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

[fusion/tr1] Include path problem?
by Kimon Hoffmann 23 Jul '07

23 Jul '07

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi all, while setting up the include paths as specified in the TR1 documentation I stumbled across a Boost.Fusion related compile error (included indirectly from <xpressive/proto/proto.hpp> in my code). After stepping through the include hierarchy, I was able to extract this minimal example, that demonstrates the problem: - -------- START: Test.cpp -------- // Originally included from <boost/xpressive/proto/proto_fwd.hpp> #include <boost/spirit/fusion/sequence/tuple_forward.hpp> // Originally included from #include <boost/xpressive/proto/op_base.hpp> #include <boost/spirit/fusion/sequence/tuple.hpp> - -------- END: Test.cpp -------- With MSVC++ 8.0 SP1 This file fails to compile with certain include path orderings while it compiles fine with others. The error messages are: ...\boost\spirit\fusion\sequence\make_tuple.hpp(23) : error C2027: use of undefined type 'boost::fusion::tuple<>' ...\boost\spirit\fusion\sequence\make_tuple.hpp(23) : error C2079: 'boost::fusion::make_tuple' uses undefined struct 'boost::fusion::tuple<>' ...\include\boost\spirit\fusion\sequence\make_tuple.hpp(24) : error C2514: 'boost::fusion::tuple<>' : class has no constructors Working include path orderings: 1. Any include path ordering that doesn't contain Boost.TR1 2. VC++ Standard Library, regular Boost, Boost.TR1 3. VC++ Standard Library, Boost.TR1, regular Boost Non-working include path orderings: 1. Boost.TR1, regular Boost, VC++ Standard Library [*] 2. Regular Boost, Boost.TR1, VC++ Standard Library [*] Preferred Boost.TR1 setting Generally speaking I would not consider this a problem, as I can simply order the include directories accordingly, but, as far as I understand it, Boost.TR1 can not be expected to work correctly with any of these settings. Regards, Kimon -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.4 (MingW32) iD8DBQFGn3ceIISObEqaPr4RAiszAJ9ipiY+qK+lLg8uDPnHEZejyA5ShwCfQ0fC 1vOMnumqPnguAgtv7co2N8c= =MIsY -----END PGP SIGNATURE-----

3 12

[IBD] Weekly news: 2007-07-11 to 2007-07-17
by Matias Capeletto 22 Jul '07

22 Jul '07

================================================== Hi Boost, Once a week, we will send you a brief of our work for the Improving Boost Docs project. We need more hands, join us! We are waiting you on the boost-docs list. Project Wiki: http://svn.boost.org/trac/boost/wiki/ImprovingBoostDocs RSS news feed: http://beta.boost.org/development/ibd_news.rss Community Blog: http://improving-boost-docs.blogspot.com ================================================== A bussy week ------------------------------------------------------------------------------------- Sorry for the delay, It was a busy week. I have started to promote IBD in Argentina... we may get some friendly hands from my country in the future. Kudos to all IBDers... the boost-docs list activity is truly impressive. 2007-07-17 ------------------------------------------------------------------------------------- * Proofreading of the IBD Wiki is almost complete Kudos to Jake and Paul! more... http://tinyurl.com/3234qk * Kai joins IBD He will be working at the "Trac Syntax Coloring Project". Thanks you so much for jumping in! more... http://tinyurl.com/3d6mwe 2007-07-16 ------------------------------------------------------------------------------------- * IBD welcomes Kai Brüning It will be great to count with his help. Lets work to make him feel comfortable in the Boost community. more... http://tinyurl.com/2to8bg 2007-07-15 ------------------------------------------------------------------------------------- * Translation project moved to Helping Boost Authors The translation project is now part of the new section "Helping Boost Authors". We hope this section will become the main communication channel between Boost Authors and IBD developers. more... http://svn.boost.org/trac/boost/wiki/HelpingBoostAuthors 2007-07-14 ------------------------------------------------------------------------------------- * Helping Boost Authors We have a new resource at IBD: a section in the project Wiki called "Helping Boost Authors". This section is a bazaar of friendly hands. We are teachers, experienced users, native English speakers and artists. We want to offer our help to Boost authors to improve their documentation. more... http://svn.boost.org/trac/boost/wiki/HelpingBoostAuthors 2007-07-13 ------------------------------------------------------------------------------------- * Documentation Best Practices The first guideline was added to the section "Documentation Best Practices". There are currently more documents to add here, we will try to fill this section with useful information for Boost Authors. more... http://svn.boost.org/trac/boost/wiki/DocumentationBestPractices 2007-07-12 ------------------------------------------------------------------------------------- * Proofreading of IBD Wiki starts Jake and Paul first project will be proofread the IBD Wiki pages. more... http://tinyurl.com/3786jf 2007-07-11 ------------------------------------------------------------------------------------- * Glyn joins the community Blog We are eagerly waiting to read you. more... http://improving-boost-docs.blogspot.com

