August 2006 - Boost-users - lists.preview.boost.org

Design for boost::serialization of STL iterators.
by Carlo Wood 23 May '09

23 May '09

On Sat, Mar 11, 2006 at 11:45:36PM +0200, Peter Dimov wrote: > Carlo Wood wrote: > > > 1) WHY does boost::serialization not support serialization > > of iterators? > > It's very hard, if not impossible, in general. It is either impossible or possible. When it is possible, it can't be hard :p I think that at the moment of serialization, the actual container needs to be known. So, only concentrating on the syntax of the code that serializes an iterator, the following cannot work: ar & iter; However, the container will be known at that point (or at least, we might as well assume that). Therefore, I could imagine a syntax like: ar & foobar(container, iter); where 'foobar' is some template function (that will be part of boost::serialization in the future). Lets start with the last one (have to start somewhere). In order to deserialize the iterator (and taken into account that we already have the container) -- which makes me think that it might be better to pass a functor to operator& above. In other words, (at least for writing), use: template<class Container> struct Foobar { Foobar(Container const&, typename Container::const_iterator& iter) : M_container(container), M_iter(iter); private: Container const& M_container; Container::const_iterator& M_iter; }; and then assign to M_iter, whatever we find in M_container. > If you have access to the container to which the iterator i refers: > > a) if the container is a map, you can serialize i->first, then on > deserialization look up the key in the map using .find(); > > b) if the container is a sequence, and it has the same contents on > deserialization as it does on serialization, you can serialize the index of > the element - distance( c.begin(), i ). I don't think that it is a requirement that the whole container has to be read: either it's a map or set, in which case later insertions will not invalidate the container, or it is a vector, in which case we use an index, from which we can construct the end iterators before the whole container is reconstructed. Therefore, the following syntax can work imho: ar & FooBar(container, iter); Would you agree? There are a zillion questions and implementation details open - but I'd like to get anyones feedback about this type of interface for iterators before continuing. Current questions: What is a good name for Foobar? Am I right when I think that Foobar needs to be declared in boost::serialization? Now the hard part (for me) is that part that involves the internals of the boost library. At the moment one tries to serialize an iterator, it is possible (though not necessarily) that the object that the iterator points to is already serialized. In the case that we are dealing with an associative container (map or set etc), we don't want to store the contents of the object (or key part, in the case of a map) again: I'd like to just store that once, and store a reference in the archive at the other place. How can I make the library store an object only the first time and the store a reference the second time? This would be for an object of arbitrary type thus. I think that it is only feasible to do this when it's actually the same object (same memory address), which is the case; but while storing the object element of the container, the address of it isn't stored as well, is it? Is there a way to make: std::set<SomeData> s; [...] ar << s; Store the address of s (&s) along with the contents of s? And would that then automatically cause the library to not store that same object again when I do: std::set<SomeData>::iterator iter = s.find(...); [...] ar << *iter; Because then, &*iter would again be the same memory address as what was stored before. -- Carlo Wood <carlo(a)alinoe.com>

3 2

Boost Graph Library and Boehm Garbage Collector
by FBSL 08 Sep '08

08 Sep '08

Hello I am having trouble to garbage-collect adjacency_list<>. I have instanciated it with correctly garbage-collected vectors and lists (I have copied vecS and listS definitions, and added Boehm GC's default allocator so that they are garbage-collected). In my application I need to dynamically allocate many graphs, but it does not work, there is a (big) memory leak. Unfortunately, there is no template argument in adjacency_list<> to specify Boehm GC's allocator. Does anyone have a hint ? Snippet code "main.cpp" follows, compiled with Boost Graph Library 1.32, Boehm GC 6.4 and G++ 3.3.5, with command line : $ g++ -o main main.cpp -lgc -lgccpp -- #include <boost/graph/adjacency_list.hpp> #include <gc/gc_allocator.h> #include <gc/gc_cpp.h> #include <iostream> #include <vector> #include <list> using namespace std; using namespace boost; namespace boost { struct collected_vecS {}; struct collected_listS {}; template <class ValueType> struct container_gen<collected_vecS, ValueType> { typedef vector<ValueType,gc_allocator<ValueType> > type; }; template <class ValueType> struct container_gen<collected_listS, ValueType> { typedef list<ValueType,gc_allocator<ValueType> > type; }; template <> struct parallel_edge_traits<collected_vecS> { typedef allow_parallel_edge_tag type; }; template <> struct parallel_edge_traits<collected_listS> { typedef allow_parallel_edge_tag type; }; namespace detail { template <> struct is_random_access<collected_vecS> { enum { value = true }; typedef true_type type; }; } } typedef adjacency_list<collected_vecS, collected_vecS, directedS, no_property, no_property, no_property, collected_listS> graph; int main() { // leaks... while(true){ graph * g=new (UseGC) graph; for(unsigned int i=0; i<100; ++i){ add_vertex(*g); } for(unsigned int i=0; i<100; ++i){ for(unsigned int j=0; j<100; ++j){ add_edge(i,j,*g); } } } }

