September 2009 - Boost-users - lists.preview.boost.org

[Boost.MPL]is it possible to generate types with all combinations of template arguments?
by Tan, Tom (Shanghai) 30 Sep '09

30 Sep '09

I have a templated class template<class U, class V, class W> class S { //... implementations }; and some stock type implementations for type U, V and W: typedef boost::mpl::vector<U0, U1> u_types; typedef boost::mpl::vector<V0, V1, V2, V3, V4> u_types; typedef boost::mpl::vector<W0, W1, W2, W3, W4> w_types; I want to test class S with all possible combinations of the template arguments, typedef boost::mpl::vector< S<U0,V0,W0>, S<U0,V0,W1>, // ... S<U1,V4,W4>, > s_types; like this: boost::mpl::for_each<s_types>(test_func). The only problem is there are 2*5*5 = 50 combinations that I do not wish to type in one by one. Is there a way to generate all the combinations with boost::mpl or Boost.Preprocessor? thanks.

2 1

Fix for intrusive on Sun CC
by Chris Hite 30 Sep '09

30 Sep '09

I'm not sure if it fixes everything, but I was able debug some of the meta-programming and find a discrepancy between gcc and Sun CC which causes normal usage to break. http://www.boost.org/doc/libs/1_39_0/boost/intrusive/detail/utilities.hpp template <class T> struct internal_base_hook_bool { template<bool Add> struct two_or_three {one _[2 + Add];}; template <class U> static one test(...); template <class U> static two_or_three<U::boost_intrusive_tags::is_base_hook> test (detail::bool_<U::boost_intrusive_tags::is_base_hook>* = 0); static const std::size_t value = sizeof(test<T>(0)); }; On gcc it does what it's suppose to; value = 1 - iff T has no boost_intrusive_tags::is_base_hook 2 - iff T::boost_intrusive_tags::is_base_hook=false 3 - iff T::boost_intrusive_tags::is_base_hook=true On Sun CC it's always 1. So this is a usage of SFINAE which doesn't seem to work on SunCC, but other examples do work just fine: class_has_move_assign here: http://www.boost.org/doc/libs/1_38_0/boost/unordered/detail/move.hpp Note how they use a second test(int) function. This works on Sun CC! 1) I've never submitted a change. Will someone else take this over or do I need to dive in and submit changes to the repository? Do I need to talk this library's owner first? 2) Is this a definite CC compiler bug?

3 3

[mpl] Can we generate this type in a simpler way?
by vicente.botet 30 Sep '09

30 Sep '09

Hi, I want to generate the following type R1::mth1_<T1, R3::mth1_<T3, C::mth1_> > > from mpl::vector<mpl::pair<R1, T1>, mpl::pair<R3, T3>, C and mth1_? I have started with this template <typename SequenceOfPairs, typename E, ??? MTH> struct generator { typedef ??? type; }; but I don't know what MTH'can be, because we have 3 different mth1_. So I have reached to make a specific mth1_generator (for the specific symbol 'mth1_') template <typename SequenceOfPairs, typename E> struct mth1_generator { typedef ??? type; // The details of the type are ommited using 'mth1' }; satisfying mth1_generate<mpl::vector<mpl::pair<R1, T1>, mpl::pair<R3, T3>, C>::type is equal to R1::mth1_<T1, R3::mth1_<T3, C::mth1_> > > ? As aal the mthX_generator follow the same structure, I have defined a macro DEFINE_GENERATOR which will define the specific generator for a symbol MTH #define DEFINE_GENERATOR(MTH) \ template <typename SequenceOfPairs, typename E> struct MTH##_generator { typedef ??? type; // The details of the type are ommited suing now MTH. } So the user will just have to do DEFINE_GENERATOR(mth1); Can I do it in a simpler way using some kind of traits or this is an inherent preprocessor problem? Thanks for clarifications, _____________________ Vicente Juan Botet Escribá

1 0

[Thread] Naming threads
by Kenneth Porter 30 Sep '09

30 Sep '09

Some platforms offer a way to name threads, for use in a debugger or in process information tools. It would be nice if Boost.Thread encapsulated this. Here's a page with links to methods for Windows, Linux, and BSD: http://stackoverflow.com/questions/778085/how-to-name-a-thread-in-linux

1 0

atomic_inc32 for fast integers
by Archie14 30 Sep '09

30 Sep '09

This boost::interprocess::detail function atomically increments fixed width integer returning old value. Questions: 1. Is there a version that uses fast integers defined in cstdint.hpp? 2. I don't think that it is safe to cast fixed width ints to fast ints. I did not find conversion functions/classes. Do they exists?

