October 2010 - Boost-users - lists.preview.boost.org

Proto domains shenanigans
by joel falcou 13 Oct '10

13 Oct '10

OK, I have a situation in which proto domain/subdomain could help me but i can't wrap my head around a proper solution. Here is the deal, i have various sub EDSL that goes like: - constants EDSL provides named consatnts (like zero_, one_, pi_ etc) that are meant to be terminal usable in any other EDSL en provide a type based way to know which constant is beging used and provide optimization in came it helps (e.g detecting x < zero_ in SIMD EDSL to replace with the SSSE3 corresponding fused operation) - a SIMD vector EDSL - various container based EDSL : table (multidim array), vector, matrix and tensor, polynom and rigid_transform. The rules I want to have is : - constant can mix with any other EDSL - SIMD don't mix with anybody - table mix with every other container and make them evaluates as table (e.g matrix * table == elementwise product and not matrix product) - vector, matrix, tensor can mix - polynom can mix with table but not with vector, matrix, tensor - rigid_transform van mix with table and matrix However i can't find any combination of domain definition that satisfy all. Is there some systematic methodologies to not fail at that ?

2 3

[asio] DNS query
by for-gmane 13 Oct '10

13 Oct '10

Hi, is it possible to do DNS queries with asio? (for example getting the MX record(s) of a domain)

4 3

[units] quantities with system abstraction "any<length>"
by alfC 13 Oct '10

13 Oct '10

Hi, In order to avoid reinventing the wheel I was wondering if there is any effort or there was any attempt to further abstract boost::units::quantities from its unit system but with a given dimension. That is, an object that can hold any boost::units::quantity<unit<Dimension, System> > with a given Dimension fir any System. For example (names are not ideal) any<length_dimension> a1 = 2.*si::meter; // RHS is type quantity<si::length> any<length_dimension> a2 = 2.*cgs::centimeter; //RHS is type quantity<cgs::length> any<length_dimension> b1 = 2.3 * si::second; // ERROR. or can hold any other quantity with length dimension (from any system). A sort of boost::variant<quantity<Dimension, system1> , quantity<Dimension, system2>, ... >, (but since the list is of type is not well defined maybe variant is not an options.) I know that there are some unresolved issues, for example, what is the result of a1 + a2, for which I have some solutions in mind. Anyway I just was wondering if this was tried before. And if not, what would be the best way to achieve it (boost::any? boost::variant?). (Note that all quantity<Unit> types has always the sizeof a double.) Alfredo

2 1

Boost serialization through base class
by Joe Drumm 13 Oct '10

13 Oct '10

Hi, I've tried using the most basic example I can find in the documentation to serialize a class using a pointer to a polymorphic base. However, it only serializes the base class. I'm currently on 1.34.1. Am I missing something obvious? Sample code below: class base { public: friend class boost::serialization::access; template<class Archive> void serialize(Archive& ar, const unsigned int) { } }; class derived : public base { public: derived() : foo(5) { } friend class boost::serialization::access; template<class Archive> void serialize(Archive& ar, const unsigned int) { boost::serialization::base_object<base>(*this); ar & BOOST_SERIALIZATION_NVP(foo); } int foo; }; BOOST_CLASS_EXPORT_GUID(base, "base") BOOST_CLASS_EXPORT_GUID(derived, "derived") TEST(OMSTestSuite, BasicDeserializeTest) { std::ofstream ofs("test.out"); boost::archive::xml_oarchive toa(ofs); base *b = new derived; toa & BOOST_SERIALIZATION_NVP(b); }

2 1

[String algo] Breaking change between 1.41 and 1.44 in split
by Christophe Henry 13 Oct '10

13 Oct '10

Hi, I noticed a breaking change in the split function when the input string is empty. Consider the following example: #include <string> #include <vector> #include <iostream> #include <boost/algorithm/string/split.hpp> #include <boost/algorithm/string/classification.hpp> int main() { std::vector<std::string> result; std::string empty_string; boost::algorithm::split( result, empty_string,boost::algorithm::is_any_of( "\t" ),boost::algorithm::token_compress_off); std::cout << "result size: " << result.size() << std::endl; return 0; } With boost 1.41 I get: result_size: 1 With 1.44 I get: result_size: 0 Meaning that splitting an empty string now has a new meaning. All I found to this subject was that 1 seems to be the correct answer (http://lists.boost.org/Archives/boost/2005/07/90266.php) Furthermore, there has been, according to the release notes, no change to string algo since the 1.41. This brings me to the question, before I start updating my code, is this change a design decision or an accident? Thanks, Christophe

1 0

[Accumulators] How to define the initial value ?
by Adrien Todeschini 13 Oct '10

13 Oct '10

Hi everyone, As the code of weighted_sum_impl constructor is [ in header <boost/accumulators/statistics/weighted_sum.hpp> ] template<typename Args> weighted_sum_impl(Args const &args) : weighted_sum_( args[parameter::keyword<Tag>::get() | Sample()] * numeric::one<Weight>::value ) { } and the template parameter Tag is defined as tag::sample namespace tag { struct weighted_sum : depends_on<> { typedef accumulators::impl::weighted_sum_impl<mpl::_1, mpl::_2, tag::sample> impl; }; } How can I define this tagged argument when initializing the accumulator_set ? ( Indeed, weighted_sum_impl constructor is called at this point ) Thanks in advance for your help. Adrien

2 6

[serialization] Deserialization of a float whose value is infinite throws an archive_exception ("stream error")
by Luca 12 Oct '10

12 Oct '10

Hello, using the Boost.Serialization library version 1.43, I saved to an XML archive a float number which has a value of infinite; then loading that XML file leads to a boost::archive::archive_exception, exception.what() return "stream error". The source code to reproduce the problem is the following: #include <boost/serialization/serialization.hpp> #include <sstream> #include <boost/archive/xml_woarchive.hpp> #include <boost/archive/xml_wiarchive.hpp> #include <iostream> struct XXX { friend class boost::serialization::access; template<class Archive> void serialize(Archive & ar, const unsigned int version) { ar & BOOST_SERIALIZATION_NVP(mValue); } float mValue; };//struct int main() { std::wstringstream lSS; boost::archive::xml_woarchive oa(lSS); XXX xxx; float zero = 0.f; xxx.mValue = 1 / zero; oa << BOOST_SERIALIZATION_NVP(xxx); std::wcout<<"saved:"<<lSS.str() << std::endl; boost::archive::xml_wiarchive ia(lSS); ia >> BOOST_SERIALIZATION_NVP(xxx); std::wcout<<"loaded:"<<lSS.str() << std::endl; return 0; } Using gcc 4.4.4 on Linux/Debian the mValue member is saved as: <mValue>inf</mValue> while on Visual Studio 2005 on Windows XP sp2 the mValue member is saved as: <mValue>1.#INF</mValue> I see that the value of infinite has a different representation, (and this could be the main motivation behind the choice to not support them), but this lead to a situation in which a properly saved archive is not loadable by the library itself. As far I can see, the issue is at this line 209 of \libs\serialization\src\basic_xml_grammar.ipp parse_info<BOOST_DEDUCED_TYPENAME std::basic_string<CharType>::iterator> result = boost::spirit::parse(arg.begin(), arg.end(), rule_); return result.hit; where the parse function fails since result.hit is equal to false. I resolved the issue by checking all the data preventing accurately any infinite value before it is saved to an xml archive, but I wonder if this problem belongs to the library itself or not. Greetings, Luca

2 1

[asio]How to insure async handles not be called after calling deadline_timer::cancel()?
by 鑫炜廖 12 Oct '10

12 Oct '10

The document says, If the timer has already expired when cancel() is called, then the handlers for asynchronous wait operations will: have already been invoked; or have been queued for invocation in the near future. These handlers can no longer be cancelled, and therefore are passed an error code that indicates the successful completion of the wait operation. Does this mean that async handlers may be called after calling deadline_timer::cancel() (say, the second curcumstance) ? If yes, it seems make things unintuitive, eg. delete the resources that associate with handlers after cancel(). Is there any way to suppress or identify these handlers? thx.

