February 2007 - Boost-users - lists.preview.boost.org

Boost.Numeric.Interval: Please specify rounding control mechanism.
by Andreas Fabri 09 Feb '07

09 Feb '07

Hello, I use boost_1_33_1 with VC8 on a XP64 Turion system, and get an error message C:/cygwin/usr/include/boost-1_33_1\boost/numeric/interval/hw_rounding.hpp(34) : fatal error C1189: #error : Boost.Numeric.Interval: Please specify rounding control mechanism. What shall I do? Any help would be appreciated. andreas

1 0

[boost-users] [interprocess] Using semaphores to synchronize startup
by Lars Hagström 09 Feb '07

09 Feb '07

Hi, I am trying to use Boost.Interprocess to get N processes (senders) to send data to one receiver through shared memory. The receiver has to be the one that initializes the shared memory, so no senders can be "let in" before the receiver has completed initialization. I can't guarantee the order that the processes are starting in. To do this I have both the receiver and the senders create_or_open a named_semaphore with 0 as initial value. The senders then immediately do a wait on the semaphore, but the receiver goes ahead and initializes the memory and then does a post on the semaphore. This lets the first sender in, which can then initialize itself and then do a post on the semaphore and start running. The second sender is then let in (because of the post) and it can get going and so on. This approach works very well if the semaphore has been manually deleted from /dev/shm/ or c:\temp\boost_interprocess\, but will not work if there is an old semaphore around. I can get around this on Windows by doing a named_semaphore::remove, since that will fail if a process has got it loaded. But this same code will not work on linux, since the semaphore will be unlinked and then all the processes will be able to delete the semaphore and will then be using different semaphores! I am used to semaphores working so that if it is opened when noone has it loaded it will be initialized to some known value, but this does not appear to be the case with Boost.Interprocess semaphores? Is there something obvious that I've missed, or some other way of accomplishing the same thing? Cheers Lars

2 6

Serialization of pointers with custom allocators
by Anders Sundman 09 Feb '07

09 Feb '07

Hi! Is it possible to serialize a class A by pointer using a custom allocator? (Not the custom in-place allocator approach). I'm trying to serialize the (OpenCV) class IplImage. The class is used like this: IplImage* img = cvCreateImage(...); ... cvReleaseImage(&img); I've tried the following: template<class Archive> inline void load_construct_data(Archive & ar, IplImage * img, const unsigned int file_version){ ... IplImage* tmpImg = cvCreateImage(...); memcpy(img, tmpImg, sizeof(IplImage)); } But I get exceptions later, when releasing the loaded image. As far as I can tell, the image class data is identical before and after serialization, so I guess that the cvCreateImage/cvReleaseImage keep track of pointers somehow and things go bad when the ReleaseImage is called by a pointer not returned by the CreateImage function? Can the allocation of the IplImage used as input to load_construct_data be customized somehow? I also tried using an IplImage * & as input to load_construct_data: template<class Archive> inline void load_construct_data(Archive & ar, _IplImage * & img, const unsigned int file_version){ _IplImage* tmpImg = cvCreateImage(...); delete img; img = tmpImg; } But that screws up the (de)serialization. Any suggestions or thoughts anyone? // Anders Sundman

2 5

Re: [Boost-users] [Serialization] Linking Error in the Latest CVS
by Wu Yinghui, Freddie 09 Feb '07

