrClass& optr =*(res.first); std::cout << "CHILD: 'optr'::getMyint() = " << optr.getMyint() << std::endl; std::cout << "CHILD: 'optr'::SetMyint(100)" << std::endl; optr.setMyint(100); } catch (...) { std::cerr << "CHILD: Something failed..." << std::endl; throw; } } else if (pid <0){ std::cerr << "---program.cc: Failed to fork" << std::endl; exit(1); } else { std::cout << "PARENT: getpid() = " << getpid() << std::endl; ip::shared_memory_object::remove(shName.c_str()); try{ std::cout << "PARENT: creating segment" << std::endl; ip::managed_shared_memory segment(ip::create_only, shName.c_str(), 65536); std::cout << "PARENT: creating OffsetPtrClass 'optr' instance" << std::endl; OffsetPtrClass& optr = *(segment.construct<OffsetPtrClass>("optr")()); //If we allocate the int here we do not have a segmentation fault optr.myint_ = static_cast<int*>(segment.allocate(sizeof(int))); std::cout << "PARENT: 'optr'::SetMyint(333)" << std::endl; optr.setMyint(333); std::cout << "PARENT: 'optr'::getMyint() = " << optr.getMyint() << std::endl; std::cout << "PARENT: 'optr'::*myint_() = " << *(optr.myint_) << std::endl; std::cout << "PARENT: press return to continue..." <<std::endl; getchar(); std::cout << "PARENT: 'optr'::getMyint() = " << optr.getMyint() << std::endl; } catch(...){ std::cerr << "PARENT:something failed... Everything is destroyed" << std::endl; throw; } std::cout << "PARENT: Shared memory destroyed" << std::endl; } }

2 2

Re: [Boost-users] Boost serialization: is there a way to perform selective read
by Pavlo Korzhyk 16 Jul '08

16 Jul '08

> It's not there. boost::archive works sequentially on a stream and that's > it, so if you need to skip large portions of data you will have to > divide your data into multiple files and/or multiple chunks in one > stream. I did the latter and it works fine if you build some kind of > meta-information structure on where chunks begin and end. This enables > you to seek a specific chunk of data and process that with an archive. > > In any case, if you want to improve performance of your app use > (portable) binary instead of xml archives! You could take a look at http://code.google.com/p/hiberlite It will do what you need storing objects in sqlite database. There were no new releases for a couple of months, but I'll return to this project in a week. And there are _already_ enough features to solve the "selective access" problem. Yours sincerely, Pavlo Korzhyk ICQ#155870780

1 0

boost.program_options - passing string or vector<string>
by Aljaz 16 Jul '08

16 Jul '08

Hello I want to use boost.program_options to parse commands for my program (console application) What I want to do is to be able to pass std::vector<std::string> of arguments or std::string (line) to parse_command_line(ac, av, desc); instead of char *argv, int argc. How is that possible? I've been looking at basic_command_line_parser class but couldnt figure out how it would be possible to do that, since converting string/vector of strings to array of char pointers would be really stupid. Many thanks for help Aljaz

2 4

SVN tree : Issue with libstdcpp3.hpp
by Mathieu Peyréga 16 Jul '08

16 Jul '08

Hello, starting from line 66, I can find : #if defined(__GLIBCXX__) || (defined(__GLIBCPP__) && __GLIBCPP__>=20020514) // GCC >= 3.1.0 # define BOOST_STD_EXTENSION_NAMESPACE __gnu_cxx # define BOOST_HAS_SLIST # define BOOST_HAS_HASH # define BOOST_SLIST_HEADER <ext/slist> # if defined(__GLIBCXX__) && __GLIBCXX__ < 20071018 // GCC <= 4.3.0 # define BOOST_HASH_SET_HEADER <ext/hash_set> # define BOOST_HASH_MAP_HEADER <ext/hash_map> # else # define BOOST_HASH_SET_HEADER <backward/hash_set> # define BOOST_HASH_MAP_HEADER <backward/hash_map> # endif #endif on ubuntu 8.04 (64 bits) the 4.2.3 gcc version define __GLIBCXX__ as #define __GLIBCXX__ 20080329 so boost code prevents from using the ext headers when the backward headre do not exist (on my system)... Shouldn't the < be a > ? regards, Mathieu -- http://www.incub.net/

1 0

serialization - How to retain protected access of non-default ctors
by George Verdad 16 Jul '08

