- Boost - lists.boost.org

[random] example code fails inside struct
by Beman Dawes 16 Apr '11

16 Apr '11

struct X { boost::mt19937 rng; boost::uniform_int<> million(1,1000000); // VC++ 10 error C2059: syntax error : 'constant' boost::variate_generator<boost::mt19937&, boost::uniform_int<> > random_value(rng, million); }; Why does the example code from the docs fail to compile when placed inside a struct or class? What am I missing? --Beman

3 4

Changesign function and nonfinite facets added to Boost.Math
by Paul A. Bristow 16 Apr '11

16 Apr '11

I've just added changesign function and nonfinite facets to Boost.Math in boost-trunk. These functions were part of the original floating-point collection written by Johan Rade but were previously only available from Spirit in boost-sandbox. Making them First Class citizens in Boost.Math will allow portable input/output of infinity and NaN, including use of lexical cast, archiving and serialization. Multiplatform testing of these new features should start soon. I hope this will not have any unwanted side effects - I'm sure you'll shout if it does. Paul --- Paul A. Bristow, Prizet Farmhouse, Kendal LA8 8AB UK +44 1539 561830 07714330204 pbristow(a)hetp.u-net.com

1 0

[serialization] Add UTF-8 BOM support to xml_warchive
by Tijmen van Voorthuijsen 15 Apr '11

15 Apr '11

Hi, Recently I ran into the problem that the boost::serialization library could not handle XML files which contain the three UTF-8 BOM (Byte Order Mark) bytes. The serialization library creates XML files without the BOM bytes but when saving such files in an external Windows program, for example XML Notepad, these bytes are automatically added. Thereafter the XML file cannot be read anymore by the boost::serialization library. According to Wikipedia http://en.wikipedia.org/wiki/Byte_order_mark the UTF-8 BOM is optional and therefore creating XML files without the BOM bytes is all right. However, the reading should extended and be able to handle both types, files with and without the BOM. I propose to enhance the xml_warchive and text_warchive for reading with support of the BOM bytes. Example: namespace { const wchar_t g_cchUtf8Bom1 = 0xEF; const wchar_t g_cchUtf8Bom2 = 0xBB; const wchar_t g_cchUtf8Bom3 = 0xBF; } void CheckAndCorrectUtf8Bom(std::wifstream* pifs) { _ASSERT_POINTER(pifs); wchar_t chUtf8Bom1 = 0; wchar_t chUtf8Bom2 = 0; wchar_t chUtf8Bom3 = 0; chUtf8Bom1 = pifs->peek(); if (chUtf8Bom1 == g_cchUtf8Bom1) { *pifs >> chUtf8Bom1; _ASSERT(chUtf8Bom1 == g_cchUtf8Bom1); *pifs >> chUtf8Bom2; _ASSERT(chUtf8Bom2 == g_cchUtf8Bom2); *pifs >> chUtf8Bom3; _ASSERT(chUtf8Bom3 == g_cchUtf8Bom3); } else { // Reset to start of the stream pifs->seekg(0, std::ios_base::beg); } } Kind regards, Tijmen van Voorthuijsen ----------------------------------------------------------------------------------- T. van Voorthuijsen Senior System Engineer Noldus Information Technology bv Nieuwe Kanaal 5 P.O. Box 268 6700 AG Wageningen The Netherlands Phone: +31-(0)317-473300 Fax: +31-(0)317-424496 E-mail: T.van.Voorthuijsen(a)Noldus.nl<mailto:T.van.Voorthuijsen@Noldus.nl> Web: www.noldus.com<http://www.noldus.com>

3 2

[MSM] Interrupt anonymous transitions or blocking process_events?
by Mathias A 15 Apr '11

