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

[Serialization] unregistered void cast - revisited
by Soumen 05 Aug '09

05 Aug '09

Hi, I'm getting "unregistered void cast" exception while serialization. I've followed the header file order recommended for serialization. i.e. #include <boost/archive/binary_oarchive.hpp> #include <boost/archive/binary_iarchive.hpp> #include <boost/serialization/serialization.hpp> #include <boost/serialization/base_object.hpp> #include <boost/serialization/is_abstract.hpp> //only in files where class is ABC #include <boost/serialization/export.hpp> //Followed by other headers like utility, map, vector, string Do I need to do something else as well? I've few derived classes for which I don't have any data. I've tried exporting those as well. But I still get the error. Regards, ~ Soumen -- View this message in context: http://www.nabble.com/-Serialization--unregistered-void-cast---revisited-tp… Sent from the Boost - Users mailing list archive at Nabble.com.

1 0

[date_time] unexpected behaviour from microsec_clock::local_time on 32 bit system.
by werasm＠gmail.com 05 Aug '09

05 Aug '09

Hi all, I have the following little program (I'm using mingw 3.4.2 (32bit) ): #include <boost/date_time/posix_time/posix_time.hpp> #include <iostream> int main() { using namespace boost; using namespace boost::posix_time; //Time now (since epoch in seconds)... static const ptime secSinceEpoch( second_clock::local_time() ); //Time now (since epoch in micros)... static const ptime microsecSinceEpoch( microsec_clock::local_time() ); //Elapsed should be close to zero... time_duration elapsed = microsecSinceEpoch - secSinceEpoch; std::cout << "elapsed: " << elapsed.total_milliseconds() << std::endl; return 0; } It outputs: "elapsed: -1240428644900", whereas I expected it to output a positive value. It therefore seems that microsec_clock::local_time() wraps. Is 64-bits a pre-requisite for this library to work correctly or is there some remedy? Kind regards, Werner

1 0

time_facet format flags %F and %f and "."
by Mark Stapper 05 Aug '09

05 Aug '09

Hello, I've been using time_facet to create this foolish timestamp I have to send to another program. in doing so i discovered that %f returns the fractional seconds without a leading ".", and %F does this WITH a leading ".". bug or feature? Greetz, Mark

1 0

Spring layout algorithm for animation
by Kaspar Fischer 04 Aug '09

04 Aug '09

Dear list, I am interested in spring layout algorithms and have seen this: http://www.boost.org/doc/libs/1_36_0/libs/graph/doc/kamada_kawai_spring_lay… I would like to show an animation of the process of laying out the graph and wonder whether this is possible with the Kamada-Kawai algorithm, and with the Boost implementation in particular? Or does/will Boost provide another spring layout algorithm that is suitable for animation? I intend to use such an animation in a GUI where the user initially sees a small graph (a center node with neighbours, shown in a circular layout) and can double-click on nodes to add their neighbours -- at this point I want to visually refine the graph (i.e., change its layout) using a spring-based algorithm. Many thanks, Kaspar

1 0

Re: [Boost-users] prim_minimum_spanning_tree
by Line Blander Reinhardt 04 Aug '09

04 Aug '09

Well Prims is not supposed to have problems with negative cycles. However Dijkstra does. I believe the problem lies in prim_minimum_spanning_tree use of Dijkstra. Thanks Line ________________________________________ Fra: me22.ca(a)gmail.com [me22.ca(a)gmail.com] På vegne af Scott McMurray [me22.ca+boost(a)gmail.com] Sendt: 4. august 2009 18:14 Til: Line Blander Reinhardt Emne: Re: [Boost-users] prim_minimum_spanning_tree 2009/8/4 Line Blander Reinhardt <lbre(a)man.dtu.dk>: > Have done thanks. Was just wondering if it had been fixed in newer versions. > I think it fundamentally cannot be fixed, as the algorithm doesn't provide a way to detect the problematic conditions. ~ Scott

