- Boost - lists.preview.boost.org

Re: [boost] [interprocess] file_mutex class proposal
by mateusz janek 14 Mar '18

14 Mar '18

Hey, any thoughts on that? Thanks, Mateusz 2018-03-04 3:42 GMT-08:00 mateusz janek <stryku2393(a)gmail.com>: > Hello boost::interprocess community, > > During my pet projects development I found that a file access > synchronization is a common problem so, I wrote a class that wraps it and > thought that it can be useful for wider community. I would like to ask you > what do you think about such functionality in boost::interprocess? I'm > aware that class from the repo is not ready to release. Just wanted to show > about what functionality I'm talking about. > > Here you can find the repo: https://github.com/stryku/file_mutex > You can find there the proposed class and a simple usage example. > > If you will find it interesting I will prepare more formal merge proposal, > update docs, maybe more examples etc. > > Looking forward to hearing from you what do you think about it. > > Thank you, > Mateusz Janek > >

2 2

[release] Boost 1.67.0 Beta 1 Release Candidate 1
by Daniel James 14 Mar '18

14 Mar '18

The release candidates for the first 1.67.0 beta release are now available at: <https://dl.bintray.com/boostorg/beta/1.67.0.beta.1/source/> The release notes can be previewed at: <http://www.boost.org/users/history/in_progress.html> The documentation is available at: <http://www.boost.org/doc/libs/1_67_0_beta1/> The SHA256 checksums are as follows: 3d52af11ea6c45c73e18bf9b386056d31dab5092923cc9ed8efa29a083ffb9de boost_1_67_0_b1_rc1.7z 680336bd9a56dccda359061660348c5656cab71cb76713b157ce9a6f840659f0 boost_1_67_0_b1_rc1.tar.bz2 a4608b3ee0b4fadad7b2a2a9efa7c4ad8ce120ee94232e0ee86796f38cced319 boost_1_67_0_b1_rc1.tar.gz 3d1faa96ddda6f0a864f3292c014420457159f0960ff82efcca9feb898d21c1a boost_1_67_0_b1_rc1.zip 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.

11 29

[gsoc18] Boost.Detour & Mentor Search
by Michael Kiros 14 Mar '18

14 Mar '18

Hello, I'm a C++ developer and have been developing in C++ for over 6 years now in my personal time. I wanted to accomplish two things with this post: 1. Propose the idea of boost::detour 2. Find a mentor in the process Boost.Detour is not something new that I'd like to implement but a concept in programming that I have been using for a very long time now and would love to have it easily accessible with the inclusion of a simple <boost/detour.hpp>. The idea is to have function hooking easily accessible to everyone using C++. Though it may seem like a niche feature, function hooking is more commonly used than most people think (e.g. modifying binaries after compilation for feature addition, reverse engineering proprietary software). It's a beautiful concept when you see it working. In the process I need a mentor to guide me through the process of putting down my ideas to paper and implementing them according to the standards that Boost holds. This includes replacing components from my library with Boost concepts (e.g. boost::process). I'm also a student at Texas Tech University studying Computer Science. Can't wait to hear from someone. Michael

2 1

[type_traits] is_list_constructible regression failures
by Lorenzo Caminiti 13 Mar '18

13 Mar '18

Hello all, I noticed the following failure in Boost.TypeTraits regression tests on GCC 4.6 C++11. Any idea why this is failing? http://www.boost.org/development/tests/master/developer/type_traits.html Test output: igaztanaga-master-gcc-4.6c++11 - type_traits - cxx14_aliases_test / gcc-4.6c+ Rev cb62bbf20f4a0000ee6f16835b0c2dc92edea8cd / Sun, 11 Mar 2018 03:16:04 +0000 Compile [2018-03-11 10:15:18 UTC]: fail "C:/Programs/mingw-builds/x32-4.6.4-posix-dwarf-rev0/mingw32/bin/gcc-4.6.bat" -pipe -fmax-errors=1 -std=c++0x -m32 -O0 -fno-inline -Wall -pedantic -Werror -march=i686 -Wextra -Wno-bool-operation -Wno-int-in-bool-context -Wno-uninitialized -DBOOST_ALL_NO_LIB=1 -I".." -I"..\libs\type_traits\test\libs\tt2\light\include" -c -o "c:\boost\master\results\boost\bin.v2\libs\type_traits\test\cxx14_aliases_test.test\gcc-4.6c+\dbg\dbg-symbl-off\cxx14_aliases_test.o" "..\libs\type_traits\test\cxx14_aliases_test.cpp" In file included from ../boost/type_traits.hpp:110:0, from ..\libs\type_traits\test\cxx14_aliases_test.cpp:10: ../boost/type_traits/is_list_constructible.hpp:36:61: error: expected class-name before 'decltype' compilation terminated due to -fmax-errors=1. I am asking because a couple of Boost.Contract tests that #include <boost/type_traits.hpp> fail with the same Boost.TypeTraits error on this compiler. Thanks. --Lorenzo

