- Boost-users - lists.boost.org

[Graph] Defining coordinates for vertices
by Maxime van Noppen 17 Jul '09

17 Jul '09

Hi, I can't figure out if it is possible (and if it is, how) to assign coordinates to graph vertices on planar graphs. I've tried setting the property vertex_index_t to std::pair<int, int> but it doesn't seem to work (make_maximal_planar doesn't use them for example). I'm trying to represent a 2D terrain via a triangular mesh. So far it is easy : I add all the points of the mesh as vertices and create the appropriated edges. But now I'm trying to make this graph evolve, which means adding new vertices (at some precise coordinates) and having them automagically linked to all other vertices possible while keeping the graph planar. This is what I do for now but that doesn't work (full source code is attached) : ------------------ typedef adjacency_list< vecS, vecS, undirectedS, property< vertex_index_t, std::pair<int, int> >, property<edge_index_t, int> > graph_t; typedef graph_traits<graph_t>::edge_descriptor edge_t; typedef graph_traits<graph_t>::vertex_descriptor vertex_t; // Create the graph graph_t g; // Add the three vertices of a triangle add_vertex(std::make_pair(0, 0), g); add_vertex(std::make_pair(10, 0), g); add_vertex(std::make_pair(5, 10), g); // Create the three edges add_edge(0, 1, g); add_edge(0, 2, g); add_edge(1, 2, g); // Add some random points add_vertex(std::make_pair(5, 5), g); add_vertex(std::make_pair(0, 5), g); // The goal is now to make g planar w.r.t. the coordinates // 1- Either make_connected or add at least one edge to make the graph // connected // 2- make_biconnected_planar // 3- make_maximal_planar ------------------ The input graph is (using boost::print_graph) : 0 <--> 1 2 1 <--> 0 2 2 <--> 0 1 3 <--> 4 <--> The output graph is : 0 <--> 1 2 3 4 1 <--> 0 2 3 2 <--> 0 1 3 4 3 <--> 2 4 0 1 4 <--> 3 2 0 It isn't palanar w.r.t. the given coordinates because 3 can't be connected with 4 without crossing one of the edges of the triangle (3 is supposed to be inside the triangle whereas 4 is outside). I suppose that either I'm doing something wrong or that it might not be possible to use the library this way. Any help appreciated, Thanks ! -- Maxime

2 4

[numeric_cast] Warning with MSVC
by Alexander Lamaison 17 Jul '09

17 Jul '09

I'm using boost::numeric_cast to convert the size of a wstring (presumably a size_t) to an int. const int size = boost::numeric_cast<int>(my_string.size()); MSVC still gives me a warning about possible loss of data: c:\program files\boost\boost_1_38\boost\numeric\conversion\ cast.hpp(41) : warning C4267: 'argument' : conversion from 'size_t' to 'unsigned int', possible loss of data c:\users\awl03\documents\visual studio 2005\projects\swish_trunk (padd)\swish\utils.hpp(91) : see reference to function template instantiation 'Target boost::numeric_cast<int,__w64 unsigned int>(Source)' being compiled with [ Target=int, Source=__w64 unsigned int ] c:\users\awl03\documents\visual studio 2005\projects\swish_trunk (padd)\swish\utils.hpp(121) : see reference to function template instantiation 'std::basic_string<_Elem,_Traits,_Ax> swish::utils::ConvertString<`anonymous-namespace'::Narrow>(const std::basic_string<wchar_t,std::char_traits<wchar_t>,std:: allocator<_Ty>> &)' being compiled with [ _Elem=char, _Traits=std::char_traits<char>, _Ax=std::allocator<char>, _Ty=wchar_t ] Unless I misunderstand, isn't the idea of numeric_cast that warnings aren't needed because the casts are checked for loss of data? Any advice is greatly appreaciated. I am using the MSVC 8.0 compiler with Boost 1.38.0 on 32-bit Windows Vista. Thanks. Alex Lamaison

3 3

BOOST_CLASS_TRACKING compilation error : tracking_level not a class or a namespace
by G, Benjamin (NSN - IN/Bangalore) 17 Jul '09

17 Jul '09

Hi, I want to disable tracking because of the now somewhat famous (STATIC_ASSERT compilation errors). I have a few classes within a namespace in a header file. For serialization of these, I want to disable tracking so I placed the macros: BOOST_CLASS_TRACKING(mynamespace::myClassA, boost:serialization::track_never); BOOST_CLASS_TRACKING(mynamespace::myClassB, boost:serialization::track_never); outside mynamespace. For this I am getting compilation errors saying : 'struct boost::serialiazation::tracking_level<mynamespace::myClassA>::type' is not a class or a namespace. I've included the boost header files, etc. in my header file. What am I missing? All help appreciated. - Benjamin Jacob.