09 Feb '07

> OK - I checked in a small change in Jamfile.v2 of the build directory which > I believe my address this. > > Robert Ramey Thanks, Robert, for checking in the update. But I don't think that fixes the problem. (Indeed, rebuilding Boost with the update still gave me exactly the same errors.) The problems, I think, is that codecvt_null.cpp was compiled, and linked to produce boost_serialization_....dll/.lib. But I found that the codecvt_null class's methods are mysteriously missing from the resulting .dll/.lib. So when boost_wserialization_....dll/.lib links with boost_serialization....lib, all codecvt_null-related symbols are not found. I don't understand why, but turning /VERBOSE on for the linker showed that the linker somehow discarded the codecvt_null-related symbols when producing the boost_serialization....dll. > Wu Yinghui, Freddie wrote: >> Hi all, >> >> I posted a question regarding usage of BJam V2 in building the latest >> Boost checked out from the CVS >> (http://lists.boost.org/boost-users/2007/02/25201.php) But there >> seems to be no interest in explaining that at all so far. >> >> But I've sort of worked out most of the solution so far (though mostly >> hacking around in the .jam files, which I suppose could have been >> done a lot easier by someone more familiar with the building system). >> Still I failed to build the Boost.Serialization library out-of-box. >> >>> My system: Windows XP Pro SP2 >>> Compiler: MSVC 7.1 >> The errors I met was something like this: >> =====snip===== >> Creating library >> bin.v2\libs\serialization\build\msvc-7.1\release\instruction >> -set-pentium4\qt3support-on\stdlib-stlport\threading-multi\boost_wserialization- >> vc71-mt-p-1_35.lib and object >> bin.v2\libs\serialization\build\msvc-7.1\release\i >> nstruction-set-pentium4\qt3support-on\stdlib-stlport\threading-multi\boost_wseri >> alization-vc71-mt-p-1_35.exp >> basic_text_wiprimitive.obj : error LNK2001: unresolved external symbol >> "private: >> virtual enum stlp_std::codecvt_base::result __thiscall >> boost::archive::codecvt_ >> null<wchar_t>::do_in(int &,char const *,char const *,char const * >> &,wchar_t *,wc >> har_t *,wchar_t * &)const " >> (?do_in@?$codecvt_null@_W@archive@boost@@EBE?AW4resu >> lt@codecvt_base@stlp_std@@AAHPBD1AAPBDPA_W3AAPA_W@Z) >> basic_text_woprimitive.obj : error LNK2001: unresolved external symbol >> "private: >> virtual enum stlp_std::codecvt_base::result __thiscall >> boost::archive::codecvt_ >> null<wchar_t>::do_in(int &,char const *,char const *,char const * >> &,wchar_t *,wc >> har_t *,wchar_t * &)const " >> (?do_in@?$codecvt_null@_W@archive@boost@@EBE?AW4resu >> lt@codecvt_base@stlp_std@@AAHPBD1AAPBDPA_W3AAPA_W@Z) >> binary_wiarchive.obj : error LNK2001: unresolved external symbol >> "private: virtu >> al enum stlp_std::codecvt_base::result __thiscall >> boost::archive::codecvt_null<w >> char_t>::do_in(int &,char const *,char const *,char const * &,wchar_t >> *,wchar_t >> *,wchar_t * &)const " >> (?do_in@?$codecvt_null@_W@archive@boost@@EBE?AW4result@cod >> ecvt_base@stlp_std@@AAHPBD1AAPBDPA_W3AAPA_W@Z) >> binary_woarchive.obj : error LNK2001: unresolved external symbol >> "private: virtu >> al enum stlp_std::codecvt_base::result __thiscall >> boost::archive::codecvt_null<w >> char_t>::do_in(int &,char const *,char const *,char const * &,wchar_t >> *,wchar_t >> *,wchar_t * &)const " >> (?do_in@?$codecvt_null@_W@archive@boost@@EBE?AW4result@cod >> ecvt_base@stlp_std@@AAHPBD1AAPBDPA_W3AAPA_W@Z) >> =====snip===== >> >> I tried >> dumpbin /all >> bin.v2\libs\serialization\...\boost_serialization-vc71-mt-p-1_35.lib >> >> And indeed, it seems that codecvt_null object's methods were missing >> from the library. Any hints on how I can fix this problem? >> >> Cheers, >> >> Freddie -- Wu Yinghui, Freddie Senior Software Engineer Advanced Technology Research and Development Group Volume Interactions Pte Ltd 1 Kim Seng Promenade #12-01 Great World City East Tower Singapore 237994 Tel: +65 62226962 (Ext 216) Fax: +65 62226215 Email: yhwu(a)volumeinteractions.com URL: http://www.volumeinteractions.com http://www.dextroscope.com Important: This message is intended for the recipient(s) addressed above. It contains privileged and confidential information. If you are not the intended recipient, please notify the sender immediately by replying to this message and then delete it from your system. You must not read, copy, use, or disseminate this communication in any form. Thank you.

1 0

[Serialization] Linking Error in the Latest CVS
by Wu Yinghui, Freddie 09 Feb '07