3 2

Re: [Boost-users] boost 1.44.0 undefined reference to boost::filesystem2::no_check
by Asif Fayyaz 12 Oct '10

12 Oct '10

No, that issue is a bit different and I have already included <boost/filesystem.hpp> I am again posting the code: --------------------------------Sample application source: #include <stdio.h> #include <boost/filesystem.hpp> int main() { std::string m_FileName(boost::filesystem::path("testfile1.txt", boost::filesystem::no_check).native_file_string()); printf("Testboost.....%s\n", m_FileName.c_str()); return 0; } -------------------------------- Asif ---------- Replied message ---------- From: Igor R <boost.li...(a)gmail.com> Date: Oct 11, 5:57 pm Subject: boost 1.44.0 undefined reference to boost::filesystem2::no_check To: BOOST Archives > Any one?? > any idea? > Waiting for response in this regard. There was a similar issues here: http://permalink.gmane.org/gmane.comp.lib.boost.user/62789 Doesn't it apply your case? do you include <boost/filesystem.hpp>? _______________________________________________ Boost-users mailing list Boost-us...@lists.boost.orghttp:// lists.boost.org/mailman/listinfo.cgi/boost-users

3 2

Distance/size of boost ranges for std::set.
by Matthew Gwynne 12 Oct '10

12 Oct '10

Hi, I'm working within a C++ library which makes extensive use of the range concept, and also of boost::size. The current version of the library makes use of boost 1_38, however, updating to the newer versions of boost, I've had to replace boost::size with boost::distance as some of the ranges involved, such as std::set don't have random access iterators. However, this presents a problem - it's not really acceptable within our application to have linear time complexity for the size operation on sets when the size member function of the set container is constant time, so I'm looking for some advice. Are we misusing the range concept here and should be using something else? Everything other than this matches up nicely, and it's only the time complexity with regards to boost::distance that will cause problems. Should we simply be extending the boost::size function with additional overloaded versions specific to std::set if we wish to use it in this way (is this reasonable?)? Thanks in advance! Matthew Gwynne

3 3

[shared_ptr] Better support for allocators
by Alex Dubov 12 Oct '10

12 Oct '10

Greetings. shared_ptr<> can be constructed with both custom allocator and deleter. However, there's an unfortunate shortcoming: if the held object needs to do some unusual housekeeping on itself during destruction, it needs to store an additional copy of the allocator or rely on some other custom functionality. It would be considerably more convenient, if shared_ptr<> could invoke a custom deleter with allocator (optionally) supplied: virtual void sp_counted_impl_pda::dispose() { d_(p_, a_); } Another handy alternative is to be able to obtain the allocator out of shared_ptr<> explicitly, in a manner similar to get_deleter(). Tr1 implementation actually constructs a proxy deleter object around the user supplied one, which inherits from allocator, but it's impossible to obtain a reference to it from the outside.

1 0

Variadic Macro Data library in the Boost sandbox
by Edward Diener 12 Oct '10

12 Oct '10

This may interest end-users who use macros/Boost PP in their own code to do preprocessor metaprogramming. I have put in the Boost sandbox the Variadic Macro Data library, or VMD. This library supplies variadic macro functionality and integrates variadic macros with the Boost PP library without changing the latter in any way. The library in the sandbox is in the variadic_macro_data folder. The library is a header-only library of macros. There is documentation and tests and the appropriate jamfiles for both. To build the documentation you need doxygen but I have also provided documentation in an html subfolder. Access to the Boost sandbox is explained on the Boost web site at http://www.boost.org/community/sandbox.html. The main use of the library is to be able to pass variadic data to macros, and in particular to extract variadic data elements or convert variadic data to Boost PP data types for use by Boost PP. The library also enhances the functionality of Boost PP tuple data. Installation is simply to take the boost and lib subfolders of the library and copy them to a Boost installation, preferable the trunk where support for variadic macros has been added in Boost config in the form of a BOOST_NO_VARIADIC_MACROS which is defined for compilers/versions which do NOT support variadic macros. To use the macros in the library, just "#include <boost/variadic_macro_data/VariadicMacroData.hpp>". Compilers which support variadic macros are gcc 3.0 and above and VC8 and above among a few others. You could also use the library on any installed Boost installation if the compiler you are using supports variadic macros. There is a discussion in the doc of why you might choose to use variadic macros rather than a Boost PP data type in a macro tailored for end-users. Comments, suggestions, and bug reports are welcome. Edward Diener

1 0

[array][bind] Binding boost array to a boost function
by Ryan McConnehey 11 Oct '10

11 Oct '10

I'm trying to bind a boost array to be called at a later time. This is giving me a lot of errors. Is what I'm attempting possible? Ryan template<typename T, std::size_t U> class Sorter { private: typedef boost::function<void(boost::array<T, U> & a)> t_SortFunction; t_SortFunction m_SortFunction; public: Sorter(t_SortFunction func) : m_SortFunction(func) {} bool sort(boost::array<T, U> & a) { m_SortFunction(a); } }; template<typename T, std::size_t U> void quicksort(boost::array<T, U> & a) { } int main(int argc, char* argv[]) { Sorter shell(boost::bind(quicksort, _1)); return 0; }

3 4

[filesystem][v3] remove_all and system::error_code error?
by valblade 11 Oct '10

11 Oct '10

Hi everyone, First of all, I want to thanks boost::filesystem contributors for their amazing work. I have a question in the recent filesystem V3. when using boost::uintmax_t remove_all(const path& p, system::error_code* ec=0); like this : ----------- listing ----------------------- remove_all("C:\\test_path_to_delete"); ------------- end listing --------------- the "C:\\test_path_to_delete" (and its descendent) is successfully deleted (as described in documentation) but when I use it like this : ----------- listing ----------------------- boost::system::error_code ec; remove_all("C:\\test_path_to_delete", ec); ------------- end listing --------------- The directory (and it descendent) aren't removed at all. I have investigated and, I found the following : boost::uintmax_t remove_all_aux(const path& p, fs::file_status sym_stat, error_code* ec) { boost::uintmax_t count = 1; if (!fs::is_symlink(sym_stat)// don't recurse symbolic links && fs::is_directory(sym_stat)) { for (fs::directory_iterator itr(p); itr != end_dir_itr; ++itr) { fs::file_status tmp_sym_stat = fs::symlink_status(itr->path(), *ec); HERE=> if (ec != 0 && ec) return count; count += remove_all_aux(itr->path(), tmp_sym_stat, ec); } } remove_file_or_directory(p, sym_stat, ec); return count; } This function is called by remove_all. Here, ec is passed by address. When remove_all("C:\\test_path_to_delete"); is called, &ec = 0 (default argument) so the check if (ec != 0 && ec) will never hit. But when called with remove_all("C:\\test_path_to_delete", ec); ec is allocated and &ec != 0, the check if (ec != 0 && ec) is allways true, and the function returns at the first iteration, I'm maybe wrong, but, isn't this what you meant to do? if (ec != 0 && *ec) return count; Sorry if I have misunderstood something... Thanks again.

