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

[serialization] Creating new archives derived from polymorphic archive
by Robert S. Grimes 11 Mar '06

11 Mar '06

Hi, How do I get started? I want to create a set of custom archives that include such features as a small header, CRC, digital signatures, etc. I want to put this into a DLL, so I want to go the polymorphic route. There are directions for creating new archives from the template versions, but not for the polymorphic versions. I'm not sure the starting point is the same, so I thought I'd ask first, as this is stretching my abilities a bit. Thanks, -Bob

2 4

Re: [Boost-users] Syntax of get for Variant and any_cast for Any
by Sohail Somani 10 Mar '06

10 Mar '06

> -----Original Message----- > From: boost-users-bounces(a)lists.boost.org > [mailto:boost-users-bounces@lists.boost.org] On Behalf Of Scott Meyers > Sent: Friday, March 10, 2006 2:04 PM > To: boost-users(a)lists.boost.org > Subject: Re: [Boost-users] Syntax of get for Variant and > any_cast for Any > My only other knowledge of "get" is from tuple, where "get" > takes a compile-time > constant, not a type, as an argument. With that background, > I found the use of > "get" for variant initially confusing. I think it'd be > better to find a name > with the "_cast" suffix and to explicitly specify pointer and > reference types > (as with dynamic_cast) and adopt that convention. I don't > know enough about the > various Boost libraries to know whether that is a reasonable > suggestion. I think shared_ptr has dynamic_pointer_cast or something. This is much easier to read than get

1 0

[BGL] graph copy after Kruskal's algorithm?
by Alejandro Aragón 10 Mar '06

10 Mar '06

Hi all, After running Kruskal's algorithm I end up with the list of the edges that form the minimum cost. However, no new graph is created. I need the resulting graph for further computations. How do you create a graph from an existing one but only selecting the edges contained in the minimum spanning tree vector? I've seen that there are some functions to accomplish this but I couldn't find any example of how to do it: template <class EdgeIterator, class EdgePropertyIterator> adjacency_list(EdgeIterator first, EdgeIterator last, EdgePropertyIterator ep_iter, vertices_size_type n, vertices_size_type m = 0, const GraphProperty& p = GraphProperty()) The existing graph has properties in edges and vertices that I also need to keep. Can anyone help? Thank you all, A.Aragon

4 6

Re: [Boost-users] Help needed with random exponential_distribution
by dahaverk＠rockwellcollins.com 10 Mar '06

10 Mar '06

It seems my problem is with the rand48 function. With Boost_1_29_0 ---> sizeof(boost::rand48)=8 e.g. the size of a double. With Boost_1_33_1 ---> sizeof(boost::rand48)=16 e.g. twice the size of a double. Because of the truncation that happened with the boost_1_29_0 We were seeing a correctly randomized distribution for the exponential_distribution. Is this a config problem? If so, where do I fix the "boost::rand48" to be a "double" again? I did run the boost ./configure on the Boost_1_33_1 directory before I started using it on the Sun Sparc Solaris8 machine. Thanks, Dave

1 0

Boost Statechart: general questionss about what the library can do
by RIVASSEAU Jean Noel 10 Mar '06

10 Mar '06

Hi there, I am almost entirely new to state machines, finite automata (by almost I mean that I know the concept and very little of the theory). However I need to evaluate if the Boost library can meet the requirements of a project I plan to use it on. Basically I have two questions: First and most important: Is it possible to *ask* the state machine to go from a state A to a state B *automatically* ?? I mean, I am in state A. I need to go to State B, but I cannot go straight from A to B, I must go first through state C, D, E and finally B for example. Can the library automatically compute a path and generate the correct transitions (event)?? If it is not possible, I would like to know if it is a fundamental problem (theorical with state automata, common to any library implementation) or just something that Boost Statechart does not provide, but other implementations could. Note that it is not necessarily for me that the automata computes the "best" or "shortest" path (set of transitions) it must just generate correct transitions that I can log. My second question is easier, more implementation related, and I think the answer is yes, but anyway: When a transition is executed (automatically, or explicitely), can I call a function? That is, when an event is processed, can I implement some code that will be executed each time an event is processed ? Thank you!! Jean-Noël Ps: Is this the correct place to discuss Boost Statechart, since it is not yet part of the official Boost package (distribution) ?

