- Boost - lists.preview.boost.org

[multiprecision]: 15% slowdown using -std=c++11 compiler flag
by Kim Walisch 20 Nov '17

20 Nov '17

I have been investigating a 15% performance regression in my C++ primesum program (https://github.com/kimwalisch/primesum/tree/256-bit) over the last 2 days. By lots of benchmarking I was able to identify the boost multiprecision library together with -std=c++11 (or -std=gnu++11) as the culprit for the performance regression because I have also a version of the primesum program which does not use the boost multiprecision library and in this version there is no performance regression when compiling using -std=c++11. I have tested using multiple versions of the boost multiprecision library including the latest 1.65.1. The slowdown happens on both GCC (versions: 5.4, 6.4, 7.2) and Clang (version 3.8) on x86_64 Linux. I am only using the int256_t and uint256_t types (hence cpp_int backend) and I am doing only simple integer arithmetic: +, - and *. Is this a known issue and is there a known workaround e.g. special compiler flag? I could revert to C++98 but I really don't want to do that... Thanks, Kim Walisch

2 2

[contract] _NOEXCEPT errors in <ostream> for Clang 5.0 C++1z on Teeks Linux
by Lorenzo Caminiti 20 Nov '17

20 Nov '17

Hello all, Most of Boost.Contract regression test fail for toolset clang-gnu-linux-5.0~c++1z~lc on platform teeks99-02-dc5-1z-lc-Docker-64on64 because of the following error on #include <ostream> done by Boost.Range (and sometimes by Boost.IOStreams): "clang++-5.0" -c -x c++ -Wno-c99-extensions -std=c++1z -stdlib=libc++ -isystem/usr/include/libcxxabi -O0 -g -fno-inline -Wall -g -fPIC -m64 -DBOOST_ALL_NO_LIB=1 -DBOOST_CONTRACT_DISABLE_THREADS -DBOOST_CONTRACT_DYN_LINK -DBOOST_SYSTEM_DYN_LINK=1 -I".." -I"../libs/contract/test/function" -o "/var/boost/run/results/boost/bin.v2/libs/contract/test/function-smoke.test/clang-gnu-linux-5.0~c++1z~lc/debug/threadapi-pthread/function/smoke.o" "../libs/contract/test/function/smoke.cpp" In file included from ../libs/contract/test/function/smoke.cpp:9: In file included from ../libs/contract/test/function/../detail/oteststream.hpp:10: In file included from ../boost/iostreams/tee.hpp:18: In file included from ../boost/iostreams/detail/adapter/device_adapter.hpp:22: In file included from ../boost/iostreams/detail/call_traits.hpp:15: In file included from ../boost/iostreams/traits.hpp:38: In file included from ../boost/range/iterator_range.hpp:14: In file included from ../boost/range/iterator_range_io.hpp:32: In file included from /usr/include/c++/v1/ostream:138: In file included from /usr/include/c++/v1/ios:216: In file included from /usr/include/c++/v1/__locale:15: /usr/include/c++/v1/string:1942:14: error: expected ';' at end of declaration _NOEXCEPT ^ /usr/include/c++/v1/string:1947:9: error: expected unqualified-id _NOEXCEPT ^ /usr/include/c++/v1/__config:368:21: note: expanded from macro '_NOEXCEPT' # define _NOEXCEPT noexcept ^ 2 errors generated. Of course, Boost.Range and Boost.IOStreams show the same error on this toolset/platform independently of Boost.Contract. Any idea what the issue might be? Thanks, --Lorenzo

5 9

[operators] A modern SFINAE-based version of boost::operators?
by Beman Dawes 19 Nov '17

19 Nov '17

Peter Sommerlad, committee member and C++Now presenter who often proposes additions to fill in holes in the standard library, asked me: Are you aware of anybody who tried to provide a boost::operators style of automatically providing additional operators with a single base class that through SFINAE injects all possible operators based on the ones defined in the template parameter? This won't give you the control of the current boost::operators, but would be much easier to teach. For example struct Me : make_operators_for<Me>{ Me& operator+=(Me const&); // You get + bool operator<(Me const&) const; // You get all relops (<=> will make that obsolete) Me& operator++(); // you get postfix //etc. }; Today we have the facilities and compilers to make that happen. What do you think? Who should I ask? Anyone doing any work on operators or have any thoughts about updating boost::operators? --Beman

