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

link error using boost serialization library
by Y Fang 10 Nov '07

10 Nov '07

Hi, I am trying to use boost serialization library, but run into link errors that I don't know how to resolve. I created a very simple data structure with a 'string' and 'int' in it. Then I try to serialize it to a stringstream and recover from a stringstream. The compile failed during linking. The first few errors are: % make g++ -O0 -g3 -Wall -D_REENTRANT -pthread -lboost_program_options -lboost_filesystem -lboost_iostreams -lboost_thread -lboost_regex -lboost_serialization -lpthread -o a.out main.o SimpleData.o main.o: In function `boost::archive::text_oarchive& boost::smart_cast_impl::reference<boost::archive::text_oarchive&>::polymorphic::cross::cast<boost::archive::detail::basic_oarchive>(boost::archive::detail::basic_oarchive&)': /usr/include/boost/smart_cast.hpp:76: undefined reference to `typeinfo for boost::archive::detail::basic_oarchive' main.o: In function `void boost::archive::save_access::save_primitive<boost::archive::text_oarchive, std::basic_string<char, std::char_traits<char>, std::allocator<char> > >(boost::archive::text_oarchive&, std::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)': /usr/include/boost/archive/detail/oserializer.hpp:91: undefined reference to `boost::archive::detail::basic_oarchive::end_preamble()' /usr/include/boost/archive/detail/oserializer.hpp:92: undefined reference to `boost::archive::text_oarchive_impl<boost::archive::text_oarchive>::save(std::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)' main.o: In function `void boost::archive::text_oarchive_impl<boost::archive::text_oarchive>::save<int>(int const&)': /usr/include/boost/archive/text_oarchive.hpp:53: undefined reference to `boost::archive::basic_text_oarchive<boost::archive::text_oarchive>::newtoken()' main.o: In function `void boost::archive::save_access::save_primitive<boost::archive::text_oarchive, int>(boost::archive::text_oarchive&, int const&)': /usr/include/boost/archive/detail/oserializer.hpp:91: undefined reference to `boost::archive::detail::basic_oarchive::end_preamble()' main.o: In function `void boost::archive::text_oarchive_impl<boost::archive::text_oarchive>::save<boost::archive::class_id_reference_type>(boost::archive::class_id_reference_type const&)': /usr/include/boost/archive/text_oarchive.hpp:53: undefined reference to `boost::archive::basic_text_oarchive<boost::archive::text_oarchive>::newtoken()' ...... I attached the source code of the 4 files I use: main.cpp, SimpleData.h, SimpleData.cpp, and Serialize.cpp. Also, here is the g++ version I use: % g++ -v Using built-in specs. Target: i486-linux-gnu Configured with: ../src/configure -v --enable-languages=c,c++,fortran,objc,obj-c++,treelang --prefix=/usr --enable-shared --with-system-zlib --libexecdir=/usr/lib --without-included-gettext --enable-threads=posix --enable-nls --program-suffix=-4.1 --enable-__cxa_atexit --enable-clocale=gnu --enable-libstdcxx-debug --enable-mpfr --enable-checking=release i486-linux-gnu Thread model: posix gcc version 4.1.2 (Ubuntu 4.1.2-0ubuntu4) Is there something I missed in the include list? I tried '#include <boost/archive/detail/basic_oarchive.hpp>' in Serialize.h file, that would trigue compile error even before linking. Thanks, Yu main.cpp: #include <string> #include <iostream> #include "SimpleData.h" #include "Serialize.h" int main () { // test serialization of data structure SimpleData sd("hello", 1); std::string sdstr = saveData<SimpleData>(sd); std::cout << "Serialized simple data is:\n" << sdstr << std::endl; return 0; } SimpleData.h: #ifndef SIMPLEDATA_H_ #define SIMPLEDATA_H_ #include <string> class SimpleData { public: SimpleData(); SimpleData(std::string id_, int code_); virtual ~SimpleData(); std::string id; int code; }; #endif /*SIMPLEDATA_H_*/ SimpleData.cpp: #include "SimpleData.h" SimpleData::SimpleData () { id = "default"; code = 0; return; } SimpleData::SimpleData (std::string id_, int code_) : id(id_), code(code_) { return; } SimpleData::~SimpleData () { return; } Serialize.h: #ifndef SERIALIZE_H_ #define SERIALIZE_H_ #include <sstream> #include <boost/archive/text_oarchive.hpp> #include <boost/archive/text_iarchive.hpp> #include <boost/serialization/base_object.hpp> #include <boost/serialization/utility.hpp> #include <boost/serialization/string.hpp> #include "SimpleData.h" namespace boost { namespace serialization { // add a serialize function below for each data structure that needs // to be serialized // serialize simple data template<class Archive> void serialize (Archive & _ar, SimpleData & _sd, const unsigned int _version) { _ar & _sd.id; _ar & _sd.code; } } } // serialize data structure and save into a string // note: there must be a serialize function defined // above for the data structure template<class Data> std::string saveData (const Data& _d) { std::ostringstream oss; boost::archive::text_oarchive oa(oss); oa << _d; return oss.str(); } // deserialize and recover data structure from a string template<class Data> Data loadData (const std::string& _s) { std::istringstream iss(_s); boost::archive::text_iarchive ia(iss); Data d; ia >> d; return d; } #endif /*SERIALIZE_H_*/