2 2

[Serialization] Questions / Comments
by Rutger ter Borg 17 Jul '09

17 Jul '09

After having used the serialization library for quite a while, it seems that I'm approaching a point where I are seeing lots duplicate code which could be inside a serialization function. In turn, this head lead me to some questions / comments. The use of serialization in my case is both file and network oriented. This means reading and writing a great deal of short-lived archives. Works fine in between c++ programs, but outside that it's quickly becoming an infeasible task. Creating a custom archive could be an option, however, I would need to spend serious time on understanding the code due to the heavy use of macros. The question to be answered first is whether the current serialization architecture is able to fulfill my needs, or whether I should consider using other options. Considering the following two scenarios / questions: 1) Suppose I would like to create a custom XML archive which can both read shuffled XML data (its attributes/values in arbitrary order), and serialize to child attributes as well as named child nodes. Currently, name-value pairs are in the Serializable concept. I think they belong in the Archive Concept. Otherwise, it may be impossible to read XML- formatted data where the nodes are not in the same order as in the C++ code? In addition, suppose serializing a program structure is like a tree or graph, then name-value-pairs do not allow for any other nesting than the C++ code (i.e., it assumes everything is a child node). To allow for making distinction between nesting levels, one could think of, e.g., template< typename Archive > void serialize( Archive& ar, const unsigned int version ) { ar.serialize_node( "subtree", m_contained_class ); ar.serialize_attribute( "use_count", m_int ); } Is something like this possible? To this end, would it be possible to plug- in another parser, e.g., RapidXML? 2) I would like to create an archive that speaks the Action Message Format (AMF). Although I do have trouble finding a clear distinct parsing layer in the archives, I think this is possible to do. However, perhaps you think archive/serialization should not be used in conjunction with these kinds of uses. Many thanks in advance, Kind regards, Rutger ter Borg

2 3

Phoenix 2, 3 and related stuff
by joel 17 Jul '09

17 Jul '09

I'm currently fiddling with Phoenix 2 and i ended up needing look over my lambda statement to reorder/optimize them. AFAIK, Phoenix 2 is not proto-powered but phoenix 3 is not out yet. What should be the best strategy for me right now : waiting for Phoenix 3 or is there a way to hack inside Phoenix 2 so i can have a rough look at whixh type to match in my meta-function ? -- ___________________________________________ Joel Falcou - Assistant Professor PARALL Team - LRI - Universite Paris Sud XI Tel : (+33)1 69 15 66 35

3 6

serialization std::map
by Adam Szeliga 17 Jul '09

17 Jul '09

Why I can not input from archive std::map ? line: inputArchive >> inputContainer causes compile error: Thanks Adam #include <string> #include <iostream> #include <fstream> #include <sstream> #include <map> #include <boost/serialization/map.hpp> #include <boost/serialization/serialization.hpp> #include <boost/serialization/string.hpp> #include <boost/serialization/level.hpp> #include <boost/serialization/tracking.hpp> #include <boost/serialization/split_free.hpp> #include <boost/serialization/base_object.hpp> #include <boost/archive/text_iarchive.hpp> #include <boost/archive/text_oarchive.hpp> #include <boost/archive/text_wiarchive.hpp> #include <boost/archive/text_woarchive.hpp> class A { public: A() {}; A(std::string const & s): s(s) { } virtual ~A() {}; private: std::string s; friend std::ostream & operator<<(std::ostream & os, A const & a); friend class boost::serialization::access; template<class Archive> void serialize(Archive & archive, unsigned int const version) { archive & s; } }; std::ostream & operator<<(std::ostream & os, A const & a) { return os << a.s; } typedef std::map<unsigned int const, A * const> Container; BOOST_CLASS_TRACKING(Container, boost::serialization::track_never); std::ostream & operator<<(std::ostream & os, Container const & container) { Container::const_iterator end = container.end(); for (Container::const_iterator i = container.begin(); i != end; ++i) { os << "(" << i->first << "," << *i->second << ")" << std::endl; } return os; } int _tmain(int argc, _TCHAR* argv[]) { std::cout << "serialization map test starts" << std::endl; Container outputContainer; outputContainer.insert(std::make_pair(1, new A("one"))); outputContainer.insert(std::make_pair(4, new A("four"))); outputContainer.insert(std::make_pair(2, new A("two"))); std::cout << outputContainer << std::endl; { std::wofstream outputFile("data.txt"); boost::archive::text_woarchive outputArchive(outputFile); outputArchive << outputContainer; } std::cout << "---------------------------" << std::endl; Container inputContainer; { std::wifstream inputFile("data.txt"); boost::archive::text_wiarchive inputArchive(inputFile); inputArchive >> inputContainer; // this line causes compile error } std::cout << inputContainer << std::endl; std::cout << "serialization map test ends" << std::endl; return 0; }