14 31

[python] Incorrect link
by Christian Henning 19 Nov '17

19 Nov '17

Hi all, on page http://boostorg.github.io/python/doc/html/tutorial/tutorial/embedding.html the link to http://www.python.org/doc/api/exceptionHandling.html is dead. Perhaps, it should be this link: https://docs.python.org/3/tutorial/errors.html Regards, Christian

1 0

[release] Boost 1.66.0 Beta 1 Release Candidate 2
by Daniel James 19 Nov '17

19 Nov '17

The release candidates for the 1.66.0 beta release are now available at: <https://dl.bintray.com/boostorg/beta/1.66.0.beta.1/source/> You need to scroll past the first release candidates to find the downloads. The SHA256 checksums are as follows: e7b23e638e8c3a9da1512439385b04850829cea7bd56d8398677c096857d830b boost_1_66_0_b1_rc2.7z 5d0405eb73bd609ebcd54bbc96d4f1e2d97c51b96c4703706695ddd94f942d6b boost_1_66_0_b1_rc2.tar.bz2 f0f455e39bc47b2b670a17cb221687506e9a22dc478956d3d8aabb9cb835a7dc boost_1_66_0_b1_rc2.tar.gz 694c8b9e902b7981670bce166f675f3875c19326936db51aabd98d2ddfea9b45 boost_1_66_0_b1_rc2.zip The in progress release notes are at: <http://www.boost.org/users/history/in_progress.html> And the documentation is at: <http://www.boost.org/doc/libs/1_66_0_beta1/> As always, the release managers would appreciate it if you download the candidate of your choice and give building it a try. Please report both success and failure, and anything else that is noteworthy.

3 2

Re: [boost] Adding polymorphic_value to boost
by Tom 19 Nov '17

19 Nov '17

> Message: 10 > Date: Thu, 16 Nov 2017 02:14:30 -0700 > From: Jonathan Coe <jonathanbcoe(a)gmail.com> > To: boost(a)lists.boost.org > Cc: Matt Calabrese <calabrese(a)google.com> > Subject: [boost] Adding polymorphic_value to boost > Message-ID: > <CAAbBDD9Ck+LhQ4iUraBtBkxexVKDt167qxyfTpDQqYM9bcMqLQ(a)mail.gmail.com> > Content-Type: text/plain; charset="UTF-8" > > I'd like to propose adding the class-template `polymorphic_value` to boost. > > `polymorphic_value` is designed to allow the compiler to correctly generate > special member functions for classes with polymorphic components (It's > `any` with an interface). > > `polymorphic_value` has been proposed for standardisation in P0201 and has > advanced through the Library Evolution group in Albuquerque (Nov 5-10). > https://wg21.link/p0201 > > A working draft of the proposal and implementation can be found at > https://github.com/jbcoe/polymorphic_value > > I gave a presentation about polymorphic_value at C++::London here: > https://skillsmatter.com/skillscasts/9539-a-polymorphic-value-type-for-c-pl… > > Matt Calabrese has kindly offered to be review manager. > > Kind regards, > > Jon Hi, I'd be interested in such a utility. Mathias Gaunard proposed something similar on the mailing list a long while ago, I implemented a simple version myself based on that. I have a small number of questions. Given that this is a value type, does the current approach over-expose pointer (ownership) in the interface (in essence it still appears to be a glorified smart pointer) and understate its more fundamental purpose as a regular value with dynamic type? There might be other things which can usefully be added for such a value type, for example permitting direct construction of an object of specific (derived) type, by passing the type as well as the arguments for the constructor of that type (or wrapping the arguments in some kind of in-place factory with explicit type). Something like: // Simple wrapper to enable a static type to be passed through a forwarding constructor. template <typename T> struct Type {}; // Forwarding constructor - actual type is deduced from type wrapped in (unnamed) first argument's type template <typename D, typename... Args> explicit poly_obj(Type<D>, Args&&... args); // Also could have an emplace member function template <typename D, typename... Args> void emplace(Type<D>, Args&&... args); In theory you wouldn't need to expose pointers in the interface at all? In fact, should a type-erasure library be preferred (e.g., Boost.TypeErasure or Adobe.Poly) as it provides a more general solution for manipulating polymorphic value types (though typically harder to use)? Would you be considering supporting C++98 via using the boost::move library? Regards, Tom

