- Boost-users - lists.boost.org

wstring in serialization?
by dennis＠conus.info 16 Apr '08

16 Apr '08

Hello, >> c:/boost/boost/serialization/access.hpp:109: error: 'struct >> std::basic_string<wchar_t, std::char_traits<wchar_t>, >> std::allocator<wchar_t> >' has no member named 'serialize' >> >> So, is there wstring support for serialization, or I should write my >> own? > > Hi Dennis, > sure there is, did you include <boost/serialization/wstring.hpp>? At least in latest boost distribution, this file (wstring.hpp) is not exist.

2 2

Building Boost as Universal Binary on Mac OS X
by Thomas Hauk 16 Apr '08

16 Apr '08

I would like to build 1.35.0 Boost libs (especially date_time, filesystem, regex, system, and thread) on my G5 as Universal Binaries with bjam 3.1.16, so I can use them in a project with Xcode on both my G5 and my Intel-based MacBook Pro. (I was surprised they did not build that way by default). The command I used was: $> bjam --toolset=darwin --build-type=complete --with-thread --with- date_time --with-filesystem --with-regex --with-system I know they're just built PPC because I used the `file` command, e.g.: $> file libboost_system-mt-d-1_35.dylib libboost_system-mt-d-1_35.dylib: Mach-O dynamically linked shared library ppc Can anyone help? TIA, T -- "It's a very sobering feeling to be up in space and realise that one's safety factor was determined by the lowest bidder on a government contract.'' -- Alan Shepherd

2 2

[MPI] compilation error with msvc 64bit due to size_t datatype
by Andreas Buhr 16 Apr '08

16 Apr '08

Hello, I cannot compile boost.mpi with msvc 8.0 64bit. The complete error message is attached. The reason is that is_mpi_datatype<std::size_t> derives from mpl::true_ with the 32 bit compiler while it derives from mpl::false_ with the 64bit compiler. Unfortunately, I do not understand the reason, so I cannot fix it. I would be grad if someone could help. best regards, Andreas Buhr -- Ist Ihr Browser Vista-kompatibel? Jetzt die neuesten Browser-Versionen downloaden: http://www.gmx.net/de/go/browser

3 3

Unhandled exception using directory_iterator
by Silvio Reis Junior 16 Apr '08

16 Apr '08

Hello friends! I'm trying to use directory_iterator but when I try to list recursivelly an entire drive or a big directory I get an unhandled exception. I'm trying to figure out if I did something wrong but the code is very similar to the example on the boost web page. Here is the code: //------------------------------------------------------------------- VOID CFileSearch::InternalSearch( path RootDir ) { directory_iterator EndDirItr; for ( directory_iterator DirItr( RootDir ); DirItr != EndDirItr; ++DirItr ) { if ( is_regular( DirItr->status() ) ) { _tprintf( "FILE: %s\n", DirItr->path().string().c_str() ); } else if ( is_directory( DirItr->status() ) ) { _tprintf( "DIR : %s\n", DirItr->path().string().c_str() ); InternalSearch( DirItr->path() ); } } } //------------------------------------------------------------------- VOID CFileSearch::Search( string RootDir ) { path TempRootDir = system_complete( path( RootDir, native ) ); if ( exists( TempRootDir ) == true ) { InternalSearch( TempRootDir ); } } //------------------------------------------------------------------- CFileSearch FileSearch; FileSearch.Search( "c:\\" ); Is there something wrong in the above code? Thank you in advance. Silvio.

4 8

wstring in serialization?
by dennis＠conus.info 16 Apr '08

16 Apr '08

Hello, I have a simple piece of code: void save_tbl (std::map<std::wstring, std::map<WORD, std::map<WORD, std::map<WORD, int> > > > & tbl) { std::ofstream ofs ("prj12_tbl.dat", std::ios::trunc | std::ios::binary ); boost::archive::binary_oarchive oa(ofs); assert (ofs.good()); oa << tbl; ofs.close(); }; Got this errors: c:/boost/boost/serialization/access.hpp: In static member function `static voidboost::serialization::access::serialize(Archive&, T&, unsigned int) [with Archive = boost::archive::binary_oarchive, T = std::basic_string<wchar_t, std::char_tr aits<wchar_t>, std::allocator<wchar_t> >]': c:/boost/boost/serialization/serialization.hpp:81: instantiated from `void boost::serialization::serialize(Archive&, T&, unsigned int) [with Archive = boost::archive::binary_oarchive, T = std::basic_string<wchar_t, std::char_traits<wchar_ t>, std::allocator<wchar_t> >]' c:/boost/boost/serialization/serialization.hpp:140: instantiated from `void boost::serialization::serialize_adl(Archive&, T&, unsigned int) [with Archive = boost::archive::binary_oarchive, T = std::basic_string<wchar_t, std::char_traits<w char_t>, std::allocator<wchar_t> >]' c:/boost/boost/archive/detail/oserializer.hpp:151: instantiated from `void boost::archive::detail::oserializer<Archive, T>::save_object_data(boost::archive::detail::basic_oarchive&, const void*) const [with Archive = boost::archive::binar y_oarchive, T = std::basic_string<wchar_t, std::char_traits<wchar_t>, std::allocator<wchar_t> >]' collect.cpp:149: instantiated from here c:/boost/boost/serialization/access.hpp:109: error: 'struct std::basic_string<wchar_t, std::char_traits<wchar_t>, std::allocator<wchar_t> >' has no member named 'serialize' So, is there wstring support for serialization, or I should write my own?