1 0

[date_time] absolute value for time_duration?
by Michael Fawcett 09 Nov '07

09 Nov '07

What is the best way to ensure that one has a positive time_duration? I'm dealing with hundreds of thousands of data points and need to gather the points that are temporally near a given time (based on a duration). For example, find all points in the collection that are within 1 hour of Feb 2, 5:00pm. I used the following code to verify that simply multiplying by -1 would give the desired result, but it would probably be nicer to provide a std::abs overload. boost::posix_time::ptime lhs1(boost::gregorian::date(2007, 2, 1), boost::posix_time::hours(1)); boost::posix_time::ptime rhs1(boost::gregorian::date(2007, 2, 2), boost::posix_time::hours(1)); boost::posix_time::time_duration elapsed1 = lhs1 - rhs1; elapsed1 *= -1; boost::posix_time::ptime lhs2(boost::gregorian::date(2007, 2, 2), boost::posix_time::hours(1)); boost::posix_time::ptime rhs2(boost::gregorian::date(2007, 2, 1), boost::posix_time::hours(1)); boost::posix_time::time_duration elapsed2 = lhs2 - rhs2; assert(elapsed1 == elapsed2); and the actual code is something like (where duration would be 1 hour in this example): boost::posix_time::time_duration elapsed = pt.timestamp - rhs.timestamp; if (elapsed.is_negative()) elapsed *= -1; return elapsed <= duration; I realize that one could also use std::max and std::min, but that seems a bit inefficient, especially since this is a very performance critical area of our application. Thanks, --Michael Fawcett

2 2

Derive from adjacency_list
by Olivier Tournaire 09 Nov '07

09 Nov '07

Hi all, I want to derive a class from adjacency_list : template <class NodesContainer = boost::vecS, class EdgesContainer = boost::vecS, class Directed = boost::undirectedS, class NodesAttributes = boost::no_property, class EdgesAttributes = boost::no_property > class BGL_Graph : public boost::adjacency_list<NodesContainer, EdgesContainer, Directed, NodesAttributes, EdgesAttributes > { /// ... many typedefs /// ... }; I also have an other derived class from BGL_Graph : class MPPRectangleGraph : public BGL_Graph<boost::vecS,boost::vecS,boost::undirectedS,MPPRectangleGraphNode,Weight> { public: MPPRectangleGraph() {is_init_ = false;}; ~MPPRectangleGraph(); void InitDisplay(std::string name = "MPP", unsigned int width = 200, unsigned int height = 200); void Draw( unsigned int flags = DRAW_NOTHING); void Display( unsigned int flags = DRAW_NOTHING ); protected: private: bool is_init_; IdFenetreCalques window_; IdCalqueVecteur layer_rectangle_; IdCalqueVecteur layer_profiles_; IdCalqueVecteur layer_neighborhood_; IdCalqueVecteur layer_edges_; }; My problem is how can I iterate on the edges in a method of MPPRectangleGraph ? void MPPRectangleGraph::Draw( unsigned int flags ) { // On recupere les differents flags d'affichage bool draw_rect = false, draw_profiles = false, draw_neighborhood = false, draw_edges = false; if ( flags&DRAW_RECTANGLE ) draw_rect = true; if ( flags&DRAW_PROFILES ) draw_profiles = true; if ( flags&DRAW_NEIGHBORHOOD ) draw_neighborhood = true; if ( flags&DRAW_EDGES ) draw_edges = true; // On vide tous les calques layer_rectangle_.Clear(); layer_profiles_.Clear(); layer_neighborhood_.Clear(); layer_edges_.Clear(); // On fait l'affichage en fonction des flags recuperes // Les aretes MPPRectangleGraph::edge_iterator first, last; for(boost::tie(first, last)=edges(this);first!=last;++first) // does not work ! { //std::cout << boost::source(*first, g) << " to " << boost::target(*first, g) << " = "; } } Regards -- Le temps des cerises reviendra. Dans l'immédiat, c'est le temps des noyaux. Courage.

