June 2006 - Boost - lists.boost.org

[system] - system_message and gai_strerror function
by Oliver.Kowalke＠qimonda.com 26 Jun '06

26 Jun '06

Hello, the system_message function is implemented with ::strerror. How to handle systems erros related like EAI_NONAME (host not found), EAI_MEMORY (allocation failure) etc.? This erros occure in the context of getaddrinfo (related to network address and service translation) and requier ::gai_strerror in system_message. regards, Oliver

2 1

[asio] new version, prototype iostreams support
by Christopher Kohlhoff 26 Jun '06

26 Jun '06

Hello all, asio version 0.3.7 has been released. It may be downloaded from the asio sourceforge page at http://asio.sourceforge.net. This release corresponds to what is currently checked into Boost CVS. A list of the main changes in this version since 0.3.6 is available at: <http://sourceforge.net/project/shownotes.php?group_id=122478&release_id=426…> As previously discussed, in this version I have included some prototype iostreams and streambuf support. Feedback on this would be appreciated. An example use of the iostreams classes in a client program can be seen here: <http://asio.sourceforge.net/boost_asio_0_3_7/libs/asio/doc/examples/iostrea…> The iostreams classes may also be used in simple servers, as illustrated here: <http://asio.sourceforge.net/boost_asio_0_3_7/libs/asio/doc/examples/iostrea…> To better support line-based (or other delimiter-based) protocols, a new asio::streambuf class, read_until() and async_read_until() functions have been added. An example showing these in a synchronous use case (an HTTP 1.0 client) can be seen here: <http://asio.sourceforge.net/boost_asio_0_3_7/libs/asio/doc/examples/httpcli…> And the asynchronous equivalent is available at: <http://asio.sourceforge.net/boost_asio_0_3_7/libs/asio/doc/examples/httpcli…> Cheers, Chris

8 11

[Boost-bugs] [ boost-Support Requests-1512679 ] Boost.Multiindex + Boost.Shmem
by SourceForge.net 26 Jun '06

26 Jun '06

Support Requests item #1512679, was opened at 2006-06-26 05:46 Message generated for change (Tracker Item Submitted) made by Item Submitter You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=207586&aid=1512679&group_… Please note that this message will contain a full copy of the comment thread, including the initial issue submission, for this request, not just the latest update. Category: None Group: None Status: Open Resolution: None Priority: 5 Submitted By: Nobody/Anonymous (nobody) Assigned to: Nobody/Anonymous (nobody) Summary: Boost.Multiindex + Boost.Shmem Initial Comment: Hi, I am trying to get a multiindex to work on a shared memory using boost.shmem. After 2 weeks I have already succeeded in creating it and placing it in a shared but for some strage reason it won't work when retrieving values. I think it has to do with the fact that the multiindex is not using offset_ptr put raw ptrs, though this shoulnd't be a problem as I am using a fixed address... Please find the code attached. You can contact me at bblasi at jblasi.com or post here as I will chech regulary. Too much for me this one! ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=207586&aid=1512679&group_… Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642 _______________________________________________ Boost-bugs mailing list Boost-bugs(a)lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/boost-bugs

1 0

Re: [boost] [test] test/included min max macro issue
by Martin Bonner 26 Jun '06

26 Jun '06

----Original Message---- From: boost-bounces(a)lists.boost.org [mailto:boost-bounces@lists.boost.org] On Behalf Of Paul A Bristow Sent: 24 June 2006 14:34 To: boost(a)lists.boost.org Subject: Re: [boost] [test] test/included min max macro issue > A further quick question on this problem. > > It is my custom (having a strong view that it aids readers - > including me - > of my programs) to write > > #include <limits> > using std::numeric_limits; > > How do I avoid min/max macro problems when writing, for example, > > return numeric_limits<double>::max(); ? > > (numeric_limits<double>)::max(); ??? No. >From the website: >> > # If you want to call std::numeric_limits<int>::max(), use >> > (std::numeric_limits<int>::max)() instead.> The basic rule is that you must never have the (pp-)tokens "max" and "(" one after the other. If you do, the preprocessor will try to expand the max macro. The solution is to wrap the function name in parens, so what you would write is: return (numeric_limits<double::max)(); That means the preprocessore sees "max" and ")", and doesn't try to expand the macro. -- Martin Bonner Martin.Bonner(a)Pitechnology.com Pi Technology, Milton Hall, Ely Road, Milton, Cambridge, CB4 6WZ, ENGLAND Tel: +44 (0)1223 203894

4 4

[guid] - serialisation support
by Oliver.Kowalke＠qimonda.com 26 Jun '06

26 Jun '06

Hello, wouldn't it be suitable to add serialasation support to boost::guid? regards, Oliver #include <boost/serialization/split_member.hpp> class BOOST_GUID_DECL guid : public boost::totally_ordered< guid > { private: ... friend class boost::serialization::access; template< class Archive > void save( Archive & ar, const unsigned int version) const { std::string str = to_string(); ar & str; } template< class Archive > void load( Archive & ar, const unsigned int version) { std::string str; ar & str; construct( str); } BOOST_SERIALIZATION_SPLIT_MEMBER(); ... };