1 0

[gil][test] HOWTO set up the Boost testing process for projects with extra prerequisites ?
by Stefan Seefeld 17 Nov '17

17 Nov '17

Hi, I just merged some work to GIL's `develop` branch. This notably involves improvements to the testing infrastructure. Testing GIL is a bit involved, as it requires various backend libraries (PNG, JPEG, TIFF, etc.) to be installed. While the stand-alone testing works fine (https://github.com/boostorg/gil/blob/develop/.travis.yml#L304-L306) the integrated testing will thus fail, until the testing logic is augmented by the necessary instructions to a) install the prerequisites and b) invoke the build system with the appropriate flags (https://github.com/boostorg/gil/blob/develop/io/test/Jamfile#L13-L22) Thus my question: do we have any documentation that explains how to do this ? What do I (as GIL maintainer) or any testers need to do for the integrated testing to work ? (I can't resist: all these libraries that have special requirements really shouldn't be built and tested in such a monolithic process. It's a burden to all tester, maintainers, and packagers. It would be far better to handle them more independently.) Stefan -- ...ich hab' noch einen Koffer in Berlin...

3 8

[density] paged memory and concurrency
by Giuseppe Campana 17 Nov '17

17 Nov '17

Hi, in the last year I’ve been experimenting concurrent data structures on density, a C++11 library focused on: - Page-based memory management. - Heterogeneous and function queues. Single-threaded, mutex-ed, lock-free and spin-locking versions are available. Lock-free and spin-locking queues support any combination of single/multiple producers/consumers. - Lifo memory management, mainly to provide a safe alternative to alloca with almost the same performances. - Parametric control of the progress guarantee in concurrent algorithms. I have to do a major update to fix the known issues, so I’d like to receive any feedback before it, especially about the design. The document http://giucamp.github.io/density/doc/html/implementation.pdf explains the implementation of almost everything, though it’s quite long. Documentation: http://giucamp.github.io/density/doc/html/index.html Repository: https://github.com/giucamp/density Is there any interest on it? Giuseppe Campana.

1 0

Re: [boost] [Boost-users] Could guys help me on a boost MPI link error
by degski 16 Nov '17

16 Nov '17

On 16 November 2017 at 07:33, Renfan Luo <Renfan.Luo(a)perpetuum.com> wrote: > Thanks a lot! After changing X64 to X86 on the top of drop-down window > of IDE, the problem has been fixed. > You're welcome, degski

1 0

bjam boost build on windows - changing the path strategy (stage)
by James E. King, III 16 Nov '17

16 Nov '17

I would like to build boost into a prefix directory (.\b2.exe --prefix=C:\prefix) containing "C:\prefix\include\boost\..." and "C:\prefix\lib", just like the linux build does. On Windows it builds into ""C:\prefix\include\boost-1_66\boost\..." instead - is there a bjam control to get rid of the version in the prefix include path for header copies? - Jim

5 5

[release] 1.66.0 fully closed for beta
by Daniel James 15 Nov '17

15 Nov '17

I've stopped updating the master branch of the super project, and will try to get the beta release out tomorrow, so there's only a little delay. You can still add your release notes at: <https://github.com/boostorg/website/blob/master/feed/history/boost_1_66_0.q…> Thanks, -- The Boost Release Team

1 1

Re: [boost] seeking endorsement for histogram library
by Hans Dembinski 15 Nov '17

15 Nov '17

Dear Fabian, I am CCing boost, because maybe some others want to read this as well and/or correct me. > On 15. Nov 2017, at 14:21, Fabian Bösch <boeschf(a)student.ethz.ch> wrote: > > Thanks for clarifying. While it is true that Boost.Accumulators does > this, it is for the uni-variate case only, afaik. Please correct me, if > I'm wrong. I believe you can compute a multi-dimensional mean, if you feed an appropriate Boost.Accumulator with std::valarrays. Maybe there is an easier way. > Now you have multi-variate histograms for which one would probably also > like to know the Covariance (rank-2 tensor), for example. Maybe this is > out of scope but a general solution for this would be nice. But perhaps > this would rather be addressed in another generalized accumulator > library. What do you think? I think you are right. You can compute the covariance of two variables with Boost.Accumulators: http://www.boost.org/doc/libs/1_65_0/doc/html/accumulators/user_s_guide.htm… but it does not give you a full matrix for N covariates. Maybe you could contact the maintainer of Accumulators, Eric Niebler. It is a reasonable request that would fit better into Accumulators than in Histogram. Best regards, Hans