1 1

Invoke Bjam on the Command Line
by chandra_india＠webdunia.com 09 Nov '07

09 Nov '07

2 1

macro redefinition errors
by Vivek Nallur 09 Nov '07

09 Nov '07

Hi, My project uses boost 1.34.1 libraries for asio and also includes windows.h (which in turn includes winsock.h). This results in a macro redefinition error for FD_SETSIZE and then goes on to spew a bucketload of related errors. The easiest way to reproduce this is to have a simple main.cpp with just: #include<boost/asio.hpp> #include<windows.h> int main() { return 0; } Any ideas on what I'm doing wrong? Or what I could try? I've tried defining WIN32_LEAN_AND_MEAN, IPPROTO_V6 in the makefile, but to no avail. I'm using VC 8 on a windows XP. ======================== My comand line looks like this: C:\perforce\some\path\>make main.o -- Compile main.cpp... c:/perforce/toolchain/win32/winsdk-6.1.6000/VC/bin/cl.exe -nologo -Fdc:/perforce/some/path somelib.pdb -Gm -EHsc -c -Zi -TP -GR -D_WIN32 -D_USRDLL -D_WIN32_WINNT -DWINVER=0x0600 -D_WINDLL -D_MBCS -W3 -D_CRT_NO NSTDC_NO_DEPRECATE -D_CRT_SECURE_NO_WARNINGS -DUSERLEVEL -MDd -Od -D_DEBUG -RTC1 -I"C:\perforce\toolchain\win32\boost-1.34.1\include" -Ic:/perforce/toolchain/win32/winsdk-6.1.6000/Include -Ic:/perforce/toolchain/win32/winsdk-6.1.6000/VC/include -Ic:/perforce/toolchain/win32/winsdk-6.1.6000/VC/PlatformSDK/Include And the error message looks like: C:\perforce\toolchain\win32\boost-1.34.1\include\boost/asio/detail/socket_types.hpp(44) : warning C4005: 'FD_SETSIZE' : macro redefinition c:\perforce\toolchain\win32\winsdk-6.1.6000\Include\winsock.h(65) : see previous definition of 'FD_SETSIZE' c:\perforce\toolchain\win32\winsdk-6.1.6000\Include\ws2def.h(91) : warning C4005: 'AF_IPX' : macro redefinition c:\perforce\toolchain\win32\winsdk-6.1.6000\Include\winsock.h(460) : see previous definition of 'AF_IPX' c:\perforce\toolchain\win32\winsdk-6.1.6000\Include\ws2def.h(116) : warning C4005: 'AF_MAX' : macro redefinition c:\perforce\toolchain\win32\winsdk-6.1.6000\Include\winsock.h(479) : see previous definition of 'AF_MAX' c:\perforce\toolchain\win32\winsdk-6.1.6000\Include\ws2def.h(163) : warning C4005: 'SO_DONTLINGER' : macro redefinition c:\perforce\toolchain\win32\winsdk-6.1.6000\Include\winsock.h(402) : see previous definition of 'SO_DONTLINGER' c:\perforce\toolchain\win32\winsdk-6.1.6000\Include\ws2def.h(206) : error C2011: 'sockaddr' : 'struct' type redefinition c:\perforce\toolchain\win32\winsdk-6.1.6000\Include\winsock.h(485) : see declaration of 'sockaddr' c:\perforce\toolchain\win32\winsdk-6.1.6000\Include\ws2def.h(384) : error C2143: syntax error : missing '}' before 'constant' c:\perforce\toolchain\win32\winsdk-6.1.6000\Include\ws2def.h(437) : error C2143: syntax error : missing ';' before '}' c:\perforce\toolchain\win32\winsdk-6.1.6000\Include\ws2def.h(437) :error C4430: missing type specifier - int assumed. Note: C++ does not support default-int ========================

1 0

Re: [Boost-users] [bind] binding templated operator with vs 7.1
by Stjepan Rajko 09 Nov '07

09 Nov '07

On Nov 8, 2007 4:20 PM, MOSS Sebastian <Sebastian.MOSS(a)murex.com> wrote: > Hi, > > I'm having trouble with the following design with vs 7.1 and wondering if anyone could offer a workaround. > > Particularly it seems that the ".template operator" syntax isn't liked. (This works on gcc 4.0.3). Removing the > 'template' keyword from the bind arguments gives an error novel. > I was having a similar issue with msvc8. Not sure what the underlying problem is but I ended up just using the template keyword conditionally for non-MSVC compilers. I.e., something like #include <boost/config.hpp> ... boost::bind(&bound:: #ifndef BOOST_MSVC template #endif operator&<T>, instance, obj); It would probably be cleaner to define something like TEMPLATE_OPERATOR to template for non-msvc and to nothing for msvc, and use that instead. Maybe something like that is already defined somewhere. HTH, Stjepan

2 1

[Boost Users] Serialization -- serializing a map with shared_ptr values
by Oliver Kania 09 Nov '07

09 Nov '07

I have the following two data structures: std::map<const std::string, boost::shared_ptr< C >, jedox::util::CaseInsensitiv > m_LookupName; std::map<unsigned int, boost::shared_ptr< C > > m_LookupId; Before serialization, the shared pointer - values of the maps do point to sth. meaningful. Afterwards, however, they point to data-structures with meaningless content. The keys (strings / ints) are serialized / unserialized correctly. I guess its a general problem with serializing shared pointers contained in a std::map as values !? BTW: the structures the shared pointers point to do have a serialization method themselves and I a assume that it is called when serializing the map. I serialize / unserialize as follows: template<class Archive> void save( Archive &ar, unsigned int version) const{ m_offlineMode = true; bool oldCachedAll = m_CachedAll; m_CachedAll = true; ar & m_CachedAll; ar & m_offlineMode; // do not serialize the lock -- use the default constructor. // we assume that no iterators that lock the cache do exist when serializing. // ar & m_Lock; ar & m_LookupId; // 1st structure ar & m_LookupName; // 2nd structure ar & m_SequenceNumber; //reset old state for the case we want to continue working m_CachedAll = oldCachedAll; m_offlineMode = false; } template<class Archive> void load( Archive &ar, unsigned int version) { ar & m_CachedAll; ar & m_offlineMode; // do not serialize the lock -- use the default constructor. // we assume that no iterators that lock the cache do exist when serializing. // ar & m_Lock; ar & m_LookupId; ar & m_LookupName; ar & m_SequenceNumber; //reset old state for the case we want to continue working } kind regards, Oliver

