August 2008 - Boost-users - lists.boost.org

G++ compilation warnings in 1.36 beta
by Robert Mecklenburg 12 Aug '08

12 Aug '08

I'm getting these compilation warnings in serialization /home/mecklen/work/boost-install/include/boost-1_36/boost/serialization/array.hpp: In instantiation of 'void boost::serialization::serialize(Archive&, boost::array<T, N>&, unsigned int) [with Archive = boost::archive::binary_oarchive, T = double, unsigned int N = 3u]': /home/mecklen/work/boost-install/include/boost-1_36/boost/serialization/serialization.hpp:133: instantiated from 'void boost::serialization::serialize_adl(Archive&, T&, unsigned int) [with Archive = boost::archive::binary_oarchive, T = boost::array<double, 3u>]' /home/mecklen/work/boost-install/include/boost-1_36/boost/archive/detail/oserializer.hpp:144: instantiated from 'void boost::archive::detail::oserializer<Archive, T>::save_object_data(boost::archive::detail::basic_oarchive&, const void*) const [with Archive = boost::archive::binary_oarchive, T = boost::array<double, 3u>]' src/LocationHistory.cpp:73: instantiated from here /home/mecklen/work/boost-install/include/boost-1_36/boost/serialization/array.hpp:127: warning: unused parameter 'version' >From this code in include/boost-1_36/boost/serialization/array.hpp: template <class Archive, class T, std::size_t N> void serialize(Archive& ar, boost::array<T,N>& a, const unsigned int version) { ar & make_nvp("elems",a.elems); } Could we comment out the unused variable? Thanks, -- Robert

1 0

Re: [Boost-users] [python] partially resolved: the never ending
by Jean-Francois Bastien 12 Aug '08

12 Aug '08

> 1. With VS2005 it is NOT a GoodIdea(TM) to mix different > types of binaries for one executable, since the (stupid) > linker of vs2005 succeeds without error sometimes creating > executables that fail to be debugged and/or crash at random > points during execution. Summary: VC8 linker sucks in many > aspects, so don't do that. FWIW it is possible to mix debug/release DLLs with Visual Studio, I do it with a lot of DLLs with success. You just have to be very careful about a few things, for example: - Watch out for the one definition rule, especially across DLLs. Most STL types in the standard Visual Studio library don't have the same size in debug as in release! That includes std::string as well as the containers (std::vector and others), and non-STL types with members that exist depending on _DEBUG/NDEBUG. This can be mitigated by using the proprocessor macros _SECURE_SCL=0 and _HAS_ITERATOR_DEBUGGING=0, but this is a tradeoff. Violating that causes hard to debug stack corruptions. - Make sure that allocation/deallocation is symetric. If you do a new/malloc with one runtime and a delete/free with another you're asking for trouble. Don't forget about the global new/delete too! Basically you have to analyze what passes through DLL APIs. JF

1 0

asio udp bad_alloc?
by Lewis, Ambrose J. 12 Aug '08

12 Aug '08

Hi All: I'm trying (somewhat unsuccessfully) to use a udp::socket in my program. My system is running Ubuntu 7.04 and boost 1.35 After a while, the program dies with the following message: terminate called after throwing an instance of 'std::bad_alloc' The line that causes the problem is: (*m_socket).send_to(boost::asio::buffer(xmlData.m_xmlString), m_remoteEndpoint); I think that the client program isn't keeping up and my send buffer is overflowing. Is there any way to confirm this thought? Could I use something like a select statement to see if this UDP socket is ready for writing? Please note that I set the size of the send buffer to 8192: m_socket->set_option(udp::socket::send_buffer_size(8192)); Thanks for the help and/or suggestions! THANXS amb