1 0

[Proto] Forwarding evaluation via tag
by Joel Falcou 30 Sep '09

30 Sep '09

I have the following grammar that handles the evaluation of various matrix classes: struct eval_xpr : bp::or_< bp::when< bp::terminal< block<bp::_,bp::_> > , _load(bp::_) > , bp::when< bp::plus< eval_xpr,eval_xpr > , bp::call< functor<bp::tag::plus>(eval_xpr(bp::_left ), eval_xpr(bp::_right) ) > > > {}; As you can see, the functor<T> class is a Callable object but not a primitive Transform so I use call to be able to forward the proper things to it. Currently I only play with + but I wanted to know how i can use proto::tag_of and a clever way to conserve the recursive look so it works for any operator of any tag. Basically what should I do to have : bp::when< bp::nary_expr< bp::_, bp::vararg< eval_xpr > > , bp::call< functor< ???? >( eval_xpr(????), ???, ??? ) > > to work for any tag and any number of parameters while typiong the less code possible. I *think* I should make a new Primitive Transform that extarct tags and arguments from bp::_expr and then perform the call to functor but doesn't this mean I have to explicitly write the code for 1,2,3 etc arguments ?

3 10

[Assign] Error using list_of() with Visual C++ 2010 Beta 1
by Paul Baker 30 Sep '09

30 Sep '09

In the Boost.Assign documentation, the following example of list_of() is given: const list<int> primes = list_of(2)(3)(5)(7)(11); However, I have tended to write this instead: const list<int> primes(list_of(2)(3)(5)(7)(11)); This is causing problems with the Microsoft Visual C++ 2010 Beta 1 compiler. Using Visual C++ 2008, both lines compile fine but with 2010 Beta 1, the second produces the following error: error C2668: 'std::list<_Ty>::list' : ambiguous call to overloaded function with [_Ty=int] could be 'std::list<_Ty>::list(std::list<_Ty> &&)' with [_Ty=int] or 'std::list<_Ty>::list(unsigned int)' with [_Ty=int] while trying to match the argument list '(boost::assign_detail::generic_list<T>)' with [T=int] Is this: a) My fault? Should I be using copy initialization rather than direct initialization? b) A problem with the 2010 beta compiler? or c) A problem with Boost.Assign? Thanks

3 3

Re: [Boost-users] request for help re:, brandes_betweenness_centrality for weighted graph
by Emily Greenfest-Allen 30 Sep '09

30 Sep '09

Thank you very much; I had a hunch it had something to do w/default parameters but was unsure how the library handled them. I appreciate the help, Emily Message: 3 Date: Tue, 29 Sep 2009 17:07:30 -0400 (EDT) From: Jeremiah Willcock <jewillco(a)osl.iu.edu> To: boost-users(a)lists.boost.org Subject: Re: [Boost-users] request for help re: brandes_betweenness_centrality for weighted graph Message-ID: <alpine.OSX.1.10.0909291705020.77663(a)marvin.osl.iu.edu> Content-Type: TEXT/PLAIN; charset=US-ASCII; format=flowed On Tue, 29 Sep 2009, allenem(a)pcbi.upenn.edu wrote: > > Hi- > > > > I've just recently started to use the BGL and was wondering if someone could > > provide an example (and/or just let me know where I am going wrong) in trying > > to execute the brandes_betweenness_centrality algorithm for a weighted graph. > > Below are the relevant code snippets from my program. I am able to compile > > and run the code, and have successfully iterated over all elements (vertices, > > edges, and "interaction_weights" to verify that the data are stored > > correctly) but the calculated betweenness values are the same whether or not > > I use the edge weight map, which suggests that they are not being considered > > in the calculation even when passed as a parameter to the > > brandes_betweenness_centrality function. What am I missing? > The three-parameter version of brandes_betweenness_centrality has its third parameter as the (output) edge centrality map, not the weight map which you are trying to provide. Try using the named parameter version (as shown on <URL:http://www.boost.org/doc/libs/1_40_0/libs/graph/doc/betweenness_central…>). How to provide named parameters is described at <URL:http://www.boost.org/doc/libs/1_40_0/libs/graph/doc/bgl_named_params.ht…>. -- Jeremiah Willcock