1 0

[program_options] incorrect documentation
by Bryan Green 22 Jul '07

22 Jul '07

In the documentation for program_options, under the "Library Overview"/"Syntactic Information", the documentation appears to be incorrect. For the following option description: ("verbose", value<string>()->zero_tokens(), "verbosity level") It states that: "the user may either provide a single token for the value, or no token at all." I wrote a test program based on the options given in this section of the manual, and I get different behavior for the 'verbose' option. In fact, the actual behavior is almost intuitive: the option is unusable. No matter what you do, its an error. Seems natural since you're saying the option has a value associated with it, yet no tokens may be given to specify the value. It also turns out that 'multitoken()' does not allow you to pass multiple tokens, but it does raise an exception if you specify anything after the first value, i.e.: ./a.out --email blah@blah me@somewhere exception: in option 'email': multiple values not allowed ./a.out --email blah@blah --help exception: in option 'email': multiple values not allowed Here's my test program: <code> #include <boost/program_options.hpp> #include <string> #include <iostream> using namespace boost::program_options; using namespace std; int main(int argc,char *argv[]) { options_description desc; desc.add_options() ("help", "produce help message") ("compression", value<string>(), "compression level") ("verbose", value<string>()->zero_tokens(), "verbosity level") ("email", value<string>()->multitoken(), "email to send to") ; variables_map vm; try { store(parse_command_line(argc, argv, desc), vm); notify(vm); } catch (exception &e) { cout << "exception: " << e.what() << endl; return -1; } if (vm.count("help")) cout << desc << endl; if (vm.count("compression")) cout << "compression " << vm["compression"].as<string>() << endl; if (vm.count("verbose")) cout << "verbose " << vm["verbose"].as<string>() << endl; if (vm.count("email")) cout << "email " << vm["email"].as<string>() << endl; return 0; } </code> -Bryan

1 0

[boost.python] "Thin wrappers" for member functions
by Marcin Kalicinski 21 Jul '07

21 Jul '07

It's possible to expose free functions through "thin wrappers": void f() { //... } void f_wrapper() { // ... some extra functionality to be done when called from Python only f(); } def("f", f_wrapper); But is it possible to write "thin wrappers" for member functions (without modifying the wrapped class)? struct foo { void f(); }; // f_wrapper defined somehow somewhere class_<foo>("foo") .def("f", f_wrapper???) Thanks, Marcin

2 1

Serialization valgrind reports errors in load() ...
by Sascha Ochsenknecht 21 Jul '07

