April 2011 - Boost - lists.preview.boost.org

[locale] Review of the proposed Boost.Locale library
by Volker Lukas 17 Apr '11

17 Apr '11

Hello, this is my review of the Locale library proposed for inclusion in Boost. I vote to accept the proposed Boost.Locale as a Boost library. - What is your evaluation of the design? The library is nicely structured into individual components, as far as I can infer from the documentation and having a glance at some of the source files. I like that it is possible to use different backends, utilizing the large ICU library when available. - What is your evaluation of the implementation? I was having only a superficial look at some of the source files, so I can not comment much. What I have seen looks clean to me. While reading the source code I think I could benefit from even more code comments. There are a lot already, but on the other hand there exists for example a source file libs/locale/src/shared/mo_lambda.cpp apparently containing a whole state machine or something without any word how it is used. I also stumbled over an odd comment in file libs/locale/src/std/numeric.cpp in line 163: "// workaround common bug - substitute NBSP with ordinary space". As a reader who did not already write the code I can not infer much from comments such as this one - I mean "What is the common bug then?" If this comment refers to the situation that e.g. the locale implementation of GCCs standard library implementation generates invalid UTF-8 when using NBSP as thousands seperator, then the comment should say that. Well, my point about comments is a minor issue of course. I do not suggest that the author goes over all the code again just to add or change comments. I am a bit worried over the size of the libraries binary. On GNU/Linux x86_64 the built libboost_locale.so is almost 10MB large and codesize according to the "size" utility is almost 1MB. This comes in addition to the underlying ICU library. Is there any potential for reducing codesize? Due to the same concern about codesize I have also one concrete suggestion relating to the comparator template of the collation component. I suggest to not make the default comparision level a template parameter, instead to store it as a member. This way, when instantiating container templates the number of instantiations can be kept down. Or am I mistaken in thinking this would lead to more compact code? - What is your evaluation of the documentation? It is nicely structured. I could easily find the documentation for any specific component I was using and the documentation is also very helpful if you have a task at hand and want to find out which component to use (and whether such a component is available in the proposed Boost.Locale). There are some issues I have: Phrasing is sometime odd. For example right in the beginning, under the headline "What is Boost.Locale" there is a sentence which reads "C++ offers a very good base for localization via the existing C++ locale facets std::num_put, std::ctype, std::collate etc. But these are very limited and sometimes buggy by design." I can only guess what that means. Superficially it seems like a contradiction to me, something which is excellent can not be buggy by design, not even sometimes. Okay, minor issue again, it just stands out because it is only the fourth or fifth sentence or so. There are some typos (I guess) where it is important to be correct: Under "Messages Formatting (Translation)", subheading "Indirect Message Translation" all functions are documented as "The source text is not copied.". From the emphasis in the original document I think I can conclude that some of these functions copy and some do not. The reference section of the documentation should document the requirements on the CharType which many functions/classes are parameterized upon (can it be char/wchar_t only?). The reference documentation for token_iterator and break_iterator should not have individual entries for operator++, operator--, operator== and other operators which are always required for iterators, unless the documentation adds information. Instead, it should just be stated to which iterator concepts these templates conform. Documentation for the dereferening operators (operator*) of these templates states that the underlying sequences iterated over can not be modified with these operators. As this is the case, could a usage hint be incorporated stating that users should always use const_iterators as arguments to these templates? If a user sometimes employs for example std::string::iterator and sometimes std::string::const_iterator this would lead to unnecessary code bloat. Also, operator-> is missing from these templates. - What is your evaluation of the potential usefulness of the library? It is potentially very useful. The only objection I could conceive of is that ICU already provides much of the functionality of Boost.Locale. This library also duplicates functionality of Matthew Wilsons excellent Fastformat library (http://fastformat.sourceforge.net/) though the latter does not feature localization support as complete as that of the proposed Boost.Locale. - Did you try to use the library? With what compiler? Did you have any problems? I used the CMake way to build this library on an Opensuse system on x86_64. Compiler is GCC 4.5.1 (or a Svn snapshot roughly corresponding). The Boost version built against is 1.46.1, as delivered by Opensuse. I did not encounter any problems building the library. I played around with some of the components of the library and ran the supplied test cases. There is one inconsistency I noted, the following program generates output "Currency: $100.11" when the environment variable LANG is set to "en_US.UTF-8" and variable LC_MONETARY is set to "de_DE.UTF-8". Shouldn't the library honor LC_MONETARY? If LANG is set to "de_DE.UTF-8" "Currency: 100,11 €" is put out. Program: ------------------------------------------------- #include <ios> #include <iostream> #include <ostream> #include <locale> #include "boost/locale.hpp" int main() { namespace loc = boost::locale; using std::cout; loc::generator gen; cout.imbue(gen("")); cout << "Currency: " << loc::as::currency << 100.11 << '\n'; } ------------------------------------------------- - How much effort did you put into your evaluation? A glance? A quick reading? In-depth study? Reading the tuturial section of the Documentation. Glancing over the reference section. I also had a look at some of the source files, but only superficially. - Are you knowledgeable about the problem domain? Only marginally. With best regards, Volker Lukas

