March 2009 - Boost-users - lists.preview.boost.org

[serialization] nans and infinite floats in text based archives...
by François Mauger 08 Mar '09

08 Mar '09

Hi serializers, archivers, boosters and Robert, I've used boost:serialization (boost 1.34.1 under Linux ubuntu and/or SL4,, gcc 4.2) for one year. Thanks to this excellent library, our team store critical experimental data (3 TB using zipped text archives). Up to now, I met no problem with text/XML archives as all the stored double values pass the 'isnormal' check. Now my new data analysis programs needs to (de)serialize doubles even if they are NaNs or infinites (for it makes sense for me). Unfortunately, the load method for double fails while reading 'nan' (or 'inf') from the input for this is the standard behavior of the ostream with "! isnormal" doubles. After studying the source code for boost text oriented I/O archives in the hope to find a fix for my problem, I made an attempt to make boost aware of nans and +/-infs in a pragmatic way (I mean trying to save time and storage but keep the information and get rid of the stream error exception). I attached a tiny tarball to show the hack I've done. I'm not sure at all this is be a definitively good and elegant approach to handle nans within text based archives. So I'd appreciate to get some feedback and comments from Robert or any expert in this field. As I work only with Linux, gcc and 1.34.1, this approach is likely not to be portable... but maybe this could initiate some work to fix it in an official way. thanks for your attention and for all this very very useful material. regards frc -- Francois Mauger Laboratoire de Physique Corpusculaire de Caen et Universite de Caen ENSICAEN - 6, Boulevard du Marechal Juin, 14050 CAEN Cedex, FRANCE e-mail: mauger(a)lpccaen.in2p3.fr tel.: (0/+33) 2 31 45 25 12 fax: (0/+33) 2 31 45 25 49

2 3

Re: [Boost-users] [Proto] Annoying MSVC bug
by Vial, Florent 08 Mar '09

08 Mar '09

>> I'd like to encourage Microsoft to reopen this bug report and fix it. >> Could any of you with logins to Microsoft Connect go here >> https://connect.microsoft.com/VisualStudio/feedback/ViewFeedback.aspx?Fe edbackID=282942 >> and answer their 3 question survey expressing your dissatisfaction and >> encourage them to fix this. For those lazy not wanting to browse back like i did: the connect ID is 282942. >https://connect.microsoft.com/VisualStudio/feedback/ViewFeedback.aspx?F eedbackID=283021 And here 283021. Cheers, Florent Arnold & Richter Cine Technik GmbH & Co. Betriebs KG Sitz: Munchen - Registergericht: Amtsgericht Munchen - Handelsregisternummer: HRA 57918 Personlich haftender Gesellschafter: Arnold & Richter Cine Technik GmbH Sitz: Munchen - Registergericht: Amtsgericht Munchen - Handelsregisternummer: HRB 54477 Geschaftsfuhrer: Prof. Dr. Dejan Ilic; Franz Kraus; Jens Schafer

2 1

[mem_fn] Ambiguous call to mem_fn with stdcall function
by Conrad Scott 08 Mar '09