21 Jul '07

Hello, I have following class which I want to serialize: template < class ValueType > class Values { private: vector< ValueType > _values; public: Values(unsigned p_nr, const ValueType& p_value) : _values(p_nr, p_value) {} Values() {} private: friend class boost::serialization::access; template < class Archive > void load(Archive& p_ar, unsigned /*p_version*/) { p_ar & _values; } template < class Archive > void save(Archive& p_ar, unsigned /*p_version*/) const { p_ar & _values; } BOOST_SERIALIZATION_SPLIT_MEMBER(); }; My main() looks very simple: ... // save const Values<std::string> a(3, "fritz"); { std::ofstream ofs("fritz.txt"); boost::archive::text_oarchive ar(ofs); ar << a; } // restore Values<std::string> b; { std::ifstream ifs("fritz.txt"); boost::archive::text_iarchive ar(ifs); ar >> b; } valgrind reports me following errors (the load() seems to be the problem): ==13202== Conditional jump or move depends on uninitialised value(s) ==13202== at 0x4E59F72: boost::archive::detail::basic_iarchive::reset_object_address(void const*, void const*) (in /usr/lib64/libboost_serialization.so.1.33.1) ==13202== by 0x408A2C: boost::serialization::stl::archive_input_seq<boost::archive::text_iarchive, std::vector<std::string, std::allocator<std::string> > >::operator()(boost::archive::text_iarchive&, std::vector<std::string, std::allocator<std::string> >&) (collections_load_imp.hpp:54) ==13202== by 0x408ABD: void boost::serialization::stl::rebuild_collection<boost::archive::text_iarchive, std::vector<std::string, std::allocator<std::string> >, boost::serialization::stl::archive_input_seq<boost::archive::text_iarchive, std::vector<std::string, std::allocator<std::string> > >, boost::serialization::stl::reserve_imp<std::vector<std::string, std::allocator<std::string> > > >(boost::archive::text_iarchive&, std::vector<std::string, std::allocator<std::string> >&) (collections_load_imp.hpp:149) ==13202== by 0x408AF7: void boost::serialization::stl::load_collection<boost::archive::text_iarchive, std::vector<std::string, std::allocator<std::string> >, boost::serialization::stl::archive_input_seq<boost::archive::text_iarchive, std::vector<std::string, std::allocator<std::string> > >, boost::serialization::stl::reserve_imp<std::vector<std::string, std::allocator<std::string> > > >(boost::archive::text_iarchive&, std::vector<std::string, std::allocator<std::string> >&) (collections_load_imp.hpp:171) ==13202== by 0x408B1E: void boost::serialization::load<boost::archive::text_iarchive, std::string, std::allocator<std::string> >(boost::archive::text_iarchive&, std::vector<std::string, std::allocator<std::string> >&, unsigned) (vector.hpp:58) ==13202== by 0x408B5D: boost::serialization::free_loader<boost::archive::text_iarchive, std::vector<std::string, std::allocator<std::string> > >::invoke(boost::archive::text_iarchive&, std::vector<std::string, std::allocator<std::string> >&, unsigned) (split_free.hpp:58) ==13202== by 0x408B87: void boost::serialization::split_free<boost::archive::text_iarchive, std::vector<std::string, std::allocator<std::string> > >(boost::archive::text_iarchive&, std::vector<std::string, std::allocator<std::string> >&, unsigned) (split_free.hpp:74) ==13202== by 0x408BB1: void boost::serialization::serialize<boost::archive::text_iarchive, std::string, std::allocator<std::string> >(boost::archive::text_iarchive&, std::vector<std::string, std::allocator<std::string> >&, unsigned) (vector.hpp:76) ==13202== by 0x408BEF: void boost::serialization::serialize_adl<boost::archive::text_iarchive, std::vector<std::string, std::allocator<std::string> > >(boost::archive::text_iarchive&, std::vector<std::string, std::allocator<std::string> >&, unsigned) (serialization.hpp:140) ==13202== by 0x408C29: boost::archive::detail::iserializer<boost::archive::text_iarchive, std::vector<std::string, std::allocator<std::string> > >::load_object_data(boost::archive::detail::basic_iarchive&, void*, unsigned) const (iserializer.hpp:160) ==13202== by 0x4E5A5B9: boost::archive::detail::basic_iarchive::load_object(void*, boost::archive::detail::basic_iserializer const&) (in /usr/lib64/libboost_serialization.so.1.33.1) ==13202== by 0x4095F7: boost::archive::detail::load_non_pointer_type<boost::archive::text_iarchive, std::vector<std::string, std::allocator<std::string> > >::load_standard::invoke(boost::archive::text_iarchive&, std::vector<std::string, std::allocator<std::string> >&) (iserializer.hpp:375) ==13202== ==13202== Conditional jump or move depends on uninitialised value(s) ==13202== at 0x4E59FA1: boost::archive::detail::basic_iarchive::reset_object_address(void const*, void const*) (in /usr/lib64/libboost_serialization.so.1.33.1) ==13202== by 0x408A2C: boost::serialization::stl::archive_input_seq<boost::archive::text_iarchive, std::vector<std::string, std::allocator<std::string> > >::operator()(boost::archive::text_iarchive&, std::vector<std::string, std::allocator<std::string> >&) (collections_load_imp.hpp:54) ==13202== by 0x408ABD: void boost::serialization::stl::rebuild_collection<boost::archive::text_iarchive, std::vector<std::string, std::allocator<std::string> >, boost::serialization::stl::archive_input_seq<boost::archive::text_iarchive, std::vector<std::string, std::allocator<std::string> > >, boost::serialization::stl::reserve_imp<std::vector<std::string, std::allocator<std::string> > > >(boost::archive::text_iarchive&, std::vector<std::string, std::allocator<std::string> >&) (collections_load_imp.hpp:149) ==13202== by 0x408AF7: void boost::serialization::stl::load_collection<boost::archive::text_iarchive, std::vector<std::string, std::allocator<std::string> >, boost::serialization::stl::archive_input_seq<boost::archive::text_iarchive, std::vector<std::string, std::allocator<std::string> > >, boost::serialization::stl::reserve_imp<std::vector<std::string, std::allocator<std::string> > > >(boost::archive::text_iarchive&, std::vector<std::string, std::allocator<std::string> >&) (collections_load_imp.hpp:171) ==13202== by 0x408B1E: void boost::serialization::load<boost::archive::text_iarchive, std::string, std::allocator<std::string> >(boost::archive::text_iarchive&, std::vector<std::string, std::allocator<std::string> >&, unsigned) (vector.hpp:58) ==13202== by 0x408B5D: boost::serialization::free_loader<boost::archive::text_iarchive, std::vector<std::string, std::allocator<std::string> > >::invoke(boost::archive::text_iarchive&, std::vector<std::string, std::allocator<std::string> >&, unsigned) (split_free.hpp:58) ==13202== by 0x408B87: void boost::serialization::split_free<boost::archive::text_iarchive, std::vector<std::string, std::allocator<std::string> > >(boost::archive::text_iarchive&, std::vector<std::string, std::allocator<std::string> >&, unsigned) (split_free.hpp:74) ==13202== by 0x408BB1: void boost::serialization::serialize<boost::archive::text_iarchive, std::string, std::allocator<std::string> >(boost::archive::text_iarchive&, std::vector<std::string, std::allocator<std::string> >&, unsigned) (vector.hpp:76) ==13202== by 0x408BEF: void boost::serialization::serialize_adl<boost::archive::text_iarchive, std::vector<std::string, std::allocator<std::string> > >(boost::archive::text_iarchive&, std::vector<std::string, std::allocator<std::string> >&, unsigned) (serialization.hpp:140) ==13202== by 0x408C29: boost::archive::detail::iserializer<boost::archive::text_iarchive, std::vector<std::string, std::allocator<std::string> > >::load_object_data(boost::archive::detail::basic_iarchive&, void*, unsigned) const (iserializer.hpp:160) ==13202== by 0x4E5A5B9: boost::archive::detail::basic_iarchive::load_object(void*, boost::archive::detail::basic_iserializer const&) (in /usr/lib64/libboost_serialization.so.1.33.1) ==13202== by 0x4095F7: boost::archive::detail::load_non_pointer_type<boost::archive::text_iarchive, std::vector<std::string, std::allocator<std::string> > >::load_standard::invoke(boost::archive::text_iarchive&, std::vector<std::string, std::allocator<std::string> >&) (iserializer.hpp:375) I don't see any mistakes in my client code. Seems to be a problem of the boost::serialization library. (I'm using boost 1.33.1) Should I create a bug entry for that, or is it already known? Regards, Sascha