3 4

[statechart] history transitions make state-local-storage "un-cool"
by Mick Hollins 26 Jun '08

26 Jun '08

Apologies if this has been discussed before, but I couldn't find any mention of it: I am currently using the history facility of boost::fsm. I've discovered that when I re-enter a state, X, via a history transition I don't actually re-enter the same instance of X that I was in before, but instead enter a newly constructed instance of X. I just checked the UML spec regarding history transitions and it says that "Any necessary entry actions are performed", so given that boost::fsm uses constructors to represent entry actions I guess a new instance of X has to be constructed to ensure UML compliance. My actual concern is the effect this has on the "coolness" of state-local-storage (see http://boost-sandbox.sourceforge.net/libs/statechart/doc/faq.html#Whats_so_…) It seems to me that state-local-storage, which I agree is generally cool, is particularly un-cool (and bug-inducing) in the presence of history transitions. I'm not sure what a good alternative is, but just thought I'd raise the discussion. cheers, mick PS: does the history facility display the same semantics in boost::statechart? I plan to move to statechart real soon now ...

4 19

BGL: INTERNAL COMPILER ERROR in VC7.1
by Yariv Tal 10 Jun '07

10 Jun '07

This one was already posted by Nicholas Burlett awhile back (see http://lists.boost.org/MailArchives/boost-users/msg06534.php) but I have some new info that may help. The problem: ========= When compiling code that uses the boost.graph library the VC7.1 compiler shouts: c:\boost_1_32_0\boost\mpl\aux_\preprocessed\plain\apply_wrap.hpp(48) : fatal error C1001: INTERNAL COMPILER ERROR As Nicholas already noted, this seems to be due to some compiler problem with the boost.graph out_edge_iterators since it is caused by the following lines in my code: (FilteredGraphGenerator::FilteredGraph is a typedef of a boost::filtered_graph) typedef boost::graph_traits<FilteredGraphGenerator::FilteredGraph> ::out_edge_iterator oedge_iterator; oedge_iterator edgesItr, edgesEnd; boost::tie(edgesItr, edgesEnd) = boost::out_edges(itr->first, m_g.m_filteredGraph); while (edgesItr != edgesEnd) { // do anything! } In fact, it seems to be related to the operator!= between iterators (if the while loop is commented out it compiles). How ever, the error persists even after changing the operator!= to operator== : while (edgesItr == edgesEnd) { // do anything! } ========================================================================== Some findings: =========== * We used to use boost_1_30_0 and the code used to compile PERFECTLY. The problem only appeared after migrating to boost_1_32_0. * Calling directly to the "equal" function in iterator_core_access instead of the operator== works. typedef boost::graph_traits<FilteredGraphGenerator::FilteredGraph> ::out_edge_iterator oedge_iterator; oedge_iterator edgesItr, edgesEnd; boost::tie(edgesItr, edgesEnd) = boost::out_edges(itr->first,m_g.m_filteredGraph); while (boost::iterator_core_access::equal(edgesItr, edgesEnd, boost::mpl::true_())) { // do somthing } * The error message location is at the mpl::apply_wrap2 function. Any relation to the mpl::apply2 function used for declaring the operator== in the BOOST_ITERATOR_FACADE_INTEROP_HEAD macro? Have these functions changed since v1.30.0?

