- Boost-users - lists.boost.org

[Graph] filtered_graph of IncidenceGraph is not an IncidenceGraph
by Luke Meyers 20 Mar '12

20 Mar '12

Hello, list. First time poster, sometime reader. Today I finally achieved a milestone I've been working towards in my BGL-oriented project: I got astar_search_no_init working on an implicit graph. Now I want to filter the input graph to restrict the search, and I'm running into trouble with filtered_graph. My implicit graph class is a model of IncidenceGraph, as required by astar_search_no_init, but when I apply filtered_graph to it I find that the result is no longer an IncidenceGraph (according to BOOST_CONCEPT_ASSERT). The filtered_graph documentation saith: If the underlying Graph type models VertexAndEdgeListGraph and > PropertyGraph then so does the filtered graph. If the underlying Graph type > models fewer or smaller concepts than these, then so does the filtered > graph. >From this, I would assume that if I run filtered_graph on an IncidenceGraph, the result will be a model of IncidenceGraph, not just of Graph. Also, if my input is an IncidenceGraph and not a VertexAndEdgeListGraph, I'd hope compilation would not fail if I don't have graph_traits defined for types required by VertexAndEdgeListGraph, but it does: c:\src\boost_1_49_0\boost\graph\filtered_graph.hpp(181) : error C2039: > 'in_edge_iterator' : is not a member of 'boost::graph_traits<XYGraph>' When I provide a bunch of dummy typedefs to get around this error, I can get to the point of doing a BOOST_CONCEPT_ASSERT((IncidenceGraphConcept<* FilteredGraphType*>)). This fails: c:\src\boost_1_49_0\boost\graph\graph_concepts.hpp(94) : error C2679: > binary '=' : no operator found which takes a right-hand operand of type > 'XY' (or there is no acceptable conversion) XY here is my vertex descriptor type. The edge descriptor is a pair<XY,XY>, but for some reason it's getting XY out of the graph_traits instead and failing in this strange way. Am I missing something? Are my expectations incorrect? Is this actually a bug (or two)? At the bottom is a decently small example that fails for me (some function bodies excised for brevity -- my concerns are pre-link). Is there a way to work around this without making my graph into a model of VertexAndEdgeListGraph? Since my graph is infinite, this would hardly seem sensible. Thanks in advance for any assistance. Incidentally, if anyone's interested to see a working example of astar_search_no_init over an implicit graph, I'd be happy to clean up the code a little and post it. Luke Project blog (SnargleQuest, a Roguelike game): http://snarglequest.blogspot.com/ // Code listing #include <iostream> #include <list> #include <map> #include <set> #include <utility> #include <boost/graph/filtered_graph.hpp> #include <boost/graph/graph_concepts.hpp> #include <boost/graph/graph_traits.hpp> #include <boost/operators.hpp> using namespace boost; using namespace std; namespace Direction { enum id { MIN = 0, N = MIN, S, E, W, NW, NE, SE, SW, NONE, NUM_DIRECTIONS }; } struct XY : public boost::additive<XY, boost::totally_ordered<XY, boost::equivalent<XY> > > { typedef int X; typedef int Y; XY(X x = 0, Y y = 0); XY & operator=(XY const& that); XY & operator+=(XY const& that); bool operator<(XY const& that) const; X x; Y y; }; std::ostream & operator<<(std::ostream & os, XY const& xy); struct neighbor_iterator; /* * Model of: * * Graph * * IncidenceGraph */ struct XYGraph { XYGraph(); // Graph concept requirements typedef XY vertex_descriptor; typedef std::pair<XY, XY> edge_descriptor; typedef undirected_tag directed_category; typedef disallow_parallel_edge_tag edge_parallel_category; typedef incidence_graph_tag traversal_category; // IncidenceGraph concept requirements typedef neighbor_iterator out_edge_iterator; typedef int degree_size_type; }; // IncidenceGraph concept requirements std::pair<XYGraph::out_edge_iterator, XYGraph::out_edge_iterator> out_edges(XYGraph::vertex_descriptor v, XYGraph const& g); XYGraph::degree_size_type out_degree(XYGraph::vertex_descriptor v, XYGraph const& g); XYGraph::vertex_descriptor source(XYGraph::edge_descriptor e, XYGraph const& g); XYGraph::vertex_descriptor target(XYGraph::edge_descriptor e, XYGraph const& g); // Iterator struct neighbor_iterator : public boost::forward_iterator_helper<neighbor_iterator, XY, std::ptrdiff_t, XY *, XY> { public: neighbor_iterator(); neighbor_iterator(XY xy, Direction::id direction); neighbor_iterator & operator=(neighbor_iterator const& that); std::pair<XY,XY> operator*() const; void operator++(); bool operator==(neighbor_iterator const& that) const; }; namespace boost { template <> struct graph_traits<XYGraph> { typedef XYGraph G; typedef G::vertex_descriptor vertex_descriptor; typedef G::edge_descriptor edge_descriptor; typedef G::out_edge_iterator out_edge_iterator; typedef G::directed_category directed_category; typedef G::edge_parallel_category edge_parallel_category; typedef G::traversal_category traversal_category; typedef G::degree_size_type degree_size_type; // Shouldn't have to do this! typedef void in_edge_iterator; typedef void vertex_iterator; typedef void vertices_size_type; typedef void edge_iterator; typedef void edges_size_type; }; } struct orthogonal_only { typedef pair<XY,XY> Edge; bool operator()(Edge const& edge) { return edge.first.x == edge.second.x || edge.first.y == edge.second.y; } }; int main(int argc, char **argv) { BOOST_CONCEPT_ASSERT((IncidenceGraphConcept<XYGraph>)); // This fails BOOST_CONCEPT_ASSERT((IncidenceGraphConcept< filtered_graph<XYGraph, orthogonal_only> >)); return 0; }