09 Feb '07

Hi all, I posted a question regarding usage of BJam V2 in building the latest Boost checked out from the CVS (http://lists.boost.org/boost-users/2007/02/25201.php) But there seems to be no interest in explaining that at all so far. But I've sort of worked out most of the solution so far (though mostly hacking around in the .jam files, which I suppose could have been done a lot easier by someone more familiar with the building system). Still I failed to build the Boost.Serialization library out-of-box. > My system: Windows XP Pro SP2 > Compiler: MSVC 7.1 The errors I met was something like this: =====snip===== Creating library bin.v2\libs\serialization\build\msvc-7.1\release\instruction -set-pentium4\qt3support-on\stdlib-stlport\threading-multi\boost_wserialization- vc71-mt-p-1_35.lib and object bin.v2\libs\serialization\build\msvc-7.1\release\i nstruction-set-pentium4\qt3support-on\stdlib-stlport\threading-multi\boost_wseri alization-vc71-mt-p-1_35.exp basic_text_wiprimitive.obj : error LNK2001: unresolved external symbol "private: virtual enum stlp_std::codecvt_base::result __thiscall boost::archive::codecvt_ null<wchar_t>::do_in(int &,char const *,char const *,char const * &,wchar_t *,wc har_t *,wchar_t * &)const " (?do_in@?$codecvt_null@_W@archive@boost@@EBE?AW4resu lt@codecvt_base@stlp_std@@AAHPBD1AAPBDPA_W3AAPA_W@Z) basic_text_woprimitive.obj : error LNK2001: unresolved external symbol "private: virtual enum stlp_std::codecvt_base::result __thiscall boost::archive::codecvt_ null<wchar_t>::do_in(int &,char const *,char const *,char const * &,wchar_t *,wc har_t *,wchar_t * &)const " (?do_in@?$codecvt_null@_W@archive@boost@@EBE?AW4resu lt@codecvt_base@stlp_std@@AAHPBD1AAPBDPA_W3AAPA_W@Z) binary_wiarchive.obj : error LNK2001: unresolved external symbol "private: virtu al enum stlp_std::codecvt_base::result __thiscall boost::archive::codecvt_null<w char_t>::do_in(int &,char const *,char const *,char const * &,wchar_t *,wchar_t *,wchar_t * &)const " (?do_in@?$codecvt_null@_W@archive@boost@@EBE?AW4result@cod ecvt_base@stlp_std@@AAHPBD1AAPBDPA_W3AAPA_W@Z) binary_woarchive.obj : error LNK2001: unresolved external symbol "private: virtu al enum stlp_std::codecvt_base::result __thiscall boost::archive::codecvt_null<w char_t>::do_in(int &,char const *,char const *,char const * &,wchar_t *,wchar_t *,wchar_t * &)const " (?do_in@?$codecvt_null@_W@archive@boost@@EBE?AW4result@cod ecvt_base@stlp_std@@AAHPBD1AAPBDPA_W3AAPA_W@Z) =====snip===== I tried dumpbin /all bin.v2\libs\serialization\...\boost_serialization-vc71-mt-p-1_35.lib And indeed, it seems that codecvt_null object's methods were missing from the library. Any hints on how I can fix this problem? Cheers, Freddie -- Wu Yinghui, Freddie Senior Software Engineer Advanced Technology Research and Development Group Volume Interactions Pte Ltd 1 Kim Seng Promenade #12-01 Great World City East Tower Singapore 237994 Tel: +65 62226962 (Ext 216) Fax: +65 62226215 Email: yhwu(a)volumeinteractions.com URL: http://www.volumeinteractions.com http://www.dextroscope.com Important: This message is intended for the recipient(s) addressed above. It contains privileged and confidential information. If you are not the intended recipient, please notify the sender immediately by replying to this message and then delete it from your system. You must not read, copy, use, or disseminate this communication in any form. Thank you.

2 1

FW: Question about Boost.Test
by Ranji Raghavan 09 Feb '07

09 Feb '07

Folks - I would appreciate any help for the following question .. _____ From: Ranji Raghavan [mailto:ranji.raghavan@openspirit.com] Sent: Tuesday, February 06, 2007 10:23 AM To: 'boost-test(a)emailaccount.com' Subject: Question about Boost.Test Hello Gennadiy - I have been asked recently to investigate an UNIT testing tool for our C++ framework we have and have taken a quick stab at cppUnit, CxxTest and I am currently looking into Boost.Test. I have a fair bit of experience in JUnit and MbUnit/NUnit that we have used to unit test the JAVA & .NET version of the same application. What we are trying to accomplish (and have accomplished for JAVA & .NET versions at least) is to have all our test cases defined in XML files - this way we have the same inputs for the .NET, JAVA & C++ versions of our application. JUnit & MbUnit have the capability that allows me to feed different data sets (1000's of them in some case) to the same test method. Delving into cppUnit & CxxTest, I could not find any pragmatic way to feed disparate data sets to the same test method - or at least that's what I gather from what little I could decipher. But looking at Boost.Test I did find BOOST_PARAM_CLASS_TEST_CASE( test_class_method_pointer, first_parameter, last_parameter ) method that caught my attention. The question I have for you is this: Can I pass an instance of a class (that contains all the requisite data I parsed from the XML file) to a test method. For example - class my_complex_test { void test_assignment( can this be an instance of a class? ) { ... } }; ... std::list<can this be a list that holds instances of a class> possible_xmlvalues; ts->add( BOOST_PARAM_CLASS_TEST_CASE( &my_complex_test::test_assignment, possible_xmlvalues.begin(), possible_xmlvalues.end() ) ); Any help you can render will be greatly appreciated. Thanks Ranji Raghavan