2 2

Re: [Boost-users] Something simple is wrong I am sure
by Daniel Lord 09 Nov '07

09 Nov '07

> > Daniel Lord wrote: > > >> So how can this be? The build was for Darwin and yet its the wrong > > >> architecture?!?!?! > > >> > > >> I don't understand the error messages frankly. > > > Me neither: but after a quick bit of Googling this appears to be a > common problem if you mix "universal" and "non-universal" binaries. So my > guess is > > that the Boost libraries and your executable were built with different > > architechture options: one universal one not. But I can't tell you which > is > > which ! :-) > > Sure enough none of the libraries are universal binaries and only the dynamic libraries are for the i386 architecture (see lipo output below). $lipo -detailed_info libboost_regex* input file libboost_regex-1_34_1.a is not a fat file input file libboost_regex-1_34_1.dylib is not a fat file input file libboost_regex-d-1_34_1.a is not a fat file input file libboost_regex-d-1_34_1.dylib is not a fat file input file libboost_regex-d.a is not a fat file input file libboost_regex-mt-1_34_1.a is not a fat file input file libboost_regex-mt-1_34_1.dylib is not a fat file input file libboost_regex-mt-d-1_34_1.a is not a fat file input file libboost_regex-mt-d-1_34_1.dylib is not a fat file input file libboost_regex-mt-d.a is not a fat file input file libboost_regex-mt.a is not a fat file input file libboost_regex.a is not a fat file Non-fat file: libboost_regex-1_34_1.a is architecture: ppc7400 Non-fat file: libboost_regex-1_34_1.dylib is architecture: i386 Non-fat file: libboost_regex-d-1_34_1.a is architecture: ppc7400 Non-fat file: libboost_regex-d-1_34_1.dylib is architecture: i386 Non-fat file: libboost_regex-d.a is architecture: ppc7400 Non-fat file: libboost_regex-mt-1_34_1.a is architecture: ppc7400 Non-fat file: libboost_regex-mt-1_34_1.dylib is architecture: i386 Non-fat file: libboost_regex-mt-d-1_34_1.a is architecture: ppc7400 Non-fat file: libboost_regex-mt-d-1_34_1.dylib is architecture: i386 Non-fat file: libboost_regex-mt-d.a is architecture: ppc7400 Non-fat file: libboost_regex-mt.a is architecture: ppc7400 Non-fat file: libboost_regex.a is architecture: ppc7400 I made the assumption that Boost would, as a default, build for the native architecture it was building one since most of the other libraries I use have been doing that for nearly 2 years. My mistake. It is building universal binaries that usually requiring customization of the build. But to build ppc on an intel platform--that was completely unexpected at this late date. Searching the Boost site for the key word "universal binaries" found nothing. Searching for "universal OS X" gave four generic uses of the word "universal" but nothing of help. Anybody have any experience building Boost as a universal library set? If not... It's been over a year since I had to hack a build so I need to refresh my memory. It my well be time to pour another cup of java, take the red pill, and see how deep the rabbiit hole goes.

1 0

Re: [Boost-users] [asio] Problems with Boost.System
by Damjan Miklic 09 Nov '07

09 Nov '07

Thaks for the quick reply. Unfortunately i still can't compile the system library. My default installation doesn't even have a libs\system folder. I have downloaded system-0.2.1.zip from the boost vault and copied the 'system' folder to boost_1_34_1\libs but running bjam --toolset=msvc --with-system results with the same error: wrong library name 'system' in the --with-<library> option. Running bjam in boost\libs\system\build gives a lengthy error message and doesn't seem to build anything. Further assistance would be appreciated. Best regards, Damjan Miklic Simon Pickles wrote: > Hi, I had this problem recently. > > I had to abandon the stage argument and just use: > > bjam --toolset=msvc --with-system > > or alternatively, just build the system part by running bjam in: > > boost\libs\system\build > > like this: > > bjam --toolset=msvc threading=multi link=static > > > Good luck! > > Simon > > > > Damjan Miklic wrote: >> Hi, >> >> I am running into some problems when trying to build a simple Asio >> program (the Timer2 tutorial) with MSVC 8.0. I have installed Boost >> using the installer from Boost Consulting and copied the Asio headers to >> the boost_1_34_1\boost directory. The program compiles (with three >> warnings about deprecated unsafe functions in asio headers) but linking >> fails with the error: >> >> fatal error LNK1104: cannot open file >> 'libboost_system-vc80-mt-sgd-1_34_1.lib' >> >> I have tried building Boost.System with the command: >> >> bjam --with-system stage >> >> according to the instructions found on: >> >> http://asio.sourceforge.net/boost_asio_0_3_8/libs/asio/doc/html/boost_asio/… >> >> but i get the following error: >> error: wrong library name 'system' in the --with-<library> option. >> >> I have searched the Boost website and mailing list archives but couldn't >> find any further information on building the System library. >> >> Any help in resolving this problem will be greatly appreciated. >> >> Best Regards, >> >> Damjan Miklic >>