1 0

Python Boost Graph Library API
by Claudio Martella 04 Aug '09

04 Aug '09

Hello list, I'm currently trying to implement my program, which makes extensive use of graph operations, with the py bgl module. I haven't found any API in the documentation that relates to the py language bindings. Could you please point me to the API of the python bindings of BGL? Or to the proper mailling list to ask for? TIA /CM -- Claudio Martella claudio.martella(a)gmail.com

2 1

prim_minimum_spanning_tree
by Line Blander Reinhardt 04 Aug '09

04 Aug '09

Hi all Is it really true that prim_minimum_spanning_tree cannot handle negative cycles?? I was surprised to find that when I have negative cycles the boost function crashes in Dijkstra. Has this error been fixed?? Or is it just viewed as a feature?? Thanks Line

2 2

Re: [Boost-users] [boost users][tuple] foreach with tuple
by pete＠pcbartlett.com 04 Aug '09

04 Aug '09

>typedef vector<tuple<int,int> > vec_type; >vec_type l_listLeftLines; >foreach (vec_type::reference i_pair, l_listLeftLines) >{...} It is a bit of a pain to have to resort to the typedef. Because I iterate over std::maps quite a lot, I've semi-seriously toyed with the idea of a submitting a feature request for BOOST_FOREACH_PAIR( std::pair<T1,T2> const& my_pair , my_container_with_a_pair_value_type ) which through a bit of PP_CAT and PP_COMMA somehow does the "right" thing. Given the potential for confusion though, I assumed Eric would reject the request :-).

2 1

iterator_range !is_singular assertion
by Bradley Russell 04 Aug '09

04 Aug '09

We recently upgraded our software to 1.36. Still using Visual Studio 2003. Started getting these ! is_singular assertions. I think I have tracked it down to being NDEBUG. For example if my exe does not have NDEBUG defined and I call a dll that does then the assertion happens. I have found a little bit of information about it online http://www.nabble.com/Bug-in-iterator_range-when-NDEBUG-not-defined-but-boos t-asserts-are-enabled-td17161844.html This link has a proposed patch replacing #ifndef NDEBUG with #ifndef BOOST_DISABLE_ASSERTS in the range/iterator_range.hpp In limited testing this patch seems to fix the issue. I am wondering if this is a good fix or is it something more fundamental. Thanks, Bradley Russell Programmer NPC International (620)231-3390 x 122 ---------------------------------------------------------------------------- -------- NOTICE: This electronic mail message and any attached files are confidential. The information is exclusively for the use of the individual or entity intended as the recipient. If you are not the intended recipient, any use, copying, printing, reviewing, retention, disclosure, distribution or forwarding of the message or any attached file is not authorized and is strictly prohibited. If you have received this electronic mail message in error, please advise the sender by reply electronic mail immediately and permanently delete the original transmission, any attachments, and any copies of this message from your computer system.

1 0

setsockopt SO_RCVTIMEO for udp::socket in boost::asio
by Евгений Шубин 04 Aug '09