2 4

[Serialization] Problems Getting Started
by Jeffrey Brent McBeth 09 Feb '07

09 Feb '07

It seems that most people are getting past this, but I have been unable to get serialization to get to work on even simple classes. A search of the last three months' archives and google didn't turn up anything obvious, so I am relying on your good humor. When I try to compile a (I think simple) class like #include <fstream> #include <boost/serialization/string.hpp> #include <boost/archive/text_oarchive.hpp> #include <boost/archive/text_iarchive.hpp> class Card { public: Card():name("Icky"){} private: std::string name; friend class boost::serialization::access; template <class Archive> void serialize(Archive &ar, const unsigned int version) { ar & name; } }; int main(int argc, char *argv[]) { const Card Howdy; std::ofstream ofs("Foo"); boost::archive::text_oarchive oa(ofs); oa << Howdy; } I get myriad errors mcbeth@sirocco ~/tmp/blah $ g++ blah.cpp -lboost_serialization /usr/include/boost/archive/detail/oserializer.hpp: In constructor 'boost::archive::detail::oserializer<Archive, T>::oserializer() [with Archive = boost::archive::text_oarchive, T = Card]': /usr/include/boost/archive/detail/oserializer.hpp:135: instantiated from 'static boost::archive::detail::oserializer<Archive, T>& boost::archive::detail::oserializer<Archive, T>::instantiate() [with Archive = boost::archive::text_oarchive, T = Card]' /usr/include/boost/archive/detail/oserializer.hpp:264: instantiated from 'static void boost::archive::detail::save_non_pointer_type<Archive, T>::save_standard::invoke(Archive&, const T&) [with Archive = boost::archive::text_oarchive, T = Card]' /usr/include/boost/archive/detail/oserializer.hpp:322: instantiated from 'static void boost::archive::detail::save_non_pointer_type<Archive, T>::invoke(Archive&, const T&) [with Archive = boost::archive::text_oarchive, T = Card]' /usr/include/boost/archive/detail/oserializer.hpp:536: instantiated from 'void boost::archive::save(Archive&, const T&) [with Archive = boost::archive::text_oarchive, T = Card]' /usr/include/boost/archive/basic_text_oarchive.hpp:78: instantiated from 'void boost::archive::basic_text_oarchive<Archive>::save_override(T&, int) [with T = const Card, Archive = boost::archive::text_oarchive]' /usr/include/boost/archive/detail/interface_oarchive.hpp:78: instantiated from 'Archive& boost::archive::detail::interface_oarchive<Archive>::operator<<(T&) [with T = const Card, Archive = boost::archive::text_oarchive]' blah.cpp:26: instantiated from here /usr/include/boost/archive/detail/oserializer.hpp:107: error: incomplete type 'boost::serialization::extended_type_info_null<Card>' used in nested name specifier /usr/include/boost/archive/detail/oserializer.hpp: In member function 'bool boost::archive::detail::oserializer<Archive, T>::is_polymorphic() const [with Archive = boost::archive::text_oarchive, T = Card]': blah.cpp:26: instantiated from here /usr/include/boost/archive/detail/oserializer.hpp:131: error: invalid use of undefined type 'struct boost::serialization::extended_type_info_null<Card>' /usr/include/boost/serialization/type_info_implementation.hpp:26: error: declaration of 'struct boost::serialization::extended_type_info_null<Card>' /usr/include/boost/archive/detail/oserializer.hpp:132: error: invalid use of undefined type 'struct boost::serialization::extended_type_info_null<Card>' /usr/include/boost/serialization/type_info_implementation.hpp:26: error: declaration of 'struct boost::serialization::extended_type_info_null<Card>' The relevant bit seems to be that extended_type_info_null<Card> is an incomplete type, which would point to me missing an important header or something. Relevant details Fails on (with slightly different wording but same errors) Gentoo Linux, g++-4.1.1, Boost 1.33.1 Gentoo Linux, g++-3.4.6, Boost 1.33.1 Max OSX (Tiger), g++-4.0.1, Boost 1.33.1 Max OSX (Tiger), g++-3.3, Boost 1.33.1 If needed, I can try it out on a WinXP box with VS2003 in about 7 hours. Thanks again for your patience Jeff -- ---------------------------------------------------------------------------- Computer Science is as much about computers as astronomy is about telescopes -- Edsger Wybe Dijkstra (1930-2002) ----------------------------------------------------------------------------

3 5

container traits information
by abir basak 08 Feb '07

08 Feb '07

Hi, Is there any boost utility which statically determines if a template parameter is const or not? ( MPL library ?) I need to design something like boost::sub_range, but little different (It will iterate over arbitrary sequence stored in a vector form, rather than a sequence stored in pair form). like, template<typename C> class index_range{ private: C* container_; std::vector<size_type> index_; }; container & index will be passed through constructor. Now if C is const container it will return const_reference etc. Thus what is needed is something like, container_traits<const std::vector>::iterator is a const_iterator while container_traits<std::vector>::iterator is a iterator etc. Thank you for any kind of help. -- Abir Basak, Member IEEE Software Engineer, Read Ink Technologies B. Tech, IIT Kharagpur email: abir(a)abirbasak.com homepage: www.abirbasak.com

3 3

Re: [Boost-users] [Serialization] Undefined reference to... (header or compilation problem)
by misha＠gildia.pl 08 Feb '07

08 Feb '07

> add the "const" > > Read the "rationale" section of the documentation for an explanation. > > Robert Ramey > > > int main() { > > > > > > > > > > const Vertex v1(1.5, 2.5); // add "const" here > > std::ofstream ofs("myfile.vtx"); > > > > boost::archive::text_oarchive oa(ofs); > > > > oa << v1; > > ofs.close(); > > > > > > > > return 0; > > } > > I'm sorry, "const" did not help. I'm almost positive it's missing libs or something similar, cause I get same errors compiling "bus stops" demo (demo.cpp) from boost tutorial. Thx for your time ! Regards Misha