16 Jul '08

Hi All, I am getting compiler errors using mingw 3.4.5 on the code below: error: `BaseClass::BaseClass(const std::string&)' is protected I tried to declare as friend the load_construct_data() override in BaseClass but gave me the following error: error: call of overloaded load_constuct_data(..) is ambiguous. What am I doing wrong? TIA, Gvv ... #include <boost/serialization/base_object.hpp> #include <boost/serialization/export.hpp> #include <boost/serialization/shared_ptr.hpp> #include <boost/serialization/list.hpp> #include <boost/archive/text_oarchive.hpp> #include <boost/archive/text_iarchive.hpp> template <class T> void Serialize(const T& t, std::string& OutData) { std::ostringstream ArchiveStream; boost::archive::text_oarchive Archive(ArchiveStream); Archive << t; OutData = ArchiveStream.str(); } template <class T> void Deserialize(T& t, const std::string& InData) { std::istringstream ArchiveStream(InData); boost::archive::text_iarchive Archive(ArchiveStream); Archive >> t; } class BaseClass : private boost::noncopyable { public: virtual ~BaseClass() {} const std::string& GetName() const { return Name; } protected: BaseClass(const std::string& name) : Name(name) {} private: std::string Name; template<class Archive> void serialize(Archive& ar, const unsigned int version) { ar & Name; } friend class boost::serialization::access; }; typedef boost::shared_ptr<BaseClass> BaseClassPtr; class DerivedClass : public BaseClass { public: DerivedClass(const std::string& name) : BaseClass(name) {} private: template<class Archive> void serialize(Archive& ar, const unsigned int version) { ar & boost::serialization::base_object<BaseClass>(*this); } friend class boost::serialization::access; }; BOOST_CLASS_EXPORT(DerivedClass) namespace boost { namespace serialization { template<class Archive> inline void save_construct_data(Archive& ar, const BaseClass* t, const unsigned int version) { ar << t->GetName(); } template<class Archive> inline void load_construct_data(Archive& ar, BaseClass* t, const unsigned int version) { std::string name; ar >> name; ::new(t)BaseClass(name); } template<class Archive> inline void save_construct_data(Archive& ar, const DerivedClass* t, const unsigned int version) { ar << t->GetName(); } template<class Archive> inline void load_construct_data(Archive& ar, DerivedClass* t, const unsigned int version) { std::string name; ar >> name; ::new(t)DerivedClass(name); } } } int main() { BaseClassPtr b(new DerivedClass("My name is gvv")); std::string data; std::cout << "Serialize" << std::endl; Serialize(b,data); std::cout << data << std::endl; b.reset(); std::cout << "Deserialize" << std::endl; Deserialize(b,data); std::cout << b->GetName() << std::endl; return 0; }

1 0

[preprocessor] BOOST_PP_IF with BOOST_PP_SEQ_ENUM as then-argument
by Maximilian Matthe 16 Jul '08

16 Jul '08

Hello! #define seq (int)(int) BOOST_PP_IF(1, BOOST_PP_SEQ_ENUM(seq), BOOST_PP_EMTPY) does not expand to what i want: it should be int, int . G++ gives the following error: test.cpp:20:1: macro "BOOST_PP_IIF" passed 4 arguments, but takes just 3 BOOST_PP_IIF which is logical, because SEQ_ENUM expands to int, int and so if gets 4 arguments. I tried to surround SEQ_ENUM with IDENTITY but it's the same problem. Only surrounding SEQ_ENUM with () works, but then I have the () in the code, which I do not want. How can I solve this problem? Max

2 4

Basic Conversion to Boost Graph Library Data Structure
by Lex Fridman 16 Jul '08

16 Jul '08

I know that what I'm asking is in the documentation, but I haven't been able to figure it out from just looking through it on a couple passes. I would like to convert my graph adjacency list to the BGL adjacency_list. Each node is represented with the Node data structure, and so is each edge. I also have "global" properties for the graph as a whole. So here are the three data structures my graph uses: GraphNode { int foo; double bar; } GraphEdge { double weight; bool fun; } GraphGlobals { string graphName; } How do I declare an adjacency_list that will uses these representations of the node, edge, and global parameters? Thanks. -- View this message in context: http://www.nabble.com/Basic-Conversion-to-Boost-Graph-Library-Data-Structur… Sent from the Boost - Users mailing list archive at Nabble.com.

2 2

Boost serialization: is there a way to perform selective read
by ari vol 16 Jul '08

16 Jul '08

Hey, I would like to know whether there is a way to preform selective read using Boost serialization, thus reading serialized instance with its id/key. The implementation that i know is saving instances to a file and then loading them all according to the saving order. Is there a way to read only one selected instance from the file. Thanks, -Ariella

3 2

Facilities for view
by Peng Yu 16 Jul '08

16 Jul '08

Hi, I'm wondering if there is any facility to help define view in boost. There is view for static types http://www.boost.org/doc/libs/1_35_0/libs/fusion/doc/html/fusion/view.html. And there is a package to help define iterators http://www.boost.org/doc/libs/1_35_0/libs/iterator/doc/index.html. Due to the similarity between view and iterator, I'm wondering if there is a similar package for view? Thanks, Peng