04 Aug '09

> Hello. > I tried a simple program with clear sockets: > #include <stdlib.h> > #include <sys/socket.h> > #include <netinet/in.h> > #include <arpa/inet.h> > #include <cstring> > #include <unistd.h> > #include <iostream> > > using namespace std; > > /* > * > */ > int main(int argc, char** argv) > { > if (argc < 2) > { > cout << "Few parameters" << endl; > return 1; > } > > unsigned int taddr = inet_addr(argv[1]); > if(INADDR_NONE == taddr) > { > cout << "wrong address" << endl; > return 1; > > } > > int sock = socket(AF_INET, SOCK_DGRAM, 0); > if( sock < 0 ) > { > cout << "socket error" << endl; > return 1; > } > > sockaddr_in sin; > memset(&sin, 0, sizeof(sin)); > sin.sin_family = AF_INET; > > unsigned short port = 3334; > sin.sin_port = htons(port); > > sin.sin_addr.s_addr = taddr; > int ret = connect( sock, reinterpret_cast<const sockaddr *>(&sin), sizeof(sin) ); > if ( ret < 0 ) > { > cout << "connect error" << endl; > return 1; > } > > timeval tv; > memset( &tv, 0, sizeof(tv) ); > tv.tv_sec = 5; > > ret = setsockopt( sock, SOL_SOCKET, SO_RCVTIMEO, &tv, sizeof(tv) ); > if( ret < 0 ) > { > cout << "sockopt error" << endl; > return 1; > } > > char hello[] = "hello"; > send(sock, hello, sizeof(hello), 0); > > char msg[256]; > cout << "Recived "<< recv(sock, msg, sizeof(msg), 0) << endl; > > close(sock); > > return (EXIT_SUCCESS); > } > > This program just send buffer via connected udp socket and then wait for response. Recv may block only for 5 seconds. All work fine. > When I try to make same with boost.asio recieve method does not return if there is no response. > I use boost 1.37/ > > #include <boost/asio.hpp> > #include <iostream> > #include <vector> > using boost::asio::ip::udp; > using namespace std; > > int > main() > { > try > { > boost::asio::io_service serv; > udp::socket s( > serv, > udp::endpoint( > udp::v4(), > 3333 > ) > ); > > udp::resolver::query query(udp::v4(), "127.0.0.1", "3334"); > udp::resolver resolver(serv); > udp::resolver::iterator iterator = resolver.resolve(query); > s.connect(*iterator); > int sock = s.native(); > cout << sock << endl; > timeval tv; > memset( &tv, 0, sizeof(tv) ); > tv.tv_sec = 5; > cout << setsockopt( sock, SOL_SOCKET, SO_RCVTIMEO, &tv, sizeof(tv) ) << endl; > > string h("hello"); > s.send(boost::asio::buffer(h)); > vector<char> v(12); > cout << s.receive(boost::asio::buffer(v)) << endl; > cout << v.size() << endl; > } > catch(boost::system::system_error &err) > { > cout << err.what() << endl; > } > } > I tested both programs with nc -lu 3334 > ooh! I use Linux. > > >

1 0

[boost users][tuple] foreach with tuple
by Oliver Kania 04 Aug '09

04 Aug '09

Hello , I have a problem getting boost foreach to work with a vector of tuples of type tuple<int,int>: #define foreach BOOST_FOREACH .... vector<tuple<int,int> > l_listLeftLines; .... foreach (boost::tuple<int,int>& i_pair, l_listLeftLines) { ..... } The visual studio compiler gives me the following : 1>d:\projects\software\lib_msvc\ptw2ddatalib20\source\ptwfieldcheckalgorithm.cpp(227) : warning C4002: too many actual parameters for macro 'BOOST_FOREACH' 1>d:\projects\software\lib_msvc\ptw2ddatalib20\source\ptwfieldcheckalgorithm.cpp(227) : error C2143: syntax error : missing ')' before '>' ..... and a long tail of more errors best regards, Oliver

3 2

Re: [Boost-users] [boost users][tuple] foreach with tuple
by pete＠pcbartlett.com 04 Aug '09

04 Aug '09

On Tue 04/08/09 09:02 , Matthieu Brucher <matthieu.brucher(a)gmail.com> wrote: >Hi, > >Try with: >foreach ((boost::tuple<int,int>& i_pair), l_listLeftLines) > >The issue is that < and > are not delimiters for the macro, so you >have actually 3 parameters if you don't add the parenthesis. >From memory the extra parentheses solution isn't good enough either. You have to go with one of the workarounds described in the "pitfalls" section of the documentation: http://www.boost.org/doc/libs/1_39_0/doc/html/foreach/pitfalls.html Pete

1 0

[boost users][tuple] foreach problem resolved
by Oliver Kania 04 Aug '09

04 Aug '09

Sry for the trouble folks. This is actually documented: one solution is to predeclare the loop variable : tpl<int,int> i_pair; //4. for both the right and the left spike, get the biggest set. foreach (i_pair, l_listLeftLines) { ... } best regards, Oliver