2 1

Building on Windows
by Stephen Torri 08 Feb '07

08 Feb '07

Does someone have a tutorial or instructions that step-by-step telling me how to compile boost for Windows XP? I have Visual Studio 2005 installed. Stephen

3 2

[BGL] dijkstra visitor for examine edge event
by Alejandro Marcos Aragón 08 Feb '07

08 Feb '07

Hi everyone, I am trying to store all the vertices with their respective distances from a source within the Dijkstra shortest path algorithm. For that I plan to use a std::vector<std::queue > >. I was able to have the visitor working but I don't know how to access to those values that are computed automatically by the algorithm (d_v, d_u, for example). The code I wrote is as follows: template <class Tag> struct vis : public default_dijkstra_visitor { vis() { } template <class Edge, class Graph> void edge_relaxed(Edge e, Graph& g){ cout<<"VISITOR!!!"<<endl; cout<<"Edge -> "<<e<<endl; } private: std::vector<std::queue<Vertex> > storage; }; template <class Tag> vis<Tag> target_visit(Tag) { return vis<Tag>(); } void Dijkstra() { // maps needed by the algorithm IndexMap indexmap = get(vertex_index, *gPtr); WeightMap weightmap = get(edge_weight, *gPtr); // define containers for the predecessor vertices and // for distances std::vector<Vertex> pred(num_vertices(*gPtr)); std::vector<double> dist(num_vertices(*gPtr)); // vertex to compute Dijkstra's algorithm on Vertex source = vertex(get_source(), *gPtr); // Dijkstra's algorithm dijkstra_shortest_paths(*gPtr, source, predecessor_map(&pred[0]).distance_map(&dist[0]).visitor(target_visit(on_examine_edge()))); ... Any help will be appreciated. aaragon

1 0

serialization question
by Aljaz 07 Feb '07

07 Feb '07

Hello.. I have a derived class serialization: for instance: class client_request_base { public: client_request_base() : _type( 0 ) {} private: friend boost::serialization::access; template< class ARCHIVE > void serialize( ARCHIVE& ar, const unsigned int version ) { ar & _type; } unsigned int _type; }; class calc_skip_images_request : public client_request_base { public: calc_skip_images_request() : client_request_base( 1 ), _images_client( 0 ) {} private: friend boost::serialization::access; template< class ARCHIVE > void serialize( ARCHIVE& ar, const unsigned int version ) { ar & boost::serialization::base_object<client_request_base>( *this ); ar & _images_client; } unsigned int _images_client; }; Now I wonder.. If I want to use serialization do I have do EXPORT second class? Or set first class to abstract? Like: BOOST_IS_ABSTRACT( client_request_base ) BOOST_CLASS_EXPORT(calc_skip_images_request) Im confused which of these two things is needed? When should I use export and when is abstract macro?

3 5

[signals] Automatic connection management and MT
by Timmo Stange 07 Feb '07

07 Feb '07

I'm working on a preliminary solution for an interface-compatible but thread-safe implementation of Signals for my project so that I will be able to switch to Boost.Signals should Doug Gregor find the time to add thread-safety there. While deciding what parts to include and which to leave out, I noticed that the automatic connection management is probably not usable in a multi-threaded context in the current form. Using a base class destructor (signals::trackable) to monitor an observer's lifetime is bound to result in race condition trouble with the order of destruction in a class hierarchy. Has that been discussed somewhere and are there any proposed solutions/alternatives? Regards Timmo Stange

7 55

Re: [Boost-users] Boost.Thread contained within a class?
by Sohail Somani 07 Feb '07

07 Feb '07

> -----Original Message----- > From: boost-users-bounces(a)lists.boost.org > [mailto:boost-users-bounces@lists.boost.org] On Behalf Of > Joshua ChaitinPollak > > Hello, > > I have a question about Boost.Thread. I am trying to create a class > which upon instantiation creates a thread based upon itself, > and then > on destruction, stops the thread and cleans up after itself. > A simple > example is below. Is this possible, and if not, what alternatives do > I have? It should be possible the way you've outlined it except for the fact that you try to initialize a pointer with an unrelated type and also never initialize the background thread. Are there specific problems that you are having?