1 0

[multi_array][cw-8_3] Fixes for CW 8.3
by Joaquín Mª López Muñoz 26 Jun '06

26 Jun '06

I'm proposing patches for two problems CW 8.3 is currently having with multi_array: 1. CW 8.3 does not support the default arg in the following declaration of a member function of index_range (boost/multi_array/index_rang.hpp): index get_start(index low_index_range = index_range::from_start()) const unless index_range::from_start() has been previously declared, which has not in the current codebase. The fix consists simply in moving from_start upward inside the definition of index_range, see attached file index_range.hpp.diff. This fix is absolutely innocent and cannot break other compilers. 2. In CW 8.3, std::ptrdiff_t is *not* int, which, combined with the fact that this compiler does not support SFINAE, makes the line 243 of libs/multi_array/test/generative_tests.hpp A.reindex(1); to unintendedly resolve to a call to this overload of const_multi_array_ref: // disabled for integral types in SFINA-capable compilers template <class BaseList> void reindex(const BaseList& values); The fix consists in modifying generative_tests.hpp so as to make sure we are calling the overload of reindex taking a boost::multi_array_types::index value: #ifdef BOOST_NO_SFINAE typedef boost::multi_array_types::index index; A.reindex(index(1)); #else A.reindex(1); #endif See the attached patch file generative_tests.hpp.diff. Curiously enough, the typedef above was already defined, but not used, in the CVS version of generative_tests.hpp, which leads me to think that the author was thinking of this problem but somehow forgot about it before making effective use of the typedef. Is there any objection to these two patches being commited to the trunk and RC_1_34_0 branch? Thank you, Joaquín M López Muñoz Telefónica, Investigación y Desarrollo 33a34,40 > private: > static index from_start() > { return (std::numeric_limits<index>::min)(); } > > static index to_end() > { return (std::numeric_limits<index>::max)(); } > 135,140d141 < private: < static index from_start() < { return (std::numeric_limits<index>::min)(); } < < static index to_end() < { return (std::numeric_limits<index>::max)(); } 51a52 > #include <boost/config.hpp> /* BOOST_NO_SFINAE */ 241a243 > #ifdef BOOST_NO_SFINAE 243c245,248 < A.reindex(1); --- > A.reindex(index(1)); > #else > A.reindex(1); > #endif

1 1

[multi_array][vc-6_5/7_0] compiler problem with dependent typename
by Joaquín Mª López Muñoz 26 Jun '06

26 Jun '06

MSVC 6.5 and 7.0 have a hard time with line 391 of boost/multi_array.hpp, which involves a dependent typename keyword: ranges.ranges_[i] = typename gen_type::range(0,extents[i]); (See for instance http://tinyurl.com/pjxxl .) The attached patch fixes the problem (confirmed in MSVC 6.5, only an educated guess for MSVC 7.0.) Any objection to this being commited to the trunk and RC_1_34_0? Thank you, Joaquín M López Muñoz Telefónica, Investigación y Desarrollo 391c391,392 < ranges.ranges_[i] = typename gen_type::range(0,extents[i]); --- > typedef typename gen_type::range range_type; > ranges.ranges_[i] = range_type(0,extents[i]);

1 1

[system] - bug in system_message
by Oliver.Kowalke＠qimonda.com 26 Jun '06

26 Jun '06

Hello, I think there are some bugs in the implementation: void system_message( error_code ec, std::string & target ) { target += std::strerror( sys_err_code ); // shoudn't this be target += std::strerror( ec); } In the documentation the signature of the function should be std::string& system_message(error_code err, std::string& target); std::string& system_message( error_code ec, std::string & target ) { target += std::strerror( ec); return target; } regards, Oliver

1 0

[asio] WSASend needs reinterpret_cast<WSAOVERLAPPED*>.
by Felipe Magno de Almeida 26 Jun '06

26 Jun '06

Hi Christopher, The VC7.1 complains that win_iocp_operation cant be converted to WSAOVERLAPPED when tried to pass it to WSASend. I inserted this on win_iocp_socket_service.hpp:526 - &bytes_transferred, flags, ptr.get(), 0); +&bytes_transferred, flags, reinterpret_cast<LPWSAOVERLAPPED>(static_cast<LPOVERLAPPED>(ptr.get())), 0); And it worked here. -- Felipe Magno de Almeida

2 1

Re: [boost] problems using build 2
by Reece Dunn 26 Jun '06

26 Jun '06

Arkadiy Vertleyb wrote: > Now I configured both: > > ---------- > using gcc ; > using msvc : 7.1 ; > ---------- > > But, even if I want to use msvc, it seems to go with gcc anyway: > > C:\ark\boost-cvs\boost\libs\typeof\test>bjam --v2 -a -s"TOOLS=msvc" You are using the BoostBuild V1 way of specifying toolsets. For V2, use the simpler: bjam --v2 -a msvc or bjam --v2 -a msvc-7.1 if you want to specify the version. V2 will use the first toolset initialized (gcc in this case) if no toolset is passed to bjam, which is why you were getting the gcc build. > Also how would I configure different versions of Visual C++ ? Easy: using msvc : 7.1 ; using msvc : 8.0 ; or if you want to auto-detect them: using msvc : all ; The BBv2 documentation at http://www.boost.org/boost-build2 should help you out with further queries. HTH, - Reece _________________________________________________________________ Try Live.com: where your online world comes together - with news, sports, weather, and much more. http://www.live.com/getstarted

1 0