1 0

Something simple is wrong I am sure
by Daniel Lord 09 Nov '07

09 Nov '07

OS X 10.5 Macbook Pro (Intel Dual Core x86) Boost builds and installs fine (except for Unicode Regex) and recognizes the toolset is Darwin. $./configure -n Building Boost.Jam with toolset darwin... tools/jam/src/bin.macosxx86/bjam -n Detecting Python version... 2.5 -n Detecting Python root... /Library/Frameworks/Python.framework/Versions/2.5 -n Unicode/ICU support for Boost.Regex?... not found. Backing up existing Boost.Build configuration in user-config.jam.4 Generating Boost.Build configuration in user-config.jam... Generating Makefile... and make and install proceeded fine. So I tried the examples on the "Getting Started Page" First one works fine. Second one (using the regex libraries does not): "ld: warning in /usr/local/lib/libboost_regex-mt-d.a, file is not of required architecture Undefined symbols: "boost::match_results<__gnu_cxx::__normal_iterator<char const*, std::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::allocator<boost::sub_match<__gnu_cxx::__normal_iterator<char const*, std::basic_string<char, std::char_traits<char>, std::allocator<char> > > > > >::maybe_assign(boost::match_results<__gnu_cxx::__normal_iterator<char const*, std::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::allocator<boost::sub_match<__gnu_cxx::__normal_iterator<char const*, std::basic_string<char, std::char_traits<char>, std::allocator<char> > > > > > const&)", referenced from: boost::re_detail::perl_matcher<__gnu_cxx::__normal_iterator<char const*, std::basic_string<cha" So how can this be? The build was for Darwin and yet its the wrong architecture?!?!?! I don't understand the error messages frankly.

2 1

Re: [Boost-users] shared_ptr and covariant return types
by lists.boost.org＠xmlizer.biz 09 Nov '07

09 Nov '07

Hi Ovanes Thanks for your reply. Yes, I understand that do_clone is private. I wonder what Peter Dimov's idea is. I believe the point of this free function would be that you would only need to write it once and use it for any class that provides the do_clone() method. Which I'd love to see in action. Or be proven wrong. Thanks Bernhard On Thu, 8 Nov 2007, Ovanes Markarian wrote: > Hi! > > do_clone is really private ;) Your derived class has public clone > member and you should use it from the free function, but no the > do_clone. > > > Regards, > Ovanes > > > > On 8 Nov 2007 14:53:49 +0100, Bernhard Wagner boost > <lists.boost.org(a)xmlizer.biz> wrote: > In this post Peter Dimov shows a solution for the covariant return type > problem when using shared_ptr: > http://lists.boost.org/boost-users/2003/02/2996.php > > I'm interested in the free function solution he provides: > > > [snip] you can use the usual pattern from the no-covariant-return > days: > > > > class B > > { > > virtual shared_ptr<B> do_clone() const = 0; > > }; > > > > class X: public B > > { > > virtual shared_ptr<B> do_clone() const { return shared_ptr<X>(new > > X(*this)); } > > > > public: > > > > shared_ptr<X> clone() const > > { > > shared_ptr<X> px = dynamic_pointer_cast<X>(this->do_clone()); > > assert(px); > > return px; > > } > > }; > > > > It buys you some extra safety, too. > > > > You could also use a free function: > > > > template<class X> shared_ptr<X> clone(X const & x) > > { > > > > shared_ptr<X> px = dynamic_pointer_cast<X>( x.do_clone()); > > assert(px); > > return px; > > } > > > > So, to clone an object using the clone method, I could use: > > X* x; > shared_ptr<X> spX(x = new X()); > shared_ptr<X> spX2 = spX->clone(); > // or: > shared_ptr<X> spX3 = x->clone(); > > But how is the free function meant to be used? > > This provokes the compiler error: '... do_clone() const is private ...': > shared_ptr<X> spX4 = clone(*spX); > > Thank you > Bernhard > > _______________________________________________ > Boost-users mailing list > Boost-users(a)lists.boost.org > http://lists.boost.org/mailman/listinfo.cgi/boost-users > > _______________________________________________ > Boost-users mailing list > Boost-users(a)lists.boost.org > http://lists.boost.org/mailman/listinfo.cgi/boost-users > -- Bernhard Wagner bernhard.wagner(a)iaeth.ch Milchbuckstr. 14 VOICE +41-1-364 50 03 CH-8057 Zurich, Switzerland http://listen.to/bernard

