- Boost-users - lists.boost.org

Replacing task_io_service (using io_service with an external threadpool)
by Juan Ramírez 23 Dec '12

23 Dec '12

Hi! I created a (yet another) threapool library a few month ago and, at my company, we are starting to use it widely in different services. Using the library has lots of advantages for us, mainly because of it performance compared with similar libraries. The problem we are having is that we are using 'two' different kinds of thread pool in the same service: * The first one is the one I wrote above, the pool provided by my library, which is used to execute by the service to execute long or expensive operations. * And the second one, is the one created a the network level, where about a hundred threads are created in order to attend network requests. ** Actually calling new thread(io_service::run, ...) I wanted to move the thread-handling at the network level to my library, so I started to look at documentation about extending boost::asio and adapting my library to the io_service requirements. What I want to do now is to make the io_service object actually use the threadpool library for task execution. I have taken a look to use_service, add_service, and the io_service::service class, and I think I already have a pretty good implementation of a 'poor man's adaptor' between my library and boost asio. The question is, how do I force the io_service object to use my service implementation instead of the on provided by boost::asio::detail::task_io_service? I mean, I haven't tested yet but I have no idea about how does the io_service know which service should be used in each case. Do I have to force the removal of task_io_service? Do I have to call pool() somewhere or re-implement run() functions? Thanks in advance Best Regards __________ Saludos! Juan

2 1

[spirit][qi] on_error handler
by Diederick C. Niehorster 22 Dec '12

22 Dec '12

Hi All, I have struggled through getting a parser set up with qi, but in the end it all worked out. Its really a nice library once you get the hang of it a bit! I have one leftover question. I now have an inline function with phoenix for the error handler as per the example, but as i'm using it multiple times, i would like to make its own function that i can reuse. I have tried a function object, wrapped in a phoenix function or not, but never got it to work (either got number of argument mismatches or errors that const-volatile qualifiers were being dropped). So basically, how would i rewrite the below into a free function (object/etc)? qi::on_error<qi::fail> ( stimSpecVec // rule name , std::cout // the function << phoenix::val("Error! Expecting \"") << qi::_4 // what failed? << phoenix::val("\", but got: \"") << phoenix::construct<std::string>(qi::_3, qi::_2) // iterators to error-pos, end << phoenix::val("\"") << std::endl ); Thanks! Dee

2 2

Boost.MSM: Exit a composite state after an event has been fired from that composite state?
by RaRi 21 Dec '12

21 Dec '12

Hi list, I'm using boost.msm from the boost 1.52 library and trying to fire an event from a composite state to enter a state of a top level state machine. Unfortunatly this doesn't work. Please consider here for the following code: namespace { namespace msm = boost::msm; namespace msmf = boost::msm::front; namespace mpl = boost::mpl; // ----- Events struct Event1 {}; struct Event2 {}; struct EventSub1 {}; struct EventSub2 {}; // ----- State machine struct OuterSm_:msmf::state_machine_def<OuterSm_> { struct State1_:msmf::state_machine_def<State1_> { template <class Event,class Fsm> void on_entry(Event const&, Fsm&) const { std::cout << "State1::on_entry()" << std::endl; } template <class Event,class Fsm> void on_exit(Event const&, Fsm&) const { std::cout << "State1::on_exit()" << std::endl; } struct SubState1:msmf::state<> { template <class Event,class Fsm> void on_entry(Event const&, Fsm& fsm) const { std::cout << "SubState1::on_entry()" << std::endl; std::cout << "SubState1::fire event2()" << std::endl; fsm.process_event(Event2()); // Produces an assertion -> It is not possible to leave the submachine although the "Event2()" is defined as transition in the transition table of the outermachine. } template <class Event,class Fsm> void on_exit(Event const&, Fsm&) const { std::cout << "SubState1::on_exit()" << std::endl; } }; struct SubState2:msmf::state<> { template <class Event,class Fsm> void on_entry(Event const&, Fsm& a) const { std::cout << "SubState2::on_entry()" << std::endl; } template <class Event,class Fsm> void on_exit(Event const&, Fsm&) const { std::cout << "SubState2::on_exit()" << std::endl; } }; struct Exit1:msmf::exit_pseudo_state<msmf::none> {}; // Set initial state typedef mpl::vector<SubState1> initial_state; // Transition table struct transition_table:mpl::vector< // Start Event Next Action Guard msmf::Row < SubState1, EventSub1, SubState2, msmf::none, msmf::none > // Is there a possibility in MSM to exit a composite state if an event is fired from that // msmf::Row < SubState1, Event2, Exit1, msmf::none, msmf::none > I don't want this! > {}; }; struct State2:msmf::state<> { template <class Event,class Fsm> void on_entry(Event const&, Fsm&) const { std::cout << "State2::on_entry()" << std::endl; } template <class Event,class Fsm> void on_exit(Event const&, Fsm&) const { std::cout << "State2::on_exit()" << std::endl; } template <class Event, class Fsm, class SourceState, class TargetState> void operator()(Event const&, Fsm& f, SourceState&, TargetState&) const { std::cout << "Do function" << std::endl; } }; typedef msm::back::state_machine<State1_> State1; // Set initial state typedef State1 initial_state; // Transition table struct transition_table:mpl::vector< // Start Event Next Action Guard msmf::Row < State1, Event2, State2, State2, msmf::none > > {}; }; // Pick a back-end typedef msm::back::state_machine<OuterSm_> Osm; void test() { Osm osm; osm.start(); // osm.process_event(Event2()); // This works fine -> Leaving state "Substate1" of the submachine } } int main() { test(); return 0; } You see that the code calls f*sm.process_event(Event2())* within the state *SubState1* of the submachine to make a transition from the top-level state *State1* to *State2* of the outer statemachine. Unfortunately this doesn't work although the *Event2* is defined in the transition table of the outer-statemachine. On the other hand it a transition from *State1* to state *State2* works in case the event *Event2* is fired outside the submachine ( please refer to the void test() function in the code above). I actually don't want to define a transition in the transition table of submachine to enter an pseudo exit point like: msmf::Row < SubState1, Event2, Exit1, msmf::none, msmf::none > Is there a possibility in MSM to exit a composite state if an event is fired from that composite state? Thank you in advance for your time and help! Grüße ;), RaRi -- View this message in context: http://boost.2283326.n4.nabble.com/Boost-MSM-Exit-a-composite-state-after-a… Sent from the Boost - Users mailing list archive at Nabble.com.