1 0

FW: Boost Build Variations: One Missing, One Extra!
by Bilokon, Paul 12 Aug '08

12 Aug '08

Forwarding this to Boost.Users (at the moderator's discretion), as this is a Boost usage question, as well as a Boost.Build question. Many thanks, Paul > ______________________________________________ > From: Bilokon, Paul > Sent: Monday, August 11, 2008 4:41 PM > To: 'boost-build(a)lists.boost.org' > Subject: Boost Build Variations: One Missing, One Extra! > > Hi, > > Apologies if the answer to this question is supposed to be obvious. I > am trying to build Boost using the following command: > > "E:\dev\external\boost\boost_1_35_0\bjam\bjam.exe" > --build-type=complete > --build-dir="E:\dev\external\boost\boost_1_35_0\build" > --toolset=msvc-8.0 stage > > Now, as far as I understand, --build-type=complete should give me all > combinations of > > variant=debug,release; threading=single,multi; link=static,shared > > I'm expecting to see 2^3 = 8 combinations in total. As a result I get > this (I do hope the original formatting gets preserved): > > +-- > E:\dev\external\boost\boost_1_35_0\build\boost\bin.v2\libs\date_time\b > uild\msvc-8.0 > | > +-- debug > | | > | +-- link-static > | | | > | | +-- runtime-link-static: > libboost_date_time-vc80-sgd-1_35.lib > | | | | (variant=debug, > threading=single, link=static) > | | | | > | | | +-- threading-multi: > libboost_date_time-vc80-mt-sgd-1_35.lib > | | | (variant=debug, threading=multi, > link=static) > | | | > | | +-- threading-multi: > libboost_date_time-vc80-mt-gd-1_35.lib > | | (variant=debug, threading=multi, > link=static) > | | > | +-- threading-multi: > boost_date_time-vc80-mt-gd-1_35.dll > | (variant=debug, threading=multi, > link=shared) > | > +-- release > | > +-- link-static > | | > | +-- runtime-link-static: > libboost_date_time-vc80-s-1_35.lib > | | | (variant=release, > threading=single, link=static) > | | | > | | +-- threading-multi: > libboost_date_time-vc80-mt-s-1_35.lib > | | (variant=release, > threading=multi, link=static) > | | > | +-- threading-multi: > libboost_date_time-vc80-mt-1_35.lib > | (variant=release, > threading=multi, link=static) > | > +-- threading-multi: boost_date_time-vc80-mt-1_35.dll > (variant=release, > threading=multi, link=shared) > > Indeed, I do get 8 combinations, *but* I get two of > > variant=debug, threading=multi, link=static > > (under different names -- one with -s, the other one without) and I > get two of > > variant=release, threading=multi, link=static > > (under different names -- one with -s, the other one without) but I > never get > > variant=debug, threading=single, link=shared > variant=release, threading=single, link=shared > > Would you please explain to me what's going on? > > Many thanks, > Paul - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - This message is intended only for the personal and confidential use of the designated recipient(s) named above. If you are not the intended recipient of this message you are hereby notified that any review, dissemination, distribution or copying of this message is strictly prohibited. This communication is for information purposes only and should not be regarded as an offer to sell or as a solicitation of an offer to buy any financial product, an official confirmation of any transaction, or as an official statement of Lehman Brothers. Email transmission cannot be guaranteed to be secure or error-free. Therefore, we do not represent that this information is complete or accurate and it should not be relied upon as such. All information is subject to change without notice.

2 2

Multi-index container (Modifying non-unique key in the container)
by Ramesh 12 Aug '08

12 Aug '08

Hi. I am using boost multi-index container for storing objects of type class A, in the class key1 & Key 3 are unique and key2 is non-unique. I try to insert objects of type A and it works fine. No issues. At some point after the insertion, I wish to modify the key2 to 20 in the container. Can I do this with modify? If yes How do I define the iterator (since I need to do a find based on key1 but modify key2 in the container) Here is the code snippet: ----------- class A { private: std::string key1; long key2; long key3; < Other class members > }; struct DataMap { std::string key1; long key2; long key3; A *pObj; }; typedef multi_index_container< DataMap, indexed_by< ordered_unique< tag <Key1>, BOOST_MULTI_INDEX_MEMBER(DataMap,std::string,Key1)>, ordered_non_unique< tag<Key2>,BOOST_MULTI_INDEX_MEMBER(DataMap,long,Key2)>, ordered_unique< tag<Key3>,BOOST_MULTI_INDEX_MEMBER(DataMap,long,Key3)> > > tDMap; class myclass { private: tDMap Map; public: insert(std::string, long , long, A *); }; myclass::insert(std::string key1, long key2, long key3, A *pObj) { } main() { std::string key1 = "Test String"; A abc; abc.key1 = key1; abc.key2 = 10; abc.key3 = 400; insert(key1, key2, key3, &abc); } ------------------- Thanks in advance! Ramesh