1 0

variant: bounded types derive from same base class
by Hicham Mouline 04 Aug '09

04 Aug '09

Hello, I have a library with an abstract base class ABC and derived classes D1... Dn. These classes cannot be changed. User code wants to write code that depends on which concrete type the object being passed. User writes: void f( const ABC& abc ) { Tag t = abc.GetTag() ; // Tag is an enum for all the derived types, delivered by the lib Switch (t) { case: case: case: } } /// SO ugly, nevertheless if some case is missing, g++ for e.g. prints warnings I am told this is the visitor pattern I then looked at boost::variant<> The lib could deliver typedef boost::variant<D1, ... , Dn> ABCDerived; and the user could write visitors to visit ABCDerived objects, here the user is enforced to implement all cases in the visitor. but then the function f expects ABC objects. Can I still do: void f( const ABC& abc ) { // how to write code to use the visitors with abc ? } Rds,

3 8

[serialization - newbie] serializing multiple objects in one archive
by Soumen 04 Aug '09

04 Aug '09

Hi, I've to serialize and deserialize multiple objects. Can I serialize them in one archive and deserialize from the same correctly? Say something like following is my save/restore function: void save(boost::archive::binary_oarchive &archive) { string version = tool.getversion(); MyData *data = MyData::instance(); MyDifferentData *differentData = MyDifferentData::instance(); if (data && differentData) { archive << version; archive << data; archive << differentData; } } void restore(boost::archive::binary_iarchive &archive) { string version = tool.getversion(); string serializedToolversion; MyData::cleanup(); MyDifferentData::cleanup(); archive >> serializedToolVersion; if (version == serializedToolVersion) { archive >> (MyData *) MyData::instance(); archive >> (MyDifferentData *) MyDifferentData::instance(); } } Or do I need to serialize each of these into different archives? Regards, ~ Soumen -- View this message in context: http://www.nabble.com/-serialization---newbie--serializing-multiple-objects… Sent from the Boost - Users mailing list archive at Nabble.com.

1 0

bimap intellisense problems in VC++ 2008
by Donald Adu-Poku 03 Aug '09

03 Aug '09

Hi all, I'm having some intellisense problems with with boost::bimaps::bimap of the current boost build (boost 1.39) . The problem is that no intellisense on bimap is generated at all, intellisense goes as far as the boost::bitmaps namespace. My programme however compiles just fine. I have the boost library included in my project (..Programme Files\boost\boost_1_39), linked needed libraries (Programme Files\boost\boost_1_39\lib) and have my include statement (#include <booost/bimap.hpp>). I have also forced the regeneration of the project's intellisense database on suspicion that it might have been corrupted, this however did not resolve the problem . Is there anything I have missed? Thank you //Donald

3 6

Boostcon Keynote (finally!) available.
by Marshall Clow 03 Aug '09

03 Aug '09

at http://boostcon.blip.tv Thank you, everyone, for your patience. More videos from Boostcon will be going up there in the near future. -- Marshall

1 0

reference_wrapper
by er 03 Aug '09

03 Aug '09

Hi All, 1) I don't understand this part of reference_wrapper: template<class T> class reference_wrapper { operator T& () const { return *t_; } }; Could someone please explain it? 2) Is there already a reference wrapper for n-ary functors which exposes the same result_type (if it exists) as the n-ary functor, but whose arguments types are arbitrary (deduced from the arguments)?