1 0

Re: [Boost-users] boost 1.44.0 undefined reference to boost::filesystem2::no_check
by Asif Fayyaz 11 Oct '10

11 Oct '10

Any one?? any idea? Waiting for response in this regard. Asif -- Muhammad Asif Fayyaz Address: 42A:10 Kista Allevag. 16455 Kista Stockholm Sweden. Mobile: (+46)0707446631 Profile: http://www.isk.kth.se/~mafayyaz/

2 1

[asio] accepting all connections in a seperate thread - how?
by for-gmane 11 Oct '10

11 Oct '10

My server app needs to be "highly responsive". Ie. it shall handle connection attemps with a higher priority. Currently it has the following problem: if a handler function like the one below takes too much time, then unfortunately no new connections will be handled as long as the operation below lasts (ie. new connections have to wait). Which alternatives are there to solve this problem? (I was thinking of doing the accept in a seperate thread, but not sure how and whether that would work?) void session::handle_read(const boost::system::error_code& error, size_t bytes_transferred) { if (!error) { // PROBLEM HERE: // ... a lengthy job here blocks the whole app (ie. new connections have to wait) // ... which coding/design alternatives would solve this problem? boost::asio::async_write(socket_, boost::asio::buffer(data_, bytes_transferred), boost::bind(&session::handle_write, this, boost::asio::placeholders::error)); } else { delete this; } }

3 5

AIX and Boost.PropertyTree
by Will Mason 10 Oct '10

10 Oct '10

Hello, I'm using Boost 1.42, IBM XL C++ 10.1, and AIX 5.3. I'm running into trouble building code that uses Boost.PropertyTree. I get errors compiling the following code, which works fine on numerous other platforms: boost::property_tree::ptree pt; pt.put(LOG_FILE_KEY, props.log_file_name()); pt.put(LOG_LEVEL_KEY, props.log_level()); pt.put(LOG_NAME_KEY, props.log_name()); pt.put(LOG_FILE_SIZE_MAX_KEY, props.log_file_max_size()); pt.put(LOG_TO_CONSOLE_KEY, props.log_to_console()); pt.put(MESSAGE_MAX_KEY, props.message_max_size()); pt.put(SCAN_THREADS_KEY, props.scan_threads()); std::ostringstream prop_info; boost::property_tree::ini_parser::write_ini(prop_info, pt); The errors, all of which are attributed to the last line of code above, are: "/home/wmason/src/cap-aix/fim/../opt/include/aix/boost/v1.42/boost/property_tree/detail/ptree_implementation.hpp", line 55.28: 1540-0129 (S) The qualifier "subs" is not defined in the current scope. "/home/wmason/src/cap-aix/fim/../opt/include/aix/boost/v1.42/boost/property_tree/detail/ptree_implementation.hpp", line 78.34: 1540-0129 (S) The qualifier "subs" is not defined in the current scope. "/home/wmason/src/cap-aix/fim/../opt/include/aix/boost/v1.42/boost/property_tree/detail/ptree_implementation.hpp", line 91.42: 1540-0716 (S) The template argument "boost::iterator" does not match the template parameter "class Iterator". "/home/wmason/src/cap-aix/fim/../opt/include/aix/boost/v1.42/boost/property_tree/detail/ptree_implementation.hpp", line 101.42: 1540-0274 (S) The name lookup for "const_iterator" did not find a declaration. "/home/wmason/src/cap-aix/fim/../opt/include/aix/boost/v1.42/boost/property_tree/detail/ptree_implementation.hpp", line 101.42: 1540-1226 (I) Declarations for non-dependent names are resolved in the template definition. "/home/wmason/src/cap-aix/fim/../opt/include/aix/boost/v1.42/boost/property_tree/detail/ptree_implementation.hpp", line 101.42: 1540-1227 (I) "const_iterator" does not depend on a template argument. "/home/wmason/src/cap-aix/fim/../opt/include/aix/boost/v1.42/boost/property_tree/detail/ptree_implementation.hpp", line 116.51: 1540-0129 (S) The qualifier "subs" is not defined in the current scope. "/home/wmason/src/cap-aix/fim/../opt/include/aix/boost/v1.42/boost/property_tree/detail/ptree_implementation.hpp", line 138.45: 1540-0129 (S) The qualifier "subs" is not defined in the current scope. "/home/wmason/src/cap-aix/fim/../opt/include/aix/boost/v1.42/boost/property_tree/detail/ptree_implementation.hpp", line 78.34: 1540-0129 (S) The qualifier "subs" is not defined in the current scope. "/home/wmason/src/cap-aix/fim/../opt/include/aix/boost/v1.42/boost/property_tree/ini_parser.hpp", line 231.45: 1540-0700 (I) The previous message was produced while processing "class boost::property_tree::basic_ptree<std::basic_string<char,std::char_traits<char>,std::allocator<char> >,std::basic_string<char,std::char_traits<char>,std::allocator<char> >,struct std::less<std::basic_string<char,std::char_traits<char>,std::allocator<char> > > >::const_iterator". The release notes for Boost 1.42 at http://www.boost.org/users/news/version_1_42_0 claim that my combination of OS, compiler and Boost version are supported. I was wondering if anyone might know of a way around this problem besides not using Boost.PropertyTree. Thanks very much for any help, Will

3 3

[fusion] deduce type of resulting vector from a transformation
by alfC 10 Oct '10

10 Oct '10

Hi, (I keep struggling with Boos.Fusion.) How do I deduce the type of the vector resulting from the transformation of another vector? I have a source vector of different types, e.g. vector<quantity<si::length>, quantity<si::mass>, quantity<si::time> > and I have a transformation that is well defined and returns another type depending on the original type. The question is how do I generate a vector type that is the result of the given transformation. I though that result_of::transform was the function to obtain the result but it turns not because it returns a type transform_view instead of a vector. vector< ... > v1; T? v2(transform(v1, transformation())); i.e. what is the type of T as a fusion::vector as deduced from the type of v1 and the transformation type? -- Thank you, Alfredo-- Below is the illustrative code: #include<boost/array.hpp> #include <boost/fusion/container.hpp> #include <boost/fusion/algorithm/transformation/zip.hpp> #include <boost/units/systems/si.hpp> #include<iostream> using std::clog; using std::endl; using namespace boost::fusion; using namespace boost::units; // this is the transformation struct divide{ template <typename SignatureT> struct result; template<class Q> struct result<divide(Q const&)>{ typedef typename divide_typeof_helper<Q, quantity<si::time> >::type type; }; template<typename Q> typename result<divide(Q const&)>::type operator()(Q const& q) const { return q/quantity<si::time>(1.*si::second); } }; int main(){ typedef vector< quantity<si::length>, quantity<si::mass>, quantity<si::time> > source_type ; source_type v1(1.*si::meter, 2.*si::kilogram, 3.*si::second); clog << "v1: " << at_c<0>(v1) << ", "<< at_c<1>(v1) << ", " << at_c<2>(v1) << endl; // how can I deduce the result_type from the type "divide"? // here manually defined typedef vector< divide_typeof_helper<quantity<si::length>, quantity<si::time> >::type, divide_typeof_helper<quantity<si::mass>, quantity<si::time> >::type, divide_typeof_helper<quantity<si::time>, quantity<si::time> >::type > result_type ; result_type v2( boost::fusion::transform(v1, divide() ) ); clog << "v2: " << at_c<0>(v2) << ", "<< at_c<1>(v2) << ", " << at_c<2>(v2) << endl; return 0; }

3 4

[boost-install]update old version to new one by 'bjam install'
by hongleij 10 Oct '10