2 3

Install Boost 1.52.0 for Visual Studio 2012 on Windows 8
by Boost lzw 21 Dec '12

21 Dec '12

Hello All, I used the following Windows installer to install Boost in the past for Visual Studio 2008 Professional on Windows XP in the past: http://www.boostpro.com/download/ It seems to me that the above installer is not for Visual Studio 2012 on Windows 8. Is there anyone here who is aware of a Windows Installer for installing Boost 1.52.0 for Visual Studio 2012 on Windows 8 ? Thanks in advance. Happy Holidays, Robert

2 3

Is 'double' a signed type? (boost type traits)
by cdmh 21 Dec '12

21 Dec '12

Hi, I have found some surprising behaviour with boost type traits using MSVC2012 (v11). I used boost::is_signed<T>::value where T is double and it evaluates to false, but I expected true as double can hold negative numbers. More surprisingly, I found that, for double, is_unsigned is not the inverse of is_signed, and hence this static assert produces a compiler error. BOOST_STATIC_ASSERT(boost::is_unsigned<double>::value == !boost::is_signed<double>::value); I am missing a rationale for this, or it is a bug? Thanks -- Craig -- View this message in context: http://boost.2283326.n4.nabble.com/Is-double-a-signed-type-boost-type-trait… Sent from the Boost - Users mailing list archive at Nabble.com.

2 1

boost::bind for C function
by Philipp Kraus 21 Dec '12

21 Dec '12

Hello, I have got a a C library function call, which defines a function: typedef int (*CFunction) (State* K); So I would like to use boost::bind / boost::function to create the binding to a class method: boost::function<CFunction>( boost::bind(&MyClass::myMethod, this) ) so that I can use the class method with the CFunction signature. I would like to call a class method with the object context like a C function. Can anybody help me to create a correct binding? Thanks Phil

6 11

[Property Message] How to store variant in property tree
by Elizabeta 21 Dec '12

21 Dec '12

Hi I want to store variant object in my property tree. The variant for now will have only 3dPoint object, but it should vary multiple types in future. I tried the following code , and I am getting compiler errors. I know that maybe I need to write some kind of translation code for this to work,but don't know how to do that exactly and having hard time to grasp the concept. So how can I do this if possibly at all ? A code example would be great class 3dPoint { public: 3dPoint(double,double,double); double x; double y; double z; }; typedef boost::variant<ON_3dPoint> MyVariant; using namespace boost::property_tree; typedef basic_ptree<std::string,MyVariant> my_ptree; my_ptree pt; MyVariant point_variant = 3dPoint(1,2,3); pt.put("point",point_variant); MyVariant point_variant2 = pt.get("point"); boost::property_tree::xml_writer_settings<char> settings('\t', 1); write_xml("C:\\property.xml",pt,std::locale(),settings);

2 2

Symbol 'is_arithmetic' could not be resolved, Symbol 'is_const' could not be resolved, Symbol 'value' could not be resolved
by Tanya 21 Dec '12

21 Dec '12

Dear Boost Users, I have a code with: BOOST_STATIC_ASSERT(::boost::is_arithmetic<T>::value); and it gives error message: - Symbol 'is_arithmetic' could not be resolved - Symbol 'value' could not be resolved and, BOOST_STATIC_ASSERT(!::boost::is_const<T>::value); and it gives error message: - Symbol 'value' could not be resolved - Symbol 'is_const' could not be resolved what should I include to get rid of these error message? I use: - Boost 1.46.1. - Compiler : Mingw with gcc - IDE: Eclipse Indigo Indigo Service Release 2 - Windows XP Professional SP 3, v. 2002 Any help, hints, would be highly appreciated! Best Regards, Tanya

2 1

BOOST::ASIO load key.pem with passphrase
by Alvaro Leiva 21 Dec '12

21 Dec '12

Currently I'm working with this: ... ctx.use_certificate_chain_file("./C/cert.pem"); ctx.use_private_key_file("./C/key.pem", boost::asio::ssl::context::pem); ctx.load_verify_file("./C/ca.pem");... So far everything works perfect, but what I really need to do is load the same key.pem but with a passphrase, Looking at asio docs found ssl::context::set_password_callback<http://www.boost.org/doc/libs/1_52_0/doc/html/boost_asio/reference/ssl__con…> wich according to the info calls SSL_CTX_set_default_passwd_cb<http://www.openssl.org/docs/ssl/SSL_CTX_set_default_passwd_cb.html> and allows to handle encrypted PEM files, please remember that I'm much more familiarized with high level languages like python, so c++ is not my forte any help is appreciated, Thank you ! -- Alvaro Leiva

1 0

Link problem on Mac - tag_original_exception_type
by John M. Dlugosz 20 Dec '12

20 Dec '12

I'm getting a linker error when I build my program on the Mac, that __ZTSPN5boost27tag_original_exception_typeE is duplicate defined. This is the only symbol that is giving this problem. Any ideas what's special about that one, or what it might be?

1 0