3 4

Compiler warnings when using boost::accumulators with gcc
by Rob Wheeler 03 Aug '09

03 Aug '09

If I turn on gcc's -Wextra option when compiling code that uses an accumulator with tag::max, I get a screen full of compiler warnings: [wheeler@cib: /var/tmp/acc] cat acc.cpp #include <iostream> #include <boost/accumulators/accumulators.hpp> #include <boost/accumulators/statistics/stats.hpp> #include <boost/accumulators/statistics/max.hpp> using namespace boost::accumulators; int main() { accumulator_set<double, stats<tag::max> > acc; acc(1.2); acc(2.3); acc(3.4); acc(4.5); std::cout << "Max: " << max(acc) << std::endl; return 0; } [wheeler@cib: /var/tmp/acc] g++ -W -I /opt/ros/include/boost-1_37/ acc.cpp -o acc /opt/ros/include/boost-1_37/boost/concept_check.hpp: In instantiation of ‘boost::UnaryFunction<boost::numeric::functional::as_min<const double, void>, double, const double&>’: /opt/ros/include/boost-1_37/boost/concept/detail/has_constraints.hpp:40: instantiated from ‘const bool boost::concept::not_satisfied<boost::UnaryFunction<boost::numeric::functional::as_min<const double, void>, double, const double&> >::value’ /opt/ros/include/boost-1_37/boost/concept/detail/has_constraints.hpp:43: instantiated from ‘boost::concept::not_satisfied<boost::UnaryFunction<boost::numeric::functional::as_min<const double, void>, double, const double&> >’ /opt/ros/include/boost-1_37/boost/mpl/if.hpp:67: instantiated from ‘boost::mpl::if_<boost::concept::not_satisfied<boost::UnaryFunction<boost::numeric::functional::as_min<const double, void>, double, const double&> >, boost::concept::constraint<boost::UnaryFunction<boost::numeric::functional::as_min<const double, void>, double, const double&> >, boost::concept::requirement<boost::UnaryFunction<boost::numeric::functional::as_min<const double, void>, double, const double&> > >’ /opt/ros/include/boost-1_37/boost/concept/detail/general.hpp:19: instantiated from ‘boost::concept::requirement_<void (*)(boost::UnaryFunction<boost::numeric::functional::as_min<const double, void>, double, const double&>)>’ /opt/ros/include/boost-1_37/boost/accumulators/numeric/detail/function1.hpp:67: instantiated from ‘typename boost::detail::function1<F>::result<boost::detail::function1<F> ()(const A0&)>::type boost::detail::function1<F>::operator()(const A0&) const [with A0 = double, F = boost::numeric::functional::as_min<mpl_::arg<-0x00000000000000001>, boost::numeric::functional::tag<mpl_::arg<-0x00000000000000001> > >]’ /opt/ros/include/boost-1_37/boost/accumulators/statistics/max.hpp:36: instantiated from ‘boost::accumulators::impl::max_impl<Sample>::max_impl(const Args&) [with Args = boost::parameter::aux::arg_list<boost::parameter::aux::tagged_argument<boost::accumulators::tag::accumulator, boost::accumulators::accumulator_set<double, boost::accumulators::stats<boost::accumulators::tag::max, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na>, void> >, boost::parameter::aux::empty_arg_list>, Sample = double]’ /opt/ros/include/boost-1_37/boost/accumulators/framework/depends_on.hpp:282: instantiated from ‘boost::accumulators::detail::accumulator_wrapper<Accumulator, Feature>::accumulator_wrapper(const Args&) [with Args = boost::parameter::aux::arg_list<boost::parameter::aux::tagged_argument<boost::accumulators::tag::accumulator, boost::accumulators::accumulator_set<double, boost::accumulators::stats<boost::accumulators::tag::max, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na>, void> >, boost::parameter::aux::empty_arg_list>, Accumulator = boost::accumulators::impl::max_impl<double>, Feature = boost::accumulators::tag::max]’ /opt/ros/include/boost-1_37/boost/accumulators/framework/depends_on.hpp:215: instantiated from ‘static boost::fusion::cons<typename boost::fusion::result_of::value_of<Iterator>::type, typename boost::accumulators::detail::build_acc_list<typename boost::fusion::result_of::next<Iterator>::type, Last, boost::fusion::result_of::equal_to<typename boost::fusion::result_of::next<Iterator>::type,Last>::value>::type> boost::accumulators::detail::build_acc_list<First, Last, false>::call(const Args&, const First&, const Last&) [with Args = boost::parameter::aux::arg_list<boost::parameter::aux::tagged_argument<boost::accumulators::tag::accumulator, boost::accumulators::accumulator_set<double, boost::accumulators::stats<boost::accumulators::tag::max, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na>, void> >, boost::parameter::aux::empty_arg_list>, First = boost::fusion::mpl_iterator<boost::mpl::v_iter<boost::mpl::v_item<boost::accumulators::detail::accumulator_wrapper<boost::accumulators::impl::max_impl<double>, boost::accumulators::tag::max>, boost::mpl::vector0<mpl_::na>, 0>, 0l> >, Last = boost::fusion::mpl_iterator<boost::mpl::v_iter<boost::mpl::v_item<boost::accumulators::detail::accumulator_wrapper<boost::accumulators::impl::max_impl<double>, boost::accumulators::tag::max>, boost::mpl::vector0<mpl_::na>, 0>, 1l> >]’ /opt/ros/include/boost-1_37/boost/accumulators/framework/depends_on.hpp:234: instantiated from ‘typename boost::accumulators::detail::meta::make_acc_list<Sequence>::type boost::accumulators::detail::make_acc_list(const Sequence&, const Args&) [with Sequence = boost::mpl::v_item<boost::accumulators::detail::accumulator_wrapper<boost::accumulators::impl::max_impl<double>, boost::accumulators::tag::max>, boost::mpl::vector0<mpl_::na>, 0>, Args = boost::parameter::aux::arg_list<boost::parameter::aux::tagged_argument<boost::accumulators::tag::accumulator, boost::accumulators::accumulator_set<double, boost::accumulators::stats<boost::accumulators::tag::max, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na>, void> >, boost::parameter::aux::empty_arg_list>]’ /opt/ros/include/boost-1_37/boost/accumulators/framework/accumulator_set.hpp:145: instantiated from ‘boost::accumulators::accumulator_set<Sample, Features, Weight>::accumulator_set() [with Sample = double, Features = boost::accumulators::stats<boost::accumulators::tag::max, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na>, Weight = void]’ acc.cpp:9: instantiated from here /opt/ros/include/boost-1_37/boost/concept_check.hpp:325: warning: non-static reference ‘const double& boost::UnaryFunction<boost::numeric::functional::as_min<const double, void>, double, const double&>::arg’ in class without a constructor