1 0

[bind] binding templated operator with vs 7.1
by MOSS Sebastian 08 Nov '07

08 Nov '07

Hi, I'm having trouble with the following design with vs 7.1 and wondering if anyone could offer a workaround. Particularly it seems that the ".template operator" syntax isn't liked. (This works on gcc 4.0.3). Removing the 'template' keyword from the bind arguments gives an error novel. Any ideas would be warmly welcomed! s #include <boost/bind.hpp> #include <boost/function.hpp> #include <iostream> class tmpl_operator { public: template <typename T> void operator&( const T& rhs ) const { // do stuff std::cout << rhs; } }; template <class T, class bound> class command { public: command( T& obj, bound& instance ) : _instance(instance) { // works with gcc (4.0.3) but not cl (7.1): apply = boost::bind(&bound::template operator&<T>, instance, obj); // doesn't compile in vs: //_instance.template operator&<T>(obj); // this compiles in vs though: // _instance.operator&<T>(obj); } void execute() { apply(); } private: bound _instance; boost::function<void (void)> apply; }; int main() { double dbl = 2.71; tmpl_operator instance; command<double,tmpl_operator> test(dbl,instance); test.execute(); } ******************************* This e-mail contains information for the intended recipient only. It may contain proprietary material or confidential information. If you are not the intended recipient you are not authorised to distribute, copy or use this e-mail or any attachment to it. Murex cannot guarantee that it is virus free and accepts no responsibility for any loss or damage arising from its use. If you have received this e-mail in error please notify immediately the sender and delete the original email received, any attachments and all copies from your system.

1 0

[asio] Problems with Boost.System
by Damjan Miklic 08 Nov '07

08 Nov '07

Hi, I am running into some problems when trying to build a simple Asio program (the Timer2 tutorial) with MSVC 8.0. I have installed Boost using the installer from Boost Consulting and copied the Asio headers to the boost_1_34_1\boost directory. The program compiles (with three warnings about deprecated unsafe functions in asio headers) but linking fails with the error: fatal error LNK1104: cannot open file 'libboost_system-vc80-mt-sgd-1_34_1.lib' I have tried building Boost.System with the command: bjam --with-system stage according to the instructions found on: http://asio.sourceforge.net/boost_asio_0_3_8/libs/asio/doc/html/boost_asio/… but i get the following error: error: wrong library name 'system' in the --with-<library> option. I have searched the Boost website and mailing list archives but couldn't find any further information on building the System library. Any help in resolving this problem will be greatly appreciated. Best Regards, Damjan Miklic

2 1

Boost.preprocessor Minor Update for 1.34.1
by Andry 08 Nov '07

08 Nov '07

Hello boost-users, You can download sources here: https://sourceforge.net/project/showfiles.php?group_id=198908&package_id=24… Main idea of this update is to add new macros BOOST_PP_MACRO_IF/BOOST_PP_MACRO_IIF to use BOOST_PP_IF/BOOST_PP_IIF more complex when i need to pass in multiple arguments like what: /////////////////////////////////////////////////////////////////////////////// //The same as BOOST_PP_IDENTITY, but different #define MY_PP_EVAL1(x1) x1 #define MY_PP_EVAL2(x1,x2) x1,x2 #define MY_PP_EVAL3(x1,x2,x3) x1,x2,x3 #define MY_PP_EVAL4(x1,x2,x3,x4) x1,x2,x3,x4 #define MY_PP_EVAL5(x1,x2,x3,x4,x5) x1,x2,x3,x4,x5 #define MY_PP_EVAL6(x1,x2,x3,x4,x5,x6) x1,x2,x3,x4,x5,x6 #define n 0 //basically is argument of another macro // if n == 0 -> //1,2; // if n != 0 -> //1,2,3; BOOST_PP_MACRO_IF(n, (1,2,3), (1,2), MY_PP_EVAL3, MY_PP_EVAL2); //More complex usage // if n == 0 -> //1,2; // if n == 1 -> //-1,0,1,2,3; // if n > 1 -> //-1,0,1,2,3,4; BOOST_PP_MACRO_IF(n, (-1,0, BOOST_PP_MACRO_IF(BOOST_PP_GREATER(n,1), (1,2,3,4), (1,2,3), MY_PP_EVAL4, MY_PP_EVAL3)), (1,2), BOOST_PP_IF(BOOST_PP_GREATER(n,1), MY_PP_EVAL6, MY_PP_EVAL5), MY_PP_EVAL2); /////////////////////////////////////////////////////////////////////////////// I uses it to properly generate template classes implementations. PS: This is slightly different than BOOST_PP_IF(n, BOOST_PP_IDENTITY(x), BOOST_PP_EMPTY)() cause x can have multiple values and used as an expression to subsequent macro. -- Best regards, Andry mailto:andry@inbox.ru

1 0

[Boost Interprocess] - Shared Memory - Dynamic memory size
by Srinivas Gardas 08 Nov '07

08 Nov '07

Hi, Is there any option to automatically increase the size of the 'Shared Memory' dynamically, in the shared memory objects of Interprocess library? Thanks & Regards, Srinivas Gardas

2 1

[Boost Users] Serialization: problems when splitting serialize into load / save
by Oliver Kania 08 Nov '07

