March 2015 - Boost - lists.preview.boost.org

[1.58.0] Release Notes
by Daniel James 09 Mar '15

09 Mar '15

Hi, The beta release is coming up soon, so please can you add your release notes as necessary. Easiest way is probably to create a pull request at: https://github.com/boostorg/website/blob/master/feed/history/boost_1_58_0.q… thanks, Daniel

3 3

[gsoc15]Request for feedback: Enhanced vector and deque containers proposal
by Sandilya Jandhyala 08 Mar '15

08 Mar '15

Hello Boost, My name is Sandilya Jandhyala, and I'm interested in the 'Enhanced vector and deque containers' project idea. I've prepared a first draft of my proposal. It can be viewed here: https://github.com/jysandy/boost-gsoc-proposal/blob/master/README.md I haven't included my solution to the programming competency test, since I don't want it to be copied. I'd appreciate feedback/suggestions/criticisms, since I feel that the proposal still has some room for improvement. I'd also appreciate suggestions for additional features. Thanks, Sandilya

1 0

GSoC project " Enhanced vector and deque containers "
by ahmad samir 08 Mar '15

08 Mar '15

Good morning guys I'm Ahmad, I want to contribute in " Enhanced vector and deque containers ", I'm working on the programming competency test, but I have a question about the main idea * The idea is " Write full-fledged and enhanced C++11 versions of std::deque and std::vector with extra functionality that gives the programmer more control over buffers, segments and iteration " . * Is there specific features or new functionality you want to add or I should search for some new features in other languages like " java & python " which isn't in C++ vector and deque containers, or search for what can give the programmer more control on these containers ?.

1 0

GSoC 2015 Paritipation
by vladiksonic＠mail.ru 07 Mar '15

07 Mar '15

Dear Boost, I am eager to participate in GSoC 2015 in one of the two projects (from this page: https://svn.boost.org/trac/boost/wiki/SoC2015 , 1 or 5). As far as I understood - the first one is tougher, and (as for me) conditions are clearer, and the second one is (seemingly) simpler. I am totally new to Boost (never used it, actually, but that is because I mostly used C++ in programming contests). I am familiar with algorithms (from contests), and testing (manual, both for stability and performance), but I am not very familiar with C++11 (but I can compensate for that, because I have knowledge of other languages (C#, Javascript (nodejs)) and older C++ versions), and automated tests. At first, I'd like to ask - what should I choose? My main goals are: 1) To enter the community 2) Learn C++11 new features (I will start early on that, obviously) 3) And to learn how to code open-source programs (up till now I only have closed-source ones, apart from one project, which has not-so-good code quality - see here: https://github.com/hotab/OpticalBuilder ) 4) Plus understand better how to maintain library code (I already have some understanding of that, but I want to sharpen it). Both can give me that. I'm looking forward to hearing from you, Vladislav Babkin

1 0

[dynamic_bitset] Maintainer available?
by Riccardo Marcangelo 07 Mar '15

07 Mar '15

Hi Boost developers, Does anybody know if the maintainer of dynamic_bitset is available? I made some contributions to dynamic_bitset on GitHub last year but there doesn't appear to have been any activity in the repository for many months. My pop_back() commit was accepted into the 'develop' branch over 5 months ago and, despite passing tests, has yet to be merged to 'master'. I also implemented capacity(), reserve(), and shrink_to_fit() which have been sitting in a pull request waiting to go into 'develop' since November. It would be nice to get these into 1.58.0 if possible. Kind regards, Riccardo -- View this message in context: http://boost.2283326.n4.nabble.com/dynamic-bitset-Maintainer-available-tp46… Sent from the Boost - Dev mailing list archive at Nabble.com.

2 1

GSoC project " Enhanced vector and deque containers "
by ahmad samir 07 Mar '15

07 Mar '15

Good morning guys I'm Ahmad, I want to contribute in " Enhanced vector and deque containers ", I'm working on the programming competency test, but I have a question about the main idea * The idea is " Write full-fledged and enhanced C++11 versions of std::deque and std::vector with extra functionality that gives the programmer more control over buffers, segments and iteration " . * Is there specific features or new functionality you want to add or I should search for some new features in other languages like " java & python " which isn't in C++ vector and deque containers, or search for what can give the programmer more control on these containers ?.