2 1

[statechart] - passing data between states during state transission?
by Oliver.Kowalke＠infineon.com 10 Mar '06

10 Mar '06

Hello, I want to pass data from outside into the state machine and pass the data from one state to the next state during state transition. But it seams that statechart doesn't support this. Could this feature be added? What is the preferred way which statechart suggests? With best regards, Oliver #include <iostream> #include <stdexcept> #include <boost/statechart/custom_reaction.hpp> #include <boost/statechart/event.hpp> #include <boost/statechart/simple_state.hpp> #include <boost/statechart/state_machine.hpp> namespace sc = boost::statechart; struct ev_activate : public sc::event< ev_activate > { int data; ev_activate( int d) : data( d) {} }; struct inactive; struct sm : public sc::state_machine< sm, inactive > {}; struct active : public sc::simple_state< active, sm > { active( int i) { std::cout << "active(): " << i << std::endl; } }; struct inactive : public sc::simple_state< inactive, sm > { typedef sc::custom_reaction< ev_activate > reactions; sc::result react( ev_activate const& ev) { return transit< active >( ev.data); // passing data to new state during transition } }; int main() { try { sm fsm; fsm.initiate(); fsm.process_event( ev_activate( 1) ); // passing data from outside return 0; } catch ( std::runtime_error const& e) { std::cerr << e.what() << std::endl; } catch (...) { std::cerr << "unhandled exception" << std::endl; } return -1; }

2 1

vertex property setting
by Camille Lemen 10 Mar '06

10 Mar '06

Hi all! I'm new at using boost and my request certainly is obvious but I can't manage to get the error... I am trying to set std::vector's type vertex properties in a graph. But it does not want te set a property to more than one edge. The program is the following: /****************************************code*****************************/ #include <vector> #include <iostream> #include <list> #include <map> //boost includes #include <boost/property_map.hpp> #include <boost/graph/graph_traits.hpp> #include <boost/graph/properties.hpp> #include <boost/graph/adjacency_list.hpp> #include <boost/graph/adjacency_matrix.hpp> using namespace std; enum edge_intersec_t { edge_intersec }; enum edge_dist_t { edge_dist }; enum vertex_param_t { vertex_param}; namespace boost { BOOST_INSTALL_PROPERTY(edge, intersec); BOOST_INSTALL_PROPERTY(edge, dist); BOOST_INSTALL_PROPERTY(vertex, param); } using namespace boost; typedef property<edge_intersec_t, unsigned int, property<edge_dist_t,float> > EdgeProp; typedef property<vertex_param_t,std::vector<float> > VertexProp; typedef adjacency_list<setS,vecS, bidirectionalS,VertexProp,EdgeProp> Graph; typedef graph_traits<Graph>::edge_descriptor edge_descriptor; int main(int argc, char** argv) { int NbRegions = 9; Graph g(NbRegions); edge_descriptor e; bool inserted; tie(e,inserted) = add_edge(vertex(0,g),vertex(1,g),EdgeProp(0,20),g); tie(e,inserted) = add_edge(vertex(0,g),vertex(2,g),0,g); tie(e,inserted) = add_edge(vertex(1,g),vertex(2,g),0,g); tie(e,inserted) = add_edge(vertex(3,g),vertex(4,g),0,g); tie(e,inserted) = add_edge(vertex(3,g),vertex(5,g),0,g); tie(e,inserted) = add_edge(vertex(3,g),vertex(7,g),0,g); tie(e,inserted) = add_edge(vertex(4,g),vertex(5,g),0,g); tie(e,inserted) = add_edge(vertex(5,g),vertex(6,g),0,g); tie(e,inserted) = add_edge(vertex(1,g),vertex(4,g),50,g); tie(e,inserted) = add_edge(vertex(1,g),vertex(3,g),100,g); tie(e,inserted) = add_edge(vertex(0,g),vertex(4,g),200,g); tie(e,inserted) = add_edge(vertex(2,g),vertex(5,g),500,g); tie(e,inserted) = add_edge(vertex(2,g),vertex(6,g),20,g); tie(e,inserted) = add_edge(vertex(2,g),vertex(4,g),50,g); tie(e,inserted) = add_edge(vertex(0,g),vertex(3,g),50,g); std::vector<float> param(2); std::vector<float> propriete(2); param[0] = 800; std::cerr<<"param[0] = "<<param[0]<<std::endl; param[1] = 20; std::cerr<<"param[1] = "<<param[1]<<std::endl; put(vertex_param,g,vertex(0,g),param); propriete = get(vertex_param,g,vertex(0,g)); std::cout<<"propriete[0]: "<<propriete[0]<<", propriete[1]: "<<propriete[1]<<std::endl; param.clear(); param[0] = 2; std::cerr<<"param[0] = "<<param[0]<<std::endl; param[1] = 50; std::cerr<<"param[1] = "<<param[1]<<std::endl; put(vertex_param,g,vertex(1,g),param); propriete = get(vertex_param,g,vertex(1,g)); std::cout<<"propriete[0]: "<<propriete[0]<<", propriete[1]: "<<propriete[1]<<std::endl; return 0; } /************************************************************************/ it compiles and executes, but the property of the second vertex has not been asigned. Here is the result it gives: /***************command window*************************************/ param[0] = 800 param[1] = 20 propriete[0]: 800, propriete[1]: 20 param[0] = 2 param[1] = 50 propriete[0]: 800, propriete[1]: 20 /**************************************************************************/ I tried without using "propriete" but directly"get(vertex_param,g,vertex(1,g)) [0]", and it gives a segmentation fault during the execution which confirms that the vertex property has not been asigned to the second vertex. Thanks! Camille.