10 Oct '10

i have install 1.43 version in my system,after use “./bjam install” to intall boost1.44,i met the following error. [root@localhost gridcast]# g++ -o datetime datetime_tcp_async_server.cpp -lpthread -lboost_system -L/usr/local/lib datetime_tcp_async_server.cpp:142:2: 警告：文件未以空白行结束 /tmp/ccjG0R2d.o: In function `__static_initialization_and_destruction_0(int, int)': datetime_tcp_async_server.cpp:(.text+0x16c): undefined reference to `boost::system::get_system_category()' datetime_tcp_async_server.cpp:(.text+0x176): undefined reference to `boost::system::get_generic_category()' datetime_tcp_async_server.cpp:(.text+0x180): undefined reference to `boost::system::get_generic_category()' datetime_tcp_async_server.cpp:(.text+0x18a): undefined reference to `boost::system::get_generic_category()' datetime_tcp_async_server.cpp:(.text+0x194): undefined reference to `boost::system::get_system_category()' /tmp/ccjG0R2d.o: In function `boost::asio::error::get_system_category()': datetime_tcp_async_server.cpp:(.text._ZN5boost4asio5error19get_system_categoryEv[boost::asio::error::get_system_category()]+0x7): undefined reference to `boost::system::get_system_category()' /tmp/ccjG0R2d.o: In function `boost::system::error_code::error_code()': datetime_tcp_async_server.cpp:(.text._ZN5boost6system10error_codeC1Ev[boost::system::error_code::error_code()]+0x10):undefined reference to `boost::system::get_system_category()' collect2: ld 返回 1 so i delete /usr/local/include/boost dir ,and rerun ”./bjam install” -bash-3.2# g++ datetime_tcp_async_server.cpp -lpthread datetime_tcp_async_server.cpp:142:2: warning: no newline at end of file /tmp/ccQeKssV.o: In function `__static_initialization_and_destruction_0(int, int)': datetime_tcp_async_server.cpp:(.text+0x176): undefined reference to `boost::system::generic_category()' datetime_tcp_async_server.cpp:(.text+0x182): undefined reference to `boost::system::generic_category()' datetime_tcp_async_server.cpp:(.text+0x18e): undefined reference to `boost::system::system_category()' /tmp/ccQeKssV.o: In function `boost::system::get_system_category()': datetime_tcp_async_server.cpp:(.text._ZN5boost6system19get_system_categoryEv[boost::system::get_system_category()]+0x5): undefined reference to `boost::system::system_category()' /tmp/ccQeKssV.o: In function `boost::system::error_code::error_code()': datetime_tcp_async_server.cpp:(.text._ZN5boost6system10error_codeC1Ev[boost::system::error_code::error_code()]+0x17): undefined reference to `boost::system::system_category()' collect2: ld returned 1 exit status --------------------------------- the name changed from `boost::system::get_generic_category()' to `boost::system::generic_category()',so it seems if i already have an old version of boost,i need remove it clearly to reinstall the new one by myself.

2 3

[boost.datetime] query
by Sean Farrow 10 Oct '10

10 Oct '10

Hi: Is it possible to return a number representing the date. Like you can with COleDateTime.m_dt; Any help appreciated. Regards Sean.

3 2

[asio] How to get informed about failed connection attempts?
by for-gmane 10 Oct '10

10 Oct '10

The following code (from the examples) accepts a connection and calls the handler handle_accept(). But it calls the handler only if the connection attempt (internally) succeeds. The handler gets not called if the connection fails (for example when an ssl-connection-attempt fails due to wrong keys etc.). What should I do to get informed also about such failed connection attempts? acceptor_.async_accept(new_session->socket(), boost::bind(&server::handle_accept, this, new_session, boost::asio::placeholders::error));

1 0

Memory usage
by Olivier Tournaire 10 Oct '10

10 Oct '10

Hi all, Is there a (portable) way with boost libs to check memory usage? Best regards, Olivier

4 3

Http Server 3 example with SSL, ie. https
by for-gmane 09 Oct '10

09 Oct '10

Hi, has anybody extended the Http Server 3 example with SSL, ie. https? Are there any other simple https-server examples based on asio?

1 0

AIX large file support
by Will Mason 09 Oct '10

09 Oct '10

Hello, I'm using Boost 1.42, IBM XL C++ 10.1, and AIX 5.3. On AIX the way to enable large file support is to define _LARGE_FILES on the compiler command line. When I attempt to build Boost with this definition, I see errors like the following: "/home/wmason/src/cap-aix/fim/../opt/include/aix/boost/v1.42/boost/date_time/date_parsing.hpp", line 50.21: 1540-0215 (S) The wrong number of arguments have been specified for "tolower(int)". This causes catastrophic failure of Boost to build. Fortunately, by poking around the sources I found that I can enable large file support in Boost on AIX with IBM's compiler by defining _LARGEFILE64_SOURCE. Although this bypasses IBM's recommended way of enabling large file support, it does work to build Boost with large file support. The problem is that my product enables large file support the normal way, which leads me back to the above error when I try to build. I could just manually enable large file support in my code, but this causes portability problems. Not all operating systems support large files the same way, so trying to do it manually on every system will be a headache. Our portability strategy involves enabling large file support the normal way on every system, which in many cases, including AIX's, involves defining things on the command line. Is there any way to use normal large file support on AIX and not have Boost go down in flames? Thanks very much for any help, Will

1 0

[proto] Can't use template domain class with BOOST_PROTO_BASIC_EXTENDS
by joel falcou 09 Oct '10

09 Oct '10

Trying to use template domain class with BOOST_PROTO_BASIC_EXTENDS ends up in an error leading to the fact this ligne in the macro is incorrect when the Domain is template: BOOST_PROTO_BASIC_EXTENDS_(Expr, Derived, Domain) typedef void proto_is_aggregate_; typedef Domain::proto_generator proto_generator; Obviously this is alckign a typename before Domain::proto_generator if Domain is actually a template class. Thing is used to work before ( aorund 1.40) so I guess somethign changed around here. I guess a details meta-function extracting the domain will fix that or am I wrong ? namespace boost { namespace proto { template<class Domain> generator_of { typedef typename Domain::proto_generator type; }; } } BOOST_PROTO_BASIC_EXTENDS_(Expr, Derived, Domain) typedef void proto_is_aggregate_; typedef typename generator_of<Domain>::type proto_generator; Or is it ?

2 4

[variant][any] template parameterized variant
by alfC 09 Oct '10

09 Oct '10

Hi, I need a variant type object that can hold any type characterized by a template. For example Boost.Units have a family of types parameterized by "System". quantity< unit<dimension, System>, double> suppose dimension is fixed, e.g. typedef boost::units::length dimension is it possible to define a variant over that family of types? I could create a make_variant_over< list of all types > but that can be that list will be fixed and each time I defined a new System the list will have to grow. to simplify the question: template<class T> class dummy{ ... }; variant< "dummy<T>" > v; //for all T If it simplifies things, suppose that all dummy<T> have the same sizeof and that all are convertible to each other. Or should I be using an extension of boost::any with some adhoc restriction over the possible types on construction? Thank you, Alfredo

4 3

[fusion] .at<N>() template member in containers?
by alfC 09 Oct '10

09 Oct '10

Hi, Excuse my simple minded questions. Sometimes by asking this question (often with negative answer) I understand more on the design of the Boost libraries. Besides possible redundancy, why is that the fusion containers, for example fusion::vector, doesn't have a template member function to extract elements? fusion::vector<double, int, std::string> stuff(1.2, 3, "hola"); stuff.at<2>(); //returns "hola", just like at_c<2>(stuff); Wouldn't it be more consistent with std::vector v, v[2] or v.at(2); By the way, does anybody know where the name given "at_c" (I mean the '_c') comes from? Thank you, Alfredo

7 10

boost/preprocessor checking if argument is a constant
by Z S 09 Oct '10

09 Oct '10

All the BOOST_PP functions only work on integer constants. I want to write macros, where the same interface can be used for both constants and variables, and only call the BOOST_PP if the argument is a constant, and use runtime code if it is a variable. For example, #define TWICE(n) BOOST_PP_IF( _IS_CONSTANT(n), \ BOOST_PP_MUL(n,2), \ ( (n) * 2) ) For more details, you can also see the original question here : http://stackoverflow.com/questions/3884489/mixing-variables-and-integer-con… Is there a way to do so? The C preprocessor knows whether something is a constant, so we should be able to do this somehow. Or should I suggest this as a feature request for the Boost Preprocessor library? Thanks!