9 8

[statechart] - problem with sc::transition<> -> invalid use of 'boost::statechart::detail::no_context::no_function [with Event = jm::ffcp::ev_init]' to form a pointer-to-member-func
by Oliver.Kowalke＠qimonda.com 15 Mar '07

15 Mar '07

Hello, I get following error: /opt/boost/include/boost-1_33_1/boost/statechart/transition.hpp:33: error: invalid use of 'boost::statechart::detail::no_context::no_function [with Event = jm::ffcp::ev_init]' to form a pointer-to-member-function /opt/boost/include/boost-1_33_1/boost/statechart/transition.hpp:33: note: a qualified-id is required if I try to compile: #ifndef JM_FFCP_UNINITIALIZED_H #define JM_FFCP_UNINITIALIZED_H #include <boost/mpl/list.hpp> //#include <boost/statechart/custom_reaction.hpp> #include <boost/statechart/simple_state.hpp> #include <boost/statechart/transition.hpp> #include <jm/ffcp/events.h> #include <jm/ffcp/proto.h> namespace jm { namespace ffcp { namespace mpl = boost::mpl; namespace sc = boost::statechart; template< typename SocketDeviceT > struct initialized; template< typename SocketDeviceT > struct uninitialized : public sc::simple_state< uninitialized< SocketDeviceT >, proto< SocketDeviceT > > { typedef mpl::list< // sc::custom_reaction< ev_init > sc::transition< ev_init, initialized< SocketDeviceT > > > reactions; uninitialized() { uninitialized< SocketDeviceT >::custom_static_type_ptr("uninitialized"); } // sc::result react( ev_init const& ev) // { return this->template transit< initialized< SocketDeviceT > >(); } }; } } #endif // JM_FFCP_UNINITIALIZED_H Any hints? regards, Oliver

6 10

[Threads] Conditions and avoiding lengthy mutex locks (background loading for a game)
by Peter Mackay 11 Mar '07

11 Mar '07

Hello, I'm trying to use the Boost Threads library to help me implement a background loading system for a game I'm writing. I want to enqueue load requests while the main game thread is running, and have a separate thread load the data and pass back the appropriate objects. Currently when I enqueue a new request, I notify a 'something to do' condition, which my loading thread waits on. The trouble is, when I notify it, the loader could be loading in the background, and has a lock on the mutex connected with the condition. Would anyone be able to offer me any advice on how to solve this? Many thanks, Peter P.S. Please find attached my current implementation. begin 666 loader.cpp M(VEN8VQU9&4@(FQO861E<BYH<' B#0H-"B-I;F-L=61E(#QB;V]S="]B:6YD M+FAP<#X-"@T*;&]A9&5R.CIL;V%D97(H*0T*.B!T:')E861?<VAO=6QD7V5X M:70H9F%L<V4I#0I[#0H)=&AR96%D+G)E<V5T*&YE=R!B;V]S=#HZ=&AR96%D M*&)O;W-T.CIB:6YD*"9L;V%D97(Z.G1H<F5A9%]P<F]C+"!T:&ES*2DI.PT* M?0T*#0IL;V%D97(Z.GYL;V%D97(H*0T*>PT*"2\O(%-I9VYA;"!T:&4@=&AR M96%D('1O(&5X:70N#0H)>PT*"0EB;V]S=#HZ;75T97(a)Z.G-C;W!E9%]L;V-K M(&QO8VLH;75T97(a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` ` end begin 666 loader.hpp M(VEF;F1E9B!,3T%$15)?2%!0#0HC9&5F:6YE($Q/041%4E](4% -"@T*(VEN M8VQU9&4@/'%U975E/@T*(VEN8VQU9&4@/&)O;W-T+W-H87)E9%]P='(N:'!P M/@T*(VEN8VQU9&4@/&)O;W-T+W1H<F5A9"]C;VYD:71I;VXN:'!P/@T*(VEN M8VQU9&4@/&)O;W-T+W1H<F5A9"]M=71E>"YH<' ^#0HC:6YC;'5D92 \8F]O M<W0O=&AR96%D+W1H<F5A9"YH<' ^#0H-"F-L87-S(&QO861E<B Z('!R:79A M=&4@8F]O<W0Z.FYO;F-O<'EA8FQE#0I[#0IP=6)L:6,Z#0H-"@EL;V%D97(H M*3L-"@EV:7)T=6%L('YL;V%D97(H*3L-"@T*<')I=F%T93H-"@T*"79O:60) M=&AR96%D7W!R;V,H*3L-"@T*"2\O($%C8V5S<VEB;&4@8GD@86YY('1H<F5A M9"X-"@EB;V]S=#HZ;75T97@)"6UU=&5X.PT*"6)O;W-T.CIC;VYD:71I;VX) M<V]M971H:6YG7W1O7V1O.PT*"6)O;VP)"0D)=&AR96%D7W-H;W5L9%]E>&ET M.PT*"7-T9#HZ<75E=64\:6YT/@D)<&5N9&EN9U]L;V%D<SL-"@T*"2\O($%C M8V5S<VEB;&4@8GD@=&AE(&UA:6X@=&AE860@;VYL>2X-"@EB;V]S=#HZ<VAA M<F5D7W!T<CQB;V]S=#HZ=&AR96%D/@ET:')E860[#0I].PT*#0HC96YD:68- !"@`` ` end