2 1

[locale] Review: Build
by Gevorg Voskanyan 17 Apr '11

17 Apr '11

build/Jamfile.v2: line 18: s/Feaures/Features line 130: There should be a space between '!' and '<target-os>cygwin', otherwise the whole condition will never evaluate to true line 132: s/need/needs/ . Also it can be equivalently replaced with just: EXIT "error: Boost.Locale needs either iconv or ICU library to be built." ; line 200: doesn't posix backend work for BSDs other than darwin? line 283: <link>shared:<define>BOOST_LOCALE_DYN_LINK=1 should be present in usage requirements too. The same applies to <threading>multi as well. lines 275, 282: what's the reason for commenting build_options and having <conditional>@configure instead? line 288: Being a vim user myself I appreciate this, but in general tabs are not allowed in code per Boost's development policy. Users of other editors would not be happy with tabs here :) General remarks: As a Boost.Build user I would see it very beneficial when using Boost.Locale to be able to have ICU build from source with BB just like the other 3rd party libraries commonly are. For it to work Boost.Locale Jamfile can check if a project /icu is available, and if so use /icu//icu as source and skip ICU configuration checking together with options for linking to prebuilt ICU. I can prepare and propose a patch for this if that scheme is OK with you. Examples and performance benchmarks should have Jamfiles too. Gevorg

2 6

SFINAE yes_type and no_type
by Jeroen Habraken 16 Apr '11

16 Apr '11

Hi, Is there a specific yes_type and no_type inside boost that I should be using when writing a SFINAE test? There seem to be quite a few defined already and reinventing the wheel (albeit it being trivial) again doesn't make sense. Regards, Jeroen

6 9

[GSoC] Reviewing
by Andrew Sutton 16 Apr '11

16 Apr '11

We have a little less than a week to finalize the set of 10 accepted projects. I've made tentative assignments to the top 10 projects that have a single willing mentor. There are two proposals with two mentors, that I did not assign a mentor to. If you are one of the mentors, you'll be getting a separate email, off list :) There are two highly rated proposals without mentors. I'm trying to work out what to do with those also. If you don't like your tentative assignment, let me know. There are 15-20 proposals that haven't been rated. I know that some are waiting on further input, but we should at least have initial ratings by the end of the week. Also, please check with the students of your highest rated proposals to see if they submitted proposals to other projects. I have to sit thru the duplication meeting next week and would like to know in advance if we're in competition with other organizations. Thanks, Andrew