2 2

Beginner design question
by Hansi 16 Jul '08

16 Jul '08

Hello, I have to make a library which provides a few classes with different implementation. This classes should have a common interface. Normally I would use derived classes from the common interface, but this classes should be really fast. What is the best way to provide a common interface and to not have the overhead of polymorphic classes? Best regards Hansjörg

5 6

linking problem for self registering dll ?
by Peter Boot 16 Jul '08

16 Jul '08

I am using VS2005 Professional to build an ATL dll that is self registered by the IDE. The project uses boost::regex so I needed to ad the directory containing libboost_regex-vc80-mt-gd-1_35.lib to Linker->Additional Include Directories to get it to compile. But now I am getting the following error at the self register stage, the dll registers fine when the boost::regex code is commented out or I use boost::xpressive instead. Project : error PRJ0050: Failed to register output. Please ensure you have the appropriate permissions to modify the registry. Please tell me how to fix TIA

1 0

[Thread] Trying to share a mutex between objects
by Adam Nielsen 16 Jul '08

16 Jul '08

Hi all, I'm having some trouble getting my head around Boost mutexes and how they work. Basically I have a handful of objects, and they all use a shared resource (a C-style function.) The function isn't thread-safe, so I want to protect it to ensure only one thread at a time can call it. My idea was to create a mutex for this (lock before the call, unlock after it), but I'm having trouble working out how to share the mutex between all the objects. The boost::mutex object seems like what I want, but it's non-copyable so I can't share it among multiple objects. I tried passing around references to the mutex instead, but this just triggers an assertion failure when I try to lock it. Should I be using boost::mutex for this? How would you normally serialise access among different objects like this? I'm finding the examples somewhat tricky to work from, as most of them use global variables and C-style functions for creating threads, which doesn't really fit with the C++ model of my program - I could just make the mutex a global variable but it doesn't seem like the cleanest solution to me. Any pointers would be much appreciated. Thanks, Adam.

2 3

How can I get a reference from a boost smart pointer?
by Meryl Silverburgh 16 Jul '08

16 Jul '08

Hi, I have a pointer to a class: A* aPt; and I deference it to pass that as a reference void aFunction(A& aRef); and I call aFunction(*aPt); but if I switch to use boost smart pointer how can I de-reference it and call aFunction()? shared_ptr<A> aPt; I tried: aFunction(aPt); // does not compile aFunction(aPt->get()); // does not compile. Thank you for any help.

2 1

Fatal link error LNK1561: entry point must be defined when using boost::test
by Tjernstrom, Staffan 15 Jul '08

15 Jul '08

Folks, Hopefully this is me having finger trouble, but I can't seem to get around the problem. Boost 1.35 installed using the binary installer, all the vc8 flavours installed. What happens is that I try to compile / link any of the unit test examples in Visual Studio 2005 Pro, for example the code at http://www.boost.org/doc/libs/test/example/unit_test_example_05.cpp I keep getting the LNK1561: entry point must be defined error. I know that it's looking for, and picking up, the libboost_unit_test_framework-vc80-mt-1_35.lib as if I remove the /LIBPATH from the link command options it compains that it cannot find that library. I have my project settings set as follows: Compiler: /O2 /GL /I "C:\Program Files (x86)\boost\boost_1_35_0" /D "_MBCS" /FD /EHa /MD /Fo"Release\\" /Fd"Release\vc80.pdb" /W3 /nologo /c /Wp64 /Zi /TP /errorReport:prompt Linker: /OUT:"D:\staffan\Documents\Visual Studio 2005\Projects\CeT3x\test\AllUnitTests\Release\test_XOLibrary.exe" /NOLOGO /LIBPATH:"C:\Program Files (x86)\boost\boost_1_35_0\lib" /MANIFEST /MANIFESTFILE:"Release\test_XOLibrary.exe.intermediate.manifest" /DEBUG /PDB:"d:\staffan\Documents\Visual Studio 2005\Projects\CeT3x\test\AllUnitTests\release\test_XOLibrary.pdb" /OPT:REF /OPT:ICF /LTCG /MACHINE:X86 /ERRORREPORT:PROMPT kernel32.lib user32.lib gdi32.lib winspool.lib comdlg32.lib advapi32.lib shell32.lib ole32.lib oleaut32.lib uuid.lib odbc32.lib odbccp32.lib Any thoughts anyone? Rgds Staffan Tj. Staffan Tjernström | Sr. Software Engineer - Market Data Systems | www.esignal.com <http://www.esignal.com/> eSignal, Interactive Data Desktop Solutions (NYSE: IDC) 3955 Point Eden Way | Hayward, CA 94545-3720 | 510.723.1723 | FAX: 510.266.6100 DISCLAIMER: This message (including any files transmitted with it) may contain confidential and / or proprietary information, is the property of Interactive Data Corporation and / or its subsidiaries and is directed only to the addressee(s). If you are not the designated recipient or have reason to believe you received this message in error, please delete this message from your system and notify the sender immediately. An unintended recipient's disclosure, copying, distribution or use of this message, or any attachments, is prohibited and may be unlawful.