1 0

Re: [Boost-users] boost 1.40 beta
by Guiliani, Florent 03 Aug '09

03 Aug '09

boost-users-request(a)lists.boost.org wrote: > Date: Wed, 29 Jul 2009 14:27:39 +0100 > From: Avi Bahra <avibahra(a)googlemail.com> > Subject: [Boost-users] boost 1.40 beta > To: boost-users(a)lists.boost.org > > Does any know the schedule for release of boost 1.40 beta ? > > Best regards, > Ta, > Avi > According to: http://www.boost.org/development/ it will be the first of August. -- Florent,

3 2

[parameter] How to create make_ functions with boost::parameter returning templated classes?
by jesseperla 03 Aug '09

03 Aug '09

I want to write a make_ function for a complicated template, but can't figure out how to determine the deduced types for the return type: Say my class is: template<class F> struct templated_class { F f_; double y_; templated_class(F f, double y) : f_(i), y_(y){} }; I tried variations on the following for the template argument of templated_class: BOOST_PARAMETER_NAME(y) BOOST_PARAMETER_NAME(f) BOOST_PARAMETER_FUNCTION( (templated_class<typename boost::parameter::binding<ArgumentPack, tag::f>::type>), make_templated_class, tag, (required (f, *) ) (optional (i, *, 2.0) ) ) { return templated_class<typename boost::parameter::binding<ArgumentPack, tag::f>::type>(f, y); } This doesn't work. The parameter::binding was my attempt to extract the type, but ArgumentPack doesn't exist. Is this the right general method? What template parameter can I use for the metafunction? Thanks, Jesse

1 0

Boost's Linear Algebra Solution for y=Ax
by Peverall Dubois 03 Aug '09

03 Aug '09

Does boost have one? Where A, y and x is a matrix and vectors respectively. Either y or x can be unknown. - G.V.