2 2

Re: [Boost-users] solved + bug with mingw (?) (was: [MPL] Need help crossing runtime boundary (lengthy))
by Ames Andreas 10 Mar '06

10 Mar '06

Hi all, I've solved my problem (FWIW, I attached member_notifier.cc). While trying it out, I observed that mingw (see attached mingw.out, it's gcc 3.4.4 with boost 1.33.0) can't compile that code. VC7.1 (boost 1.33.0) and VC8 (boost 1.33.1) accept the code and the exe runs as expected. Curiously a stripped down but very "similar" example (see attached inherit_test.cc) is also compiled by mingw. Is this mingw's or boost's fault, or is it even mine? cheers, aa -- Andreas Ames | Programmer | Comergo GmbH | Voice: +49 69 7505 3213 | andreas . ames AT comergo . com "I imagine Sisiphus as happy ..." Albert Camus

1 0

Re: [Boost-users] New User With Busted Compiler
by RIVASSEAU Jean Noel 10 Mar '06

10 Mar '06

I never used uBLAS, but the documentation for it in the 1.33.1 distribution says that gcc 3.3 is supported. Jean-Noël -----Message d'origine----- De : boost-users-bounces(a)lists.boost.org [mailto:boost-users-bounces@lists.boost.org] De la part de V. Thane Norton III Envoyé : vendredi 3 mars 2006 02:27 À : Boost-users(a)lists.boost.org Objet : [Boost-users] New User With Busted Compiler I have just started using Boost, and am really enjoying using it, especially uBLAS. Unfortunately, my production code needs to use GCC 3.3 (for marketing reasons) which the Boost 1.33.1 uBLAS does not support. What is the latest version of Boost that I can use for uBLAS on GCC 3.3? Is there some hackery that I can perform to get 1.33.1 working? ----------------------------------------- V. Thane Norton III Macintosh Software Engineer Wacom Technology Corporation phone: 360.896.9833 x172 mailto:tnorton@wacom.com _______________________________________________ Boost-users mailing list Boost-users(a)lists.boost.org http://lists.boost.org/mailman/listinfo.cgi/boost-users

2 1

[Announce] Mailing list downtime this Saturday
by Doug Gregor 10 Mar '06

10 Mar '06

All Boost mailing lists hosted at lists.boost.org will be down from 8pm to midnight EST this Saturday, March 11th while we migrate to a new, dedicated mail server provided by the Open Systems Lab at Indiana University. During this down time, the mail archives at lists.boost.org will be unavailable and any mail sent to lists.boost.org will be queued and forwarded. We ask that members refrain from posting messages to lists.boost.org during this time "just in case," although we expect that all mail sent during this time will be forwarded properly. If you should have any questions, please contact me directly at dgregor(a)cs.indiana.edu. Doug Gregor Boost Moderator dgregor(a)cs.indiana.edu

1 0