4 3

Re: [Boost-users] Boost Logos: proposed high constrast logo
by Zoltan Juhasz 08 Oct '10

08 Oct '10

Follow up on the original topic: I've uploaded a high-contrast version of the Boost Logo, and moved all other logo related resources to the follow Google Docs folder: https://docs.google.com/leaf?id=0B_WA6oW88X1UZWZiNzUwYzctYzg1NS00ZmY5LTllMD… On the logo discussion: Various explanation could be made for the logo. For some these are molecular structures; for others this is a small matrix of hexagonal cells from a beehive; some may claim these are interconnected - in some places overlapping - software components etc. Probably all of these seem reasonable, maybe non of these for someone. My point is that one could come up with any justification for the hexagons, but it will not change others mindset on it. On the other hand... ... this is an abstract logo, which is fairly simple (well at least the stylized form), it does not bring any cultural or any specific technology connotations, and most importantly provides / provided a "face" for Boost in the last 4 (or 5?) years. This is a strong connection in public mind between the Boost community and the logo; where the previous one defines the later - which (the community, and the Boost project) has excellent reputation, which made the logo highly recognizable (like any famous company brand). The point is, it is not the logo which defines Boost, but the other way around. Ofc. if the logo was something like this: http://preview.tinyurl.com/2wcg757 that would be arguably a big deflection compared to the spirit of Boost, but I do not think this applies to the current case (chemical analogy). The other important point is that most of the developers that used Boost in the previous years, already made the connection between the logo and the Boost Libraries; making a radical change might be a bad idea, unless there is a really strong reason to do that (for example Boost ceases using C++, and focusing on Objective C from now), otherwise keeping the continuity is important. I hope the high contrast version of the logo is usable, let me know if you need further help with it. I also hope that I was able to give you some background on the abstract hexagons. Regards, Zoltan Juhasz

2 3

[proto] baby steps learning with proto, proxy object
by alfC 08 Oct '10

08 Oct '10

Hi, This can be an exercise for definining the scope of Boost.Proto for little things. I have a histogram class that internally looks like this class histogram{ ... bool operator()(double const& x){return impl.increment(x);} bool operator()(double const& x, double const& weight){return impl.accumulate(x, weight);} }; histogram h( .. bin spec .. ); double x = ... ; so to add a count to the bin of x, I call h(x); to add a count with weigh w to bin x, I call h(x,w); so far so good. Now suppose I want to "improve" the syntax. The desired syntax would be: h[x]++; h[x]+=w; I don't want to expose the internal representation of the bins, so I can't return a reference to something from h[x]. So I guess we can create a proxy of the expression h[x] , "subindex of historgram with double". The proxy can be used with the syntax to call either accumulate or increment depending on the application of += or + + to the proxy. The naive implementation is this: class histogram{ ... private: struct proxy_histogram_subscript{ histogram& self; double const& x; proxy_histogram_subscript(histogram& self, double const& x) : self(self), x(x){} proxy_histogram_subscript const& operator++() const{self.increment(x); return *this;} bool operator++(int) const{return self.increment(x);} bool operator+=(double const& weight) const{return self.accumulate(x, weight);} }; public: proxy_histogram_subscript operator[](double const& x){ return proxy_histogram_subscript(*this, x); } }; which allows the desired syntax. The question is, can Boost.Proto do better? What I don't understand is how proto can make such a mini language *inside* the class. Hopefully without adding to much lines of code. I can see that one could create a histogram terminal, but that will be public (e.g. from the main program) and that will allow to create all sorts of expression involving brackets of anything. Thank you, Alfredo

3 2

How to abort boost test execution on the failure of any BOOST_TEST_CASE
by Gaurav N Shah 08 Oct '10

08 Oct '10

Hi, I have sample code like this: void test1() { BOOST_REQUIRE(2 ==1); } void test2() { BOOST_CHECK(1 !=1); } boost::unit_test::test_suite* ts1 = BOOST_TEST_SUITE( "test_suite1" ); ts1->add(BOOST_TEST_CASE(&test1)); ts1->add(BOOST_TEST_CASE(&test2)); ::boost::framework::master_test_suite().add(ts1); In this case I do not want test2 function to be executed it test1 fails. How is it possible?? (I can not merge code in test1 and test2 into single function) Thank You, Gaurav This communication is for informational purposes only. It is not intended as an offer or solicitation for the purchase or sale of any financial instrument or as an official confirmation of any transaction. All market prices, data and other information are not warranted as to completeness or accuracy and are subject to change without notice. Any comments or statements made herein do not necessarily reflect those of JPMorgan Chase & Co., its subsidiaries and affiliates. This transmission may contain information that is privileged, confidential, legally privileged, and/or exempt from disclosure under applicable law. If you are not the intended recipient, you are hereby notified that any disclosure, copying, distribution, or use of the information contained herein (including any reliance thereon) is STRICTLY PROHIBITED. Although this transmission and any attachments are believed to be free of any virus or other defect that might affect any computer system into which it is received and opened, it is the responsibility of the recipient to ensure that it is virus free and no responsibility is accepted by JPMorgan Chase & Co., its subsidiaries and affiliates, as applicable, for any loss or damage arising in any way from its use. If you received this transmission in error, please immediately contact the sender and destroy the material in its entirety, whether in electronic or hard copy format. Thank you. Please refer to http://www.jpmorgan.com/pages/disclosures for disclosures relating to European legal entities.

2 1

Program Options.. Specifying "-e" as an option value.....
by david.weber＠l-3com.com 08 Oct '10

08 Oct '10

I have an application, which spawns off other applications. As input, it takes in environment variables and arguments to pass to the resultant application. For instance: mutate -E "env_var1=val" -a "." ls This will run: ls . and the environment will only have "env_var1" equal to "val" So. Something funny comes along when I try to pass "-e" as an argument. So, in the above would be: mutate -a "-e" -a "foo" xterm The program options parser attempts to parse the "-e", and says "I don't know what it is". Any thoughts on having it skip that option? Code Snippits: po::options_description visible("Program options. Usage %1%"); visible.add_options() ("env,E", po::value<string_vector_t>(), "Specify an environment variable to run the executable with. In the form of foo=bar") ("arg,a", po::value<string_vector_t>(), "Specify a command line argument to pass to the executable") ; // Set up some hidden options po::options_description hidden("Hidden options"); hidden.add_options() ("executable", po::value < std::string>(), "the executable to run") ; // Set up positional arguments. They are going to be execute po::positional_options_description pos; pos.add("executable", 1); // There can only be 1 item executed // Consolidate all the options into a single one po::options_description all_options("All Options"); all_options.add(visible); all_options.add(hidden); // Parse the command line options po::store(po::command_line_parser(argc, argv).options(all_options).positional(pos).run(), vm); po::notify(vm);

2 4

[Boost.Log] How to link with on MSVC 2008?
by RUOFF, LARS (LARS)** CTR ** 08 Oct '10

08 Oct '10

Hello, i have found the following instructions to build Boost.Log: > Boost.Log can be compiled by: > > 1. Getting 1.42 source release > 2. Unpacking Boost.Log on top of it > 3. Following the usual procedure: > > .\bootstap.bat > .\bjam --with-log I did the above starting from a 1.44 source release. This worked out well. But when trying to use the library I get a lot of linker errors in my project: 1>Linking... 1>libboost_log-vc90-mt-gd-1_44.lib(text_file_backend.obj) : error LNK2019: unresolved external symbol "class boost::filesystem2::file_status [...] (Very long error messages containing decorated names. The above is just the start) Linker path settings should be correct, since the linker correctly finds libboost_log-vc90-mt-gd-1_44.lib. Also the project is using other Boost libraries (Asio, shared-ptr, bind, posix-time, etc) without problems. Anyone managed to compile and use Boost.Log on MSVC 2008 and can help? Thank you, Lars Ruoff

4 7

[Math] Best Way to Produce Pi
by Hossein Haeri 08 Oct '10

08 Oct '10

Dear all, What's my best way to define a constant for Pi. I'm thinking of using the following <cmath> stuff: const double pi(4 * atan(1)); Does Boost provide a better solution? TIA, --Hossein