1 0

[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

[locale] trouble building
by Noah Roberts 14 Apr '11

14 Apr '11

I grabbed the ICU bin for VS2010. I extracted the content into d:\icu so that d:\icu\include has the various includes. I set up the PATH variable to include d:\icu\bin... I copied boost and lib from the review into boost_1_46_1 where the root of boost source is. The docs seem to imply that I can get a build without ICU if I'm using Windows but this doesn't seem to be the case. I tried --without-icu and I got an error about using --with and --without for the same lib. Here's my attempt to build (with ICU installed): D:\boost_1_46_1\boost_1_46_1>bjam --build-dir=.\build toolset=msvc-10 --with-loc ale --prefix=d:\boost_msvc_1461 --layout=system -sICU_PATH=d:\icu stage Performing configuration checks - iconv (libc) : no - iconv (separate) : no - icu : no D:/boost_1_46_1/boost_1_46_1/tools/build/v2/build\virtual-target.jam:1079: in vi rtual-target.register-actual-name from module virtual-target error: Duplicate name of actual target: <pstage\lib>libboost_locale.lib error: previous virtual target { common%common.copy-libboost_locale.lib.STATIC_L IB { msvc%msvc.archive-libboost_locale.lib.STATIC_LIB { msvc%msvc.compile.c++-en coding\codepage.obj.OBJ { encoding/codepage.cpp.CPP } } { msvc%msvc.compile.c++- shared\date_time.obj.OBJ { shared/date_time.cpp.CPP } } { msvc%msvc.compile.c++- shared\format.obj.OBJ { shared/format.cpp.CPP } } { msvc%msvc.compile.c++-shared \formatting.obj.OBJ { shared/formatting.cpp.CPP } } { msvc%msvc.compile.c++-shar ed\generator.obj.OBJ { shared/generator.cpp.CPP } } { msvc%msvc.compile.c++-shar ed\ids.obj.OBJ { shared/ids.cpp.CPP } } { msvc%msvc.compile.c++-shared\localizat ion_backend.obj.OBJ { shared/localization_backend.cpp.CPP } } { msvc%msvc.compil e.c++-shared\message.obj.OBJ { shared/message.cpp.CPP } } { msvc%msvc.compile.c+ +-shared\mo_lambda.obj.OBJ { shared/mo_lambda.cpp.CPP } } { msvc%msvc.compile.c+ +-util\codecvt_converter.obj.OBJ { util/codecvt_converter.cpp.CPP } } { msvc%msv c.compile.c++-util\default_locale.obj.OBJ { util/default_locale.cpp.CPP } } { ms vc%msvc.compile.c++-util\info.obj.OBJ { util/info.cpp.CPP } } { msvc%msvc.compil e.c++-util\locale_data.obj.OBJ { util/locale_data.cpp.CPP } } { msvc%msvc.compil e.c++-std\codecvt.obj.OBJ { std/codecvt.cpp.CPP } } { msvc%msvc.compile.c++-std\ collate.obj.OBJ { std/collate.cpp.CPP } } { msvc%msvc.compile.c++-std\converter. obj.OBJ { std/converter.cpp.CPP } } { msvc%msvc.compile.c++-std\numeric.obj.OBJ { std/numeric.cpp.CPP } } { msvc%msvc.compile.c++-std\std_backend.obj.OBJ { std/ std_backend.cpp.CPP } } { msvc%msvc.compile.c++-util\gregorian.obj.OBJ { util/gr egorian.cpp.CPP } } { msvc%msvc.compile.c++-win32\collate.obj.OBJ { win32/collat e.cpp.CPP } } { msvc%msvc.compile.c++-win32\converter.obj.OBJ { win32/converter. cpp.CPP } } { msvc%msvc.compile.c++-win32\lcid.obj.OBJ { win32/lcid.cpp.CPP } } { msvc%msvc.compile.c++-win32\numeric.obj.OBJ { win32/numeric.cpp.CPP } } { msvc %msvc.compile.c++-win32\win_backend.obj.OBJ { win32/win_backend.cpp.CPP } } } } error: created from ./stage-proper error: another virtual target { common%common.copy-libboost_locale.lib.STATIC_LI B { msvc%msvc.archive-libboost_locale.lib.STATIC_LIB { msvc%msvc.compile.c++-enc oding\codepage.obj.OBJ { encoding/codepage.cpp.CPP } } { msvc%msvc.compile.c++-s hared\date_time.obj.OBJ { shared/date_time.cpp.CPP } } { msvc%msvc.compile.c++-s hared\format.obj.OBJ { shared/format.cpp.CPP } } { msvc%msvc.compile.c++-shared\ formatting.obj.OBJ { shared/formatting.cpp.CPP } } { msvc%msvc.compile.c++-share d\generator.obj.OBJ { shared/generator.cpp.CPP } } { msvc%msvc.compile.c++-share d\ids.obj.OBJ { shared/ids.cpp.CPP } } { msvc%msvc.compile.c++-shared\localizati on_backend.obj.OBJ { shared/localization_backend.cpp.CPP } } { msvc%msvc.compile .c++-shared\message.obj.OBJ { shared/message.cpp.CPP } } { msvc%msvc.compile.c++ -shared\mo_lambda.obj.OBJ { shared/mo_lambda.cpp.CPP } } { msvc%msvc.compile.c++ -util\codecvt_converter.obj.OBJ { util/codecvt_converter.cpp.CPP } } { msvc%msvc .compile.c++-util\default_locale.obj.OBJ { util/default_locale.cpp.CPP } } { msv c%msvc.compile.c++-util\info.obj.OBJ { util/info.cpp.CPP } } { msvc%msvc.compile .c++-util\locale_data.obj.OBJ { util/locale_data.cpp.CPP } } { msvc%msvc.compile .c++-std\codecvt.obj.OBJ { std/codecvt.cpp.CPP } } { msvc%msvc.compile.c++-std\c ollate.obj.OBJ { std/collate.cpp.CPP } } { msvc%msvc.compile.c++-std\converter.o bj.OBJ { std/converter.cpp.CPP } } { msvc%msvc.compile.c++-std\numeric.obj.OBJ { std/numeric.cpp.CPP } } { msvc%msvc.compile.c++-std\std_backend.obj.OBJ { std/s td_backend.cpp.CPP } } { msvc%msvc.compile.c++-util\gregorian.obj.OBJ { util/gre gorian.cpp.CPP } } { msvc%msvc.compile.c++-win32\collate.obj.OBJ { win32/collate .cpp.CPP } } { msvc%msvc.compile.c++-win32\converter.obj.OBJ { win32/converter.c pp.CPP } } { msvc%msvc.compile.c++-win32\lcid.obj.OBJ { win32/lcid.cpp.CPP } } { msvc%msvc.compile.c++-win32\numeric.obj.OBJ { win32/numeric.cpp.CPP } } { msvc% msvc.compile.c++-win32\win_backend.obj.OBJ { win32/win_backend.cpp.CPP } } } } error: created from ./stage-proper error: added properties: <debug-symbols>on <inlining>off <optimization>off <runt ime-debugging>on <variant>debug error: removed properties: <debug-symbols>off <define>NDEBUG <inlining>full <opt imization>speed <runtime-debugging>off <variant>release D:/boost_1_46_1/boost_1_46_1/tools/build/v2/build\virtual-target.jam:490: in act ualize-no-scanner from module object(file-target)@962 D:/boost_1_46_1/boost_1_46_1/tools/build/v2/build\virtual-target.jam:135: in obj ect(file-target)(a)962.actualize from module object(file-target)@962 D:/boost_1_46_1/boost_1_46_1/tools/build/v2\build-system.jam:748: in load from m odule build-system D:\boost_1_46_1\boost_1_46_1\tools\build\v2/kernel\modules.jam:283: in import fr om module modules D:\boost_1_46_1\boost_1_46_1\tools\build\v2\kernel\bootstrap.jam:142: in boost-b uild from module D:\boost_1_46_1\boost_1_46_1\boost-build.jam:17: in module scope from module D:\boost_1_46_1\boost_1_46_1> Any ideas?

