March 2007 - Boost-users - lists.boost.org

[multi_index] function object as key extractor?
by Boris 22 Mar '07

22 Mar '07

As I find myself replacing all kind of C++ standard containers with multi_index_container (a really very useful container) I've been defining key extractors quite a lot and wonder if there is any shortcut like using lambda expressions or function objects. Here's some pseudo-code to explain what I mean: struct foo { std::string s; foo(std::string s) : s(s) { } }; typedef multi_index_container< foo, indexed_by< hashed_unique< boost::bind<std::size_t, &std::string::size, member<foo, std::string, &foo::s> > > > > mi_t; The idea is to call another function on the object which is returned by the key extractor member. Here in this example the multi_index_container would only store strings with a unique size. However I'm lost how to define the key extractor. The difficulty is probably that I normally don't need to specify types when I use boost::bind or lambda expressions. Now I need to figure out how to use them with multi_index_container - is there any trick or I better go on defining key extractors? Boris

6 21

Review of Intrusive library begins today March 12
by Joaquín Mª López Muñoz 22 Mar '07

22 Mar '07

The review of Ion Gaztañagas's Intrusive library begins today, March 12, and extends up to March 21. * Download: Boost vault: http://boost-consulting.com/vault/ Folder: Containers File: intrusive.2007-03-04.zip * Online docs: http://ice.prohosting.com/newfunk/boost/libs/intrusive * Description: Intrusive is a library presenting some intrusive containers to the world of C++. While intrusive containers were and are widely used in C, they became more and more forgotten in C++ due to the presence of the standard containers, which don't support intrusive techniques. Intrusive not only reintroduces this technique to C++, but also encapsulates the implementation in STL-like interfaces. Hence anyone familiar with standard containers can easily use Intrusive. Like their STL-counterparts, intrusive containers use template parameters to control the stored data types and some special aspects of intrusive containers. * Compilers tested (so far): Visual 7.1/WinXP Visual 8.0/WinXP GCC 4.1.1/MinGW Intel 9.1/WinXP GCC 4.1.2/Linux * Review questions: Here are some questions you might want to answer in your review: - What is your evaluation of the design? - What is your evaluation of the implementation? - What is your evaluation of the documentation? - What is your evaluation of the potential usefulness of the library? - Did you try to use the library? With what compiler? Did you have any problems? - How much effort did you put into your evaluation? A glance? A quick reading? In-depth study? - Are you knowledgeable about the problem domain? And finally, every review should answer this question: - Do you think the library should be accepted as a Boost library? Be sure to say this explicitly so that your other comments don't obscure your overall opinion. I invite you to submit your review, either publicly to the developers or the users lists, or else, if for whatever reason you don't want to publicize it, directly to me. Thank you for investing some of your time on reviewing Ion Gaztañaga's Intrusive library. The Review Manager, Joaquín M López Muñoz Telefónica, Investigación y Desarrollo

4 4

End of Intrusive library review period
by Joaquín Mª López Muñoz 22 Mar '07

22 Mar '07

The official period for the formal review of Ion Gaztañagas's Intrusive has finished on March 21, although there is no problem with continuing the ongoing discussions or submiting past last minute reviews during a few days. I plan to compile the contributions and publish the final result by the end of the next week. Thank you to all the participants for the quality and quantity of their contributions. The Review Manager, Joaquín M López Muñoz Telefónica, Investigación y Desarrollo