3 4

boost::noncopyable - I see
by John M. Dlugosz 20 Mar '12

20 Mar '12

The compiler (VS10) generates this when the copy constructor is used somewhere by mistake. It doesn't tell me where the offending use is, unlike putting a private dummy declaration right in the class. So it did its job, but is most unfriendly about it.

1 0

[Iterator] shared_container_iterator usability
by Szymon Gatner 19 Mar '12

19 Mar '12

Hello boosters, a design question: how often (if ever) and under which circumstances do you use shared_container_iterator? (http://www.boost.org/doc/libs/1_49_0/libs/utility/shared_container_iterator…) I don't feel too good about allocating a vector<> on the heap but I guess it was made for some reason and motivated by actuall need. What are good use cases for such iterator? Obviously returning a container from function (as presented in example) is not one of them. Cheers, Simon

3 4

Thread safety in Boost.Test
by Eric Teutsch 19 Mar '12

19 Mar '12

I found this article from 2003. http://lists.boost.org/Archives/boost/2003/02/44515.php I gather that 9 years later, Boost.Test is still not thread-safe (at least it's still on the To-Do list). Can somebody explain what BOOST_TEST_THREAD_SAFE accomplishes? Thanks, Eric T This message, and the documents attached hereto, is intended only for the addressee and may contain privileged or confidential information. Any unauthorized disclosure is strictly prohibited. If you have received this message in error, please notify us immediately so that we may correct our internal records. Please then delete the original message. Thank you.

1 0

[Geometry] Calculating distances on the earth
by Sean Farrow 19 Mar '12

19 Mar '12

Hi all: I need to calculate distances on the earth. Ideally I'd like to use the haversine formula. There is an example in the docs at: http://www.boost.org/doc/libs/1_49_0/libs/geometry/doc/html/geometry/quicks… under the non-Cartesian heading. Will this do what I need, or do I need the Geographical coordinate system, (if yes where is this available and what features does this supply?) Any help appreciated. Regards Sean.

2 5

convert a boost::iostreams::filter_streambuf to a std::wofstream
by Sean Farrow 19 Mar '12

19 Mar '12

Hi: I'm trying to pass a boost::iostreams::filtering_streambuf to a boost::archive constructor. I'm using visual studio 2010, and am getting c2664 telling me that this conversion carn't be performed. Is there anyway I can convert a boost::iostreams::filtering_streambuf. Any help appreciated. Cheers Sean.