1 0

Re: [boost] seeking endorsement for histogram library
by Hans Dembinski 15 Nov '17

15 Nov '17

Dear Francesco, > On 13. Nov 2017, at 19:09, Francesco Guerrieri <f.guerrieri(a)gmail.com> wrote: > > Hi this proposal is of great interest to me, thanks for sharing your work with us. Which compilers/OSs are supported? I tested the code on OSX with Apple clang, and via TravisCI on several versions of gcc on Linux. The CMake build works on both OSX and Linux. I wanted to also test Windows via Appveyor, but couldn't get my CMake build script to work with msvc. I am passing compiler flags directly, which works for clang and gcc, but naturally not for msvc. Debugging this non-interactively, by committing a change to github, letting Appveyor run, seeing it breaking again..., was a pain and I stopped at some point. If an expert for Windows had a look at my CMakeLists.txt, it is probably something that could be easily fixed. Best regards, Hans

2 2

seeking endorsement for histogram library
by Hans Dembinski 14 Nov '17

14 Nov '17

Dear Boost developers, I believe the histogram library is now ready to be presented here and I am looking for endorsement. Histogram is a C++11 header-only library that provides a safe, convenient, and fast multi-dimensional histogram for statistical analysis and visualisation. The library has a unique feature set, among it a safety guarantee that the counts in the histogram cannot overflow. It is easily customisable for power users, while providing defaults that just work for the occasional user. Meta-programming is used to provide an especially fast histogram implementation that can be used when the histogram configuration is known at compile-time. A dynamic implementation is also provided for the other case when the configuration is only known at run-time. The two implementations share a common interface, so it is easy to switch between them. Python bindings are included for the dynamic implementation. The Python interface supports Numpy arrays to greatly speed up the exchange of data between the Python and C++ side. I tested the performance of the library in benchmarks against other libraries, which have fewer features, and this library beats them in almost all cases. I am stealing the style of the rest of the email from Antony Polukhin. Library: https://github.com/HDembinski/histogram <https://github.com/HDembinski/histogram> Docs: https://htmlpreview.github.io/?https://raw.githubusercontent.com/HDembinski… <https://htmlpreview.github.io/?https://raw.githubusercontent.com/HDembinski…> Boost Library Incubator: http://blincubator.com/bi_library/histogram-2/?gform_post_id=1582 <http://blincubator.com/bi_library/histogram-2/?gform_post_id=1582> Library changes since last mail: * Support for efficient adding of multiple histograms and scaling * Support for reduce transformation * Re-design of category axis as a general mapping between unique values and bins * Re-design of the bin description an axis returns upon element access * Regular axis can accepted bijective (user-provided) transformations * Interface cleanup, refactoring, and simplification * Finished documentation Best regards, Hans

3 4

[test] [build] best practices for looking up data files from test executables
by Stefan Seefeld 14 Nov '17

14 Nov '17

Hi there, I'm trying to write a test that needs to access a data file. As I'm surely not the first wanting to use external data in my tests, I wonder whether there are established "best practices" or idioms when using Boost.Build and Boost.Test. My first attempt was to compute the relative path using the test source's `__FILE__`, until I realized the path was invalid due to the build logic moving executables before running them. Then I considered using a command-line option to point at the data directory, but that seems quite involved when using boost::test (and BOOST_TEST_MODULE)... Any suggestions for how to approach this ? Thanks, Stefan -- ...ich hab' noch einen Koffer in Berlin...

