March 2015 - Boost - lists.boost.org

To split, or not to split, or something else? RE: type_traits rewrite, modularization, dependencies, etc.
by Rene Rivera 20 Mar '15

20 Mar '15

I keep reading emails about the effort to reduce dependencies, and to split libraries into core/simple + whole/big.. And I keep thinking it's making things more complex than they need to be. I can see the need to compartmentalize the parts of a library that have minimal to little dependencies. So here's a suggestions for structure of managing this. I see options: 1. In each library allow for a "core" set of headers that are still in a core subdir. But that we can tell is a subset of the library that has the minimal dependencies. For example we would have: * libs/<lib>/include/boost/<lib>/core/.. * libs/<lib>/include/boost/<lib>/<the-rest>.. This of course assumes that it's headers only that would be core. Second option is to have each library have to explicit sublibs/parts. "core" and "full". Structure would be: * libs/<lib>/core/(build, doc, include, meta, test) * libs/<lib>/full/(build, doc, include, meta, test) Note, both core and full would be in the same git repo to make life easier for library authors and users that visit a library git repo on github. As it's only one place for everything for that one library. As a user, release manager, testing manager, and library author this would be easy to understand and manage than the current "some random set of libraries have core or not and in different places". -- -- Rene Rivera -- Grafik - Don't Assume Anything -- Robot Dreams - http://robot-dreams.net -- rrivera/acm.org (msn) - grafikrobot/aim,yahoo,skype,efnet,gmail

6 11

[1.58.0] Request permission to merge smart_ptr:94824c8 into master
by Peter Dimov 20 Mar '15

20 Mar '15

This https://github.com/boostorg/smart_ptr/commit/94824c807f76dc4f3d41109a75808e… fixes a problem on QNX/Blackberry by adding std:: to ptrdiff_t, in a place where size_t is already similarly qualified, so it should be very low risk. Tested locally on g++ 4.8.3, clang-3.4.2, msvc-8.0, msvc-10.0, msvc-11.0, msvc-12.0.

2 1

[metaparse] Review Manager
by Christophe Henry 20 Mar '15

20 Mar '15

Dear Review Wizards, I would like to volunteer to be Review Manager for the metaparse library, in agreement with Abel. We would like to suggest May 25 to June 07 for the review. If there are no objections, please sign me up. Regards, Christophe