4 6

How can we serialize pointer of type base abstract class pointing to Derived concrete class
by Piyush Kapadia 18 Jan '07

18 Jan '07

We have a situation where there is BaseAbstract class and Derived concrete class. Interface requires that we pass BaseAbstractClassPointer, using which user can provide any of Derived Concrete class pointer and Interface remain generic to them. However serializing BaseAbstractPointer is throwing following error - *A unregistered class* How do we handle this situation ? #include <iostream> #include <fstream> #include <boost/archive/text_iarchive.hpp> #include <boost/archive/text_oarchive.hpp> #include <boost/archive/xml_iarchive.hpp> #include <boost/archive/xml_oarchive.hpp> #include <boost/archive/binary_iarchive.hpp> #include <boost/archive/binary_oarchive.hpp> #include <boost/serialization/vector.hpp> #include <boost/serialization/map.hpp> #include <vector> #include <map> #include <boost/archive/archive_exception.hpp> #include <boost/archive/basic_xml_archive.hpp> class BaseAbstract { public : virtual void setRate(double rate) = 0; virtual double getRate(int) = 0; }; class DerivedConcrete : public BaseAbstract { std::vector<double> rateVec; friend class boost::serialization::access; template<class Archive> void serialize(Archive & ar, const unsigned int version) { ar & BOOST_SERIALIZATION_NVP(rateVec); } public: DerivedConcrete(double rate) {rateVec.push_back(rate);} DerivedConcrete(){} virtual ~DerivedConcrete(){} void setRate(double rate){ rateVec.push_back(rate); }; double getRate(int idx){ return rateVec[idx]; }; }; class AnyClass { friend class boost::serialization::access; BaseAbstract* rate; template<class Archive> void serialize(Archive & ar, const unsigned int version) { ar & BOOST_SERIALIZATION_NVP(rate); } public: AnyClass(BaseAbstract* rateInput){ rate = rateInput; }; AnyClass() {}; setRate(BaseAbstract* rateInput){ rate = rateInput; }; }; int main(int argc, char* argv[]) { try { DerivedConcrete* fr = new DerivedConcrete(12.12); fr->setRate(14.5); AnyClass ac; ac.setRate(fr); std::ofstream ofs("AC.xml"); assert(ofs.good()); boost::archive::xml_oarchive oa(ofs); oa << BOOST_SERIALIZATION_NVP(ac); ofs.close(); } catch (boost::archive::xml_archive_exception ex) { std::cerr << " X " << ex.what() << std::endl; } catch (boost::archive::archive_exception ex) { std::cerr << " A " << ex.what() << std::endl; } catch (std::exception ex) { std::cerr << " E " << ex.what() << std::endl; } catch (...) { std::cerr << "Unknown Exception " << std::endl; } return 0; }

