September 2013 - Boost - lists.preview.boost.org

[asio] Compilation failures on mingw
by Andrey Semashev 20 Sep '13

20 Sep '13

Hi, I'm seeing test failures on MinGW [1], which are probably caused by commit 85764 [2]. Could this be fixed? [1] <http://www.boost.org/development/tests/trunk/developer/MinGW-32%20jc-bell-c…> [2] <https://svn.boost.org/trac/boost/changeset/85764>

1 0

[multi_index][intel-linux-14.0.0] Help needed debugging initializer list support
by Joaquin M Lopez Munoz 19 Sep '13

19 Sep '13

Hi, The trunk version of Boost.MultiIndex introduces support for initializer list, which seems to work fine with regression tests except for intel-linux-14.0.0: http://tinyurl.com/pclssoe http://tinyurl.com/nzwzpu6 which spits error messages like this (replaced large chunks of text with *** for clarity): ../libs/multi_index/test/test_modifiers.cpp(230): error: no instance of overloaded function "boost::multi_index::detail::hashed_index<***>::insert [***]" matches the argument list argument types are: ({...}) object type is: employee_set_by_name i1.insert({{4,"Vanessa",20,9236},{5,"Penelope",55,2358}}); At first I thought this is failing because for it to work the compiler should not only support initializer lists but also C++11 unified initialization syntax, but I think this is not the problem because a) intel-linux-14.0.0 does indeed seem to support unified initialization syntax, since BOOST_NO_CXX11_UNIFIED_INITIALIZATION_SYNTAX is not listed in http://tinyurl.com/ndutn2w . b) the following test code also fails: s.assign({0,1,2,3,4,5}); with s a multi_index_container of ints, and this does not depend on unified initialization syntaxes. The error message here, FWIW, is different than before: ../libs/multi_index/test/test_copy_assignment.cpp(56): internal error: assertion failed: node_has_side_effects: bad node kind (shared/cfe/edgcpfe/il.c, line 20611) The tested code seems simple enough, for instance lines 271 and following of multi_index_container.hpp look like: #if !defined(BOOST_NO_CXX11_HDR_INITIALIZER_LIST) multi_index_container( std::initializer_list<Value> list, const ctor_args_list& args_list=ctor_args_list(), const allocator_type& al=allocator_type()) [...] and the ofending assign member function in sequenced_index.hpp is as innocent as: #if !defined(BOOST_NO_CXX11_HDR_INITIALIZER_LIST) void assign(std::initializer_list<value_type> list) { assign(list.begin(),list.end()); } #endif Besides, other libs with initializer list support such as Boost.SmartPtr and Boost.Unordered cope fine with this compiler. Clues? Any help much appreciated. Joaquín M López Muñoz Telefónica Digital

1 0

[smart_ptr] extra colon on make_shared variadic template instantiation
by Vicente J. Botet Escriba 19 Sep '13

19 Sep '13

Hi, while instantiating this template, the Intel compiler finds that there is an extra colon when Args is empty. ::new( pv ) T( boost::detail::sp_forward<Arg1>( arg1 ), boost::detail::sp_forward<Args>( args )... ); ^ Is this a bug on the compiler or on Boost.SmartPtr? Best, Vicente template< class T, class Arg1, class... Args > typename boost::detail::sp_if_not_array< T >::type make_shared( Arg1 && arg1, Args && ... args ) { boost::shared_ptr< T > pt( static_cast< T* >( 0 ), BOOST_SP_MSD( T ) ); boost::detail::sp_ms_deleter< T > * pd = static_cast<boost::detail::sp_ms_deleter< T > *>( pt._internal_get_untyped_deleter() ); void * pv = pd->address(); ::new( pv ) T( boost::detail::sp_forward<Arg1>( arg1 ), boost::detail::sp_forward<Args>( args )... ); pd->set_initialized(); T * pt2 = static_cast< T* >( pv ); boost::detail::sp_enable_shared_from_this( &pt, pt2, pt2 ); return boost::shared_ptr< T >( pt, pt2 ); }

3 4

Comparing floating point Alberto Squassabia - original article wanted.
by Paul A. Bristow 19 Sep '13

19 Sep '13

This link to this article http://www.adtmag.com/joop/crarticle.asp?ID=396 seems now dead [4] Alberto Squassabia Comparing Floats: How To Determine if Floating Quantities Are Close Enough Once a Tolerance Has Been Reached <http://www.adtmag.com/joop/crarticle.asp?ID=396> C++ Report March 2000. Referenced in http://www.boost.org/doc/libs/1_54/libs/test/doc/components/test_tools/floa… l Does anyone still have a copy of the original? Thanks Paul --- Paul A. Bristow, Prizet Farmhouse, Kendal LA8 8AB UK +44 1539 561830 07714330204 pbristow(a)hetp.u-net.com