15 Apr '11

Hi there! I have been experimenting with the MSM (Meta State Machine) and I think it is a very nice tool! After reading and trying most of the example code regarding the MSM I present to you my problem. How do I interrupt blocking calls or long running states with anonymous transitions? Ok, here is my story. Regarding "Defining a submachine" in the docs (http://www.boost.org/doc/libs/1_46_1/libs/msm/doc/HTML/ch03s02.html#d0e597) and supplied source code example (http://www.boost.org/doc/libs/1_46_1/libs/msm/doc/HTML/examples/CompositeTu…) Let's say that this was a real CD player then the player would change to the next song by itself. It would not rely on the user pushing the Next button to change song. Now, if I implement the transition table like this: struct transition_table : mpl::vector4 // Start Event Next Action Guard -----------+-------------+---------+---------------------+----------------------+ a_row < Song1 , none , Song2 , &pl::start_next_song >, a_row < Song2 , PreviousSong, Song1 , &pl::start_prev_song >, a_row < Song2 , none , Song3 , &pl::start_next_song >, a_row < Song3 , PreviousSong, Song2 , &pl::start_prev_song > +---------+-------------+---------+---------------------+----------------------+ > {}; Note the "none" for anonymous transitions. By calling p.process_event(play()); the the player will start playing Song1 and then change to the next song by itself. But now in single threaded mode it blocks until it reaches "Song3". So it is not possible to process any other events during this time. In other words, it cannot process another process_event call. The "Playing" state has a "stop" event but calling this will not help as it will not be processed until the MSM has reached "Song3". In multi threaded mode (no complete source code for this here sorry. with mutex protected access to process_event) thread 1: lock(mutex) p.process_event(play()); lock.unlock(); thread 2: wait a few ms so we are in the middle of playing a song lock(mutex) p.process_event(stop()); lock.unlock(); I start the same p.process_event(play()); in thread 1. And in thread 2 I call p.process_event(stop()); The problem with this is that the stop() event does not get processed until it reaches Song3 because of the blocking call in p.process_event(play());. Perhaps I am going about this the wrong here and please tell me if it can be done in some other much better way! I want to be able to have automatic transitions that I can interrupt. A not so elegant workaround: In thread 2 access the Playing state directly and set a bool variable. (bool externalStop_;) Add a guard to each anonymous transition checking if the source state has externalStop_ == true and stop the transition. The Playing state can then create a stop event and then call process_event(stop()) by itself. And we have successfully injected a stop event. Summary: I cannot seem to get round the blocking call using only the transition table. I really hope it can be done in a more elegant way than I did in the workaround. Any suggestions on how to do this? Thank you! Best regards, Mathias

1 0

[locale] localization_backend_manager interface issue
by Max Sobolev 15 Apr '11

15 Apr '11

> std::auto_ptr<localization_backend> > localization_backend_manager::get() const; Class localization_backend_manager is marked as BOOST_LOCALE_DECL. This is hinting at placing instances of the class in a standalone shared library without problems. But here it is: using std::auto_ptr<> in interfaces is an error (Due to possibility of placing instances of classes in dynamic libraries). Dynamic library that contain your class might be compiled in debug / release / another mode, but client's code might have an alternative heap (because it use another C++ Run Time system) with an alternative heap manager. std::auto_ptr<> doesn't carrying *deleter* with a raw pointer of the allocated object, as boost::scoped_ptr<>, boost::shared_ptr<> or c++0x's std::unique_ptr<> does. Memory allocated from one heap might be "released" by foreign heap manager! :-/ See Sutter/Alexandrescu recommendation do not use std::auto_ptr<> (in ISBN: 9780321113580, C++ Coding Standards, "Rule 60: Avoid allocating and deallocating memory in different modules") and problem description (in ISBN: 9780201604641, Douglas Schmidt, Stephen Huston - C++ Network Programming [Volume 2] Chapter 5, "Sidebar 29: Portable Heap Operations with ACE") ..so change std::auto_ptr<> to some of the more safe smart pointer. > void localization_backend_manager::select( > std::string const& backend_name, > locale_category_type category = all_categories); Why you don't use enums instead of std::string and locale_category_type? (latter is unsigned, actually). So this code is permissible: boost::locale::localization_backend_manager::global().select("...", static_cast<unsigned>(std::rand())); std::string as opposed to ((emulated) scoped) enum implies discarded static type checking plus run-time overhead (due to *run-time* "type" checking, as i guess). unsigned type provides a wider "supertype" for argument than necessary -- - Do you speak English? Мужик с глубоким вздохом: - Yes I do. А хули толку?