2 1

[Serialization: Newbie] Compilation Problem Text Archive
by Soumen 03 Aug '09

03 Aug '09

Hi, I'm facing some compilation problem when I use text archive. But the same code works fine when I use binary archive. And it does write something on disk with specified file name (I'm yet to finish the restore part. Otherwise, I could've said it works). Do I need to do something differently? The error I get is something like following: ------------------------------------------------------------------------------------------ /boost/1_35_0/include/boost/archive/detail/oserializer.hpp: In function ‘void boost::archive::save(Archive&, T&) [with Archive = boost::archive::text_oarchive, T = TableMgr*]’: /boost/1_35_0/include/boost/archive/detail/common_oarchive.hpp:64: instantiated from ‘void boost::archive::detail::common_oarchive<Archive>::save_override(T&, int) [with T = TableMgr*, Archive = boost::archive::text_oarchive]’ /boost/1_35_0/include/boost/archive/basic_text_oarchive.hpp:75: instantiated from ‘void boost::archive::basic_text_oarchive<Archive>::save_override(T&, int) [with T = TableMgr*, Archive = boost::archive::text_oarchive]’ /boost/1_35_0/include/boost/archive/detail/interface_oarchive.hpp:64: instantiated from ‘Archive& boost::archive::detail::interface_oarchive<Archive>::operator<<(T&) [with T = TableMgr*, Archive = boost::archive::text_oarchive]’ /remote/gsoumen/hiers/trunk/code/cmds/Cmds.cpp:1518: instantiated from here /boost/1_35_0/include/boost/archive/detail/oserializer.hpp:564: error: invalid application of ‘sizeof’ to incomplete type ‘boost::STATIC_ASSERTION_FAILURE<false>’ cc1plus: warnings being treated as errors ------------------------------------------------------------------------------------------ My code is something like following: #include <boost/archive/text_oarchive.hpp> #include <boost/archive/text_iarchive.hpp> //I use binary archive hpp when I use binary archive mode #include <boost/serialization/serialization.hpp> #include <boost/serialization/utility.hpp> std::ofstream ofs(fileName.c_str(), ios_base::out); // with bin archive I also use ios_base::binary boost::archive::text_oarchive oarchive(ofs); //for bin archive, oarchive is of binary_oarchive type TableMgr *tblMgr = TableMgr::instance(); if (tblMgr) { oarchive << tblMgr; } //I've TableMgr::serialize() defined ... Regards, ~ Soumen -- View this message in context: http://www.nabble.com/-Serialization%3A-Newbie--Compilation-Problem-Text-Ar… Sent from the Boost - Users mailing list archive at Nabble.com.

2 6

any C++ portable library to create pdf documents?
by Frédéric Bron 03 Aug '09

03 Aug '09

I have looked in in the vault but could not find any library to create pdf documents. Does anybody knows a good one written in portable C++? Regards, Frédéric

3 2

[asio] AES encryption/decryption
by Igor R 03 Aug '09

03 Aug '09

Hello all, Does someone have an experience with the subj? In my application I need an AES-encrypted socket. I try to accomplish this by building a custom asio socket that performs all the needed encryption (with openssl) and delegates the i/o to its service. However, I encountered an issue that seems to be unsolvable with the current approach: from one hand, every read opeartion has to return the number of *decrypted* bytes, but from the other hand, bytes_readable command returns the size of the available *encrypted* data. Sometimes this causes severe problems: void connection::read() { asio::socket_base::bytes_readable command(true); socket_.lowest_layer().io_control(command); size_t canRead = command.get(); if (canRead) // this line creates completion condition that might never be met, because read_some() would return less than canRead asio::read(socket_, input_, asio::transfer_at_least(canRead), err); } I'd appreciate any idea - probably I'm going the wrong way, and there's some more appropriate design that would give better results? Thanks.

2 1