3 4

Re: [Boost-users] Visual C++ 2005 help installing Boost
by Harry Georgakopoulos 17 Jan '07

17 Jan '07

I am having exactly the same problem as a previous user. Could you please show me how you fixed your boost build? I'm getting the following errors ...failed updating 120 targets ...skipped 80 targets... ...updated 847 targets... I don't know how to open the vsvars32.bat file in order to change the paths that Mr Kozicki suggested. Any help would be greatly appreciated. I have installed the latest SDK. Also, how do i get rid of the warnings? I know i need to define _SCL_SECURE_NO_DEPRECATE globally, but how do i do that exactly? Finally, how do i know i have a correct build install? How do i test it? Thank you Please send your response to hgeorgako(a)hotmail.com harry g.

5 5

vertex descriptors
by Luis Quesada 18 Dec '06

18 Dec '06

Dear all, Sorry for the newbie question! Given the index of a vertex, how can I get its descriptor? I guess function vertex (page 225 of The Boost Graph Library) does that but the function is not defined. In fact, what I want to do is to check whether an edge belongs to a graph in a context where descriptors get invalidated after vertex removals? Suppose I have done this: // create a typedef for the Graph type typedef adjacency_list<vecS, vecS, bidirectionalS> Graph; // Make convenient labels for the vertices enum { A, B, C, D, E, N }; const int num_vertices = N; const char* name = "ABCDE"; // writing out the edges in the graph typedef std::pair<int, int> Edge; Edge edge_array[] = { Edge(A,B), Edge(A,D), Edge(C,A), Edge(D,C), Edge(C,E), Edge(B,D), Edge(D,E) }; const int num_edges = sizeof(edge_array)/sizeof(edge_array[0]); // declare a graph object Graph g(num_vertices); Then I remove some vertices. Now I want to check whether edge <A,B> is still there. Calling edge(A,B,g) would not be OK because the descriptors are no longer referring to the same vertices.... Thanks in advance for your answer, Luis

6 15

The fibonacci heap is behaving strange
by Magnus Ekdahl 15 Dec '06

15 Dec '06

Hi, I put in 0,1,2,3,4 into a Fibonacci heap, manipulate some and get 0,2,4,1,25 back. With the relaxed heap I get 0,2,4,1,3 back, which is as expected with my code. AFAIK this is due to a bug in the Fibonacci heap? Anyway here is code adopted from the cvs (fibonacci_heap.cpp) that reproduces the result on two different machines (intel/amd), where the programs are compiled using g++ 4.0.4. Any help in clearing out this confusing result is appreciated :) #include <iostream> #include <cstdlib> #include <vector> #include <boost/pending/Fibonacci_heap.hpp> #include <boost/pending/relaxed_heap.hpp> #include <boost/graph/graph_utility.hpp> #include <boost/pending/indirect_cmp.hpp> using namespace std; using namespace boost; int main(int argc, char *argv[]) { typedef indirect_cmp<unsigned*,std::less<unsigned> > ICmp; unsigned current; const unsigned N=5; vector<unsigned> w(N); ICmp cmp(&w[0], std::less<unsigned>()); Fibonacci_heap<unsigned, ICmp> Q(N, cmp); //relaxed_heap<unsigned, ICmp> Q(N, cmp); Q.push(0); current=Q.top(); cout << current << " "; Q.pop(); w[1]=180428938; Q.push(1); w[2]=84693088; Q.push(2); w[3]=168169277; Q.push(3); w[4]=171463691; Q.push(4); current = Q.top(); cout << current << " "; Q.pop(); w[3]=164976049; Q.update(3); w[4]=59651664; Q.update(4); current = Q.top(); cout << current << " "; Q.pop(); w[1]=71988538; Q.update(1); w[3]=118964142; Q.update(3); current = Q.top(); cout << current << " "; Q.pop(); w[3]=42423833; Q.update(3); current = Q.top(); cout << current << " "; Q.pop(); return EXIT_SUCCESS; }

4 5