5 5

[locale] Formal review of Boost.Local library -- need reviews!
by Chad Nelson 15 Apr '11

15 Apr '11

The final day of the Boost.Locale formal review is scheduled to be Saturday, and though there has been a lot of discussion, so far only one complete review has been submitted. If you're planning to review it, please don't delay. Writing a review ================ If you feel this is an interesting library, then please submit your review to the developer list (preferably), or to the review manager (me). Here are some questions you might want to answer in your review: - What is your evaluation of the design? - What is your evaluation of the implementation? - What is your evaluation of the documentation? - What is your evaluation of the potential usefulness of the library? - Did you try to use the library? With what compiler? Did you have any problems? - How much effort did you put into your evaluation? A glance? A quick reading? In-depth study? - Are you knowledgeable about the problem domain? And finally, every review should answer this question: - Do you think the library should be accepted as a Boost library? Be sure to say this explicitly so that your other comments don't obscure your overall opinion. -- Chad Nelson Oak Circle Software, Inc. * * *

8 11

Fw: [locale] Formal review of Boost.Locale library
by Chad Nelson 15 Apr '11

15 Apr '11

Begin forwarded message: Date: Thu, 07 Apr 2011 08:49:20 +0900 From: Darren Cook <darren(a)dcook.org> Subject: [locale] Formal review of Boost.Locale library Hello, I'm not on the devel list but just had a look at the Boost.Locale library documentation. Feel free to forward this to the list, or discard it. > - What is your evaluation of the documentation? * No prev/next page links [This should be fixed ASAP, as it discourages evaluating the library] * Confusing first page as it seems to only talk about other libraries, not the one in question. (?) * The examples are not motivating, and just end up looking complex. * I'd have liked to see examples in more languages, in particular some Asian languages, to comfort me that the library has taken their issues into consideration. If the itch being scratched here is just for a couple of European languages then the library may have design flaws. * The backends page ( http://cppcms.sourceforge.net/boost_locale/html/using_localization_backends… ) looks interesting, and may be the start of justifying the reason for this library to exist. It could be expanded with something more concrete: complete examples of doing something useful with different backends, with timings and exe sizes we could compare. > - What is your evaluation of the potential usefulness of the > library? I would choose to use ICU directly, as I couldn't see a reason not to, and I'd be concerned that the wrapper would not have wrapped some obscure function in ICU that I find I end up needing. > - How much effort did you put into your evaluation? A glance? A > quick reading? In-depth study? A quick reading. > - Are you knowledgeable about the problem domain? I've used ICU (it is essential in some situations), though possibly not from C/C++. Darren -- Darren Cook, Software Researcher/Developer http://dcook.org/work/ (About me and my work) http://dcook.org/blogs.html (My blogs and articles)