1 0

using filesystem on Solaris 9 and 10
by Polder, Matthew J 07 Feb '07

07 Feb '07

Hi, I'm using the boost library filesystem on Windows XP and Solaris 9 and 10. I haven't had any problems with Windows so far. When linking my executable on Solaris 9, the linker warned me that the filesystem library I linked wanted to be linked with libm.so.2, which isn't on my Solaris 9 system. The executable still runs but when I use filesystem to list the files in a directory it doesn't list all of them. I was able to link on Solaris 10, but there I am getting a core dump when I run. Has anyone ever had similar problems on Solaris? thanks, matthew

1 0

Boost.Thread contained within a class?
by Joshua ChaitinPollak 07 Feb '07

07 Feb '07

Hello, I have a question about Boost.Thread. I am trying to create a class which upon instantiation creates a thread based upon itself, and then on destruction, stops the thread and cleans up after itself. A simple example is below. Is this possible, and if not, what alternatives do I have? The reason for this is to create a single external API (actually defined in a base class), and still have a background thread (hidden from the API) which can share that object's internal member variables. I'm looking to do something like this: class my_handler { public: boost::thread* background_thread; my_hander() : background_thread(NULL) { background_thread(*this); } ~my_hander() { if (background_thread != NULL) { running = false; background_thread.join(); } operator()() { while(running) { // do something } } }; -Thanks, Josh -- Joshua ChaitinPollak Software Engineer Kiva Systems

1 0

[thread_safe_signals] postconstructors and predestructors
by Frank Mori Hess 07 Feb '07

07 Feb '07

> Actually, I've realized there's no reason it has to be tied so closely to > enable_shared_from_this. There could be a completely independent base > class called boost::postconstructible which is treated similarly to > enable_shared_from_this in shared_ptr.hpp. Similarly, there could be a > boost::predestructible base class that gets its hook called before the > shared_ptr deletes its pointer (perhaps only in the default deleter). I > believe the hooks would compile to nothing if the shared_ptr isn't > holding a pointer derived from one of these classes, just like (I assume) > they do for enable_shared_from_this. > > > I think you're better off writing a specialized smart pointer if you > > want that kind of functionality. > > I'm no so keen on a specialized smart pointer, for reasons that you've > already brought up earlier. But I suppose that's what I'll do if it comes > to it. It shouldn't be too bad to come up with a new name and do a search > and replace on shared_ptr.hpp. > > My point is, enable_shared_from_this was deemed important enough for > special hooks to be put into shared_ptr.hpp for it. So there is > precedent. And postconstructors/predestructors aren't peculiar ideas that > I just came up with myself. I'm sure they could be useful in other > contexts. > Now I realize I don't need any changes to shared_ptr at all, I can do it all in a wrapper function inserted between the raw pointer and shared_ptr. I can't think of a name, so call it ptr_wrapper(): shared_ptr<MyClass>(signalslib::ptr_wrapper(new MyClass)); ptr_wrapper would call postconstruct() if its input pointer is derived from postconstructible, and return a shared_ptr. The deleter for the returned shared pointer would be (by default) customized to call the predestructor if the input pointer is derived from predestructible. A custom deleter could be accepted as a second argument to ptr_wrapper. -- Frank

2 1

[Serialization] Undefined reference to... (header or compilation problem)
by misha＠gildia.pl 07 Feb '07

07 Feb '07

Hello, I wonder if anyone could help me with my problem. I'm trying to serialize my project's data. Beeing new to the boost libs I started with simple tutorial example. Today is the 3rd day I'm unable to launch a serialization of an object of the simpliest Vertex class containing 2 double values. The errors I get follow this patern: [mack/.../]$ g++ -Wall -o main test3.cpp /tmp/ccQkfdAS.o (.gnu.linkonce.t._ZN5boost7archive18text_oarchive_implINS0_13text_oarchiveEED2Ev+0x31): In function `boost::archive::text_oarchive_impl<boost::archive::text_oarchive>::~text_oarchive_impl()': : undefined reference to `boost::archive::basic_text_oprimitive<std::basic_ostream<char, std::char_traits<char> > >::~basic_text_oprimitive()' The problem is in line : boost::archive::text_oarchive oa (ofs); Any ideas? IMHO it's either compilation issue ( should I include any additional libs to cmd line?) or header order/lacking. To be sure I didn't miss anything else let me include the whole code: //CODE #include <iostream> #include <vector> #include <functional> #include <algorithm> #include <fstream> #include <boost/archive/text_oarchive.hpp> #include <boost/archive/text_iarchive.hpp> #include <boost/serialization/vector.hpp> //using namespace std; //using namespace boost::serialization; class Vertex { private: friend class boost::serialization::access; public: double x; double y; Vertex() {} // Serialization requires a default constructor Vertex(double newX, double newY) : x(newX), y(newY) {} double getX() const { return x; } double getY() const { return y; } void dump() { std::cout << x << " " << y << std::endl; } }; BOOST_IS_ABSTRACT(Vertex) namespace boost { namespace archive { template<class Archive> void serialize(Archive & ar, Vertex & vtx, const unsigned int version) { ar & vtx.x; ar & vtx.y; } } // namespace serialization } // namespace boost int main() { Vertex v1(1.5, 2.5); std::ofstream ofs("myfile.vtx"); boost::archive::text_oarchive oa(ofs); oa << v1; ofs.close(); return 0; } // END CODE I use g++ (GCC) 3.4.4 20050721 (Red Hat 3.4.4-2) Fedora Core Boost ver. 1.32 Thanks in advance for any feedback. Regards Misha