3 4

Re: [boost] [gsoc18]Boost.SafeFloat
by Rajat Miglani 13 Mar '18

13 Mar '18

Is there any specific template that I should follow for the proposal or I can make it my own way? Thanks, Rajat Miglani On Tue, Feb 27, 2018 at 12:43 AM, Damian Vicino <damian(a)vicino.com.ar> wrote: > Hi, > From the top of my head, the big missing pieces are the safefloat > literals, some pieves of unit testing (copile and runtime), documentation, > implementing some of the policies that are already defined. Maybe a better > way to specify policies could be implemented. > However, the complete list of things being done is what your proposal is > about, you need to define the milestones and goals for the project. > > 2018-02-25 18:57 GMT+00:00 Rajat Miglani <rajatmiglani26(a)gmail.com>: > >> I have read the paper "Goldberg, David. "What every computer scientist >> should know about floating-point arithmetic." ACM Computing Surveys (CSUR) >> 23.1 (1991): 5-48."and documentation and now having a little >> understanding of what this project is about.Presently going through the >> code and concepts, please guide me of what changes I have to do in the >> previous code to make it better. >> >> On Wed, Feb 21, 2018 at 11:52 PM, Damian Vicino <damian(a)vicino.com.ar> >> wrote: >> >>> Hi Rajat, >>> For the Boost.SafeFloat project, I suggest you take a look to the >>> following links: >>> - Repository: https://github.com/sdavtaker/safefloat >>> - Documentation: https://sdavtaker.github.io/safefloat/doc/htm >>> l/index.html >>> >>> Also, try to read this paper if you didn't before: Goldberg, David. >>> "What every computer scientist should know about floating-point >>> arithmetic." ACM Computing Surveys (CSUR) 23.1 (1991): 5-48. >>> From the C++ perspective, key concepts you need to know are: >>> - Templates (a lot of it) >>> - Custom Literals >>> >>> A good proposal would be to take a read to the documentation and code >>> base and describe what is missing to complete the project. >>> Best regards, >>> Damian Vicino >>> >>> >>> >>> >>> 2018-02-21 14:11 GMT-03:00 Rajat Miglani via Boost < >>> boost(a)lists.boost.org>: >>> >>>> Dear boost community, >>>> I am a GSOC aspirant and I'd like to contribute to your organization. >>>> I have read the project ideas and is interested in working on:- >>>> 1)Boost.SafeFloat or >>>> 2)Enhancing Boost.Intrusive Library >>>> Can I get some more info about how to work on these projects and the >>>> basic >>>> requirements so that I can learn them and be able to contribute to my >>>> fullest. >>>> >>>> Thank you, >>>> >>>> _______________________________________________ >>>> Unsubscribe & other changes: http://lists.boost.org/mailman >>>> /listinfo.cgi/boost >>>> >>> >>> >> >

2 1

[gsoc2018]Guidance for Google Summer of code on Project1
by sumit_kumar 12 Mar '18

12 Mar '18

I am a third year undergraduate student. I have a good implementation skills in C++ . I know machine learning algorithms & math required for it very well. So ,I would like to contribute to the Boost.uBLAS by doing project1. Could you please guide me. -- Sent from: http://boost.2283326.n4.nabble.com/Boost-Dev-f2600599.html

2 1

Re: [boost] [gsoc18] Mentor Request for uBLAS (prefer support for tensors)
by David Bellot 11 Mar '18