2 2

[mpl] and_<true_> returns false
by Agustín K-ballo Bergé 19 Sep '13

19 Sep '13

The following snippet fails to compile (tested with 1.54.0 and trunk, several compilers), #include <boost/mpl/and.hpp> #include <boost/mpl/bool.hpp> #include <boost/mpl/or.hpp> int main() { using namespace boost::mpl; static_assert(and_<true_>::value, "U KID?"); //this one fires static_assert(or_<true_>::value, "U KID?"); //this one is ok } This issue actually showed up in a variadic template scenario. Is this a bug? I feel like I'm doing something really wrong... I guess one may not call `and_` with less than two arguments. If that's the case, it would be helpful if it would be impossible to instantiate it with a single argument. Regards, -- Agustín K-ballo Bergé.- http://talesofcpp.fusionfenix.com

4 6

NuGet packages for boost
by Ben Craig 19 Sep '13

19 Sep '13

NuGet is a package manager that has traditionally been focused on .NET, but it has recently added support for C++ libraries. Here is a presentation by the lead developer adding C++ support to NuGet: http://channel9.msdn.com/Events/GoingNative/2013/Find-Build-Share-Use-Using… I would like to see boost packages added to NuGet, and to have these packages be "official" packages. I am a contributor on another project (Thrift) and I would like to be able to produce NuGet packages for Thrift, and I would like that package to have a "magically" working Boost dependency. Sergey Shandar has made some unofficial packages. Blog describing his effort: http://sergey-shandar.blogspot.com/2013/08/boost-on-nugetorg.html Nuget packages he has submitted: https://www.nuget.org/profiles/sergey.shandar/ If someone (perhaps me) authored the boost NuGet package build process, would the person in charge of releasing Boost be willing to publish the NuGet packages that get created? There are going to be plenty of follow up questions if Boost releases are willing to publish NuGet packages, but I figured I would get the big question out of the way first.

7 14

[voronoi]medial axis
by Louis Lavery 19 Sep '13

19 Sep '13

[mods: I posted this a few days ago, and that post was acknowledged etc (it was my first post as a new subscriber) but I haven't receieved a copy myself and cannot find it in the archives - perhaps I should be more patient?] Hello, I'm hoping to use boost voronoi to construct the (internal) medial axis for closed shapes (with holes). I need to work out which vertices of the voronoi are inside, outside and on a boundary. So my first goal is to determine which vertices correspond to the ends of the input segments (boundary vertices). The way I do this at the moment, is to look at each edge and test if its source vertex, vertex0(), is equal to either end point of the segment contained by the edge's cell. (1) If it is then it's a boundary vertex otherwise it's not. Is that correct? (2) When comparing a vertex with an input point can I do an exact compare (==) or should I use a tolerance (epsilon)? Thanks, fuzzy.

2 8

[cpo-proposal] About mixout container for polymorphic objects
by Santiago Tapia 17 Sep '13

17 Sep '13

Hi all, I have design a new concept of container to be part of the cpo library. In summary, those containers will break down the inserted objects into some elementary parts. Why do that? Short answer: to save memory. Long answer: Because the alignment requirement of data members in an object could be very different, in a range from 1 byte to 16 bytes, structs or classes use padding to provide the correct location for members. That means an object size could be significantly greater than the sum of its members sizes. Even more, when dealing with polymorphic objects, the size is increased a little more due to the pointer to the table of virtual methods. I will try to outline the alternative solution, obviously, the implementation will be part of the container, the user will not need to know the details. ---- Example:----- #include <cstdlib> #include <iostream> #include <vector> class output { public: virtual void print(std::ostream& out) = 0; }; class integer { public: integer(short a_ = 0) : a(a_) { } short a; }; class character { public: character(char c_ = 'a') : c(c_) { } char c; }; class data { public: data(double x_ = 0) : x(x_) { } double x; }; class my_object : public integer, public data, public character, public output { public: my_object(int show = 1) : integer(rand() % 100) , data(1.0*rand()/RAND_MAX) , character('a' + rand() % 10) { if (show) print(std::cout); } virtual void print(std::ostream& out) { std::cout << a << "\t" << x << " \t" << c << "\n"; } }; void insert( std::vector<integer>& v_int, std::vector<data>& v_data, std::vector<character>& v_char, const my_object& o) { v_int.push_back( o ); v_data.push_back( o ); v_char.push_back( o ); } void iterate( std::vector<integer>& v_int, std::vector<data>& v_data, std::vector<character>& v_char ) { my_object o(0); unsigned i; for ( i = 0; i < v_int.size(); ++i) { static_cast<integer&>( o ) = v_int[i]; static_cast<data&>( o ) = v_data[i]; static_cast<character&>( o ) = v_char[i]; o.print(std::cout); } } int main(int, char**) { std::cout << "sizeof(my_object): " << sizeof(my_object) << "\n"; std::vector<integer> v_int; std::vector<data> v_data; std::vector<character> v_char; std::cout << "--- inserting (random) objects ---\n"; unsigned i; for ( i = 0; i < 5; ++ i ) { insert( v_int, v_data, v_char, my_object() ); } std::cout << "--- show content ---\n"; iterate(v_int, v_data, v_char); return 0; } ---- Output: (32 bit PC) ---- sizeof(my_object): 20 --- inserting (random) objects --- 83 0.394383 h 15 0.911647 f 86 0.76823 j 21 0.477397 h 90 0.513401 d --- show content --- 83 0.394383 h 15 0.911647 f 86 0.76823 j 21 0.477397 h 90 0.513401 d Final comment. In this example, all the std::vector objects will be part of the mixout container. The user will only use the insert method and the container iterator that will dereference to output& (template parameter in the container). Just in this simple example the implementation saves aproximately n* (20 - (8+2+1) = 9) bytes, being n the number of instances of my_object. Comments are wellcome. Best regards, Santiago Tapia