7 10

Exceptions vs Assertion for library API?
by Jonathan Biggar 13 Nov '17

13 Nov '17

Any advice in choosing between throwing an exception vs asserting for invalid uses of a library API? I'm building a small "transactional" library I'd like to contribute to boost where certain API calls are constrained to only be valid if the transaction is still "in-progress". Such calls will never succeed, and are easy for the implementation to diagnose. I'd like to have the library fail an assertion, which is far clearer to the user that the code logic is incorrect, but it bothers me that assertions are much more difficult to write unit tests for. (Is there even a concept of a test that passes if if fails to compile or fails with an assertion in the boost testing framework?) I suppose I could mangle the interface to ensure that the user has nothing to invoke the API calls in question unless the transaction is in-progress, but that generally requires a custom "lock" interface for my RAII implementation that makes the API more complex to grasp. -- Jon Biggar jon(a)biggar.org

5 4

[release] 1.66.0 closing for beta tomorrow
by Daniel James 13 Nov '17

13 Nov '17

Hi, I'll close the master branch for the beta release tomorrow. Sorry we've been quiet about the release and that this is a bit short notice. If you haven't written your release notes yet, you can add them to: https://github.com/boostorg/website/blob/master/feed/history/boost_1_66_0.q… Use the pencil icon at the top right of the code to edit it, or you can mail your notes to me, and I'll add them. thanks, Daniel

10 28

[release][1.66.0] Permission to merge Hana
by Louis Dionne 12 Nov '17

12 Nov '17

Hi, Being at a C++ meeting all week, I did not see the release coming and the email in time. None of the changes to Hana have been merged to master, and I'd like to get them in for the release. Can I merge them, or should I wait until after the beta? Thanks and apologies for the untimeliness, Louis -- Sent from: http://boost.2283326.n4.nabble.com/Boost-Dev-f2600599.html

2 2

[release][1.66.0] Permission to merge Predef.
by Rene Rivera 10 Nov '17

10 Nov '17

Some late fixes for Predef came in over the past two days: < https://github.com/boostorg/predef/commit/4a230ebe853980e27e7fb7cb36605fd6c… > < https://github.com/boostorg/predef/commit/74f88005cc2227b4f8fd4d640daae18c2… > < https://github.com/boostorg/predef/commit/5e8f5d5a4b9df640ffaeaf51e89c83e28… > Can I merge those to master for the beta? -- -- Rene Rivera -- Grafik - Don't Assume Anything -- Robot Dreams - http://robot-dreams.net

2 1

Seeking endorsement for the reflection library (PFR)
by Antony Polukhin 10 Nov '17

10 Nov '17

Hello, PFR is a C++14 library that provides tuple-like methods for aggregate initializables. Library has a set of predefined operators and useful functions for everyday use. Works with user-defined types without any macro or boilerplate code. Library: https://github.com/apolukhin/magic_get Docs: http://apolukhin.github.io/magic_get/ Boost Library Incubator: http://blincubator.com/bi_library/pfr-pod-flat-reflection/?gform_post_id=16… Seeking for some endorsement, as per http://www.boost.org/development/submissions.html Library changes since last mail: * Now works on latest MSVC in C++17 mode * C++14 limitations in reflection were significantly relaxed * Removed UB (no more reinterpret casting of unrelated types!) * Thaler Benedek was approved as a review manager for the library. Boost libraries that may benefit from PFR: * Convert - for out-of-the-box conversions of user types * Hana - for using user types instead of tuples * Intrusive - for reusing unused bits in user provided types * Serialization - for out-of-the-box serialization of user types * Spirit - for putting rule results into user provided types without Phoenix or Fusion macro * Test - for outputting content of user provided types * TypeTraits - for traits like has_unique_object_representations in C++14 -- Best regards, Antony Polukhin

5 11

Re: [boost] A GUI Library
by ha ha 09 Nov '17

09 Nov '17

If your library is only dependent on boost, not any other library like nana then it will be better in most of the beginner programmers. IF your library is pre-compiled then it will catch popularity soon, because "allegro 5" is popular for two reasons :- 1. It is very powerful and vast. 2. It is very very easy to deploy, than SDL and SFML. The first choice of the beginner programmers is always SDL but they quickly move to Allegro because of the availability of pre-built binaries. -- Sent from: http://boost.2283326.n4.nabble.com/Boost-Dev-f2600599.html