19 52

Re: [boost] [chrono] Compiler errors for mpl/abs.hpp
by Vicente BOTET 15 Apr '11

15 Apr '11

Hi, since Boost.Chrono was accepted I have moved to the trunk and the release branch. There are a lot of file in the sandbox that have been changed since. In particular the error comes from some compilers don't supporting nested value arithmetic in templates. To workaround this the trunk contain instead template<> struct abs_impl { #if defined(BOOST_MPL_CFG_NO_NESTED_VALUE_ARITHMETIC_2) template< typename N > struct apply : aux::abs_wknd< typename N::value_type, N::value > #else template< typename N > struct apply : integral_c< typename N::value_type, ((N::value < 0) ? (-N::value) : N::value ) > #endif { }; }; For this reason a suggest you to move to the trunk or the release branch. I will remove the Chrono sandbox as soon as the release 1.47 is out. Best, Vicente Message du 15/04/11 00:29 > De : "lcaminiti" > A : boost(a)lists.boost.org > Copie à : > Objet : [boost] [chrono] Compiler errors for mpl/abs.hpp > > Hello all, > > I am trying to compile the example below on MSVC using Boost.Chrono > (headers-only) from the sandbox but I am getting some compiler errors. > > What am I doing wrong? > > Thank you very much. > --Lorenzo > > #define BOOST_CHRONO_INLINED > #include > #include > #include > #include > #include > #include > #include > > int main(int argc, char* argv[]) { > assert(argc == 2); > unsigned long n = sqrt(double(atol(argv[1]))); > > using boost::phoenix::ref; > using boost::phoenix::arg_names::_1; > > double sum = 0.0; > int factor = 1; > > std::vector v(n); > std::fill(v.begin(), v.end(), 1.0); > > boost::chrono::duration sec; > for (unsigned long i = 0; i < n; ++i) { > boost::chrono::system_clock::time_point start = > boost::chrono::system_clock::now(); > std::for_each(v.begin(), v.end(), ( > ref(sum) += factor * _1 > )); > sec += boost::chrono::system_clock::now() - start; > } > std::cout << sec.count() << std::endl; > > assert(sum == n * n); > return 0; > } > > E:\sandbox\boost-sandbox-local\libs\local\example>cl /EHs /Ox /I..\..\.. > /I"c:\P > rogram Files\boost\boost_1_46_1" /I..\..\..\..\boost-sandbox-chrono.x > benchmark_ > boost_phoenix.cpp > Microsoft (R) 32-bit C/C++ Optimizing Compiler Version 14.00.50727.42 for > 80x86 > Copyright (C) Microsoft Corporation. All rights reserved. > > benchmark_boost_phoenix.cpp > ..\..\..\..\boost-sandbox-chrono.x\boost/mpl/abs.hpp(72) : error C2143: > syntax e > rror : missing ',' before ')' > ..\..\..\..\boost-sandbox-chrono.x\boost/mpl/abs.hpp(75) : see > reference > to class template instantiation 'boost::mpl::abs_impl::apply' being > compiled > ..\..\..\..\boost-sandbox-chrono.x\boost/mpl/abs.hpp(72) : error C2059: > syntax e > rror : ')' > ..\..\..\..\boost-sandbox-chrono.x\boost/mpl/abs.hpp(74) : error C2143: > syntax e > rror : missing ';' before '{' > ..\..\..\..\boost-sandbox-chrono.x\boost/mpl/abs.hpp(75) : error C2143: > syntax e > rror : missing ';' before '}' > ..\..\..\..\boost-sandbox-chrono.x\boost/mpl/abs.hpp(72) : error C2143: > syntax e > rror : missing ';' before '}' > ..\..\..\..\boost-sandbox-chrono.x\boost/mpl/abs.hpp(72) : fatal error > C1004: un > expected end-of-file found > > > -- > View this message in context: http://boost.2283326.n4.nabble.com/boost-chrono-Compiler-errors-for-mpl-abs… > Sent from the Boost - Dev mailing list archive at Nabble.com. > _______________________________________________ > Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost >

1 0

[move] Declarations of the copy constructor and copy-assign operator produced by BOOST_MOVABLE_BUT_NOT_COPYABLE
by Joseph Trebbien 15 Apr '11