4 4

[boost-users] [bind][function] - too many copies
by Lorimer, Stephen 08 Oct '10

08 Oct '10

For a given member function of the form void foo::f0(), passing this to boost bind results in the class instance being copied 7 times. Storing the result of bind in boost function incurs another 11 copies (code and results are here: http://boost.codepad.org/6jBmoKuj) Using tr1::bind and function has slightly better results, but still not ideal: Doing just tr1::bind results in the class instance being copied 4 times. Storing the result of bind in tr1 function incurs another 3 copies (code and results are here: http://boost.codepad.org/w1EV6JBY) Am I right in saying this is because boost::bind uses normal templates while tr1::bind uses variadic templates? Either way this is not very efficient. Is there a way to improve this? ---------------------------------------------------------------------- This message w/attachments (message) is intended solely for the use of the intended recipient(s) and may contain information that is privileged, confidential or proprietary. If you are not an intended recipient, please notify the sender, and then please delete and destroy all copies and attachments, and be advised that any review or dissemination of, or the taking of any action in reliance on, the information contained in or attached to this message is prohibited. Unless specifically indicated, this message is not an offer to sell or a solicitation of any investment products or other financial product or service, an official confirmation of any transaction, or an official statement of Sender. Subject to applicable law, Sender may intercept, monitor, review and retain e-communications (EC) traveling through its networks/systems and may produce any such EC to regulators, law enforcement, in litigation and as required by law. The laws of the country of each sender/recipient may impact the handling of EC, and EC may be archived, supervised and produced in countries other than the country in which you are located. This message cannot be guaranteed to be secure or free of errors or viruses. References to "Sender" are references to any subsidiary of Bank of America Corporation. Securities and Insurance Products: * Are Not FDIC Insured * Are Not Bank Guaranteed * May Lose Value * Are Not a Bank Deposit * Are Not a Condition to Any Banking Service or Activity * Are Not Insured by Any Federal Government Agency. Attachments that are part of this EC may have additional important disclosures and disclaimers, which you should read. This message is subject to terms available at the following link: http://www.bankofamerica.com/emaildisclaimer. By messaging with Sender you consent to the foregoing.

4 3

Boost Logos: proposed high constrast logo
by Alfredo Correa 08 Oct '10

08 Oct '10

Hi all, In my opinion there should be a high contrast version of the Boost logo . (by that I mean either black and white or high contrast with very mild gradients on white background). Particularly I am talking about the three hexagons icon. As an example, the first attachment here has the official logo in SVG format. Because of the background gradient and the gradient fill of the hexagon it is not *obvious* how to convert this to high contrast. What should be replaced by black (or solid blue)? the hexagon filling? the hexagon borders? the background? A possible high contrast logo is attached in the second file. By no means official or the best option, it is just an possibility, maybe the simplest one. *Just an example*. Note that even the darkest blue used in the gradient of the official logo is not enough to contrast with white, so I used a darker blue. The reason I think a high contrast logo is the following: If you want to have a document printing quality (or high quality PDF) you need a vector format. If the printing or visualization is done in black and white, the logo should have high contrast otherwise the appearance will be very device dependent. While everyone should take care in printing in good devices having a logo that is barely visible in black/white/grey printing in some devices is not acceptable. As a test for what makes a high contrast logo a good one, just try printing and small version (one centimeter square) of the logo in a black and white laser printer with random text around it and you will see the problem with the low contrast original logo. It will appear like a grey blob at best. Please this thread is not about a new logo design just an adaptation of the existing one. This thread is not be related with the "Powered by" logo discussion Alfredo PS: the attachments are Inkscape SVG.

6 5

[statechart] Unhandled events
by Oleg Smolsky 07 Oct '10

07 Oct '10

Hey there, I've been using SC for a while and something has been bugging me - given a state machine with: OuterSt, OneSt and TwoSt as well as OneEv and TwoEv that are handled by OneSt and TwoSt respectively, it is possible to send a wrong event to either of the last two states. Such an "unhandled" event will be forwarded to OuterSt and then quietly eaten. My question is as follows - is it possible to define some kind of "catch all" handler in the OuterSt? I have already fixed a tricky bug in my code where a submitted event was lost, and wonder if there is a simple way to handle these situations? I can certainly add a custom_reaction<>(X) for each possible event to the OuterSt and catch issues that way, but that's far from ideal. Any suggestions would be appreciated. P.S. perhaps a compromise solution would be to add something like this to OuterSt: sc::bug<OneEv> sc::bug<TwoEv> but that would effectively limit handling to issuing an assert(false)... Thanks, Oleg.

2 2

[boost-spirit qi] internal copies
by Ovanes Markarian 07 Oct '10

07 Oct '10

Hello *, I think this is more the question to spirit devlopers. I would like to know how spirit internally maintains the parsed data? Here is an example of what I mean. I have to parse some byte sequence represented as hex pairs: 00 ce ab 00 de ad be ef ... If I use the parser like this: boost::spirit::qi::uint_parser<unsigned char, 16, 2,2> hex_byte; std::string str("..."); std::vector<byte_t> vec; iter_type curr=str.begin(), end=str.end(); bool r=phrase_parse(curr, end, +hex_byte, space, vec); Is vec going to be copied somewhere inside of qi? Or are the bytes directly appended to the vec? I would like to reserver some space for the maximum possible input and avoid additional internal copies inside of qi... How does the behaviour changes std::vector<byte_t> is part of the fusion::map like fusion::map < fusion::pair<tag1, string> , fusion::pair<tag2, string> ... , fusion::pair<tagN, std::vector<byte_t> > > data_item; Are the fields of the map used directly while parsing, or are there some temporaries, which will be first created (somewhere inside qi core) and filled with data and afterwards assigned to the corresponding map field? Many thanks for the gread lib! Ovanes

2 2

[boost::math]binomial_coefficient compile error
by Suresh Kumar 07 Oct '10

07 Oct '10

Hi this code gives a compile error. #include <iostream> #include <boost/math/special_functions/binomial.hpp> using namespace std; int main(){ cout << boost::math::binomial_coefficient(10,1) << endl; } binom.cpp:8: error: no matching function for call to ‘binomial_coefficient(int, int)’ what is that I am missing? thanks suresh -- R Suresh Kumar, http://www.ee.ucr.edu/~sramachandranna Sarve bhavantu sukinaha (May everyone be happy) Sarve santu niramiyaha (May everyone be without affliction) Sarve bhadrani pashyantu (May everyone see only goodness) Ma kashchit dukha bhak bhavet (Let none be the victim of suffering)

2 2

[Python] Unable to use static library of boost.python due to undefined symbols
by Philipp Münzel 06 Oct '10

