November 2013 - Boost - lists.preview.boost.org

Data Structure and Algorithms yet to be implemented
by Manmohit Rekhi 12 Nov '13

12 Nov '13

Hi, I would really be grateful if could get to know 1)which Data Structures and Algorithms are yet to be implemented in boost? 2)How can i get started on them? Thanking you, Manmohit Rekhi

3 2

Boost 1.55.0 released!
by Marshall Clow 12 Nov '13

12 Nov '13

Release 1.55.0 of the Boost C++ Libraries is now available. These open-source libraries work well with the C++ Standard Library, and are usable across a broad spectrum of applications. The Boost license encourages both commercial and non-commercial use. This release contains one new library and numerous enhancements and bug fixes for existing libraries. === New Libraries === • Predef: This library defines a set of compiler, architecture, operating system, library, and other version numbers from the information it can gather of C, C++, Objective C, and Objective C++ predefined macros or those defined in generally available headers, from Rene Rivera. === Known bugs with Visual Studio 2013/Visual C++ 12 === Visual Studio 2013 was released quite late in the release process, so there exist several unresolved issues. These include: • Serialization can't compile because of a missing include. • Using has_member_function_callable_with from Boost.Container's allocator_traits causes a compile error (#9332). • For several libraries overloads with intializer lists are incorrectly reported as ambiguous. This is a Visual C++ bug and it isn't clear if there's a good workaround. • In libraries such as Unordered and MultiIndex, calling overloaded functions with initializer lists can result in a compile error, with Visual C++ claiming that the overloads are ambiguous. This is a Visual C++ bug and it isn't clear if there's a good workaround. This won't affect code that doesn't use initializer lists, or uses an initializer list that doesn't require an implicit conversion (i.e. an initializer list of the container's exact value type). • Thread: ex_scoped_thread compile fails (#9333). For details, including download links, see http://www.boost.org/users/news/version_1.55.0 You can also download directly from SourceForge: http://sourceforge.net/projects/boost/files/boost/1.55.0/ To install this release on your system, see http://www.boost.org/doc/libs/release/more/getting_started/index.html Thanks, --The Boost release team Vladimir Prus Rene Rivera Marshall Clow Eric Niebler Daniel James Beman Dawes

2 2

Re: [boost] sqlpp11: SQL for C++
by Brian Wood 12 Nov '13

12 Nov '13

From: Bjorn Reese > On 11/09/2013 11:03 PM, Roland Bock wrote: >> >> * define types representing tables and columns, > > There seems to be room for improvement here. You should consider an > integration with Boost.Fusion. This way you can avoid having to build > a code generator. Hmm. If you want a second opinion, these guys http://springfuse.com are building a code generator. -- Brian Ebenezer Enterprises - So far G-d has helped us. http://webEbenezer.net

2 1

[Functional/Factory] Value factory not accepts rvalues?
by Lin, Yi-Li 12 Nov '13

12 Nov '13

I found that boost::value_factory did not accept rvalues in some case. In my case, temporary value is produced and passed to value factory directly. But boost::value_factory<T>::operator() did not accept rvalue, they take lvaue reference instead. Maybe it'll be better to provide another version of operator takes const lvalue? (Additionally, use std::forward to forward universal reference in C++11?) Sample code which compiler complains: int get_value(); struct var { var( ); explicit var(int); }; void caller() { boost::value_factory<var> f; f(get_value()); }

2 2

Changes to trunk for clang using VC++RTL
by Edward Diener 12 Nov '13

12 Nov '13

I have held off making some changes to trunk in SVN for clang with VC++ RTL until Boost 1.55 was out. I can do so now based on my tests with clang but I am concerned that the changes will not be carried over to the Boost git development branch when Boost switches to git. Should I be waiting until the git changeover is made and everyone learns to deal with git and the setup for using Boost with git, or should I start making changes now and watch the regression tests cycle ? I am sure from my tests that the changes, although involved in low-level libraries like config, mpl, type_traits, will not adversely affect any other compiler and will help in testing clang using VC++ RTL for Boost as clang inches closer to full support in Windows using VC++ headers and libraries.

1 0

Re: [boost] [release][1.55.0] Visual C++ known bugs
by Daniel James 11 Nov '13

11 Nov '13

On 11 November 2013 20:40, Robert Ramey <robertmacleranramey(a)gmail.com> wrote: > > > On Monday, November 11, 2013 12:04:22 PM UTC-8, Daniel James wrote: >> >> On 11 November 2013 12:36, Sergey Cheban <s.ch...(a)drweb.com> wrote: >> > >> > Btw, what do you think about an additional SVN branch for the post-RC2 >> > VC12-related bugfixes? >> >> I just created a branch at: >> >> https://svn.boost.org/svn/boost/branches/maintenance/1_55_0 >> > > Since we've tagged the current state of the release branch with 1_55_0 do we > really have to create another branch? Can't > we just open the release branch for checkins? That would be easier. This branch is just for fixes. The release branch can have other changes, including breaking changes.

2 2

Re: [boost] [release][1.55.0] Visual C++ known bugs
by Daniel James 11 Nov '13

11 Nov '13

On 11 November 2013 21:23, Robert Ramey <robertmacleranramey(a)gmail.com> wrote: > > > On Monday, November 11, 2013 12:42:07 PM UTC-8, Daniel James wrote: >> >> On 11 November 2013 20:40, Robert Ramey <robertmac...(a)gmail.com> wrote: >> > >> > >> > On Monday, November 11, 2013 12:04:22 PM UTC-8, Daniel James wrote: >> >> >> >> On 11 November 2013 12:36, Sergey Cheban <s.ch...(a)drweb.com> wrote: >> >> > >> >> > Btw, what do you think about an additional SVN branch for the >> >> > post-RC2 >> >> > VC12-related bugfixes? >> >> >> >> I just created a branch at: >> >> >> >> https://svn.boost.org/svn/boost/branches/maintenance/1_55_0 >> >> >> > >> > Since we've tagged the current state of the release branch with 1_55_0 >> > do we >> > really have to create another branch? Can't >> > we just open the release branch for checkins? That would be easier. >> >> This branch is just for fixes. The release branch can have other >> changes, including breaking changes. > > > Soo... if we have fixes, we should check them into BOTH branches? If you want to, or just check it into release. It's a way to collect patches, make of that what you will.

1 0

Thanks to everyone for all the work on 1.55.0
by Marshall Clow 11 Nov '13

11 Nov '13

I just wanted to take off my "release manager" hat and say, personally, thank you to all the people who tried out the RCs/Betas/etc, reported bugs, and suggested fixes. You were all a big help in making the release. -- Marshall Marshall Clow Idio Software <mailto:mclow.lists@gmail.com> A.D. 1517: Martin Luther nails his 95 Theses to the church door and is promptly moderated down to (-1, Flamebait). -- Yu Suzuki

1 0

[release][1.55.0] Visual C++ known bugs
by Daniel James 11 Nov '13

11 Nov '13

Hello everyone, I'm compiling a list of known issues in boost with Visual Studio 2013/Visual C++ 12. I've made a quick start at: http://www.boost.org/users/history/version_1_55_0.html#version_1_55_0.known… Please let me know if there's anything I should add. Also, if available, tested patches against the release branch or release candidate would be very useful (I don't have the compiler myself, so I can't do that). If you can, please make them using subversion from the root of the repository. thanks, Daniel

5 9

New Version (0.4) of Boost.Application (Aspect Based) available to comments (RFC).
by Renato Forti 11 Nov '13

11 Nov '13

>> How is that supposed to work in Windows? (Does it need to?) Well, a simple analysis, it seems to be specific in POSIX, we do not have anything like it in WINDOWS. (I do not speak for sure) >> Its sensible to expose 'something' that can be integrated to native event loops, but the ideal choice under Windows is probably an event, not a pipe descriptor. I agree. I'm thinking of adding this aspect only in posix, Then if user want use it need add it and use it only on UNIX, e.g.: class myapp { public: int operator()(application::context& context) { #if defined(BOOST_ BOOST_POSIX_API) boost::shared_ptr<application:: selfpipe > selfpipe = context.get_aspect<application:: selfpipe >(); select(...,selfpipe->readfd(),...); #endif return 0; } }; // main int main(int argc, char *argv[]) { myapp app; application::context app_context; #if defined(BOOST_ BOOST_POSIX_API) app_context.add_aspect<application:: selfpipe>( boost::make_shared<application:: selfpipe>()); #endif return application::launch<application::common>(app, app_context); } Comments, Ideas, please! >> I think some analysis is needed. I need help in this subject, if any one with great experience on both sides can propose something (shine a light), it would be of great value! Thanks for help

2 2