2 1

Re: [Boost-users] [Regex] regex_search
by Etienne Philip Pretorius 16 Jul '09

16 Jul '09

Never mind, I worked around the problem. Thank you, Etienne

1 0

[Regex] regex_search
by Etienne Philip Pretorius 16 Jul '09

16 Jul '09

Hello List, I am having an issue compiling the my code - I get the following error: xml.hpp:43: error: no matching function for call to ‘regex_search(boost::asio::buffers_iterator<boost::asio::const_buffers_1, char>&, boost::asio::buffers_iterator<boost::asio::const_buffers_1, char>&, boost::match_results<boost::asio::buffers_iterator<boost::asio::const_buffers_1, char>, std::allocator<boost::sub_match<boost::asio::buffers_iterator<boost::asio::const_buffers_1, char> > > >&, boost::_bi::bind_t<boost::regex&, boost::_mfi::mf0<boost::regex&, xmpp::xml::parser>, boost::_bi::list1<boost::_bi::value<const xmpp::xml::parser*> > >, boost::regex_constants::match_flags)’ Here is the related code snippet: template <typename iterator> std::pair<iterator, bool> operator()( iterator begin, iterator end ) const { if(begin == end) { return std::make_pair(end,false); } boost::match_results<iterator> match_results; //Compiler Error Start if(boost::regex_search( begin, end, match_results, boost::bind( &parser::regex, this ), boost::match_default | boost::match_partial)) { //Compiler Error End if(match_results[0].matched) { //test return std::make_pair(match_results[0].second,true); }else{ //test return std::make_pair(match_results[0].first,false); } } return std::make_pair(end,false); }; private: boost::regex regex( ){ return XMLDecl_; }; static boost::regex S_, Eq_, EncName_, VersionInfo_, EncodingDecl_, SDDecl_, XMLDecl_, UTF8_, Char_, NameStartChar_, NameChar_, Name_, Names_, Comment_, PITarget_, PI_, Misc_, prolog_; Could someone please shed some light on this problem for me. I want to be able to change the regular expression based on a state variable so that operator() can advance to the next xml type that it should/could encounter. Thank you. Etienne

1 0

(newbie) boost thread compile problems on Fedora 8
by Kishalay Kundu 16 Jul '09

16 Jul '09

Hi, I am having trouble compiling a correct version of the boost thread library on a x86_64 Fedora Core 8 and was wondering if anyone can help me out. I've tried both version 1.37 and 1.39. I describe below what I do and what happens after that. I am on an AMD64 Fedora Core 8 platform (with gcc 4.1.2 and glibc 2.7) that already had boost 1.34 on it at /usr/lib level. I erased that version by deleting all the libraries and moving 'boost' directory in /usr/include to 'boost_1-34' directory. I download the .gz tarball, bootstrap it, make it (with make for 1.37 and bjam for 1.39) and install it to /usr/local/lib. The compilation goes without a hitch (except for some warnings for boost serialization and others, but not for thread). I have both .a and .so files generated and residing in /usr/local/lib and boost include files in /usr/local/include. I add the links libboost_thread-mt.a and libboost_thread-mt.so to point to libboost_thread-gcc41-mt_1.3x.* libs. Now, I write a small test program (given below). It gives me linker errors with undefined references to pretty much every pthread symbol. If anyone wants I can paste all the errors in the next message. When I add the -lpthread flag to the compile, it gives me an undefined reference error with regards to __vdso_clock_gettime from /lib64/libpthread.a. I'm not sure where exactly the problem lies and would greatly appreciate if someone could point me in the right direction. Thank you. regards, Kishalay My test program: compiled with: gcc -o testk test.cpp -lboost_thread-mt #include <cstdio> #include <cstdlib> #include <boost/thread.hpp> using namespace std; using namespace boost; static void somefunc( ) { printf( "thread calling this\n" ); } int main( int argc, char** argv ) { boost::thread* t = new thread( somefunc ); delete t; }

3 4

boost 1.39 hp-ux support dropped ?
by Avi Bahra 16 Jul '09

16 Jul '09

It appears that boost 1.37, 1.38 all support HP-UX as a primary compiler. However 1.39 release notes, make no mention of HP-UX at all? is this deliberate or an oversight? Best regards, Ta, Avi

2 1