1 0

[BGL] dfs on graphs with bundled-properties
by Naomi Fox 22 Mar '07

22 Mar '07

I am trying to use an adjacency list with bundled properties and running into lots of problems. I'm wondering if bundled properties is mature enough, or if I should resort to the older style using property maps. Here's a description of one problem I've run into: So I define my graph class with bundled properties as so: struct MyV { int numStars; }; struct MyE { int numStars; }; typedef adjacency_list<setS, setS, bidirectionalS, MyV, MyE> MyGraph; I would like to perform dfs on this graph. So I have a very simple visitor defined: struct star_visitor : public boost::default_dfs_visitor { template <class Vertex, class Graph> void start_vertex(Vertex v, const Graph &) { std::cout << 'start'; } template <class Vertex, class Graph> void discover_vertex(Vertex v, const Graph &) { std::cout << "*"; } template <class Vertex, class Graph> void finish_vertex(Vertex v, const Graph &) { std::cout << 'finish'; } }; Then in my main function, I make a MyGraph object and put in vertices and edges, and assign property values. That works fine. MyGraph g; MyGraph::vertex_descriptor v1 = addVertex(g); MyGraph::vertex_descriptor v2 = addVertex(g); MyGraph::edge_descriptor e2 = addEdge(v1, v2, g); g[*v1].numStars = 100; g[*v2].numStars = 2; g[*e1].numStars = 7; Then I perform dfs on my graph depth_first_search(g, visitor(star_vis)); And I get the following errors. I don't get these errors and my program compiles fine if I do a dfs with my star_visitor on a graph with no properties. Main.cpp c:\boost_1_33_1\boost\property_map.hpp(349) : error C2679: binary '+' : no operator found which takes a right-hand operand of type 'const boost::detail::error_property_not_found' (or there is no acceptable conversion) c:\program files\microsoft visual studio 8\vc\include\vector(366): could be 'std::_Vector_iterator<_Ty,_Alloc> std::_Vector_iterator<_Ty,_Alloc>::operator +(__w64 int) const' with [ _Ty=boost::default_color_type, _Alloc=std::allocator ] while trying to match the argument list '(const std::_Vector_iterator<_Ty,_Alloc>, const boost::detail::error_property_not_found)' with [ _Ty=boost::default_color_type, _Alloc=std::allocator ] c:\boost_1_33_1\boost\property_map.hpp(349) : while compiling class template member function 'boost::default_color_type boost::iterator_property_map::operator [](void *) const' with [ RandomAccessIterator=std::_Vector_iterator>, IndexMap=boost::adj_list_vertex_property_map, T=boost::default_color_type, R=boost::default_color_type & ] c:\documents and settings\istreinu\desktop\naomi\boost_1_33_1\boost\graph\depth_first_search.hpp(250) : see reference to class template instantiation 'boost::iterator_property_map' being compiled with [ RandomAccessIterator=std::_Vector_iterator>, IndexMap=boost::adj_list_vertex_property_map, T=boost::default_color_type, R=boost::default_color_type & ] c:\documents and settings\istreinu\desktop\naomi\boost_1_33_1\boost\graph\depth_first_search.hpp(332) : see reference to function template instantiation 'void boost::detail::dfs_dispatch::apply(const VertexListGraph &,DFSVisitor,Vertex,const boost::bgl_named_params &,boost::detail::error_property_not_found)' being compiled with [ VertexListGraph=MyGraph, DFSVisitor=Star_visitor, Vertex=void *, T=Star_visitor, Tag=boost::graph_visitor_t, Base=boost::no_property ] c:\documents and settings\istreinu\my documents\visual studio 2005\projects\refactorfirst\refactorfirst\main.cpp(228) : see reference to function template instantiation 'void boost::depth_first_search(const VertexListGraph &,const boost::bgl_named_params &)' being compiled with [ VertexListGraph=MyGraph, T=Star_visitor, Tag=boost::graph_visitor_t, Base=boost::no_property ]

2 1

which one is preferable vector (or similar things) or valarray (or similar things ) ?
by abir basak 22 Mar '07

22 Mar '07

Hi, I am using vector and similar containers extensively for my project, as well as boost circular_buffer_space_optimized (and adobe circular_queue which is built over vector ). They all work with iterator based concept. As I need to refer a portion of the vector (or circular buffer ) from within a class , I need to store index (not iterators, as container elements are movable ) or a pair of index within the class (using std::distance to get index from iterator, all of my containers are random access ). now from index , I create boost sub_range and use it. For arbitrary access, I store a vector<size_t>, and use boost permutation_iterator with the index. One point here, iterator is not very cheap when it is circular_buffer (and space optimized version) , as it is not merely a pointer , and it need to know container (i.e. store a ref to container) or store 3 pointers, one current pos, one for begin of buffer, one end of buffer (AFAIK boost uses former version). But after that I came to a different concept , a concept of view, which is used by boost uBLAS containers Blitz array and the STL low profile container valarray. Now it seems slice_view , range_view are cheaper for circular buffer, as only one reference to container is stored (esp when using permutation_iterator or arbitrary sequence access). Also seems cross referencing from one valarray to another is easier due to slice concept which can be directly stored. and it also provides some numeric operations. In my case all of the arrays are one dimensional, and they are array of some small struct (like point, tangent etc) most of them have some geometric meaning. And mostly I want to have some numeric operations on them (my original test implementation is in matlab and the array in matlab is just like Blitz array with little more overhead) I am little confused at this stage, which one to use. and if I use valarray kind of thing , how to use stl algorithm's with it (i.e how to get a iterator from it ? valarray is continuous so getting iterator is not a problem, but not sure whether I can sort a slice) . Can anyone help in this regard? abir -- Abir Basak, Member IEEE Software Engineer, Read Ink Technologies B. Tech, IIT Kharagpur email: abir(a)abirbasak.com homepage: www.abirbasak.com