11 Mar '18

Hi Karl, you're exactly the kind of people I want to see in GSOC: student with passion for C++ and some already implemented library. Your competency test is good. Best now is to work on the proposal and send it to us directly (no need to go through the main mailing-list from now on). Cheers, David On Wed, Mar 7, 2018 at 9:27 PM, Karl Chen <karl.chen(a)mail.utoronto.ca> wrote: > Hi, > > > Thanks for replying! Here is a link to my competency test: > https://github.com/HiImKarl/GSOC18-uBLAS. > > The link in my first email was to a different library, more of a > basic proof of concept than anything. > > > Cheers! > > Karl > ------------------------------ > *From:* mohammad sharique <1sharique1(a)gmail.com> > *Sent:* Wednesday, March 7, 2018 12:06:05 PM > *To:* boost(a)lists.boost.org; Karl Chen > *Cc:* David Bellot > *Subject:* Re: [boost] [gsoc18] Mentor Request for uBLAS (prefer support > for tensors) > > Hi Karl, > David and I are mentoring this project. Thanks for sending in the source. > We'll have a look at it, and get back to you soon. > You may start writing proposal in the meanwhile. The proposal usually > involves details about what you want to achieve > with the project, the resource required, approach(s) for the same, time division, > class designs and interactions, implementation > details etc. > > Should there be any questions, please feel free to ask. > > Regards, > Sharique > > On Wed, Mar 7, 2018 at 2:36 PM, HiImKarl via Boost <boost(a)lists.boost.org> > wrote: > > Hi, > > I am Karl, a financial engineering student that is very passionate about > C++. I think my background in applied math and stats would make me a great > candidate for writing a tensor library. I would also like to point to this > header-only multi-dimensional array template library with compile time > tensor rank that I've written in the past: > https://github.com/HiImKarl/Tensors, which may be good groundwork. > > It is a little late into March, so I was wondering if there were any > mentors > still looking for students for this project. I have already completed the > competency test. > > > > -- > Sent from: http://boost.2283326.n4.nabble.com/Boost-Dev-f2600599.html > > _______________________________________________ > Unsubscribe & other changes: http://lists.boost.org/mailman > /listinfo.cgi/boost > > >

1 0

[gsoc18] uBLAS competency test
by Arun Ramachandran 11 Mar '18

11 Mar '18

Hello, I'm Arun, a final year computer science undergrad. I'm interested in working on project 1 of the uBLAS library (adding statistics and machine learning functions). I have experience with developing in C++ from working on Code Character (https://github.com/delta/codecharacter-simulator) an online AI programming challenge that some fellow students and I hosted. Since this is an application, I've taken the competency test for uBLAS and have implemented a small matrix library using expression templates along with some unit tests and benchmarks. The repo is currently private, so if a mentor shares their GitHub handle I can add them as a collaborator for reviewing it. Thank you, Arun Ramachandran

3 3

[gsoc18] Request for Guidance
by Mohd Faisal 11 Mar '18

11 Mar '18

Hello, my name is Mohd Faisal. I want to work on Boost.uBLAS: linear algebra and matrix computations. Is there a specific reason that only the listed algorithms have to be implemented. Because there are many good ML algorithms that can also be implemented like SVM.

3 2

[gsoc18] how to share a competency test
by vardhan gupta 11 Mar '18

11 Mar '18

Hi, My name is Vardhan Gupta. I am a 2nd year undergraduate student from IIT Kanpur. I am interested in working on project #3 of Boost.uBLAS on adding multicore and gpu computations to uBLAS. I have created a private repository of it's competency test on gitlab. Now, I am having trouble in sharing the repository with the mentors since I do not have their email address. how do I share it/find their e-mail address? Thank you, Yours sincerely; Vardhan Gupta

6 10

[gsoc2018]Mentor Required
by P.H S 11 Mar '18

11 Mar '18

Hello, my name is Zhixuan Xiao, a student from BUPT, BeiJing China. I am interested to Boost.uBLAS:project 3, Add Multicore and GPU computations to uBLAS and I want to connect with a mentor. Thanks, Zhixuan Xiao

2 1

Google Summer of Code: Potential Applicant
by Christos Maris 10 Mar '18

10 Mar '18

Hey there, My name is Christos (Chris for short) and I would love to work with you via the GSoC program this summer. I am sending this mail because I am in a need for some instructions on how to find a mentor and a project (at first). Can anyone help me with that? Also is there any potential mentor here? Any info helps. Thanks a lot in advance!