06 Oct '10

Hi folks, I'm trying to build a python extension with statically linked libboost_python on Linux. I built boost_python with ./bjam -q --layout=system --with-python toolset=gcc variant=release link=static threading=multi runtime-link=static which results in a 700kb libboost_python.a Then I try to build the hello world example located at /libs/python/example/tutorial but I fail, because their remain undefined symbols, despite the fact I did define BOOST_PYTHON_STATIC_LIB. Heres what I did: $ cat hello.cpp // Copyright Joel de Guzman 2002-2004. Distributed under the Boost // Software License, Version 1.0. (See accompanying file LICENSE_1_0.txt // or copy at http://www.boost.org/LICENSE_1_0.txt) // Hello World Example from the tutorial // [Joel de Guzman 10/9/2002] #include <boost/python/module.hpp> #include <boost/python/def.hpp> char const* greet() { return "hello, world"; } BOOST_PYTHON_MODULE(hello_ext) { using namespace boost::python; def("greet", greet); } $ g++ -I/usr/local/include -I/usr/include/python -DBOOST_PYTHON_STATIC_LIB -fpic -c -o hello.o hello.cpp $ g++ -shared -Wl,-soname,"libhello.so" -L/usr/local/lib -lboost_python -fpic -o libhello.so hello.o $ nm -u libhello.so U PyString_Type w _Jv_RegisterClasses U _Py_NoneStruct U _Unwind_Resume@@GCC_3.0 U _ZN5boost6python6detail11init_moduleEPKcPFvvE U _ZN5boost6python6detail12gcc_demangleEPKc U _ZN5boost6python6detail17scope_setattr_docEPKcRKNS0_3api6objectES3_ U _ZN5boost6python7objects15function_objectERKNS1_11py_functionE U _ZN5boost6python7objects21py_function_impl_baseD2Ev U _ZN5boost6python9converter19do_return_to_pythonEPKc U _ZN5boost6python9converter8registry5queryENS0_9type_infoE U _ZNK5boost6python7objects21py_function_impl_base9max_arityEv U _ZNK5boost6python9converter12registration25expected_from_python_typeEv U _ZTIN5boost6python7objects21py_function_impl_baseE U _ZTIPKc@@CXXABI_1.3 U _ZTIc@@CXXABI_1.3 U _ZTVN10__cxxabiv120__si_class_type_infoE@@CXXABI_1.3 U _ZTVN5boost6python7objects21py_function_impl_baseE U _ZdlPv@@GLIBCXX_3.4 U _Znwm@@GLIBCXX_3.4 U __cxa_atexit@@GLIBC_2.2.5 w __cxa_finalize@@GLIBC_2.2.5 U __cxa_guard_abort@@CXXABI_1.3 U __cxa_guard_acquire@@CXXABI_1.3 U __cxa_guard_release@@CXXABI_1.3 w __gmon_start__ U __gxx_personality_v0@@CXXABI_1.3 $ python >>> import libhello Traceback (most recent call last): File "<stdin>", line 1, in <module> ImportError: ./libhello.so: undefined symbol: _ZNK5boost6python7objects21py_function_impl_base9max_arityEv The example works marvellous when using the shared lib of boost_python, but it has to be a static build for me. I found a mailing list entry from 2002 mentioning the static build of boost.python was broken, but that is for sure fixed by now, isn't it? Regards, Philipp

2 2

[Fusion] How to join more than two sequences
by Clinton Mead 06 Oct '10

06 Oct '10

Hi All I've noticed the function "join" in the boost fusion library. However, it only joins two sequences. Was wondering whether there is (or can be easily made) a function that joins any number of sequences. For example: new_join(make_vector(1,2,3), make_vector(4,5), make_vector(6)) -> is the sequence 1,2,3,4,5,6 AND/OR (pass parameters in a sequence) new_join(make_vector(make_ vector(1,2,3), make_vector(4,5), make_vector(6))) -> is the sequence 1,2,3,4,5,6 I'd prefer both the first and second versions, but even one version would be ok. Thanks, Clinton

4 12

[filesystem] V3 path::replace_extension() spans path components
by Timothy Madden 06 Oct '10

06 Oct '10

Hello It appears that in boost 1.44.0 filesystem V3 path("\Cristina.ADI\Local Settings").replace_extension(".png") just returns "\Cristina.png" instead of "\Cristina.ADI\Local Settings.png" Could this be fixed ? Thank you, Timothy Madden

1 1

[proto]Compile issue on Sun Solaris 10, Sun Studio 12.2
by Sascha Ochsenknecht 06 Oct '10

06 Oct '10

Hello, I got an error from the Sun Studio 12.2 (CC 5.11) when trying to compile my sources. The error comes out of the Boost.Proto library (actually I'm doing something with Boost.Spirit and Boost.Spirit is using Boost.Proto ..., but I don't think that the problem is in the client code, it seems to come from the buggy compiler). I'm using Boost 1.43. This is what I got from the compiler: "...include/boost/proto/detail/decltype.hpp", line 343: Error: Partial specialization parameter V is not used in the arguments. "...include/boost/proto/transform/arg.hpp", line 287: Error: Partial specialization for boost::proto::_byval::result has identical arguments. "...include/boost/proto/extends.hpp", line 600: Error: type is not a member of boost::result_of<boost::proto::domainns_::default_domain>. "...include/boost/spirit/home/support/terminal.hpp", line 237: Where: While specializing "boost::proto::exprns_::extends<boost::proto::exprns_::expr<boost::proto::tag::terminal, boost::proto::argsns_::term<boost::spirit::tag::bin>, 0>, boost::spirit::terminal<boost::spirit::tag::bin>, boost::proto::domainns_::default_domain, 0>". (the following is talking about "include/boost/proto/detail/decltype.hpp", line 343) With g++ this code compiles without any problems and runs as expected. Searching around the internet, I got the following bug entry of the Sun Studio compiler: http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6926895 It looks like a similar problem. The bug entry contains a small example which reproduces the compile error: template <typename T> struct test { T t; }; template <typename T, typename X> struct test<X T::*> { X T::* x; }; template <typename T> struct test2 {}; template <typename T, typename X> struct test2< test<X T::*> > { test<X T::*> xt; }; int main() { test<int> a; a.t = 1; int test<int>::* s = &test<int>::t; test<int test<int>::*> a2; a2.x = s; test2<test<int test<int>::* > > a3; a3.xt = a2; return a.*a3.xt.x; } $ CC example.C "example.C", line 4: Error: Partial specialization parameter T is not used in the arguments. The bug entry also provides a workaround which seems to work. Replace the template types with the following: template <typename T> struct helper { }; template <typename T> struct test2 : public test2< helper<T> > { }; template <typename T> struct test2<helper< T> > { } ; template <typename T, typename X > struct test2< helper<test<X T::* > > {test<X T::*> xt; }; I would like to try a similar workaround adapted to the Boost.Proto source code. I tried several things but since I'm not fully aware what is going on in the proto sources and what is the exact reason the compiler fails, I couldn't make it work so far. Does anybody had a similar problem or does anybody know how to adapt the Boost.Proto source to make the Sun compiler happy? BTW: For me it would be ok for now to patch my local Boost source files. Here is the snippet from proto/detail/decltype.hpp: //////////////////////////////////////////////////////////////////////////////////////////// // result_of_ is a wrapper around boost::result_of that also handles "invocations" of // member object pointers. template<typename T, typename Void = void> struct result_of_ : boost::result_of<T> {}; // THE FOLLOWING IS TROUBLE MAKER FOR THE SUN COMPILER -> template<typename T, typename U, typename V> struct result_of_<T U::*(V), typename enable_if_c<is_member_object_pointer<T U::*>::value>::type> { BOOST_STATIC_CONSTANT(bool, is_V_a_smart_ptr = 2 == sizeof(test_V_is_a_U<U>(&lvalue(make<V>())))); BOOST_STATIC_CONSTANT(bool, is_ptr_to_const = 2 == sizeof(test_ptr_to_const(BOOST_PROTO_GET_POINTER(U, make<V>())))); // If V is not a U, then it is a (smart) pointer and we can always return an lvalue. // Otherwise, we can only return an lvalue if we are given one. typedef typename mpl::eval_if_c< (is_V_a_smart_ptr || is_reference<V>::value) , mpl::eval_if_c< is_ptr_to_const , add_reference<typename add_const<T>::type> , add_reference<T> > , mpl::identity<T> >::type type; }; Thanks in advance, Sascha

2 4

boost 1.44.0 undefined reference to boost::filesystem2::no_check
by Asif Fayyaz 06 Oct '10

06 Oct '10

Hi, I am compiling a very simple application using boost filesystem. However it is giving following error message: "undefined reference to `boost::filesystem2::no_check(std::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)'" Here is the simple application source: --------------------------------Sample application source: #include <stdio.h> #include <boost/filesystem.hpp> int main() { std::string m_FileName(boost::filesystem::path("testfile1.txt", boost::filesystem::no_check).native_file_string()); printf("Testboost.....%s\n", m_FileName.c_str()); return 0; } -------------------------------- More Details: Libraries being used: libboost_filesystem-gcc41-mt-1_44.a libboost_system-gcc41-mt-1_44.a Boost version: 1.44.0 Boost was build using command: bjam --build-type=complete --layout=versioned Operating System: Linux CentOS5.2 Any help in this regard will be appreciated. Asif