1 0

ptr_sequence and transfer functions.
by Matthew Bray 22 Mar '07

22 Mar '07

I have a problem with the current ptr_sequence interface, according to the documentation point #7 at this location: http://www.boost.org/libs/ptr_container/doc/examples.html#transferring-owne… it should be possible to write something such as this ptr_list<X> list; ptr_vector<X> vec; list.transfer( list.begin(), vec.begin(), vec ); vec.transfer( vec.end(), list.begin(), list.end(), list ); i.e. given a U<T> and a V<T> I should be able to perform transfer operations in either direction. However given a suitable X this code fails to compile on VC++ 8 as well as GCC 3.4. Indeed any attempt to transfer between different ptr_sequence types (e.g. ptr_deque<X> to ptr_vector<X>) fails to compile. Looking at the implementation of ptr_sequence it appears easy to see why this is the case. However my question is is the documentation incorrect or is the current code incorrect? Right now if I wish to create containers of objects and then transfer ownership to others I'm forced to use the same container type throughout (i.e. the transfer functions provide no benefit at all if I have to choose the wrong container for the job), either that or write awkward transferal code when it appears the simple transfer function should do the job. Could someone please explain the situation regarding the transfer functions and documentation at this time? Thanks, Mat.

2 4

[Date_Time] undefined reference to as_short_string()
by Artus 22 Mar '07

22 Mar '07

Hi, I am trying to use Boost.Date_Time (and Boost) for the first time, and I failed to compile the following : ------------------------------------ #include <iostream> #include <iomanip> #include <boost/date_time/gregorian/gregorian.hpp> #include <boost/date_time/posix_time/posix_time.hpp> int main(int, char **) { using namespace std; using namespace boost::posix_time; using namespace boost::gregorian; std::cout << to_simple_string(second_clock::universal_time()); return 0; } ----------------------------------------- I get these comments : g++ -c -pipe -Wall -W -O2 -g -pipe -m32 -march=i386 -mtune=pentium4 -DQT_NO_DEBUG -DQT_SHARED -DQT_THREAD_SUPPORT -I/usr/lib/qt-3.3/mkspecs/default -I. -I/usr/lib/qt-3.3/include -o receiver.o receiver.cpp g++ -o receiver.exe receiver.o -L/usr/lib/qt-3.3/lib -L/usr/X11R6/lib -lchoo -lqt-mt -lXext -lX11 -lm receiver.o(.text+0xb96): In function `main': /usr/include/boost/date_time/date_formatting.hpp:43: undefined reference to `boost::gregorian::greg_month::as_short_string() const' I found some threads about this error but no solution. Can any one help? I am using Boost 1.32.0. Cheers, Patrick

1 0

[multi_index] problem with hashed index
by Boris 21 Mar '07

21 Mar '07

I've been using Boost Multiindex quite a lot but don't understand why the code below inserts two items. The container has a unique hash index and the two items return the same hash value. Is this is a bug in the library or in my code? Boris -- struct foo { int i, j; foo(int i, int j) : i(i), j(j) { } }; std::size_t hash_value(const shared_ptr<foo> &f) { std::size_t seed = 0; boost::hash_combine(seed, f->i); boost::hash_combine(seed, f->j); return seed; } typedef multi_index_container< shared_ptr<foo>, indexed_by<hashed_unique<identity<shared_ptr<foo> > > > > mi_t; shared_ptr<foo> f1(new foo(0, 0)); shared_ptr<foo> f2(new foo(0, 0)); mi_t mi; cout << mi.insert(f1).second << endl; // Returns 1. cout << mi.insert(f2).second << endl; // Returns again 1?! cout << mi.size() << endl; // Returns 2?!

2 4

[fusion] Weird compiler error
by Christian Henning 21 Mar '07

21 Mar '07

Hi there, I know the subject looks suspicious. Usually all compiler errors are weird when you do metaprogramming. Anyway, this example is a bit different. I'm just trying to compile a small example taking from the documentation. #include <boost/fusion/sequence.hpp> using namespace std; using namespace boost; using namespace fusion; int _tmain(int argc, _TCHAR* argv[]) { typedef map<pair<int, char>, pair<char, char>, pair<double, char> > mymap; BOOST_MPL_ASSERT((boost::is_same<result_of::at_key<mymap, int>::type, char&>)); return 0; } The error is: c:\cybermdx\Test Programs\Boost\MPL\tagged types\tagged types.cpp(15) : error C2143: syntax error : missing '}' before '(' c:\cybermdx\Test Programs\Boost\MPL\tagged types\tagged types.cpp(15) : fatal error C1903: unable to recover from previous error(s); stopping compilation I have no clue where the compiler is missing a closing bracket. I'm using MSVC 7.1 Christian

3 8

polymorphic_downcast and references
by Michael Marcin 21 Mar '07

21 Mar '07

Unless I'm missing something it looks like polymorphic_downcast doesn't provide an overload for reference types... is there a reason? This makes it a little inconvenient for use. Thanks, Michael Marcin

2 3