2 1

Error from boost
by gopal.reddy＠wipro.com 07 Feb '07

07 Feb '07

Hi All , I am using boost in Opendiameter code using boost libraray, but i am not running into problems! Details: Tool(compiler) : MSVC60 OS : Windows XP Boost Library Version: boost_1_33_1 Error : :\projects\diameter\opendiameter-1.0.7-h\include\framework.h(255) : error C2352: 'boost::shared_ptr<class AAA_GroupedJob>::get' : illegal call of non-static member function d:\projects\diameter\supporting softwares\boost_1_33_1\boost\shared_ptr.hpp(257) : see declaration of 'get' d:\projects\diameter\opendiameter-1.0.7-h\include\framework.h(255) : while compiling class-template member function 'class AAA_GroupedJob &__thiscall AAA_JobHandle<class AAA_GroupedJob>::Job(void)' memory_manager.cxx How it is used in framework.cpp (file) : template <class JOB> class AAA_JobHandle : public boost::shared_ptr<JOB> { public: AAA_JobHandle(JOB* job) : boost::shared_ptr<JOB>(job, AAA_JobDeleter()) {} JOB& Job() { return (JOB&)*boost::shared_ptr<JOB>::get(); } }; I seem to be doing some thing wrong with the above. Any help on how to go about resolving the problem would be appreciated! Thanks, Gopal The information contained in this electronic message and any attachments to this message are intended for the exclusive use of the addressee(s) and may contain proprietary, confidential or privileged information. If you are not the intended recipient, you should not disseminate, distribute or copy this e-mail. Please notify the sender immediately and destroy all copies of this message and any attachments. WARNING: Computer viruses can be transmitted via email. The recipient should check this email and any attachments for the presence of viruses. The company accepts no liability for any damage caused by any virus transmitted by this email. www.wipro.com

2 1

Re: [Boost-users] [statechart] : asynchronous_state_machine - access member functions
by Oliver.Kowalke＠qimonda.com 07 Feb '07

07 Feb '07

Hi Andreas, Your solution will not work in my application. I use an io-framework which has some requirements that could not be fullfilled with the creation process of an asynchronous_state_machine. asynchronous_state_machine is derived from state_machine - does this mean that asynchronous_state_machine provides only a thread-safe access to state_machine (acts as a wrapper)? In my application I've three event sources (io-device, UNIX-signal handler, trigger which periodically produces events). Each event source runs in ist own thread. I could use a state_machine and synchronize the insertion of events (process_event() function) via a mutex. Would asynchronous_state-machine provide a preformance benfit in this case? Would it be an adequat solution to queue the events from the three event sources in a thread-safe queue (producer-consumer pattern)? The threads providing events act as producer threads) and should not be blocked and the processing of the events (calling state_machine::process_event()) will be done snychronous (maybe by one consumer thread). Regards,Oliver

2 3

Problem Using Bjam V2
by Wu Yinghui, Freddie 07 Feb '07

07 Feb '07

Hi all, I've been using bjam v1 in 1.33 to build my Boost, and everything went smoothly. I tried lately to grab the latest Boost from CVS in order to preview its upcoming feature, but I faced problems building it using the new bjam v2. My system: Windows XP Pro SP2 Compiler: MSVC 7.1 First of all, I'd like to build Boost with STLport 5.1.0 (the latest STLport release). In the old bjam, I could use the vc_7_1-stlport toolset to activate STLport. In the new bjam, I found that the <stdlib> option was always set to "native" when I run bjam with --debug-building. (I've enabled "using stlport" in user-config.jam".) So how do I activate STLport? Secondly, I downloaded Zlib and Libbz2 source and wanted to build Boost.IOStream with zip and bzip2 support. In the old bjam, I could set ZLIB_INCLUDE, etc. environment variables and then Boost will find the necessary headers and libraries. In the new bjam, I always get an error complaining zlib.h not found even though I've already set the environment variables properly. How do I go about doing this? Lastly, due to our project requirements, I need to provide some customized compiler options when building Boost. In the old bjam, I could use -sBUILD=<cxx-flags>option to do so. But these flags doesn't seem effective with the new bjam (I don't see them when using -d+2 to check the compiler parameters). So how do I go about doing this? Thanks, Freddie -- Wu Yinghui, Freddie Senior Software Engineer Advanced Technology Research and Development Group Volume Interactions Pte Ltd 1 Kim Seng Promenade #12-01 Great World City East Tower Singapore 237994 Tel: +65 62226962 (Ext 216) Fax: +65 62226215 Email: yhwu(a)volumeinteractions.com URL: http://www.volumeinteractions.com http://www.dextroscope.com Important: This message is intended for the recipient(s) addressed above. It contains privileged and confidential information. If you are not the intended recipient, please notify the sender immediately by replying to this message and then delete it from your system. You must not read, copy, use, or disseminate this communication in any form. Thank you.