8 13

[afio][v1.30 "C++ Now edition" released]
by Niall Douglas 20 Mar '15

20 Mar '15

Dear Boost, I am very pleased to announce the v1.30 release of proposed Boost AFIO, a Boost library implementing asynchronous file input/output which has been in the review queue since 2013. This is the version of AFIO which will be overviewed in my talk at C++ Now 2015. It also demonstrates the validity of the clang AST driven BindLib standalone Boost library conversion technique which will also be overviewed at C++ Now. Source distro: https://github.com/BoostGSoC13/boost.afio/archive/v1.30_Boost_v1.57.zi p Docs and CI dashboard: https://boostgsoc13.github.io/boost.afio/ Major new features in the v1.30 release: * Performance was improved by removing the transactional memory support in the v1.2 engine which is now compile time enabled only. Throughput now reaches up to 85% of raw ASIO and average latencies were reduced by 48%, yielding a 15 microsecond average latency with +/- 0.15 microsecond at a 95% confidence interval. This should enable AFIO to handle the latest generation PCIe SSDs with their 80 microsecond latencies reasonably well. * AFIO is now capable of complete standalone usage no Boost dependencies whatsoever, including for its unit testing via a Boost.Test emulation. * Due to its use of libclang AST generated template aliasing to allow user swapping of STL 11 implementation, minimum compilers required are now VS2013, GCC 4.7 and clang 3.2. All the VS2010 workarounds baggage has been eradicated. * All things related to async_data_op_req are now fully metaprogrammed with automatic generation by the compiler of suitable ASIO scatter-gather buffers from input types and STL containers with correct scatter-gather buffer coalescing where safe to do so. This automatic generation is user customisable via free function specialisation. * Support for relative paths from an open file descriptor was added. This lets you specify a file descriptor as a base from which a path fragment operates with the path resolution atomically performed in the kernel, and therefore the path of the base file descriptor may rapidly change safely. * AFIO is now optimally filing system race condition safe on all supported platforms, with per-API documentation of how much race safety is guaranteed per platform. A soak unit test verifies per commit that perfect race safety under severe and rapid filing system change is maintained. * A new afio::path type which very thinly wraps filesystem::path was added. The new afio specific path type makes the management of NT kernel path to Win32 path transition and conversion explicit, and therefore much harder to accidentally get wrong. A new normalise_path() function can convert a NT kernel path to a traditional Win32 path, complete with drive letter. * New platforms supported: Apple OS X and Android 5.0. * New filing system features support: All platforms: sparse file support, storage extent enumeration, extent deallocation ("hole punching"), volume metadata fetching (statfs), open file handle path discovery, hard link support, atomic relink (rename) support. Windows only: file symlinking, NT lock file performance flags (lock file performance is now not far from POSIX), workaround semantics for all the ways Windows will surprise you on the filing system such as files and directories which won't delete. AFIO now does a fair job of replicating POSIX filing system semantics on Windows (and better than Mingw/Cygwin presently). Linux: symlinks are now opened as file descriptors on kernels supporting that feature. * Doubled the size of the tutorial, with all the new material being non-toy examples. There is a particularly nifty "File locking via only atomic appending and hole punching" algorithm implementation in there, and it holds up surprisingly well compared to traditional file locks except on ZFS. Plus, it's completely portable. * Some utility routines in preparation for AFIO v1.4: a cryptographically strong random filename generator, a hugepage TLB STL allocator useful for efficient file copy buffers, a very highly optimised C++ 14 constexpr constructing arbitrary Hamming code SECDED template class (an amazing 22 cycles per byte for the 32784,32768 code), and secure to-hex and from-hex validating converters. * No shortage of bug fixes, helped hugely by additional unit testing for correct error handling. Detailed changelog can be found at https://boostgsoc13.github.io/boost.afio/doc/html/afio/release_notes.h tml. I'll tag a Boost 1.58 release on github once Boost 1.58 is out. This release took far longer than I originally intended mainly due to the extensive debugging of weird filing system races, and for six weeks now I've been working till 5am every night after my day job. I cannot tell you how glad I am to see this release out! My enormous thanks to my family who have put up with a lot for this release, and to my employer for a fairly dopey employee recently. Bed, here I come! Niall -- ned Productions Limited Consulting http://www.nedproductions.biz/ http://ie.linkedin.com/in/nialldouglas/

3 6

enhanced vector and deque competency test
by Islam Taha 19 Mar '15

19 Mar '15

Hi, I am Islam Taha i’m interested in the idea of enhanced vector and deck and i’ve implemented the simple vector-like container, it would be pleasurable to review and give me your feedback on my implementation. that’s the link ( https://github.com/islam-taha/stdVector/blob/master/vector.cpp <https://github.com/islam-taha/stdVector/blob/master/vector.cpp> ). Thanks in advance.

2 1

GSoC 2015 prospective student
by Tanuj kr Aasawat 19 Mar '15

19 Mar '15

Hi Everyone, I'm Tanuj Kr Aasawat, a final year student of Information Technology (B.E.). I would like to work on implementing a graph container with at least the following features: size, insert, delete, bfs and dfs. In the later part, would like to work on optimising bfs, using parallel computing techniques. In recent years, we have seen that major research is going on optimising the search operations for real world graphs (including RMAT graph and LiveJournal graph). Anticipating positive reply from your side. Thanks and Regards, Tanuj Kr Aasawat Deptt. of I.T, Jadavpur University, India Mitacs Globalink Research Intern, University of British Columbia, Vancouver, Canada

3 4

Gsoc 15 application (boost.uBLAS)
by siddharth kumar 19 Mar '15

19 Mar '15

Hello, I am Siddharth, a Computer science student from IIT Guwahati. After doing a bit of research over the project ideas, I have found the idea of further developing boost.ublas library interesting and so I have written the competency code. Here is the link to the code: https://github.com/sid6641/boost.uBLAS-Gsoc.git I have done a basic course on Linear algebra and Discrete Mathematics, so such a project suits my needs. I need a little help for proceeding further. 1. how is the error handling done in boost library. I am using assert for error handeling (but it does not sem the right way to me), the gaussian elimination doesn't work on singular matrices. So, I would like to set the errno and display apt message to the user when a singular matrix is the input. 2. Is there any predefined format for testing codes in boost library. I have done manual testing. Sorry, I am not familiar with most of the code in boost library. Any further helpful resources are welcome. p.s. I have started writing the proposal for the project and I'll soon draft it for review. Thanks, Siddharth