1 0

[GSOC2015] uBLAS Matrix Solver Project
by Rajaditya Mukherjee 07 Mar '15

07 Mar '15

Hi, My name is Raj and I am Phd student in Computer Graphics. I am interested in tackling the problem of uBLAS Matrix Solver and in order to write my proposal, I am looking for inputs for which of the following algorithms will be most useful for prospective users in boost-numeric library. Here is a categorical list of all the prospective ones which will bring uBLAS updated to other commercial libraries like Eigen/Armadillo. Please let me know your preferences.... *David Bellot* : As a potential mentor, do you have any specific additions or deletions for this list? This could also be useful for other candidates pursuing this project. *DENSE SOLVERS AND DECOMPOSITION* : 1) *QR Decomposition* - *(Must have)* For orthogonalization of column spaces and solutions to linear systems. (Bonus : Also rank revealing..) 2) *Cholesky Decomposition* - *(Must have)* For symmetric Positive Definite systems often encountered in PDE for FEM Systems... 3) *Householder Method* - Conversion to tridiagonal form for eigen solvers. *SPARSE SOLVERS AND PRECONDITIONERS* : 1) *Conjugate Gradient* - *(Must have)* For symmetric Positive Definite systems, this is the kryvlov space method of choice. Both general and preconditioned variants need to be implemented for convergence issues. 2) *BiCGSTAB* *(Needs introspection)* - For non symmetric systems.. 3) *Incomplete Cholesky Decomposition* *(Good to have)* - For symmetric Positive definite sparse matrices, to be used as preconditioner as extension to (1) for preconditioned CG Methods ... 4) *Jacobi Preconditioner* *(Must have)* - As prerequisite for step(1). *EIGEN DECOMPOSITION MODULES (ONLY FOR DENSE MODULES)**:* 1) *Symmetric Eigen Values* - *(Must have)* Like SSYEV Module in Lapack - That is first reduction to a tridiagonal form using Householder then using QR Algorithm for Eigen Value computation. 2) *NonSymmetric Eigen Values* - *(Good to have)* Like SGEEV module in Lapack - using Schur decompositions as an intermediate step in the above algorithm. 3) *Generalized Eigen Values* - *(needs introspection)* I use this in my research a lot and its a good thing to have.. ** Computing Eigen Decomposition of sparse modules needs special robust numerical treatment using implicitly restarted arnoldi iterations and may be treated as optional extensions.

3 3

[MPL lite or MPL 2] A modest proposal
by Robert Ramey 07 Mar '15

07 Mar '15

I've been looking at another thread which address the problems of maintaining and/or finding a maintainer for MPL. I've come up with an idea for a proposal which I've laid out here: Library Idea MPL lite or MPL2 <http://rrsd.com/blincubator.com/bi_suggestion/mpl-lite-or-mpl2/> -- View this message in context: http://boost.2283326.n4.nabble.com/MPL-lite-or-MPL-2-A-modest-proposal-tp46… Sent from the Boost - Dev mailing list archive at Nabble.com.

8 18

[asio] Massive regression failures
by Aparna Kumta 06 Mar '15

06 Mar '15

I am running the regression tests on the develop branch with Oracle Solaris Studio12.4 on Solaris 11.2 OS. Tests for asio fail as follows: ../boost/system/config.hpp", line 34: Error: #error Must not define both BOOST_SYSTEM_DYN_LINK and BOOST_SYSTEM_STATIC_LINK. I looked at the developer issues on http://www.boost.org/development/tests/develop/developer/issues.html and see that asio tests fail similarly. I'm not sure what causes both BOOST_SYSTEM_DYN_LINK and BOOST_SYSTEM_STATIC_LINK. to be defined on the command line. Any idea on how this can be fixed? Thanks, Aparna

2 2

[move] unique_ptr: problems with move-assign constructor when used with base/derived classes in C++03
by Adam Romanek 06 Mar '15

06 Mar '15

Hi, There is a problem with move-assignment constructor in boost::movelib::unique_ptr when used with a pair of base/derived classes in C++98/03 mode [1] (uncomment the code to see the compiler error). Is it a known issue? Or a limitation of the emulation? I discovered the issue when I wanted to use boost::movelib::make_shared in an answer to a question on StackOverflow [2], i.e. return boost::movelib::make_shared<Derived>(); Where the return value is boost::movelib::unique_ptr<Base>, but unfortunately because of this issue, in C++03 we need to use: return unique_ptr<Base> x(new Derived()); I can create a ticket if that is desired. Nevertheles, thanks for boost::movelib::unique_ptr anyway! BTW, note how the appearance of boost::movelib::unique_ptr makes the most voted answer on the question [3] invalid. This is where my answer comes in. Best regards, Adam Romanek [1] http://melpon.org/wandbox/permlink/8Rvk2g7LJ0Ey0BQL [2] http://stackoverflow.com/a/28193068/1776942 [3] http://stackoverflow.com/q/2953530/1776942

2 2