December 2005 - Boost-users - lists.preview.boost.org

How get the size of a stream< io::array_sink > memoryStream?
by Rodrigo Baroni 13 Dec '05

13 Dec '05

Hi all, How can I know the size of a stream< io::array_sink > memoryStream( buffer, size ) , after some operations that has filled the stream (the memory) ? Is there some way ? Example: //--------------- Compress the content of buffer to compressedBuffer---------------------- char buffer = "Helloooo compressed filter example!"; int SIZE = 10000; char compressedBuffer = new char [ SIZE ]; io::stream< io::array_source > bufferStream( buffer, std::strlen( buffer ) ); io::stream< io::array_sink > compressedBufferStream( &compressedBuffer, SIZE ); io::filtering_ostream compressFilterOutput; compressFilterOutput.push( io::zlib_compressor( ) ); compressFilterOutput.push( compressedBufferStream ); io::copy( bufferStream, compressedFilterOutput ); compressedBufferStream.flush( ); compressFilterOutput.reset( ); //----------------------------- How can I know the size of the compressedBufferStream, or compressedBuffer now? std::strlen( compressedBuffer ) gives me wrong value. Thanks in advance, Rodrigo Ferreira Baroni

2 4

[serialization] Removing objects from an archive
by Przemyslaw Wojnowski 13 Dec '05

13 Dec '05

It helps a lot. But now I have another question. Could archive interface be changed/extended to support removing from an archive, so interface realization will be able to save/load/remove data selectively? If not, why? Przemek

2 1

remove_if and transform_iterator
by Josue Gomes 13 Dec '05

13 Dec '05

Hi, Can transform_iterator be used in conjunction with remove_if algorithm? The compiler (vc7.1, vc8, g++) is refusing to compile this: #include <string> #include <utility> #include <vector> #include <functional> #include <cassert> #include <iostream> #include <boost/iterator/transform_iterator.hpp> std::vector<std::pair<std::string, int> > c; struct to_num { typedef int result_type; int operator()(const std::pair<std::string, int>& p) const { return p.second; } }; int main() { using std::make_pair; c.push_back(make_pair("A", 2)); c.push_back(make_pair("A", 3)); c.push_back(make_pair("A", 4)); c.push_back(make_pair("A", 15)); c.push_back(make_pair("A", 16)); c.push_back(make_pair("A", 17)); assert(c.size() == 6); std::remove_if( boost::make_transform_iterator(c.begin(), to_num()), boost::make_transform_iterator(c.end(), to_num()), std::bind1st(std::less_equal<int>(), 10)); assert(c.size() == 6); } Regards, Josue

2 2

documentation available for forthcoming boost:fsm?
by Oliver.Kowalke＠infineon.com 13 Dec '05

13 Dec '05

Hello, where can I find the documentation fort he forthcoming fsm library? http://boost-sandbox.sf.net/Statechart.zip contains no docu Best Regards, Oliver

2 1

[mpl] remove duplicates in (non-integral) type sequences
by Karl Friedrich Walkow 13 Dec '05

13 Dec '05

i try to find a solution for eliminating all duplicates of types in a non-integral sequence. the only extisting solution i found has a set< > as output, but i want to keep the sequence type. the following code _should_ solve the problem, but there seems to be something wrong with it. if someone has an idea... #include <boost/mpl/assert.hpp> #include <boost/mpl/vector.hpp> #include <boost/mpl/fold.hpp> #include <boost/mpl/empty_sequence.hpp> #include <boost/mpl/contains.hpp> #include <boost/mpl/if.hpp> #include <boost/mpl/joint_view.hpp> #include <boost/mpl/single_view.hpp> #include <boost/mpl/size.hpp> using namespace boost::mpl; struct E {}; struct F {}; struct G {}; struct H {}; int main(int argc, char* argv[]) { typedef vector< E, E, F, E, G, H, G, G, F, G >::type with_duplicates; typedef fold< with_duplicates, empty_sequence, if_< contains< _1, _2 >::type, _1, joint_view< _1, single_view< _2 >::type >::type >::type >::type without_duplicates; BOOST_MPL_ASSERT_RELATION(size<with_duplicates>::value, ==, 10); BOOST_MPL_ASSERT_RELATION(size<without_duplicates>::value, ==, 4); return 0; } much thanks in advance, karl

2 2

Re: [Boost-users] serialization: loading std::map replaces existing mapdata objects
by Peter Whaite 13 Dec '05

13 Dec '05

Robert Ramey <ramey(a)rrsd.com> wrote: > the copy_collection was an experiment on work has been suspended and > probably not be implemented, and if it were implemented, would be > implemented differently in anycase. Ok. Thanks. > b) Just specify a serialize function for the particular map you are > using. > e.g. > > template<class Archive> > void serialize(Archive & ar, std::map<my_class> > & t, const unsigned int > version){ > // for each member in the map > ar & *it; > } > > I believe that b) would be a good solution for you. Yes thats ok, but I was serializing more than one map in the same class and I wanted to maintain the separation in xml. I guess I'd have to derive my own map I lazily trying to avoid that. > You post clarifies one point in particular. My version implements the > most commonly desired behavior - but not the only behavior people > might want. In other words - there is not always agreeement on what > it means to "restore" something. And in fact the serialization > implementation uses both and some time its confusing. Yes I agree, though in my case, now I know the limitation, its not difficult to refactor. I do find the refactoring (conditionally initialize after a load) a little forced however. Thanks again for the response. -- Peter Whaite.