3 3

[Boost.Pool] TR1?
by Phil Bouchard 14 Apr '11

14 Apr '11

Greetings, I see the Boost.Pool is not making its way up to the C++ standards yet an it is very unfortunate given its great use. Boost.Pool provides an is_from() member function that does a simple range check and returns whether a pointer is part of the pool or not. This is a very useful function which I think should be part of the standards because it could be used for the global pool as well; i.e. the one used by operator ::new and ::delete. The benefits outweigh the costs. Thanks, -Phil

8 49

[Parallel BGL] Property access problem
by Cosimo Calabrese 14 Apr '11

14 Apr '11

Hi to all, I'm working on a WinXP, VS2010, boost 1.46.1, openMPI 1.4.3. I've used in the past the BGL lib, and now I would explore the PBGL extension. I've writed a simple program (you can find it in attach), but I've some problem. When I run it with: mpirun -np 1 .\test.exe it works. But if I use 2 or 4 processors the program crash sistematically in the following way, on different machines too: num_vertices: 10 num_edges: 19 A-->B : 1 A-->C : 10 A-->D : 11 B-->D : 13 C-->D : 12 B-->E : 1 C-->A : 14 E-->D : 1 D-->B : 15 D-->A : 1 E-->B : 16 -------------------------------------------------------------------------- Sorry! You were supposed to get help about: odls-default:could-not-kill But I couldn't open the help file: D:\dev\CommesseAperte\Sviluppo\TELLUSXX_CalcServer\lib\openmpi\installed\share\openmpi\help-odls-default.txt: No such file or directory. Sorry! -------------------------------------------------------------------------- Every process allocates 1 GB. The problem disappear only when I eliminate all the property access (get() and put()). Maybe I access to the MPI framework for the first time during the property access, and so could be an MPI problem? Am I using the property in a wrong way? Thanks in advance, Cosimo Calabrese.

1 1

New, powerful way to use enable_if in C++0x
by Matt Calabrese 13 Apr '11

13 Apr '11

I'm not sure if there is much interest in this or if someone else discovered this as well, but I realized today that enable_if can be used in a new way in C++0x. Traditionally when working with enable_if with function templates you invoke it either in the return type or in an additional function parameter of the function. Each of these has advantages and disadvantages. The parameter form has the downside of augmenting the parameter list. The most notable problems with this approach are that it can't be used with overloaded operators that require a fixed number of arguments, and it can't be used after a variadic argument list. The return type use of enable_if is often preferred since it does not involve augmenting the parameter list, meaning that it can be used with most overloaded operators, but you still can't use it for conversion operations and you can't use it all with constructors as they have no explicit return type. However, in C++0x, there is a new way to use enable_if by altering neither the function parameter list nor the specification of the return type. It can be used with operators, constructors, variadic function templates, and even overloaded conversion operations (the latter of which was previously considered impossible). The way to do it is with C++0x default function template parameters. Rather than try to explain any further in English, the following code should demonstrate the approach fairly clearly (tested in GCC 4.5.2): http://codepaste.net/23yo8o There is only one drawback that I can think of off of the top of my head -- you can't refer to the function parameters by name for the enable_if condition, whereas in C++0x if you use the return type form of enable_if along with the new trailing return type form of function declarations you can. I have never personally wanted to do this and I highly doubt that it's common so this is likely entirely a non-issue, plus you can always use std::declval anyway so you aren't actually losing any high-level functionality. Are there any thoughts on this or drawbacks other than what I mentioned? Should such use be added to the docs as supported behavior in 0x compilers? -- -Matt Calabrese

16 34

[config] Local classes as template parameters
by Lorenzo Caminiti 13 Apr '11

13 Apr '11

On Sat, Feb 12, 2011 at 5:50 AM, John Bytheway <jbytheway+boost(a)gmail.com> wrote: > On 11/02/11 22:29, Lorenzo Caminiti wrote: >> On Fri, Feb 11, 2011 at 4:31 PM, Lorenzo Caminiti <lorcaminiti(a)gmail.com> wrote: >>> From: John Bytheway <jbytheway+boost(a)gmail.com> >>> 2. Might anyone care so much about performance that they absolutely must >>> have the code inlined? The answer is probably "yes" (or at least there >>> will be people who *think* they care, which is also a problem). For >>> these people you could, if you choose, provide an alternate >>> implementation which doesn't indirect through local::function<>, and >>> thus will only work in C++0x (or C++03 with non-standard extensions). >> >> Without inheriting from local::function_base<> and then using >> local::function_ref<> the local struct cannot be passed as a template >> parameter in C++. Is this different for C++0x (C++03)? > > Yes, in C++0x local structs can be passed as template parameters. > Obviously, in C++0x there are lambdas too, so you might think your > library is useless, but I'd expect most compilers to support passing > local structs as template parameters before they support lambdas, so > there is some advantage in having this intermediate implementation. > Also, your macros would allow writing code that worked in C++0x and > C++03, with the virtual function business in only those compilers/modes > where it is necessary. Hello all, Is there a Boost.Config macro that indicates if the compiler allows local classes to be passed as template parameters? Thank you. -- Lorenzo