08 Mar '09

I'm trying to use g++ (and MinGW) to compile a win32 Direct3X program that I've written under Visual C++. All is well (so far) except for one thing. I'm using the technique suggested in the Boost shared_ptr documentation to store pointers to COM objects, which works fine under Visual C++ but under g++ (either 3.4.x or 4.3.x) I get errors about the call to mem_fn being ambiguous due to stdcall / non-stdcall overloading. I can't find any references to this problem on the net so I assume I'm doing something utterly boneheaded. Any hints would be very welcome. For example: #define BOOST_MEM_FN_ENABLE_STDCALL #include <d3d9.h> #include <boost/mem_fn.hpp> #include <boost/shared_ptr.hpp> void test_mem_fn(IDirect3DDevice9 * const device) { boost::shared_ptr<IDirect3DDevice9> ptr( device, boost::mem_fn(&IDirect3DDevice9::Release) ); } test/test_mem_fn.cpp:13: error: call of overloaded 'mem_fn(ULONG (IDirect3DDevice9::*)())' is ambiguous C:/Boost/include/boost-1_38/boost/bind/mem_fn_cc.hpp:15: note: candidates are: boost::_mfi::mf0<R, T> boost::mem_fn(R (T::*)()) [with R = ULONG, T = IDirect3DDevice9] C:/Boost/include/boost-1_38/boost/bind/mem_fn_cc.hpp:15: note: boost::_mfi::mf0_stdcall<R, T> boost::mem_fn(R (T::*)()) [with R = ULONG, T = IDirect3DDevice9] C:/Boost/include/boost-1_38/boost/mem_fn.hpp:382: note: boost::_mfi::dm<R, T> boost::mem_fn(R T::*) [with R = ULONG ()(), T = IDirect3DDevice9] Many thanks, Conrad

2 3

Options for --build-type
by Juan D Guzman 07 Mar '09

07 Mar '09

I'm building boost and was wondering other than --build-type=complete what are the other options. Disk usage approaches 40 gigs with the complete option.

2 2

Boost.python - How to export shared_ptr<vector<shared_ptr<string> > >?
by blp330 07 Mar '09

07 Mar '09

Hi, I have a function that returns type boost::shared_ptr<std::vector<boost::shared_ptr<std::string> > >, I have registered that: // boost::shared_ptr<std::string> -> PyString Converter. struct StringPointerToPyString { static ::PyObject* convert( ::boost::shared_ptr<::std::string> const& p_str ){ return ::PyString_FromStringAndSize( p_str->data(), p_str->size() ); } }; bp::class_< std::vector<StringPtr> > ("vector_StringPtr") .def (bp::vector_indexing_suite< std::vector<StringPtr> >()); bp::register_ptr_to_python< boost::shared_ptr< std::vector<StringPtr> > >(); It works fine but when I try to access the returned vector, I got error: >>> v = my_function() >>> v[0] TypeError: No Python class registered for C++ class class boost::shared_ptr<class std::basic_string<char,struct std::char_traits<char>,class std::allocator<char> > > I tried to add bp::register_ptr_to_python< boost::shared_ptr< std::string > >(); But still not work. How can I make it work? Thanks you very much. -- View this message in context: http://www.nabble.com/Boost.python---How-to-export-shared_ptr%3Cvector%3Csh… Sent from the Boost - Users mailing list archive at Nabble.com.

1 0

Portable way of getting process directory, temp directory?
by Archie14 07 Mar '09

07 Mar '09

Is there a boost library to get process directory, or temp directory?

1 0

MPL - what rule am I breaking?
by Noah Roberts 06 Mar '09

06 Mar '09

Version A (works): template < typename FIELD > struct rec { typename FIELD::type value; rec() : value() {} protected: typename FIELD::type & ref() { return value; } }; template < typename VEC > struct generate_record : boost::mpl::inherit_linearly< VEC, boost::mpl::inherit< rec<pl::_2>, pl::_1> >::type { typedef VEC fields; }; Version B (fails): template < typename FIELD, bool Writable > struct rec { typename FIELD::type value; // failure point rec() : value() {} protected: typename FIELD::type & ref() { return value; } }; template < typename VEC > struct generate_record : boost::mpl::inherit_linearly< VEC, boost::mpl::inherit< rec<pl::_2, true>, pl::_1> >::type { typedef VEC fields; }; The error I get: 'type' : is not a member of 'boost::mpl::arg<2>' I think I know what's going on, rec<pl::_2, true> is getting immediately instantiated while rec<pl::_2> does not. What I don't know is why or how to fix it. Thanks.