1 0

Visual C++ 2005 error C2666 with variants.
by Bill McLean 12 Aug '08

12 Aug '08

I have spent a couple of hours trying to resolve the following error when building our system. We have some code that has been compiling and working with GCC with no issues. However building it with MS VC20005 SP1 creates the following error. We are using boost 1.34.1: Error 2 error C2666: 'boost::variant<T0_,T1,T2,T3,T4,T5,T6,T7>::convert_construct' : 2 overloads have similar conversions c:\projects\ardev\currentdev\externallibs\boost\boost\variant\variant.hpp 1342 I have searched the Boost list archives and found this post for boost 1.33.1: http://lists.boost.org/Archives/boost/2006/03/102375.php The author of this post Peter Schreegle did not receive a reply. He attached the below code and build log. It appears that this an issue for variants under visual studio 2005 SP1. Does anyone know if this is an issue for VS 2008? Is this issue for boost 1.35.0 being built by VS 2005 or VS 2008? #include <boost/variant.hpp> typedef boost::variant<char, int, double> numeric_type_variant; class numeric_type : public numeric_type_variant { public: /// Default constructor. numeric_type() : numeric_type_variant() { } template <typename T> /// Converting constructor. numeric_type(T const& t) : numeric_type_variant(t) { } // some more code }; int main(int argc, char* argv[]) { char ba; int bb; double bc; numeric_type a; numeric_type b(ba); numeric_type c(bb); numeric_type d(bc); a = ba; a = bb; a = bc; return 0; } ************************************************************* This is the build log: ------ Build started: Project: test_variant, Configuration: Debug Win32 ------ Compiling... test_variant.cpp c:\_ngi\toolkits\boost_1_33_1\boost\variant\variant.hpp(1342) : error C2666: 'boost::variant<T0_,T1,T2>::convert_construct' : 2 overloads have similar conversions with [ T0_=char, T1=int, T2=double ] c:\_ngi\toolkits\boost_1_33_1\boost\variant\variant.hpp(1327): could be 'void boost::variant<T0_,T1,T2>::convert_construct<char,int,double,boost::detail:: variant::void_,boost::detail::variant::void_,boost::detail::variant::void_,b oost::detail::variant::void_,boost::detail::variant::void_,boost::detail::va riant::void_,boost::detail::variant::void_,boost::detail::variant::void_,boo st::detail::variant::void_,boost::detail::variant::void_,boost::detail::vari ant::void_,boost::detail::variant::void_,boost::detail::variant::void_,boost ::detail::variant::void_,boost::detail::variant::void_,boost::detail::varian t::void_,boost::detail::variant::void_>(const boost::variant<T0_,T1,T2> &,long)' with [ T0_=char, T1=int, T2=double ] c:\_ngi\toolkits\boost_1_33_1\boost\variant\variant.hpp(1258): or 'void boost::variant<T0_,T1,T2>::convert_construct<const T>(T &,int,boost::mpl::false_)' with [ T0_=char, T1=int, T2=double, T=numeric_type ] while trying to match the argument list '(const numeric_type, long)' c:\_ngi\toolkits\boost_1_33_1\boost\variant\variant.hpp(1580) : see reference to function template instantiation 'boost::variant<T0_,T1,T2>::variant<T>(const T &)' being compiled with [ T0_=char, T1=int, T2=double, T=numeric_type ] c:\_ngi\toolkits\boost_1_33_1\boost\variant\variant.hpp(1590) : see reference to function template instantiation 'void boost::variant<T0_,T1,T2>::assign<T>(const T &)' being compiled with [ T0_=char, T1=int, T2=double, T=numeric_type ] c:\_schrott\test_variant\test_variant\test_variant.cpp(36) : see reference to function template instantiation 'boost::variant<T0_,T1,T2> &boost::variant<T0_,T1,T2>::operator =<numeric_type>(const T &)' being compiled with [ T0_=char, T1=int, T2=double, T=numeric_type ] Build log was saved at "file://c:\_schrott\test_variant\test_variant\Debug\BuildLog.htm" test_variant - 1 error(s), 0 warning(s) ========== Build: 0 succeeded, 1 failed, 0 up-to-date, 0 skipped ==========

2 1

ASIO - How to get mac address from interface
by Jean-Sebastien Stoezel 11 Aug '08

11 Aug '08

Hi, Is there a way to get the MAC address of the interface associated with a TCP socket, using the ASIO library? Thanks, Jean