1 0

(no subject)
by David Bergman 19 Mar '12

19 Mar '12

Hello, I just subscribed to this mailing list. I have a question that I originally sent to David Abrahams, and I'm hoping someone can help. Dear Sir or Madam, I am currently using boost in a simulator and have a few questions regarding the special functions that are (are not) available. 1. Do your special functions accept complex numbers as input and provide complex output? I am specifically interested in the Bessel functions. In most applications the argument is restricted to be positive and the index to be real. However, Bessel functions are defined for complex index and argument, the former arises in acoustic and optical wave guide problems.If the answer to 1 is no I’d suggest these capabilities as a future nice to have. 2. Does Boost plan to have hypergeometric functions in the future? They are currently listed as unsupported TR1 functions. I have not been able to find them in TR1 either but I’m not done looking. Thanks in advance for your help. David

4 6

Re: [Boost-users] boost.zlib on Windows
by László Marák 19 Mar '12

19 Mar '12

It was the question of binary files. Thank You for Your help. Message: 5 Date: Mon, 19 Mar 2012 10:57:29 -0400 From: "Andrew Holden" <aholden(a)charteroaksystems.com> To: <boost-users(a)lists.boost.org> Subject: Re: [Boost-users] boost.zlib on Windows Message-ID: < 1D97E025367AFB4A8C528590A3715BF4425F20(a)archerysummit2.willamette.charteroaksystems.com > Content-Type: text/plain; charset="iso-8859-1" On Sunday, March 18, 2012 10:05 AM, L?szl? Mar?k wrote: > > I would like to ask You what is the status of Boost.zlib on > Windows. I have a program that uses Boost.iostreams to create > a gzip file. It compiles both on Windows and Linux and on > Linux it creates a functional gz file. On Windows it creates > a gz file but when I try to decompress the file it gives me > "data error". > > When I look at Boost's regression tests both gzip and zlib is > missing on Windows. Has anybody successfully run Boost.zlib > on Windows recently? Is there some flush method that should > be called, but on Linux it does not show any symptoms? I agree with Artyom Beilis. It sounds like you're opening the files in text mode. This causes Windows to insert a carriage return (0x0D) before each line feed (0x0A), but has no effect on Linux or other Unix-like operating systems. Here is the actual code I use to open a gz or bz2 file for writing in one of my projects, which produced valid gz files with boost 1.48 (untested with 1.49) and Visual Studio 2010, both in i386 and AMD64 modes. Ignore the references to tbb::scalable_allocator. In particular, note the last line where I specify the file_sink should be binary: boost::iostreams::filtering_stream <boost::iostreams::output, char, std::char_traits <char>, tbb::scalable_allocator <char>> output_file; if (boost::iends_with (file_name, ".gz")) output_file.push (boost::iostreams::basic_gzip_compressor <tbb::scalable_allocator <char>> (boost::iostreams::zlib::best_compression)); else if (boost::iends_with (file_name, ".bz2")) output_file.push (boost::iostreams::basic_bzip2_compressor <tbb::scalable_allocator <char>> ()); output_file.push (boost::iostreams::file_sink (file_name.c_str(), std::ios::out | std::ios::binary));

1 0

boost.zlib on Windows
by László Marák 19 Mar '12

19 Mar '12

Dear Boost Users, I would like to ask You what is the status of Boost.zlib on Windows. I have a program that uses Boost.iostreams to create a gzip file. It compiles both on Windows and Linux and on Linux it creates a functional gz file. On Windows it creates a gz file but when I try to decompress the file it gives me "data error". When I look at Boost's regression tests both gzip and zlib is missing on Windows. Has anybody successfully run Boost.zlib on Windows recently? Is there some flush method that should be called, but on Linux it does not show any symptoms? Thank You in advance, UJ

5 4

Boost Windows Runtime / WinRT support
by stren tler 19 Mar '12

19 Mar '12

Hi, are there any plans to Support Windows Runtime / WinRT? For example a port of boost asio which uses Windows::Networking::Sockets::StreamSocket … ?

1 0