2 3

Re: [Boost-users] Users! Who'd like to wave goodbye to #ifdef
by Paul Kerchen 06 Oct '10

06 Oct '10

+1 for removing VC6 support. Paul Kerchen

5 4

Re: [Boost-users] Problems with boost 1.44.0 and MSVC: deque problem seems solved, but the boost::ctor problem remains
by Yana A Kireyonok (aka Iron Bug) 06 Oct '10

06 Oct '10

> Date: Tue, 5 Oct 2010 11:37:35 +0200 > From: Igor R <boost.lists(a)gmail.com> > 2010/10/5 Yana A Kireyonok (aka Iron Bug) <death.iron.bug(a)gmail.com>: >> Greetings, ppl. >> I've got strange problems with the latest stable boost (1.44.0) and >> MSVS studios 2005 and 2010. >> ... > According to your description, it seems that the issue you encountered > is only a symptom of some memory-management issue in your code >..... Well, it could happen if I wasn't a professional programmer (over 15 years of real time systems hardware and software programming). It's not a typical beginners mistake, for sure. I wouldn't bother ppl in here if I hadn't checked all possible variants and wasn't sure it was really something out of order. > Date: Tue, 5 Oct 2010 12:38:02 +0000 > From: David Ward <davidjward30(a)hotmail.com> > >> Greetings, ppl. >> I've got strange problems with the latest stable boost (1.44.0) and >> MSVS studios 2005 and 2010. >> > Hi, > You might want to check out this known bug with std::deque in MSVC. It's due > to be fixed in VC11: > https://connect.microsoft.com/VisualStudio/feedback/details/533131/bug-in-s… > Regards, > David. Yes, this is very possible! I performed some more experiments, encountering the bug issue you mentioned and in my case it seems being the pop_front method that ruins the iterators, despite of using interlocking mechanism. Perhaps, it's the matter of timings: that's why it doesn't trigger when I use other locking methods. But I should uproot using deques in my code, anyway. It may show up again on faster or slower machines. Well, I will be aware of danger of using deques in MSVS. Never liked MS, to say the truth... :) But have to deal with it sometimes. Still, remains the question what to do with constructor of boost::thread in MSVS 2010. It fails with access violation, without any specific conditions, in simple plain piece of code.

1 0

[Fusion] fusion vector transformation into another fusion vector transformation.
by alfC 05 Oct '10

05 Oct '10

I have class that is parameterized by several type (Boost::Unit units type). template<class U1, class U2, ... > class A{ A(quantity<U1> q1, quantity<U2> q2, ... ){} }; since the number of arguments is arbitrary I though it would be a good idea to use either MPL sequence and Tuples. Or better yet a fusion::vector. so the class will be parametrized by a fusion::vector or unit types. template<class FusionOfUnits> //or class MPLofUnits class A{ A(FusionOfQuantities fq){} //or TupleOfUnits }; The questions is: ¿what is the best way to transform a fusion::vector<unit1, unit2, unit3, ..., unitN > into a fusion::vector<quantity<unit1>, quantity<unit2>, quantity<unit3>, ..., quantity<unitN> > whete the N is (compile-time) arbitrary, i.e. the lenght of the fist fusion::vector is arbitrary (If you are not familiar with Boost.Units, unitN is just a type and quantity is a template type.) Thank you, Alfredo

2 1

timezone database discrepancy
by archie14 05 Oct '10

05 Oct '10

I am calculating start/end time of the daylight savings time using tz_database and default timezone database file "date_time_zonespec.csv" that comes with boost 1_44. I found discrepancy between results of the calculation for "America/Chicago" region and published data. for years 2010-2007 the results match. However, for the 2006 calendar year, published results from several sources are: Start: April 2 End: October 29 and calculated results are: Start: March 12 End November 5 I will appreciate any advise/comments on how to deal with this issue.

3 3

Re: [Boost-users] [proto] need explanation in implementing distributive law
by Mauro Bianco 05 Oct '10

05 Oct '10

Hi all, It is probably already been discovered by everyone at this point, but the code proposed by Andrew does not to work properly to apply distributivity more than once in an expression [works for a*(b+c) but not for a*(b+c*(d+c)) ] Here a modified version that seems to work recursively. How does it look? struct distr : proto::or_< proto::when< proto::multiplies< proto::_, proto::plus< proto::_, proto::_ > >, proto::call<distr( // THIS IS TO RECURSE proto::_make_plus( proto::_make_multiplies( proto::_left, proto::_left( proto::_right ) ), proto::_make_multiplies( proto::_left, proto::_right( proto::_right ) ) ) ) > >, proto::otherwise<proto::nary_expr<proto::_, proto::vararg<distr> > > > {}; Regards, Mauro. On 22.03.2010 22:11, Andrew Durward wrote: > As a follow-up, the attached code seems to compile and yield the expected > result. > > andrew > > > > _______________________________________________ > Boost-users mailing list > Boost-users(a)lists.boost.org > http://lists.boost.org/mailman/listinfo.cgi/boost-users

2 1

Serialize a class, call new() in constructor
by Michael Reichlin 05 Oct '10

05 Oct '10

Hi everybody, actually i have a big problem with serializing some of my classes. The problem is, that classA class calls new of classB in constructor and after that the load_construct_data is called and also calls a new of classB. So I have two objects of classB. Here is my code: classA.h: class CClassA { template<class Archive> friend void serialize(Archive &ar, CClassA &cCClassA, const unsigned int version); template<class Archive> friend void save_construct_data(Archive &ar, const CClassA *cCClassA, const unsigned int version); template<class Archive> friend void load_construct_data(Archive &ar, CClassA *cCClassA, const unsigned int version); // *** Aufzählungen **************************************************************************** public: CClassB* pClassB; // *** Methoden ******************************************************************************** private: protected: int iTest; public: CClassA(int iTest); // Konstruktor virtual ~CClassA(); // Destruktor // Get/Set-Methoden }; classA.cpp: CClassA::CClassA(int iTest) :iTest(iTest) { pClassB = new CClassB(iTest); } classB.h: class CClassB { template<class Archive> friend void serialize(Archive &ar, CClassB &cCClassB, const unsigned int version); template<class Archive> friend void save_construct_data(Archive &ar, const CClassB *cCClassB, const unsigned int version); template<class Archive> friend void load_construct_data(Archive &ar, CClassB *cCClassB, const unsigned int version); // *** Aufzählungen **************************************************************************** public: // *** Datenmember ***************************************************************************** private: protected: public: // *** Methoden ******************************************************************************** private: protected: int iTestInt; public: CClassB(int iTest); // Konstruktor virtual ~CClassB(); // Destruktor // Get/Set-Methoden static unsigned int count; }; serialize.h template<class Archive> void serialize(Archive &ar, CClassA &cCClassA, const unsigned int version) { ar & boost::serialization::make_nvp("ClassB", cCClassA.pClassB ); } template<class Archive> void save_construct_data(Archive &ar, const CClassA *cCClassA, const unsigned int version) { ar << boost::serialization::make_nvp("Test", cCClassA->iTest); } template<class Archive> void load_construct_data(Archive &ar, CClassA *cCClassA, const unsigned int version) { int iTest; ar >> boost::serialization::make_nvp("Test", iTest); ::new(cCClassA) CClassA(iTest); } template<class Archive> void serialize(Archive &ar, CClassB &cCClassB, const unsigned int version) { } template<class Archive> void save_construct_data(Archive &ar, const CClassB *cCClassB, const unsigned int version) { ar << boost::serialization::make_nvp("Test", cCClassB->iTestInt); } template<class Archive> void load_construct_data(Archive &ar, CClassB *cCClassB, const unsigned int version) { int iTest; ar >> boost::serialization::make_nvp("Test", iTest); ::new(cCClassB) CClassB(iTest); } Now I just want that only one ClassB will be created. How can I recognize, that ClassB is already generate through cons tructor classA. I just want to serialize the members of classB. Is there a way to solve my problem?? Best regards Michael

2 1