15 Apr '11

Looking through the sources of Boost.Move in trunk, I am wondering about a couple of lines in <boost/move/move.hpp> that are part of the expansion of BOOST_MOVABLE_BUT_NOT_COPYABLE when BOOST_NO_RVALUE_REFERENCES is defined: #define BOOST_MOVABLE_BUT_NOT_COPYABLE(TYPE)\ private:\ TYPE(TYPE &);\ TYPE& operator=(TYPE &);\ public:\ operator ::boost::rv<TYPE>&() \ { return *static_cast< ::boost::rv<TYPE>* >(this); }\ operator const ::boost::rv<TYPE>&() const \ { return *static_cast<const ::boost::rv<TYPE>* >(this); }\ private:\ // Why do the declarations of the copy constructor and copy-assign operator overload not take a const-reference to TYPE? When BOOST_NO_RVALUE_REFERENCES is not defined, they do: #define BOOST_MOVABLE_BUT_NOT_COPYABLE(TYPE)\ public:\ typedef int boost_move_emulation_t;\ private:\ TYPE(const TYPE &);\ TYPE& operator=(const TYPE &);\ // Joseph Trebbien

2 1

[chrono] Compiler errors for mpl/abs.hpp
by lcaminiti 14 Apr '11

14 Apr '11

Hello all, I am trying to compile the example below on MSVC using Boost.Chrono (headers-only) from the sandbox but I am getting some compiler errors. What am I doing wrong? Thank you very much. --Lorenzo #define BOOST_CHRONO_INLINED #include <boost/chrono.hpp> #include <boost/spirit/include/phoenix.hpp> #include #include #include #include #include int main(int argc, char* argv[]) { assert(argc == 2); unsigned long n = sqrt(double(atol(argv[1]))); using boost::phoenix::ref; using boost::phoenix::arg_names::_1; double sum = 0.0; int factor = 1; std::vector v(n); std::fill(v.begin(), v.end(), 1.0); boost::chrono::duration sec; for (unsigned long i = 0; i < n; ++i) { boost::chrono::system_clock::time_point start = boost::chrono::system_clock::now(); std::for_each(v.begin(), v.end(), ( ref(sum) += factor * _1 )); sec += boost::chrono::system_clock::now() - start; } std::cout << sec.count() << std::endl; assert(sum == n * n); return 0; } E:\sandbox\boost-sandbox-local\libs\local\example>cl /EHs /Ox /I..\..\.. /I"c:\P rogram Files\boost\boost_1_46_1" /I..\..\..\..\boost-sandbox-chrono.x benchmark_ boost_phoenix.cpp Microsoft (R) 32-bit C/C++ Optimizing Compiler Version 14.00.50727.42 for 80x86 Copyright (C) Microsoft Corporation. All rights reserved. benchmark_boost_phoenix.cpp ..\..\..\..\boost-sandbox-chrono.x\boost/mpl/abs.hpp(72) : error C2143: syntax e rror : missing ',' before ')' ..\..\..\..\boost-sandbox-chrono.x\boost/mpl/abs.hpp(75) : see reference to class template instantiation 'boost::mpl::abs_impl::apply' being compiled ..\..\..\..\boost-sandbox-chrono.x\boost/mpl/abs.hpp(72) : error C2059: syntax e rror : ')' ..\..\..\..\boost-sandbox-chrono.x\boost/mpl/abs.hpp(74) : error C2143: syntax e rror : missing ';' before '{' ..\..\..\..\boost-sandbox-chrono.x\boost/mpl/abs.hpp(75) : error C2143: syntax e rror : missing ';' before '}' ..\..\..\..\boost-sandbox-chrono.x\boost/mpl/abs.hpp(72) : error C2143: syntax e rror : missing ';' before '}' ..\..\..\..\boost-sandbox-chrono.x\boost/mpl/abs.hpp(72) : fatal error C1004: un expected end-of-file found -- View this message in context: http://boost.2283326.n4.nabble.com/boost-chrono-Compiler-errors-for-mpl-abs… Sent from the Boost - Dev mailing list archive at Nabble.com.

1 0