2 1

mathematical interval limits
by Tim Blechmann 16 Apr '08

16 Apr '08

hi all, is there a c++/boost.math way to find the highest representable floating point number lower than a certain limit ... something like boost::supremum<float>(1.f)? thanks, tim -- tim(a)klingt.org http://tim.klingt.org Avoid the world, it's just a lot of dust and drag and means nothing in the end. Jack Kerouac

2 2

empty a thread_group
by Axel 16 Apr '08

16 Apr '08

Hello I m using a boost::thread_group instance to handle a limited number of threads. When the thread_group reach a defined size, I cal join_all() on it and I wonder how I could clear it without having to keep a pointer to its threads and call remove_thread() for each one. I thought about destruct th thread_group instance but I m not sure it's a good idea, so any advice is welcome. Regards Axel

2 1

boost.function and phoenix.bind
by Sergei Politov 15 Apr '08

15 Apr '08

The following example: ============================================== #include <boost/function.hpp> #include <boost/spirit/phoenix.hpp> int main() { boost::function<void()> f; phoenix::bind(f); } ============================================== Produces compilation warning on MSVC 9.0: C:\work\cpp\bind.cpp : warning C4717: 'phoenix::functor_action<boost::function<void __cdecl(void),std::allocator<void> > >::functor_action<boost::function<void __cdecl(void),std::allocator<void> > >' : recursive on all control paths, function will cause runtime stack overflow And stack overflow in runtime. What is wrong in this example?

2 1

boost interprocess: managed_shared_memory overhead
by hgdhfdghsa＠Safe-mail.net 15 Apr '08

15 Apr '08

Hello again, I have one managed_shared_memory, which contains exactly one instance of a certain structure, allocated via MySegment.find_or_construct<TheClass>(interprocess::unique_instance)(...); How can I determine how big the segment must be? interprocess::managed_shared_memory ConfigurationManagerSegment(interprocess::open_or_create, "sharedMem", sizeof(TheClass)+x); //What should be x? Thank you

2 1

Re: [Boost-users] libboost_filesystem compile error in qnx
by Ricardo Perrone 15 Apr '08

15 Apr '08

Hi Guenther Thanks for help me, the compilation was a successfuly. Now, the libraries (regex, filesystem) are installed best regards Ricardo > Date: Fri, 11 Apr 2008 07:29:50 +0000 (UTC) > From: Guenther Stattenberger > Subject: Re: [Boost-users] libboost_filesystem compile error in qnx > To: boost-users(a)lists.boost.org > Message-ID: > Content-Type: text/plain; charset=us-ascii > > Hi, > > according to the QNX documentation, strerror itself is thread-safe, > therefore I thought it correct to add the QNX switch to > libs\system\src\error_code.cpp:86 > > # if defined(BOOST_WINDOWS_API) || defined(__hpux) || defined(__sun)\ > || (defined(__linux) && (!defined(__USE_XOPEN2K) || > defined(BOOST_SYSTEM_USE_STRERROR)))\ > || (defined(__osf__) && !defined(_REENTRANT))\ > || (defined(__vms))\ > || (defined(__QNXNTO__)) // <=================================== > > Any comments on that? > > In addition, I had to uncomment the case statement for EALREADY later in this > file. Since the QNX errno.h header defines > #define EALREADY EBUSY > it won't compile otherwise. > > Hope that helps. > > best regards > Guenther ____________________________________________________________________________________ Be a better friend, newshound, and know-it-all with Yahoo! Mobile. Try it now. http://mobile.yahoo.com/;_ylt=Ahu06i62sR8HDtDypao8Wcj9tAcJ

1 0