2 1

[serialization] Removing objects from an archive
by Przemyslaw Wojnowski 13 Dec '05

13 Dec '05

Hello. How can one remove information about object from existing archive? Do archives implement removing? I haven't seen removing from an archive in any archive interface description. Thanks for any help. Regards, Przemek

2 3

[Graph] dijkstra_shortest_paths color map
by Daniel Mitchell 12 Dec '05

12 Dec '05

I work with very large implicit graphs that use out-of-core storage for properties, including the vertex color property, but the following bit of code in dijkstra_shortest_paths_no_init (boost-1.33.0) seems to preclude the use of a custom color map. std::vector<default_color_type> color(num_vertices(g)); default_color_type c = white_color; breadth_first_visit(g, s, Q, bfs_vis, make_iterator_property_map(&color[0], index_map, c)); What is the rationale for the above code, and how should I work around it? Daniel

3 8

BGL dijksra_shortest_path and boost::on_initialize_vertex
by Andreas Fabri 12 Dec '05

12 Dec '05

Hello, I have my own graph model and want to run dikstra_shortest_path on it. I want a dijkstra_visitor to initialize the id's of my vertices, struct initialize_index : public boost::base_visitor<initialize_index> { int i; typedef boost::on_initialize_vertex event_filter; initialize_index(int i_) : i(i_) {} template <class Vertex, class Graph> void operator()(Vertex v, Graph& G) { std::cout << "i = " << i << std::endl; v->id() = i++; } }; . . . . boost::dijkstra_shortest_paths(ft, source , distance_map(distance_pmap) .predecessor_map(predecessor_pmap) .visitor(make_dijkstra_visitor(initialize_index(0)))); but the operator is never called. Is that a bug ? Shouldbn't it be called by the bfs performed in dijkstra_shortest_paths_no_init ? Also when looking in the header file dijkstra_shortest_paths.hpp I see that the color map gets initialized before the vertices get initialized. I would expect that the vertices get initialized before the color map gets initialized, because if the vertex_index_map is not set up correctly ... andreas

2 1

[bjam][mingw] signals fails to build + solution
by Paul 12 Dec '05

12 Dec '05

Hi all, I had a problem with boost 1.33.1, building it on windows XP inside the msys (MinGW) environment. The link step generates this command: "g++" "-Wl,--enable-auto-image-base" "-Wl,--exclude-symbols,_bss_end__:_bss_start__:_data_end__:_data_start__" "-Wl,--out-implib,..\..\..\bin\boost\libs\signals\build\boost_signals.dll\mingw\debug\boost_signals-mgw-d-1_33_1.lib" -g -shared -Wl,--allow-multiple-definition -mno-cygwin -o "..\..\..\bin\boost\libs\signals\build\boost_signals.dll\mingw\debug\boost_signals-mgw-d-1_33_1.dll" -L"" "..\..\..\bin\boost\libs\signals\build\boost_signals.dll\mingw\debug\trackable.obj" "..\..\..\bin\boost\libs\signals\build\boost_signals.dll\mingw\debug\connection.obj" "..\..\..\bin\boost\libs\signals\build\boost_signals.dll\mingw\debug\named_slot_map.obj" "..\..\..\bin\boost\libs\signals\build\boost_signals.dll\mingw\debug\signal_base.obj" "..\..\..\bin\boost\libs\signals\build\boost_signals.dll\mingw\debug\slot.obj" and it fails, giving a message about not being able to find a symbol for a method in trackable. I edited tools/build/v1/mingw-tools.jam and added this line: flags mingw STDLIBPATH : $(GCC_BIN_DIRECTORY) ; + flags mingw LIBPATH : $(GCC_BIN_DIRECTORY) ; } And tried again. This time it worked, with the following command: "g++" "-Wl,--enable-auto-image-base" "-Wl,--exclude-symbols,_bss_end__:_bss_start__:_data_end__:_data_start__" "-Wl,--out-implib,..\..\..\bin\boost\libs\signals\build\boost_signals.dll\mingw\release\boost_signals-mgw-1_33_1.lib" -s -shared -Wl,--allow-multiple-definition -mno-cygwin -o "..\..\..\bin\boost\libs\signals\build\boost_signals.dll\mingw\release\boost_signals-mgw-1_33_1.dll" -L"" -L"" "..\..\..\bin\boost\libs\signals\build\boost_signals.dll\mingw\release\trackable.obj" "..\..\..\bin\boost\libs\signals\build\boost_signals.dll\mingw\release\connection.obj" "..\..\..\bin\boost\libs\signals\build\boost_signals.dll\mingw\release\named_slot_map.obj" "..\..\..\bin\boost\libs\signals\build\boost_signals.dll\mingw\release\signal_base.obj" "..\..\..\bin\boost\libs\signals\build\boost_signals.dll\mingw\release\slot.obj" The difference is that there is an EXTRA -L"" in the command. I suspect that having any -L"" is a bad idea, and if there is no parameter, it should be omitted all together. I bet this command works ONLY because, to g++, it looks like g++ blah blah -L"-L" ie the second -L becomes the parameter for the first -L (spaces are allowed after the -L flag). I dont understand bjam's Jamfiles well enough to fix this properly. Can someone help put together a patch for this? Thanks Paul

2 1