08 Nov '07

Hello out there ! I have an urgent problem concerning boost::serialize. I have split the serialize operation into save / load as described in the tutorial: =================================================== template<class Archive> void save( Archive &ar, unsigned int version) { m_offlineMode = true; // we cannot load more when offline --> always cachedAll bool oldCachedAll = m_CachedAll; m_CachedAll = true; ar & m_CachedAll; ar & m_offlineMode; // do not serialize the lock -- use the default constructor. // we assume that no iterators that lock the cache do exist when serializing. // ar & m_Lock; ar & m_LookupId; ar & m_LookupName; ar & m_SequenceNumber; //reset old state for the case we want to continue working m_CachedAll = oldCachedAll; m_offlineMode = false; } template<class Archive> void load( Archive &ar, unsigned int version) { ar & m_CachedAll; ar & m_offlineMode; // do not serialize the lock -- use the default constructor. // we assume that no iterators that lock the cache do exist when serializing. // ar & m_Lock; ar & m_LookupId; ar & m_LookupName; ar & m_SequenceNumber; //reset old state for the case we want to continue working } BOOST_SERIALIZATION_SPLIT_MEMBER() =============================================================== The class I serialize is part of a deep class hierarchy that I want to serialize as a whole. With the standard serialize() it seems to work, but now the compiler complains ( when I want to perform the actual serialization): error C2662: 'void jedox::palo::AbstractCache<C,Exception>::save<Archive>(Archive &,unsigned int)' : cannot convert 'this' pointer from 'const jedox::palo::AbstractCache<C,Exception>' to 'jedox::palo::AbstractCache<C,Exception> &' AbstractCache is the class containing the above load / save operations kind regards, Oliver

6 7

[reflection] Compilation errors
by Christian Henning 08 Nov '07

08 Nov '07

Hi there, I just pulled the latest reflection code from the sandbox and tried to compile the example. But unfortunately there are compilation errors. For example: c:\boost-sandbox\boost\extension\impl\library_impl.hpp(37) : error C2664: 'LoadLibraryW' : cannot convert parameter 1 from 'const char *' to 'LPCWSTR' Is the relection lib in a usable stage? Thanks, Christian

6 10

shared_ptr and covariant return types
by Bernhard Wagner boost 08 Nov '07

08 Nov '07

In this post Peter Dimov shows a solution for the covariant return type problem when using shared_ptr: http://lists.boost.org/boost-users/2003/02/2996.php I'm interested in the free function solution he provides: > [snip] you can use the usual pattern from the no-covariant-return days: > > class B > { > virtual shared_ptr<B> do_clone() const = 0; > }; > > class X: public B > { > virtual shared_ptr<B> do_clone() const { return shared_ptr<X>(new > X(*this)); } > > public: > > shared_ptr<X> clone() const > { > shared_ptr<X> px = dynamic_pointer_cast<X>(this->do_clone()); > assert(px); > return px; > } > }; > > It buys you some extra safety, too. > > You could also use a free function: > > template<class X> shared_ptr<X> clone(X const & x) > { > > shared_ptr<X> px = dynamic_pointer_cast<X>(x.do_clone()); > assert(px); > return px; > } > So, to clone an object using the clone method, I could use: X* x; shared_ptr<X> spX(x = new X()); shared_ptr<X> spX2 = spX->clone(); // or: shared_ptr<X> spX3 = x->clone(); But how is the free function meant to be used? This provokes the compiler error: '... do_clone() const is private ...': shared_ptr<X> spX4 = clone(*spX); Thank you Bernhard

2 1

bgl-python installation problem
by Pablo Minguez 08 Nov '07

08 Nov '07

Hi, I'm having problems installing the python bindings for boost libraries, what I would like to use is boost.graph but from python. Boost is ok (programs and headers) but when trying to run bjam I have this error: [root][mem01][~/progs/bgl-python-0.9]# bjam error: Could not find parent for project at '.' error: Did not find Jamfile or project-root.jam in any parent directory. Is any other configuration required? Any help is welcome, many thanks in advance. Pablo. -- --------------------------------------------- Pablo Mínguez Paniagua PhD Student Bioinformatics Departament Centro de Investigación Príncipe Felipe, CIPF. Av. Autopista del Saler 16, 46013, Valencia, Spain Phone: +34 96 328 96 80 (Ext: 1011) http://bioinfo.cipf.es/

1 0

hashing a string
by Mark Wyszomierski 08 Nov '07

08 Nov '07

Hi, I'm trying to hash a string, something like: 1) Take string (password) from user 2) Hash it 3) Store to disk I see boost has some hashing features, but am not quite sure how they work. Do they do this type of thing? I'm looking for some really simple hashing, this is just for a simple application. Thanks

6 5

iterator_facade and gcc-4.3
by Benoit Hudson 08 Nov '07

08 Nov '07

"For example, when defining a constant lvalue iterator, the user can pass a const-qualified version of the iterator's value_type as iterator_facade's Value parameter and omit the Reference parameter which follows." This no longer works in current gcc 4.3 (previous issue I mentioned was a compiler bug, now fixed): it works out to generating a const lvalue in operator[], which makes gcc-4.3 give "error: type qualifiers ignored on function return type" This is either a bug in boost or a bug in gcc; not being clear on which it really is, I'll file it under boost. -- Benoît

3 3

Template type to constant string
by Robert Dailey 07 Nov '07

07 Nov '07

Hi, As you know, with Microsoft VS preprocessor operators you can convert a type passed in to a macro to a string. For example: #define TO_STRING( type ) #type I would like to do the same thing with a template type: template< typename t_type > void TO_STRING() { // Do something here with t_type to make it a string and then return it } Is there something in boost that can facilitate the behavior I'm describing? Thanks.