1 0

[devector] optimum growing/insertion policy
by Joaquin M López Muñoz 08 Nov '17

08 Nov '17

Hi, Much an worthy discussion has taken place with respect to growing and insertion policies of boost::double_ended::devector. I think I can justify what the optimum policy is when the overall goal is to minimize number of element movements. What follows is *not* a formal optimality proof, though it's my hunch one could be provided relatively easily. Assumptions * The cost of allocating new memory is negligible when compared to the cost of moving elements around (either intra-buffer or when migrating to a new buffer); so, the growing/insertion policy should focus on minimizing number of elements moved. This implies fully occupying free space before reallocation is *not* the primary goal of the growing/insertion policy. * For the purposes of our discussion, we consider an ever-growing devector without intervening erasures --erasures can be accomodated later as we will see. * Element insertion is modelled as a process where each new insertion happens at a random position within the current devector. The minimal characterization (i.e. its 1st moment in stochastic process theory terminology) is just the percentage R of insertions that happen within the right half of the devector --we call these *right insertions*; left insertions happen then with a frequency L=1-R. push_back is a right insertion, push_front a left insertion. Facts * When a right insertion arrives, if there's free back capacity the optimum insertion policy is to shift elements to the right, as this involves fewer movements than the other way around. The symmetrical applies for left insertions. * If a right insertion at position N arrives and there's no free back capacity (but there is free front capacity), we have only two options: A. Reallocate to get back free capacity. B. Shift elements to the left. The key observation here is that A is cheaper in the long run: as the scenario is one of an ever-growing sequence, reallocation will happen eventually and we can just amortize it in our analysis, so reallocating early saves us the extra movements incurred when left shifting on a right insertion, namely N - (size()-N) = 2*N - size() movements. In fact, we save more than that as insertion is done simultaneously with reallocation, hence no shifting occurs additionally to the (amortized) reallocation cost. All this analysis applies symmetrically to left insertions, of course. * The optimum growing policy is that which maximizes the probability that *both* front and back free capacities get exhausted at the same time. With our 1st moment characterization of the insertion random process, this implies that when reallocating a percentage R of the new space should be devoted to the right side, and a percentage L to the left side. Description of resulting policy * Fix the growing factor to G (this can be 1.5 or 2 or anything greater than one, the actual figure is orthogonal to the discussion). * Maintain a counter right_ of right insertions that gets incremented when a right insertion occurs and *decremented* when an erasure is issued past the middle of the devector. * Right insertions trigger right shifting except when there's no back free capacity, in which case a reallocation is performed. * Left insertions trigger left shifting except when there's no front free capacity, in which case a reallocation is performed. * Reallocation reserves space for G*size() elements: of the resulting G*size()-size() free space, a fraction right_/size() is devoted to back free capacity, the rest to front free capacity. In practice, we may want to keep some minimum space at each end as determined by some empirical threshold. And that's it. Comments welcome. Joaquín M López Muñoz

6 37

Including a system link library for every test, but only on mingw-w64 builds?
by James E. King, III 08 Nov '17

08 Nov '17

Is BOOST_AUTO_LINK supposed to work on mingw-w64 builds with gcc? I'm guessing not since I believe this is a MSVC concept, so I am looking for a bjam recipe that will allow me to add the link library bcrypt.lib to all of the tests in a project's test directory if the build is on a windows system and not using msvc toolset. Are these any examples out there? Thanks, Jim

6 9

[type_traits] is_nothrow_move_constructible gives false positive on GCC 4.7
by Andrzej Krzemienski 08 Nov '17

08 Nov '17

Hi, It seems that boost::is_nothrow_move_constructible sometimes gives false positives on GCC 4.7. See the Git issue for details: https://github.com/boostorg/type_traits/issues/56 Regards, &rzej;

1 0

Boost.Uuid and header-only support
by James E. King, III 08 Nov '17

08 Nov '17

Hi folks, I have an open defect related to a random seed generator that's currently in boost::uuids::detail called seed_rng. The class does a fair bit more than it really needs to, and does not support Universal Windows Platform (UWP) store targets. It is otherwise quite similar to boost::random::random_device however it is a header-only implementation (in fact seed_rng has a comment about putting it into Boost.Random). I was planning on removing it and leaving randomness to Boost.Random because that's where the code really belongs, however this would introduce a new library dependency on Boost.Random for any code using boost::uuids::random_generator. So I'm curious on how important folks believe it is to maintain the header-only status of Boost.Uuid. Thanks, Jim

8 48