1 0

get time in different zone
by Fred J. 07 Feb '07

07 Feb '07

Dear all. I am trying to get New York time in time_t epoch now. my PC is set to Sydney time. please help thanks here is my effort fro the last 4 hours. #include <fstream> #include <ios> #include <iostream> #include <string> #include "boost/date_time/local_time/local_time.hpp" #include "boost/date_time/posix_time/posix_time.hpp" #include "boost/date_time/local_time_adjustor.hpp" #include "boost/date_time/c_local_time_adjustor.hpp" using namespace std; int main(){ using namespace boost::gregorian; using namespace boost::local_time; using namespace boost::posix_time; // now time is ptime now = second_clock::local_time(); // current PC time cout << "Your computer time is: " << now << endl; // typedef boost::date_time::c_local_adjustor<ptime> local_adj; // // create a sydney time // ptime sydtime(date(2007,02,07), hours(12)+minutes(9)+seconds(32)); // cout << "Sydney local time is: " << to_simple_string( sydtime ) << endl; // load time zone conversion date form file tz_database tz_db; tz_db.load_from_file("../date_time_zonespec.csv"); time_zone_ptr nyc_tz = tz_db.time_zone_from_region("America/New_York"); time_zone_ptr syd_tz = tz_db.time_zone_from_region("Australia/Sydney"); // create a sydney time from date and time_duration date in_date(2007,02,07); time_duration td(12,25,32); local_date_time sydtime(in_date, td, syd_tz, local_date_time::NOT_DATE_TIME_ON_ERROR); cout << "Sydney time is: " << sydtime << endl; cout << "in UTC it is: " << sydtime.utc_time() << endl; // // convrt sydney to utc // ptime syd_t_utc = local_adj::local_to_utc( sydtime ); // cout << "UTC time is: " << to_simple_string( syd_t_utc ) << endl; // // conver it to new york // time_zone_ptr nyc_tz = tz_db.time_zone_from_region("America/New_York"); // ptime nytime = nyc_tz::utc_to_local( nyc_tz ); // cout << "New York local time is: " << to_simple_string( nyctime ) << endl; } --------------------------------- Any questions? Get answers on any topic at Yahoo! Answers. Try it now.

1 0

Undefined symbol
by Jojy Varghese 07 Feb '07

07 Feb '07

Hi all I am getting a link error while running my application that uses boost regex library. Here is the error message: undefined symbol: _ZN5boost9re_detail12perl_matcherIPKcN4_STL9allocatorINS_9sub_matchIS3_E EEENS_12regex_traitsIcNS_16cpp_regex_traitsIcEEEEE14construct_initERKNS_ 11basic_regexIcSC_EENS_15regex_constants12_match_flagsE Any help is appreciated. thanks in advance Jojy

1 0

[boost::fusion] set_c or vector_c in fusion?
by Ovanes Markarian 07 Feb '07

07 Feb '07

Hello all! Does someone know if the fusion library has some analog to mpl::sec_c or mpl::vector_c? Or is there any way to pass mpl::set_c as a template parameter to fusion::set class? Thanks a lot, Ovanes

2 2

Re: [Boost-users] [date_time] timezone rules for different years
by Stephen Turner 06 Feb '07

06 Feb '07

> -----Original Message----- > From: Stephen Turner [mailto:sturner@clicktracks.com] > Sent: 22 December 2006 16:39 > To: 'boost-users(a)lists.boost.org' > Subject: [date_time] timezone rules for different years > > For my application, I need to be able to get the local time > of any UTC time in the last five years in any location. > > As far as I can see, the date_time library doesn't provide > this. The documentation says that date_time_zonespec.csv is > "based on data found in the zoneinfo datebase", but whereas > the zoneinfo database contains historical information about > daylight saving time changes, date_time_zonespec.csv only > seems to allow you to specify a single rule for each > timezone. And I can't find another mechanism to do it. > > Am I missing something, or is this capability not present? > And if not, is there any plan to add it? Or a workaround? I > anticipate that this will become more of a problem when the > rules for the U.S. change by several weeks next year. > I'm not sure whether this one got lost in the holiday break, or whether there's nobody who's familiar with the internals of this library on this list, but let me try again. I've been trying to work out how to add the capability to have different timezone rules in different years to the library. It seems to me it's not too hard (although I'm nervous of making such a judgement in code I'm not familiar with). I think all that's needed is: 1) In class custom_time_zone, the rules should depend on y, rather than there just being one set of rules. 2) The date_time_zonespec.csv file would have to have extra columns for years a rule applied to. tz_db_base::parse_string would have to parse them, and instead of creating a new timezone for each line, add the rule to the appropriate timezone if it had already been seen once. Is there anything else that would need to be done? Actually, I think the hardest part would be populating the date_time_zonespec.csv file. Is there already a program to generate that file from the zoneinfo files, or was it done semi-by-hand? Thanks, -- Stephen Turner, Cambridge, UK http://homepage.ntlworld.com/adelie/stephen/ "The internet is a reflection of our society. If we do not like what we see, the problem is not to fix the mirror, we have to fix society." (Vint Cerf)

3 2