2 2

proto: analytical-only math functions
by Hicham Mouline 06 Mar '09

06 Mar '09

Hello, There are a few posts these days about generic libraries for math derivation of math functions, analytically or numerically. My objective is to start simple with analytical functions, and to expand the result gradually. I have used proto to define terminals : 1. 1D constants 2. 1D variables 3. 1D basic functions <double (double)> ( c++03 cmath + boost::math ) 4. user-defined functions ( 0 or more dimensions ) HEADER struct constant_tag {}; typedef proto::terminal< constant_tag >::type constant; struct variable_tag {}; typedef proto::terminal< variable_tag >::type variable; template < double (*basic_function_ptr)(double) > struct basic_function_tag { typedef double result_type; result_type operator()(double d) const { return basic_function_ptr(d); } }; proto::terminal< basic_function_tag<std::log> >::type const log = {{}}; // 1 example, I have one of these for each function template <size_t dim =1> struct function_tag {}; template <size_t dim> struct function : proto::terminal< function_tag<1> >::type {}; USER CODE ------------------ constant c; variable x,y,z,t,u,v; function<0> h; function<1> f; function<5> g; My grammar will be very simple for a start. All I want to allow are statements like: 1.variable = constant(proto terminal) or literal (actual number, does it have to be proto terminal), for e.g. x = 5; 2. function definition with the = operator. The LHS of the operator can one of: h = h() = f(x) = g(x,y,z,t,u) = // All arguments of f should be "variable"s only and they can't be repeated, and they have to match the dimension my LHS grammar looks like: template <size_t dim> struct function_def_lhs_grammar : proto::function< function_tag<dim>, proto::vararg< proto::terminal< variable_tag > > > {}; template <> struct function_def_lhs_grammar<0> : proto::or_< proto::terminal< function_tag<0> >, proto::function< function_tag<0>, proto::vararg< proto::terminal< variable_tag > > > {}; But vararg is not what I want at all: I want to have 0 variables (nothing) for the 0-dim function in the proto::function case, And I want to have exactly dim variables in the dim>0 proto::function case. Is it possible to "math-define" the function at the same time of the function<> object definition (in c++ terms) and have the dimension deducted, like: function f(x,y,z) = x+y+z ; The next step will be to worry about the RHS which is the much more complicated bit, regards,

4 18

boost CPDAG help
by montassar benmessaoud 06 Mar '09

06 Mar '09

Hi, I would know if it is possible to use Boost to code a partially directed graph (Completely partially DAG) CPDAG. I have a look to the class adjacency_matrix but it is used only for completely directed graph or completely undirected graph. Would you please let me know if there is any way to manipulate partially directed graph with Boost. very thank you in advance _________________________________________________________________ Découvrez Windows Live Spaces et créez votre site Web perso en quelques clics ! http://spaces.live.com/signup.aspx

2 1

[regex] How to use ICU debug libs when building regex with bjam (win32)
by Christian Hoffmann 06 Mar '09

06 Mar '09

Hi, In our debug builds, we build ICU in debug mode and it creates libraries with a "d" suffix (e.g. icuucd.lib). When I build boost regex now with: bjam.exe -sICU_PATH=<somepath>\icu -d+2 --debug-configuration toolset=msvc-9.0 threading=multi python-debugging=on link=shared variant=debug --without-mpi libs\regex\build it doesn't find the ICU libs and the linker complains. I then tried to give ICU_LINK (e.g. like this): bjam.exe -sICU_PATH=<somepath>\icu -sICU_LINK=icuucd.lib toolset=msvc-9.0 threading=multi python-debugging=on link=shared variant=debug --without-mpi libs\regex\build but it doesn't add the icuucd.lib to the linker file "bin.v2\libs\regex\build\msvc-9.0\debug\python-debugging-on\threading-multi\boost_regex-vc90-mt-gyd-1_38.dll.rsp" and I get unresolved symbols. Any idea how to do this without hacking the Jamfile.v2? Thanks a lot, Chris

1 0