1 0

boost.tokenizer (escaped_list_separator) problem
by Aljaz 15 Jul '08

15 Jul '08

Hello Why will the following code make my application crash: int main(){ std::string s = "Field 1 \"putting quotes around fields, allows commas\",Field 3"; boost::escaped_list_separator<char> sep("\"\'", " ,;", "\"\'"); boost::tokenizer<boost::escaped_list_separator<char> > tok(s, sep); for(boost::tokenizer<boost::escaped_list_separator<char> >::iterator beg=tok.begin(); beg!=tok.end();++beg){ std::cout << *beg << "\n"; } } What am I doing wrong? Many thanks in advance! Aljaz

3 3

[boost::interprocess]trouble creating a vector in a shared memory segment
by Eugen Stoianovici 15 Jul '08

15 Jul '08

I'm having some trouble creating a vector in a shared memory segment (compilation errors) The code: int main(){ //Allocator for the event list typedef boost::interprocess::allocator <int,managed_shared_memory::segment_manager> EventAllocator; //the in memory vector of events typedef boost::interprocess::vector<int, EventAllocator> EventList; //is it already created? EventList* pEventList = segment.find<EventList>(EVENTSEGMENT_NAME).first; } The error: d:\boost_1_35_0\boost\preprocessor\iteration\detail\local.hpp(37): error C2036: 'EventList *' : unknown size And the trace: DataHandle.cpp(53) : warning C4101: 'ex' : unreferenced local variable d:\boost_1_35_0\boost\interprocess\mem_algo\rbtree_best_fit.hpp(353) : see reference to class template instantiation 'boost::interprocess::rbtree_best_fit<MutexFamily,VoidPointer,MemAlignment>' being compiled d:\boost_1_35_0\boost\preprocessor\iteration\detail\local.hpp(37) : error C2036: 'EventList *' : unknown size d:\boost_1_35_0\boost\preprocessor\iteration\detail\local.hpp(37) : see reference to class template instantiation 'boost::interprocess::vector<T,A>' being compiled with [ T=int, A=EventAllocator ] d:\boost_1_35_0\boost\preprocessor\iteration\detail\local.hpp(37) : while compiling class-template member function 'void boost::interprocess::detail::Ctor1Arg<T,is_iterator,P0>::construct_n(void *,size_t,size_t &)' with [ T=EventList, is_iterator=false, P0=EventAllocator ] d:\boost_1_35_0\boost\preprocessor\iteration\detail\local.hpp(37) : see reference to class template instantiation 'boost::interprocess::detail::Ctor1Arg<T,is_iterator,P0>' being compiled with [ T=EventList, is_iterator=false, P0=EventAllocator ] DataHandle.cpp(51) : see reference to function template instantiation 'T *boost::interprocess::detail::named_proxy<SegmentManager,T,is_iterator>::operator ()<EventAllocator>(const P0 &) const' being compiled with [ T=EventList, SegmentManager=boost::interprocess::segment_manager<char,boost::interprocess::rbtree_best_fit<boost::interprocess::mutex_family>,boost::interprocess::iset_index>, is_iterator=false, P0=EventAllocator ]