5 4

[svn] pre-commit hook/mime-type
by Tim Blechmann 17 Sep '13

17 Sep '13

hi all, is it possible that the pre-commit hook had changed recently? not sure if this was intended, but it currently breaks the use of git-svn, which imo is a huge inconvenience: -- tim@moka-mac:~/Documents/boost-trunk-git/boost$ git svn dcommit Committing to https://svn.boost.org/svn/boost/trunk ... A boost/sync/detail/generic/event_cv_emulation.hpp A boost/sync/detail/windows/event.hpp A boost/sync/event.hpp A libs/sync/test/event_test.cpp ERROR from SVN: A repository hook failed: Commit blocked by pre-commit hook (exit code 1) with output: /home/subversion/boost/hooks/check-mime-type.pl: trunk/boost/sync/detail/generic/event_cv_emulation.hpp : svn:mime-type is not set trunk/boost/sync/detail/windows/event.hpp : svn:mime-type is not set trunk/boost/sync/event.hpp : svn:mime-type is not set trunk/libs/sync/test/event_test.cpp : svn:mime-type is not set Every added file must have the svn:mime-type property set. In addition text files must have the svn:eol-style property set. For binary files try running svn propset svn:mime-type application/octet-stream path/of/file For text files try svn propset svn:mime-type text/plain path/of/file svn propset svn:eol-style native path/of/file You may want to consider uncommenting the auto-props section in your ~/.subversion/config file. Read the Subversion book (http://svnbook.red-bean.com/) Chapter 7, Properties section, Automatic Property Setting subsection for more help. Also, see http://svn.boost.org/trac/boost/wiki/BoostSubversion W: b7ab541813d43b7d74b44f57eb04c2625ce305a7 and refs/remotes/git-svn differ, using rebase: :040000 040000 4b2ab78eaa8f1e5fa80761a327da5066ca15d465 73aa096ac4b16618132d6d012e964b1790542c8f M boost :040000 040000 6146f06c45f590b38f76d123656cea13f24471d1 7a95e843ffc2f67fff874b6bd6c142b040423414 M libs Current branch master is up to date. ERROR: Not all changes have been committed into SVN, however the committed ones (if any) seem to be successfully integrated into the working tree. Please see the above messages for details. -- this used to work fine just a few days ago. was this done intentionally? if so, an announcement on the list with a comment on the implications for the git-svn users would have been good to see. best, tim

3 4

[functional] New functor adaptors
by Tigran Hayrapetyan 17 Sep '13

17 Sep '13

Hello, Suppose we want to transform sequence of integers by doubling and adding 5 to each value (x = 2*x + 5 for all x in sequence). If using std::transform, then we need a functor which will for given "x" return "2*x + 5". My question is - is it possible to do it in C++98 style (without lambda expressions), and without writing additional functor class? As I understand, it's not. For doing that we need something like nesting functors into each other. Here we have "std::bind2nd( std::multiplies<int>(), 2 )" functor, and "std::bind2nd( std::plus<int>(), 5 )" functor, which must be nested to provide necessary functionality. I suggest something like "unary_result_translator" functor adapter, which is unary functor itself, and which will apply unary functor to result of another unary functor. The implementation of "unary_result_translator" can be something like this: template< typename Fn1, typename FnTrans > class unary_result_translator { public: typedef typename Fn1::argument_type argument_type; typedef typename FnTrans::result_type result_type; unary_result_translator( const Fn1& fn1, const FnTrans& fn_trans ) : fn1_( fn1 ), fn_trans_( fn_trans ) {} result_type operator()( const argument_type& arg ) const { return fn_trans_( fn1_( arg ) ); } const Fn1& fn1_; const FnTrans& fn_trans_; }; So, we will create our needed functor which converts "x" to "2*x + 5" by this: make_unary_result_translator( std::bind2nd( std::multiplies<int>(), 2 ), std::bind2nd( std::plus<int>(), 5 ) ) Of'course, unary_result_translator can be also used for more complex functors construction. Recently I used it for case-insensitive comparision of strings only with help of ::tolower. Similarly, there can be "binary_result_translator" which is binary functor itself, and which will apply unary functor to result of binary functor. Can that functor adaptors be useful? Or may be there is another way to do such things? Thanks for reading! Best Regards, Tigran Hayrapetyan

6 6