3 2

[GSoC] Boost.Tree and Boost.BloomFilter
by Alejandro Cabrera 13 Apr '11

13 Apr '11

Hello all, I've filed two proposals for this Google Summer of Code. Boost.Tree: http://www.google-melange.com/gsoc/proposal/review/google/gsoc2011/cppcabre… <http://www.google-melange.com/gsoc/proposal/review/google/gsoc2011/cppcabre…> Boost.BloomFilter: http://www.google-melange.com/gsoc/proposal/review/google/gsoc2011/cppcabre… I'd be glad to work on either project. Feedback is welcome. Thanks, -Alej

3 5

Installation fails if env VMS is set
by Dreamcat4 12 Apr '11

12 Apr '11

Hi, Over on the Boost-Users mailinglist we have found that this environment variable (and perhaps unknown but maybe others) will cause bootstrap.sh / Jam failure during the Boost libraries compile and install. Heres the full thread: http://lists.boost.org/boost-users/2011/04/67716.php You (or Jam developement team) may wish to consider taking some kind of action to avoid this as VMS is a very common acronym for "Virtual Machines". Its usage elsewhere is widespread and certainly not Boost-specific. Other choices of course - maybe look towards documenting such usag(es) if they cannot be easily avoided. Or consider ways to detect / warn / clean out the users shell environment variables if they are conflicting. Sorry I cant help any further however as a new user I simply have no experience with the Boost library yet. Many thanks. dreamcat4 dreamcat4(a)gmail.com

2 4

[GSOC]SIMD Library
by Faiçal Tchirou 12 Apr '11

12 Apr '11

Hi everyone, I recently ask documentation about the SIMD Library project here and I receive some guides and references links from Joel Falcou. Now I have some questions about the library itself. Is the goal of the project to build a library similar to libSIMDx86 but more focusing on the AltiVec instruction set ? Joel also advises me to take a look at Proto. I read that Proto is used to build DSEL. Will Proto be used to map high level classes and methods of the library to native SIMD instructions ? What about the library itself ? Which modules have to be developped ? Thanks.

8 37

[trac] Misuse of trac system by spammer
by Joachim Faulhaber 12 Apr '11

12 Apr '11

Hi, I received a ticket today: https://svn.boost.org/trac/boost/ticket/5457 that's only purpose is to distribute a commercial link. Has anyone experienced a similar misuse of the trac system? Are there ideas how to prevent this? Regards, Joachim -- Interval Container Library [Boost.Icl] http://www.joachim-faulhaber.de

3 5

[GSoC] Proposal reviewing
by Andrew Sutton 12 Apr '11

12 Apr '11

Mentors, We need to establish some guidelines for reviewing proposals. They won't be complicated :) I'm opting not to use the mentors mailing list for this announcement. I know that all the mentors are subscribed here. Here's how this should work: 1. Identify the set of proposals for projects that you are interested in mentoring. 2. Evaluate the proposal and write your evaluation as a PRIVATE comment. 3. If you have questions of the student, you can write PUBLIC comments or send them an email. 4. Rank the proposal (I think 5 is the best this year). That would be the project that you want to mentor. Towards the end of the review process, we will select the 10 best proposals. This doesn't guarantee that we'll *fund* 10 proposals, just that I've requested 10 slots. I would like, this year, if all students got feedback regarding their proposals even if they are clearly rejects. The reason for writing evaluations privately, for now, is that it will help me write a public summary for proposals that are not accepted. I'll figure out how this is going to work over the course of the week :) Specific evaluation criteria are up to you (the mentors). You may ask students to modify good proposals. You do not need to ask *all* students to improve their proposals. At the end of the week, you should have a list of 1 or 2 proposals that *you yourself* will mentor. If you are not offering to mentor the project, please don't review it as a 5. Otherwise, we end up with a case where we have to assign a mentor to a project that they aren't interested in. I don't think that this has worked out very well in the past. Also, would prefer that comments regarding specific proposals not be sent to the list. I don't think it would be a good idea for a flood of "this proposal is AWFUL" emails to start popping up. I'm all for transparency, but I don't think airing value judgements would be a beneficial for community development :) Questions or comments involving evaluation criteria and process are fine. Andrew

4 8

[convert] incomplete reference documentation
by Vicente BOTET 11 Apr '11

11 Apr '11

Hi, I was taking a look at the documentation and it seems to me that the reference documentation is incomplete and not ready for review. Do you plan to complete it before review? Thanks, Vicente

4 9