2 2

MPL apply on some exotic system
by Joel Falcou 15 Jul '08

15 Jul '08

I'm compiling some code on the Cell processor main CPU using its dedicated gcc. The code is calling mpl::apply1 onto a custom meta-function. The compiler -v is : ppu-g++ -v Using built-in specs. Target: ppu Configured with: ../toolchain/gcc/configure --prefix=/usr --mandir=/usr/share/man --infodir=/usr/share/info --with-as=/usr/bin/ppu-as --with-ld=/usr/bin/ppu-ld --enable-threads --with-system-zlib --disable-checking --enable-__cxa_atexit --disable-libunwind-exceptions --enable-languages=c,c++,fortran,ada --disable-nls --enable-clocale=gnu --enable-version-specific-runtime-libs --with-long-double-128 --program-prefix=ppu- --disable-bootstrap --host=ppu --build=powerpc64-unknown-linux-gnu --target=ppu Thread model: posix gcc version 4.1.1 The code : template<class T> struct scalar_mode { typedef mpl::false_ status; struct base { template<class Type> struct apply : mpl::if_< is_same<T,none>, Type, T> {}; }; }; typedef mpl::apply1< scalar_mode<none>::base, float>::type type; The compiler errors are : ../../../lib/include/boost-1_35/boost/mpl/aux_/preprocessed/gcc/template_arity.hpp: In instantiation of 'const int boost::mpl::aux::template_arity_impl<nt2::scalar_mode<nt2::none>::base, 1>::value': ../../../lib/include/boost-1_35/boost/mpl/aux_/preprocessed/gcc/template_arity.hpp:93: instantiated from 'const int boost::mpl::aux::template_arity<nt2::scalar_mode<nt2::none>::base>::value' ../../../lib/include/boost-1_35/boost/mpl/aux_/preprocessed/gcc/template_arity.hpp:98: instantiated from 'boost::mpl::aux::template_arity<nt2::scalar_mode<nt2::none>::base>' ../../../lib/include/boost-1_35/boost/mpl/aux_/preprocessed/gcc/apply.hpp:48: instantiated from 'boost::mpl::apply1<nt2::scalar_mode<nt2::none>::base, float>' essais/simd.cpp:13: instantiated from here ../../../lib/include/boost-1_35/boost/mpl/aux_/preprocessed/gcc/template_arity.hpp:85: error: call of overloaded 'arity_helper(boost::mpl::aux::type_wrapper<nt2::scalar_mode<nt2::none>::base>, boost::mpl::aux::arity_tag<1>)' is ambiguous ../../../lib/include/boost-1_35/boost/mpl/aux_/preprocessed/gcc/template_arity.hpp:30: note: candidates are: char (& boost::mpl::aux::arity_helper(...))[1] ../../../lib/include/boost-1_35/boost/mpl/aux_/preprocessed/gcc/template_arity.hpp:37: note: char (& boost::mpl::aux::arity_helper(boost::mpl::aux::type_wrapper<F<T1> >, boost::mpl::aux::arity_tag<1>))[2] [with F = nt2::scalar_mode<T>::base, T1 = nt2::none] ../../../lib/include/boost-1_35/boost/mpl/aux_/preprocessed/gcc/template_arity.hpp:37: note: char (& boost::mpl::aux::arity_helper(boost::mpl::aux::type_wrapper<F<T1> >, boost::mpl::aux::arity_tag<1>))[2] [with F = nt2::scalar_mode<T>::base, T1 = nt2::none] I know that "officially" boost is not supported on this kidn of platform but most of my code compiles and works flawlessly except for this. Any hints on what to do to fix this even locally ? Thanks in advance

3 3

Boost/Python: stdin not found.. not sure why it is being looked for
by Martin_Thomas＠McAfee.com 15 Jul '08

15 Jul '08

I am attempting to wrap a C++ library (wrapper included below) and all seems good until I call the processDoc method around which I have put a thin wrapper. I get "RuntimeError: St9exception" so to start trouble-shooting, launched python with strace -e file and then I see the following: Traceback (most recent call last): open("<stdin>", O_RDONLY|O_LARGEFILE) = -1 ENOENT (No such file or directory) open("<stdin>", O_RDONLY|O_LARGEFILE) = -1 ENOENT (No such file or directory) open("/usr/lib/python2.4/site-packages/setuptools-0.6c8-py2.4.egg/<stdin >", O_RDONLY|O_LARGEFILE) = -1 ENOENT (No such file or directory) open("/usr/lib/python2.4/site-packages/lxml-2.1beta1-py2.4-linux-i686.eg g/<stdin>", O_RDONLY|O_LARGEFILE) = -1 ENOENT (No such file or directory) open("/usr/lib/python2.4/site-packages/pydot-1.0.2-py2.4.egg/<stdin>", O_RDONLY|O_LARGEFILE) = -1 ENOTDIR (Not a directory) open("/usr/lib/python2.4/site-packages/pyparsing-1.5.0-py2.4.egg/<stdin> ", O_RDONLY|O_LARGEFILE) = -1 ENOENT (No such file or directory) open("/usr/lib/python2.4/site-packages/pygccxml-0.9.5-py2.4.egg/<stdin>" , O_RDONLY|O_LARGEFILE) = -1 ENOTDIR (Not a directory) open("/usr/lib/python24.zip/<stdin>", O_RDONLY|O_LARGEFILE) = -1 ENOENT (No such file or directory) open("/usr/lib/python2.4/<stdin>", O_RDONLY|O_LARGEFILE) = -1 ENOENT (No such file or directory) open("/usr/lib/python2.4/plat-linux2/<stdin>", O_RDONLY|O_LARGEFILE) = -1 ENOENT (No such file or directory) open("/usr/lib/python2.4/lib-tk/<stdin>", O_RDONLY|O_LARGEFILE) = -1 ENOENT (No such file or directory) open("/usr/lib/python2.4/lib-dynload/<stdin>", O_RDONLY|O_LARGEFILE) = -1 ENOENT (No such file or directory) open("/usr/lib/python2.4/site-packages/<stdin>", O_RDONLY|O_LARGEFILE) = -1 ENOENT (No such file or directory) open("/usr/lib/python2.4/site-packages/Numeric/<stdin>", O_RDONLY|O_LARGEFILE) = -1 ENOENT (No such file or directory) open("/usr/lib/python2.4/site-packages/gtk-2.0/<stdin>", O_RDONLY|O_LARGEFILE) = -1 ENOENT (No such file or directory) File "<stdin>", line 1, in ? Any clues ? Thanks // Martin ProcessorWrapper.cpp: #include "Processor.h" #include <boost/python.hpp> using namespace boost::python; using namespace Fiddle; void pDoc(Processor* instance, const std::wstring& inFile, const std::wstring& outFile) { instance->processDoc(inFile, outFile); } BOOST_PYTHON_MODULE(Processor) { class_<Processor, boost::noncopyable>("Proc", no_init) .def("instance", &Processor::Instance, return_value_policy<reference_existing_object>()) .staticmethod("instance") .def("setSchemaDir", &Processor::setSchemaDir) .def("getSchemaDir", &Processor::getSchemaDir) .def("processDoc", pDoc) ; }