6 10

Boost on Android
by Dmitry Moskalchuk 19 Mar '15

19 Mar '15

Hi all, As someone of you already know, we've started regular regression testing of Boost on Android. After some (bit longer than I expected) time of debugging, we finally made it working in automatic mode so you can see our results of Boost testing on http://www.boost.org/development/tests/develop/developer/summary.html. We also automatically publish Android-only testing results on https://boost.crystax.net/master/developer/summary.html and https://boost.crystax.net/develop/developer/summary.html. Currently, we have numerous of fails. Some of them are bugs of Boost libraries, some - bugs of CrystaX NDK. We're strongly aim to continue support Boost on Android and fix our NDK to make Android as POSIX-conforming as possible; however, it's not easy to review all fails and distinguish NDK bugs from Boost bugs. So we're asking Boost libraries authors to start looking on Android regression testing results and file tickets to our tracker (https://tracker.crystax.net/projects/ndk) if it become clear that root of fail is bug of CrystaX NDK. CrystaX NDK is open-source project started as my personal project in 2009 to add C++ support to the Google's Android NDK (Google published first Android NDK release in 2009, and there was no support of C++ at all). My patches were published immediately but Google have added the same level of C++ support two years after first CrystaX NDK release, being constantly requested within two years to do that by Android developers community. To that time CrystaX NDK was much more than just "Android NDK with C++ support". Nowadays CrystaX NDK differs very much from Google's one - mostly because Google isn't interested in evolving of native development for Android; but we're _very_ interested. Our main goal is to create powerful comprehensive toolkit for native development for Android, and in fact we're making it much better than Google do. We're strongly aim to continue working on CrystaX NDK and support Boost on Android - just because it's helpful to our NDK too, not only to the Boost community. However, since CrystaX NDK is fully open sourced and we're working on it without being paid by any organization, we need help from peoples interested in evolving of native development for Android. Since 2009 I was working on the project in partial time fashion, making money other way (being paid contractor and working on many commercial projects); but last eight months we have switched to the full time scheme, and have achieved significant results in last release. Now we want to continue working on it full time, improving it more and more, and making native development for Android easier. We've started fundraiser on https://www.bountysource.com/teams/crystaxndk/fundraiser to collect funds needed for us to continue working on CrystaX NDK full time. We're calling everyone interested in Boost-on-Android to help us - either by sponsoring us or providing contributions to our project. I hope I did not violated rules of this mailing list because my message is directly related to the Boost development. We just want to make both CrystaX NDK and Boost better. Some links: https://www.crystax.net/android/ndk - CrystaX NDK project page https://tracker.crystax.net/projects/ndk - our bug tracker https://www.crystax.net/contact - here you could contact us directly -- Dmitry Moskalchuk

3 13

A question about appling GSoC2015
by jianwei li 19 Mar '15

19 Mar '15

Hi I'm a graduate student of Northwestern Polytenical University in China. Just as listed in the web,based on the proposal requirement of 'Boost Documant Library development' ,we should copy and past a programming test at the end of the proposal.But i'm not sure,do you just need the result picture or the all code. Thank you Best Regard Jianwei Li

2 1

GSoC2015
by raul lopez turcios 19 Mar '15

19 Mar '15

Hello, I wish to apply to the GSoC program, i have already read the wiki, and i'm currently working in a proposal. I just wanted to make a few questions as this is my first time applying for Boost Libraries and GSoC in general. First of all, i would like to know how much of a chance one gets as being "starter" working on boost. Second i would like to know if there is a way for someone to check out my proposal to see if i could improve anything, or should i just ask a teached in regards to that. And also i would like to know if any code i had already made in C++ is valid to be added to the proposal, or if there is anything else to improve your chances to be selected. Thanks for yor time.

2 1