1 0

Preprocessor problems
by Robert Dailey 11 Aug '08

11 Aug '08

Hi, I'm using Boost.Preprocessor to create specializations of a class I have named InputSource. in the iteration loop, I have the following: # define i BOOST_PP_ITERATION() # define INPUT_repeat(z, n, user) user template<typename t_return, BOOST_PP_ENUM_PARAMS(i, typename t_param)> class InputSource< t_return, BOOST_PP_ENUM_PARAMS(i, t_param) BOOST_PP_COMMA_IF( BOOST_PP_AND(BOOST_PP_NOT_EQUAL(i, 0), BOOST_PP_NOT_EQUAL(i, MAX_PARAMETERS)) ) BOOST_PP_ENUM(BOOST_PP_SUB(MAX_PARAMETERS, i), INPUT_repeat, detail::no_param) > The part that isn't working is the BOOST_PP_COMMA_IF line. When I look at the preprocessor output generated by MSVC9, I get this: template<typename t_return, typename t_param0> class InputSource< t_return, t_param0 BOOST_PP_IIF_BOOST_PP_BOOL_BOOST_PP_BITAND_BOOST_PP_BOOL_BOOST_PP_NOT_EQUAL(1, 0)BOOST_PP_BOOL_BOOST_PP_NOT_EQUAL(1, 5)(BOOST_PP_COMMA, BOOST_PP_EMPTY)() detail::no_param , detail::no_param , detail::no_param , detail::no_param > My iteration limits has been defined as follows: # ifndef MAX_PARAMETERS # define MAX_PARAMETERS 5 # endif # define BOOST_PP_ITERATION_LIMITS (1, MAX_PARAMETERS) I would have expected the preprocessor to show a comma or a blank line, but instead it looks like some macros were not expanded. Any solutions to this problem?

1 2

Concept check vs enable_if
by joel falcou 11 Aug '08

11 Aug '08

Considering the scope of these two libraries, I still wonder which one to use in some cases. For example, let's say I have a set of functions that can only take integers as parameters, what's the correct practice : use enable_if to have SFINAE makes the compiler outputs something like "no such fonction" or have a concept check for Integer and have a concept related error indicating people what happens. Is there any policies or is this a matter of tastes ?

5 16

Re: [Boost-users] weak_ptr for intrusive_ptr
by Alan M. Carroll 11 Aug '08

11 Aug '08

At 09:00 PM 8/10/2008, you wrote: >On Sun, Aug 10, 2008 at 9:47 AM, Alan M. Carroll ><amc(a)network-geographics.com> wrote: >> I greatly prefer intrusive_ptr to shared_ptr, but one drawback of that is there is no equivalent of weak_ptr. I have started work on a class to do that and my biggest question, honestly, was what to name it. > >Have you considered the allocator support in shared_ptr? It makes it >possible to allocate the control block in the same memory block as the >managed object. Between this and the aliasing support in shared_ptr, >I've found intrusive_ptr's use rather limited. I haven't really consider the shared_ptr allocator support. I use primarily intrusive_ptr for historical and implementation reasons. I started using intrusive pointers over a decade ago (my own version at first, later Boost, when it became available). Back then shared_ptr had a much poorer and less usable interface (IIRC, it didn't support casting very well which was a big thing for me). I have long since built support classes to make use of intrusive_ptr very simple, just a 1 line inherit, plus the personal familiarity. Asking why I don't use shared_ptr and its allocator support is effectively asking "why don't you got to a lot of effort to get the same functionality you have now?". Not to disparage shared_ptr, but that's just not an attractive option for me. I might view it differently if I were a newbie, but I'm not, so there it is. Implementing my own intrusive_weak_ptr was less work overall *for me* than switching to shared_ptr. Moreover, shared_ptr doesn't solve the issue that is most important to me, safe conversion to and from raw pointers. This shows up in a couple of places 1) Self registration. 2) Handle from this. 3) Some other very esoteric places that I haven't the energy to go in to and no one else would find interesting. Yes, there are work arounds for shared_ptr, but why not just solve it up front with intrusive_ptr? If, as the flip side of me, you use shared_ptr everywhere else and this doesn't come up much, sure. But my situation is the other way around. Boost takes the right approach, in that it adapts to the peculiarities of the client, it doesn't force the client to adapt to Boost's peculiarities.

1 0