1 0

ip::tcp::socket - How to check that a connection is still alive
by Jean-Sebastien Stoezel 15 Jul '08

15 Jul '08

Hello, What's the usual way of checking that a TCP/IP connection initiated on a ip::tcp::socket is still alive? Thanks, Jean

2 1

How to I specify boost::fusion::vector<something> as the type of an argument?
by Peng Yu 15 Jul '08

15 Jul '08

Hi, template <typename T, typename R> class A; I would like a template function, whose argument should be of the type like. boost::fusion::vector<A<T, R1>, A<T, R2>,...> Basically, it is a vector of A. The first template arguments of A's should be the same, but the second ones are not necessarily the same. Would you please let me know how to do this? Thanks, Peng

2 1

[graph] edge index in properties
by Mojmir Svoboda 15 Jul '08

15 Jul '08

hello, i tried to implement something like gps with usage of bgl, but i made a mistake - i am storing graph::traits<>::vertex_desctriptor and graph_traits<>::edge_descritor in vertex/edge properties as a shortcut between data and the graph. this brings a heavy dependency of the property data on the chosen graph and i am wondering how to assign some numeric vertex/edge index into these properties (because edge_descriptor is in fact pair of vertices etc). my idea is to have adjacency list that is converted into compressed sparse row graph afterwards. btw: do you use pointers in your V/E properties? i tried to and now i quite regret it because of the serialization issues. many thanks for your attention, mojmir

1 0

Building Boost on 64bit with -fPIC
by Puneet Jethalia 15 Jul '08

15 Jul '08

HI, I need to link in Boost into my shared library application. This works fine for 32bit, but is not working for 64bit. It appears that Boost is not getting built properly with -fPIC on 64bit and leads to following errors: /usr/bin/ld: vpi2.o: relocation R_X86_64_32 can not be used when making a shared object; recompile with -fPIC 'vpi2.o' has the boost .a files included within it using 'ld -r' Can someone help identify if this is a boost problem or gcc problem? I am using Boost 1.34.1, gcc 3.2.3 on RHEL4. Please let me know if any other detail is needed. Any help is highly appreciated. thanks Puneet

1 0

which regex was matched
by Peter Boot 15 Jul '08

15 Jul '08

Is there functionality included in boost::xpressive::regex_search(...) that would allow you to know which of an alternated group of expressions e.g. "aaa|bbb|ccc" matched a substring in the input ?

2 2

Boost with Darwin (Mac) gcc 4.0.1
by Mathieu Peyréga 14 Jul '08

14 Jul '08

Hello, I have an issue with i686-apple-darwin9-gcc-4.0.1 (GCC) 4.0.1 (Apple Inc. build 5465) it seem to appear with the keyword "check" for example on files : detail/is_dereferencable.hpp detail/is_incrementable.hpp bzip2.hpp I have some compilation like : "error: declaration does not declare anything" wich disappears if i rename the check in both first files to check_ (everywhere in the file) and check_error in the third one... (and of course in the bzip2.cpp file too then) It does not happen on the very same code with gcc-4.3.0 (mingw version) nor vc8.0 does this renaming have possible side effects I have not seen (my code compiles, links, and seems to run with the modification...) Any explainations for this ? regards, Mathieu -- http://www.incub.net/

3 5

__declspec(dllexport) cannot be used with /clr:pure or /clr:safe
by Kofi Deh 14 Jul '08

14 Jul '08

I am trying to integrate a command-line program, that uses the BOOST UBLAS library, with a Windows Forms GUI. I get the following error when I attempt to compile the code: Error 1 error C3389: __declspec(dllexport) cannot be used with /clr:pure or /clr:safe C:\Program Files\boost\boost_1_35_0\boost\serialization\void_cast_fwd.hpp 26 Error 2 error C3395: 'boost::serialization::void_cast_register' : __declspec(dllexport) cannot be applied to a function with the __clrcall calling convention C:\Program Files\boost\boost_1_35_0\boost\serialization\void_cast_fwd.hpp 30 I am not linking the program to the Boost Serialization library. I realize there have already been discussions on the use of BOOST_DLLEXPORT in the serialization library. Unfortunately, it was not stated how to solve this problem.