2 1

[gsoc18] Regarding Boost uBLAS Google Summer of Code 2018
by Varun Mehrishi 09 Mar '18

09 Mar '18

Hi, I am Varun Mehrishi, a third year engineering student from Manipal University Jaipur (MUJ), India. I am currently pursuing Bachelors of Technology in Computer Science and Engineering. I am most comfortable with C++(STL) and Python, and also follow the new and experimental features introduced in the new standards with additional help of cppcon and BoostCon. I also have interests in Data Science, Machine Learning. I wish to participate and contribute to Boost.uBLAS. I've read the project idea and description from the suggested GSoC project proposals. I am really interested in working on the project "Project 1 : Add statistics and machine learning functions" which is focused on extending Boost.uBLAS functionality towards statistics/ML. I have implemented a basic template version of the mentioned Programming Competency Test (basic-matrix-lib <https://github.com/varunmehrishi/basic-matrix-lib>) and used C++14 generic lambdas wherever necessary. Your feedback and suggestions will be extremely valuable as they will help me improve my implementation and add more functionality. Looking forward to your guidance -- With warm regards, Varun Mehrishi GSoC 2018 Aspirant

1 0

Re: [boost] How to set LDFLAGS in 1.66.0?
by Nana Sakisaka 09 Mar '18

09 Mar '18

> I'm not looking to find an external library, I'm trying to set the RUNPATH on the boost shared libraries There's no need to use an external library. You just pass options to b2. Are you sure that you've tried my suggestion on previous reply? > Ordinarilly, one just adds LDFLAGS to the configure script with the -Wl,rpath ... Using boost libraries does not prevent you to do so. One could either specify RPATH for your boost library files, application executable, or both. BTW, looking at your pasted output of ldd indicates that you've set your RPATH for wrong value, but that should not be boost specific; try googling without 'boost' keyword? 2018-03-09 9:59 GMT+09:00 Dean Anderson <dean(a)av8.com>: > Thanks for the reply. > > I'm not looking to find an external library, I'm trying to set the RUNPATH > on the boost shared libraries. For example: > > The path on libstdc++ is set by the ld, so it finds the right compiler > runtime... > > libboost_wave.so needs a RUNPATH to where its going to be installed. > > Ordinarilly, one just adds LDFLAGS to the configure script with the > -Wl,rpath ... > > --Dean > > ldd libboost_wave.so.1.66.0 > linux-vdso.so.1 => (0x00007ffebf1b6000) > libboost_filesystem.so.1.66.0 => not found > libboost_thread.so.1.66.0 => not found > libboost_date_time.so.1.66.0 => not found > libboost_chrono.so.1.66.0 => not found > libboost_system.so.1.66.0 => not found > librt.so.1 => /lib64/librt.so.1 (0x00007f5082f0a000) > libstdc++.so.6 => /efs/dist/gnu/gcclib/4.9.4/.ex > ec/x86-64.rhel.7/lib/libstdc++.so.6 (0x00007f5082c00000) > libm.so.6 => /lib64/libm.so.6 (0x00007f50828fd000) > libgcc_s.so.1 => /efs/dist/gnu/gcclib/4.9.4/.ex > ec/x86-64.rhel.7/lib/libgcc_s.so.1 (0x00007f50826e6000) > libpthread.so.0 => /lib64/libpthread.so.0 (0x00007f50824ca000) > libc.so.6 => /lib64/libc.so.6 (0x00007f5082108000) > /lib64/ld-linux-x86-64.so.2 (0x00007f508343e000) > > > > On Fri, 9 Mar 2018, Nana Sakisaka via Boost wrote: > > You could try passing this option to b2: >> >> dll-path="'$ORIGIN/../lib'" >> >> Make sure to escape $ and " appropriately for your CI scripts. >> Note that there are actually two types of flags, i.e. >> (1) the flag for your application executable (the RPATH) >> and >> (2) a flag for your build system's sanity check (e.g. the "checking if >> *whatever* works..." internal binary flag for ./configure) >> >> p.s. Invoke command below to check whether the flags are set: >> objdump -p ./a.out|grep RPATH >> >> >> >> 2018-03-09 8:33 GMT+09:00 Dean Anderson via Boost <boost(a)lists.boost.org >> >: >> >> I need to set an rpath on the shared libraries. Any ideas how to do that? >>> >>> Alternatively, is there a fork of boost that use the gnu build system? >>> >>> Thanks! >>> >>> --Dean >>> >>> Av8 Internet Prepared to pay a premium for better service? >>> www.av8.net faster, more reliable, better service >>> 617 256 5494 >>> >>> >>> _______________________________________________ >>> Unsubscribe & other changes: http://lists.boost.org/mailman >>> /listinfo.cgi/boost >>> >>> >> _______________________________________________ >> Unsubscribe & other changes: http://lists.boost.org/mailman >> /listinfo.cgi/boost >> >>