4 9

[boost-users] [filesystem, format] ambgious overload error.
by chun ping wang 07 Nov '07

07 Nov '07

{ boost::scoped_ptr<boost::filesystem::ofstream> myfile(new boost::filesystem::ofstream(fileOutput)); BOOST_FOREACH(int element, (getKeys<std::vector, int, int>(pMap))) { if (pMap[element] >= min_sup) { std::vector<int> row(1, element); itemsContainer.push_back(row); // (*myfile) << "(" << element << ":" << pMap[element] << ")\n"; (*myfile) << boost::format("(%1%, %2%)\n") % element % pMap[element]; } } } ./cs522/c++/aprioriMain.cpp(79) : error C2593: 'operator <<' is ambiguous C:/boost_msvc/include\boost/format/free_funcs.hpp(33): could be 'std::basic_ostream<_Elem,_Traits> &boost::operator <<<char,std::char_traits<char>,std::allocator<_Ty>>(std::basic_ostream<_Elem,_Traits> &,const boost::basic_format<Ch> &)' [found using argument-dependent lookup] with [ _Elem=char, _Traits=std::char_traits<char>, _Ty=char, Ch=char ] C:/boost_msvc/include\boost/filesystem/path.hpp(617): or 'std::basic_ostream<_Elem,_Traits> &boost::filesystem::operator <<<boost::basic_format<Ch>>(std::basic_ostream<_Elem,_Traits> &,const Path &)' [found using argument-dependent lookup] with [ _Elem=char, _Traits=std::char_traits<char>, Ch=char, Path=boost::basic_format<char> ] while trying to match the argument list '(boost::filesystem::ofstream, boost::basic_format<Ch>)' with [ Ch=char ] note the commented out version works.. this only happens on msvc 2005. it passes on mingw devc++.

1 1

Compile issue with boost::archive::xml_oarchive
by Gerald Fordyce 07 Nov '07

07 Nov '07

Hi, I am attempting to use boost::serialization to serialize a simple derived class to an XML archive. I have no problem creating a text archive - code compiles and the text archive contains both base class and derived class members. If I switch to the boost::archive::xml_oarchive it no longer compiles. Here is the code creating the archive: const DerivedClass obj; std::string filename = "fooBar.xml"; std::ofstream ofs(filename.c_str()); assert(ofs.good()); { boost::archive::xml_oarchive oa(ofs); oa << obj; } Here is the gcc-4.1.2 compiler output (Ubuntu): /usr/local/include/boost-1_34/boost/archive/basic_xml_oarchive.hpp: In member function ‘void boost::archive::basic_xml_oarchive<Archive>::save_override(T&, int) [with T = const DerivedClass, Archive = boost::archive::xml_oarchive]’: /usr/local/include/boost-1_34/boost/archive/detail/interface_oarchive.hpp:79: instantiated from ‘Archive& boost::archive::detail::interface_oarchive<Archive>::operator<<(T&) [with T = const DerivedClass, Archive = boost::archive::xml_oarchive]’ mkevents.cpp:81: instantiated from here /usr/local/include/boost-1_34/boost/archive/basic_xml_oarchive.hpp:83: error: invalid application of ‘sizeof’ to incomplete type ‘boost::STATIC_ASSERTION_FAILURE<false>’ So I'm wondering if the requirements for using txt_oarchive are different than for xml_oarchive? It seems they should both work the same. Thanks Gerald

2 1

[boost] [review] Exception - result
by Tobias Schwinger 07 Nov '07

07 Nov '07

I'm pleased to announce that the Exception library has been accepted into boost. Thanks to all formal and informal reviewers and to Emil Dotchevski for the library! Summary of possible improvements: ================================= Design ------ Nearly all reviewers made suggestions for syntactic improvements. The "tag(value) syntax" (as suggested by John Torjo) for instance provides sufficient syntactic sugar and seems lightweight to implement. Suggested features in this area are (those are implementable on top of the current design - but will keep users from reinventing the wheel if provided by the library): - Support for grouping information, and - support for generating stack traces. I personally still have a bad feeling about 'enable_error_info' - let's just revisit and discuss it on the list before changing boost::throw_exception. The related discussions about clone/rethrow support is still ongoing. Documentation ------------- Best practices for the catch-site should be clarified in the documentation. Same goes for using the same tag multiple types with one exception object. Other comments should be addressed (I won't put much energy into summarizing documentation comments, guessing most of them might have been taken care of, already). Deployment ---------- There was a suggestion (from Pavel Vozenilek) to put #ifdef around the code of boost::exception that could be put into a separate translation unit. There should probably also be a 'build' directory and a Jamfile so there is smooth integration with the Boost build system. Of course, Jamfiles for test/examples should be added. Implementation -------------- Several reviewers noted that the library is rather allocation-heavy. It pretty much comes down to shared_ptr overuse (causing many unnecessary allocations of small objects and synchronized instructions to be executed) though not in performance critical code possibly affecting other threads and future allocations. -- Tobias Schwinger - Review Manager - _______________________________________________ Unsubscribe & other changes: http://lists.boost.org/mailman/ listinfo.cgi/boost

2 1