1 0

[preprocessor] Comparing strings in BOOST_PP_NOT_EQUAL
by Maximilian Matthe 14 Jul '08

14 Jul '08

Hi! I'm developing with preprocessor a class generator, that generates the wrapper classes needed by boost.python when you want to derive in python from polymorph c++-base classes. I would need a Macro that expands to "return" if its argument is not void. I tried the following, but that obiously does not work: #define Return(type) \ BOOST_PP_EXPR_IF(BOOST_PP_NOT_EQUAL(type, void), return) I made a workaround by adding an integer value to my Array that defines the methods signature. But that is redundant. Is there a way to achieve the behaviour above?

2 2

[preprocessor] Comparing strings in BOOST_PP_NOT_EQUAL
by Maximilian Matthe 14 Jul '08

14 Jul '08

Hi! I'm developing with preprocessor a class generator, that generates the wrapper classes needed by boost.python when you want to derive in python from polymorph c++-base classes. I would need a Macro that expands to "return" if its argument is not void. I tried the following, but that obiously does not work: #define Return(type) \ BOOST_PP_EXPR_IF(BOOST_PP_NOT_EQUAL(type, void), return) I made a workaround by adding an integer value to my Array that defines the methods signature. But that is redundant. Is there a way to achieve the behaviour above?

1 0

typo in http://www.boost.org/doc/libs/1_35_0/libs/filesystem/doc/index.htm
by Faheem Mitha 14 Jul '08

14 Jul '08

Hi, the changelog for 1.34 refers to the function is_file. However, there is no such function. Take care, Faheem. ************************************************************************ Version 1.34.0 * Internationalization, provided by class templates basic_path, basic_filesystem_error, and basic_directory_iterator. * Simplification of the path interface, including elimination of distinction between native and generic formats, and separation of name checking functionality from general path functionality. * Rationalization of predicate function design, including the addition of several new functions. * Preservation of existing user code whenever possible. Deprecated features (symbolic_link_exists(), etc.) remain available for legacy code. * Clearer specification, by reference to [POSIX-01], the ISO/IEEE Single Unix Standard, with provisions for Windows and other operating systems. * New functions status, symlink_status, is_file, is_symlink, create_hard_link, create_symlink, path member and non-member swap, path inserter, path extractor, additional path relational and "/" operator overloads, additional path member template functions taking iterator arguments. * More efficient operations when iterating over directories. * Separation within the implementation of the distinction between the native operating system path syntax and API. This is important for CYGWIN users because it allows them to build for either the Windows or POSIX API's. * Numerous small implementation fixes.

1 0

How to use boost's shared_ptr.hpp?
by Meryl Silverburgh 14 Jul '08

14 Jul '08

Hi, I am tying use boost's shared_ptr.hpp, but it gives me the following compile errors: /usr/include/boost/detail/sp_counted_impl.hpp: In member function 'virtual void* boost::detail::sp_counted_impl_pd<P, D>::get_deleter(const std::type_info&)': /usr/include/boost/detail/sp_counted_impl.hpp:150: error: cannot use typeid with -fno-rtti /usr/include/boost/detail/sp_counted_impl.hpp: In member function 'virtual void* boost::detail::sp_counted_impl_pda<P, D, A>::get_deleter(const std::type_info&)': /usr/include/boost/detail/sp_counted_impl.hpp:220: error: cannot use typeid with -fno-rtti I have these in my .h file: using namespace boost; class Rect; typedef shared_ptr<Rect> RectPtr; My question is why I need to enable 'rtti' in order to use boost shared_ptr ? And if there is any other smart pointer library which does not require me to enable rtti ? Thank you.

3 6

[boost::function] Retrieving function objects / Determining function type
by Ruediger Berlich 14 Jul '08

14 Jul '08

Hi there, is it possible to get direct access to a function object encapsulated in boost::function ? And is it possible to determine whether boost::function has indeed stored a function object rather than, say, a function pointer ? Thanks and Best Regards, Ruediger