1 0

[Boost Statechart] Execute Function
by Georg Gast 30 Sep '09

30 Sep '09

Hi, i try to build a easy statechart to learn and understand how to use boost::statechart. All i used until today was "homemade" statemachines. At some states i need a "Execute" function. In the FAQ i read, that therefore i should use a a thread and send there an event to the statemachine and process this event. So i wrote a timerclass which uses a thread to process an event via signals2. <code> ----------- timeout.h #ifndef __timeout_h__ #define __timeout_h__ #include <boost/thread.hpp> #include <boost/signals2.hpp> #include <boost/date_time/posix_time/posix_time.hpp> class Timeout { public: typedef boost::signals2::signal<void ()> signal_t; Timeout() : mp_thread(0) { lock lk(m_monitor); } ~Timeout() { lock lk(m_monitor); Terminate(); } boost::signals2::connection connect(const signal_t::slot_type &subscriber) { lock lk(m_monitor); return m_sig.connect(subscriber); } void Start(boost::posix_time::time_duration delta_time, bool repeat = false) { lock lk(m_monitor); if (mp_thread) { Terminate(); } m_repeat = repeat; m_delta_time = delta_time; mp_thread = new boost::thread( boost::bind( &Timeout::Execute, this)); } void Stop() { lock lk(m_monitor); if (mp_thread) { Terminate(); } } private: typedef boost::mutex::scoped_lock lock; void Execute() { try { do { boost::this_thread::sleep(m_delta_time); { lock lk(m_monitor); m_sig(); } } while (m_repeat); } catch(boost::thread_exception& /* e */) { // we got interrupted } } void Terminate() { if (mp_thread) { mp_thread->interrupt(); delete mp_thread; mp_thread = 0; } } signal_t m_sig; boost::thread* mp_thread; boost::mutex m_monitor; boost::posix_time::time_duration m_delta_time; bool m_repeat; }; ----------- main.cpp // SimpleBoostSCProblem // main.cpp #include "Timeout.h" #include <boost/statechart/event.hpp> #include <boost/statechart/transition.hpp> #include <boost/statechart/custom_reaction.hpp> #include <boost/statechart/state_machine.hpp> #include <boost/statechart/simple_state.hpp> #include <boost/statechart/state.hpp> #include <boost/mpl/list.hpp> namespace MyTest { namespace sc = boost::statechart; namespace mpl = boost::mpl; // Our objects struct Obj1 {}; // -------------------------------------------------- // Events // -------------------------------------------------- struct EvWork : sc::event< EvWork > {}; struct EvExit : sc::event< EvExit > {}; // -------------------------------------------------- // Forward deklaration of states // -------------------------------------------------- struct Active; struct DoStuff1; struct DoStuff2; struct Exit; // -------------------------------------------------- // Statemachine // -------------------------------------------------- struct TestMachine : sc::state_machine< TestMachine, Active > { TestMachine() { } ~TestMachine() { } }; // -------------------------------------------------- // States // -------------------------------------------------- // TestMachine / Active struct Active : sc::state<Active, TestMachine, mpl::list< DoStuff1 > > { typedef mpl::list< sc::transition< EvExit , Exit > > reactions; Active(my_context ctx) : my_base(ctx) { m_worker_connection = m_worker.connect(boost::bind(&Active::ProcessEvent<EvWork>, this)); } ~Active() { m_worker.Stop(); m_worker_connection.disconnect(); } template<typename ev> void ProcessEvent() { outermost_context().process_event(ev()); } Timeout m_worker; private: boost::signals2::connection m_worker_connection; }; // TestMachine / Active / DoStuff1 struct DoStuff1 : sc::state<DoStuff1, Active > { typedef mpl::list< sc::transition< EvWork, DoStuff2 > > reactions; DoStuff1(my_context ctx) : my_base(ctx) { // start work timeout : Repeat every 10 ms context< Active >().m_worker.Start(boost::posix_time::milliseconds(100)); } ~DoStuff1() { } }; // TestMachine / Active / DoStuff2 struct DoStuff2 : sc::state<DoStuff2, Active > { typedef mpl::list< sc::custom_reaction< EvWork > > reactions; DoStuff2(my_context ctx) : my_base(ctx) { // start work timeout : Repeat every 10 ms context< Active >().m_worker.Start(boost::posix_time::milliseconds(1),true); } ~DoStuff2() { } sc::result react( const EvWork & ) { // do stuff // .. // Nothing more to do with this event return discard_event(); } }; // TestMachine / Exit struct Exit : sc::state<Exit, TestMachine > { Exit(my_context ctx) : my_base(ctx) { } ~Exit() { } }; }; int main(int argc, char* argv[]) { MyTest::TestMachine machine; machine.initiate(); boost::this_thread::sleep(boost::posix_time::seconds(10)); machine.process_event(MyTest::EvExit()); return 0; } </code> It the statemachine exits, i get constantly crashes in the process_event<EvWork>() function, because the statemachine itself got destructed and outermost_context() is no more vaild, but the state which holds my timeout object (and the thread) did not yet got destructed and therefore the thread stopped. I dont have an idea to solve that problem. There is an assertion: "get_pointer( pOutermostUnstableState_ )==0" in state_machine.hpp. It does only happen if there was a transition from one state (DoStuff1) to DoState2. Thanks in advance! Georg Gast P.S.: My system: Windows XP / Visualstudio 2008 / Boost 1.40.0

2 1

Any complier can support lambda expression? Can any one with kindness help me?
by fmingu 30 Sep '09

30 Sep '09

I am using lambda expression under Dev-C++ for scientific computing. And the program run well. But the time consumed by the program is too much for data execution. I think this may be happened by the complier of Dev-C++(for compatibility). I am using C++ builder2009 for RAD. And this is said not suooprt Lambda expression. My computer is too limited for Visual C++ studio being installed ,which is supposed to support lambda expression. Can anyone with kindness tell me which complier can be used for lambda expression?

2 1