4 9

Boost multi_array declaration inside class header
by GQ Chen 21 Jul '07

21 Jul '07

Hi, I have some troubles in compiling the following simple code, tboost.hpp, tboost.cpp The problem is always from the declaration of array_type A(boost::extents[2000000][1]); in the header file. Error message from compiling, In file included from tboost.cpp:2: tboost.h:14: error: `extents' has not been declared boost.h:14: error: ISO C++ forbids declaration of `parameter' with no type If I define array_type A(boost::extents[2000000][1]); in the source code tboost.cpp, it will compile successfully. Could anyone help me to solve it? Thanks a lot, gq tboost.hpp #include <cassert> #include <stdio.h> #include <iostream> #include "boost/multi_array.hpp" typedef boost::multi_array<double, 2> array_type; class tboost { public: tboost(){}; ~tboost(){}; public: array_type A(boost::extents[2000000][1]); }; #endif tboost.cpp #include "tboost.h" int main () { char c =getchar(); return 0; }

2 1

boost::format convert snprintf.
by chun ping wang 21 Jul '07

21 Jul '07

Hmm is there a way for boost::format to function like snprintf. Example char str[64]; std::cout << boost::format("%s") % str;

2 2

program_options, fvisibility=default, and Mac OS X
by James Jackson 20 Jul '07

20 Jul '07

Hi, Following on from this posting: http://lists.boost.org/boost-users/2007/02/25361.php I am having the same problem with program_options using std::string, compiling my code with gcc-4.0, using Xcode 2.2.1 on Mac OS X 10.4.9. Forcing -fvisibility=default onto the command line does not fix the problem. Do I need to rebuild the libraries with this flag (and if so, how?), or can somebody offer any other suggestions? Regards, James Jackson.

1 0

[Serialization] is_serializable trait?
by Sohail Somani 20 Jul '07

20 Jul '07

Hi, I know there is the tracking level trait. Is that sufficient to be able to say that T is serializable? struct T { T(some_stuff); private: some_non_default_constructing_members; }; In this case, I think tracking level will be set to track selectively but its not serializable as is. Thanks, Sohail

3 2

Graph Permutations of Order n
by Brian Stadler 20 Jul '07

20 Jul '07

Hi list, I'm attempting to create all graphs of order n, isomorphic graphs and all. Basically, with vertices V and Edges E, permutate through all possible variations. Does anyone on this list know of a simple way to do this with the boost graph library or, can point to some location that may help. thanks

2 1