2 1

multi threaded resource repository
by Michael Marcin 14 Jul '08

14 Jul '08

I'm attempting to use boost thread's synchronization primitives to implement an efficient resource repository. The resource repository identifies immutable resources by an unique id. If the resource is cached it just returns a shared_ptr to the resource. If the resource is not yet loaded it does a potentially expensive construction on the resource stores it in the cache and then returns a shared_ptr to the resource. Resources can potentially request other resources from the repository during their construction so that case must be considered. The single-threaded version of this class is fairly straightforward but I don't have much experience in making things parallel so I'm struggling. I was hoping someone could offer suggestions on how to make an efficient multi threaded implementation. Single threaded version below: class ResourceStream; class Repository { typedef std::map< ResourceIdentifier, boost::weak_ptr<const void> > ResourceCache; public: template< typename Resource > boost::shared_ptr<const Resource> Acquire( ResourceIdentifier id ) { boost::shared_ptr<const Resource> resource; ResourceCache::iterator it = m_resourceCache.find(id); if( it != m_resourceCache.end() ) { boost::weak_ptr<const void> cachedResource = it->second; try { resource = boost::shared_static_cast<const Resource>( cachedResource.lock() ); } catch( boost::bad_weak_ptr& ) { } } if( !resource ) { resource.reset( new Resource( *this, Fetch(id) ) ); m_resourceCache.insert( ResourceCache::value_type( id, resource ) ); } return resource; } private: // magic that retrieve an unprocessed resource based on an id ResourceStream Fetch( ResourceIdentifier id ); ResourceCache m_resourceCache; }; Thanks, Michael Marcin

1 0

Re: [Boost-users] How to build 32-bit Boost libraries on a 64-bit Linux machine using bjam?
by Gehua Yang 14 Jul '08

14 Jul '08

Hi Folks, I am using Boost.Serialization and Boost.Thread in my project. Because current Boost release had problems with GCC 4.3, I had to build my own Boost libraries from a svn checkout. Fortunately, with the help from "Getting Started Guide", I accomplished the mission with ease. The commands I used are: bjam --build-dir=/tmp/build-boost --toolset=gcc --build-type="release <threading>multi <link>static <runtime-link>static" stage bjam --build-dir=/tmp/build-boost --toolset=gcc --build-type="release <threading>multi <link>static <runtime-link>static" --prefix=/software/boost-20080710 install The above produced 64-bit libraries. My question is, how I can build 32-bit Boost libraries on the same machine? Does it take command line variables, such as CXXFLAGS=-m32? Thanks. Gehua Yang

3 2

serialization of boolean values hardcoded to '1' and '0'
by Andrew McDonald 14 Jul '08

14 Jul '08

I was wondering why in the archive library the input and output text streams are hardcoded with the boolalpha bit as unset, e.g. (is >> std::noboolalpha;)? Specifically the code is in the directory: boost_1_35_0/boost/archive/impl/ In basic_text_iprimitive.ipp template<class IStream> BOOST_ARCHIVE_OR_WARCHIVE_DECL(BOOST_PP_EMPTY()) basic_text_iprimitive<IStream>::basic_text_iprimitive( IStream &is_, bool no_codecvt ) : is(is_), flags_saver(is_), precision_saver(is_), archive_locale(NULL), locale_saver(is_) { if(! no_codecvt){ archive_locale.reset( add_facet( std::locale::classic(), new codecvt_null<BOOST_DEDUCED_TYPENAME IStream::char_type> ) ); is.imbue(* archive_locale); } is >> std::noboolalpha; } and basic_text_oprimitive.ipp template<class OStream> BOOST_ARCHIVE_OR_WARCHIVE_DECL(BOOST_PP_EMPTY()) basic_text_oprimitive<OStream>::basic_text_oprimitive( OStream & os_, bool no_codecvt ) : os(os_), flags_saver(os_), precision_saver(os_), archive_locale(NULL), locale_saver(os_) { if(! no_codecvt){ archive_locale.reset( add_facet( std::locale::classic(), new codecvt_null<BOOST_DEDUCED_TYPENAME OStream::char_type> ) ); os.imbue(* archive_locale); } os << std::noboolalpha; } Andrew McDonald Software Engineer "Zufall ist nur der Ausdruck unserer Unfähigkeit, den Dingen auf Grund zu kommen" - Albert Einstein

2 2

Boost Python : Generic Function Call
by William Marié 14 Jul '08

14 Jul '08

Hi, I would like to find a way to call Python functions from C++ in a generic way. With Python you can use PyObject_CallObject( PyObject* pFun, PyObject* pyArg ) where pyArgs is a tuple, this allows the user to genericaly call Python functions assuming that you create the right tuple before calling PyObject_CallObject. With boost python i tried to do the same thing but i can't find a way to do it, using the call function or directly the () operator i always have to pass all the arguments. Is there a easy way to do that using boost python ? -- View this message in context: http://www.nabble.com/Boost-Python-%3A-Generic-Function-Call-tp18431821p184… Sent from the Boost - Users mailing list archive at Nabble.com.

2 1

Making a range from a single element?
by Mathias Gaunard 14 Jul '08

14 Jul '08

Is there an utility within Boost.Range (or Range_ex) that allows making a range of Ts from a single T? It seems quite useful and practical. Basically a std::make_pair(&foo, &foo+1) works but requires foo to be an lvalue.

2 1