1 0

How to set LDFLAGS in 1.66.0?
by Dean Anderson 09 Mar '18

09 Mar '18

I need to set an rpath on the shared libraries. Any ideas how to do that? Alternatively, is there a fork of boost that use the gnu build system? Thanks! --Dean Av8 Internet Prepared to pay a premium for better service? www.av8.net faster, more reliable, better service 617 256 5494

2 1

[foreach][polygon][xpressive][bimap][ptr_container][gil][assign][typeof][msm] PRs for removing executable attribute for files
by Edward Diener 08 Mar '18

08 Mar '18

The following libraries still have PRs, which I have previously submitted, for removing the executable attribute for one or more files: foreach polygon xpressive bimap ptr_container gil assign typeof msm None of these PRs affect anything other than the file attribute, so no testing is needed AFAICS since no code changes. If the maintainers of the above libraries, or someone else with write access to these libraries, can merge the appropriate PR to 'develop', and then merge 'develop' to 'master', we can get rid of having any files with their executable attribute set in the next release.

5 13

[gsoc18] Boost.Intrusive mentor request
by BogdanCodreanu 08 Mar '18

08 Mar '18

Hello, I've read about the Boost.Intrusive library and it matches my desires of programming. Is there a mentor available for this project? How should I prove myself worthy of this project? -- Sent from: http://boost.2283326.n4.nabble.com/Boost-Dev-f2600599.html

2 1

Re: [boost] [system] [teeks99] Mass test failures on MSVC-14.0
by Andrey Semashev 08 Mar '18

08 Mar '18

On 02/26/18 17:20, Tom Kent wrote: > On Mon, Feb 26, 2018 at 6:56 AM, Andrey Semashev via Boost > <boost(a)lists.boost.org <mailto:boost@lists.boost.org>> wrote: > > On 02/26/18 15:43, Tom Kent via Boost wrote: > > On Sat, Feb 24, 2018 at 3:45 PM, Andrey Semashev via Boost < > boost(a)lists.boost.org <mailto:boost@lists.boost.org>> wrote: > > Hi, > > Most of Boost.System tests are failing on the > teeks99-09-f-win2016-64on64 > tester, MSVC-14.0[1]. The test logs don't show any errors > and I can't > reproduce it locally. I'm seeing similar test failures in > other libraries, > like Boost.Atomic for example, and the links point to > Boost.System. > > Is it possible to get the error log from the tester? > > >> > > Here's all 82MB of it: > > http://boost.teeks99.com/temp/teeks99-09-f-results-bjam-2018-02-26.log > <http://boost.teeks99.com/temp/teeks99-09-f-results-bjam-2018-02-26.log> > > > > But be warned, I just moved the -09 runner onto a new (Win 2016) VM about a > > week ago, and there have been some teething issues. If all goes well, the > > -f run should hit again in the next 24-48hrs. > > The link returns 403: Forbidden. > > Try again now. Sorry about that. I got the log now, thanks. The relevant part seems to be this: msvc.link.dll D:\teeks99-09\run\results\boost\bin.v2\libs\system\build\msvc-14.0\dbg\adrs-mdl-64\thrd-mlt\boost_system-vc140-mt-gd-x64-1_67.dll call "D:\teeks99-09\run\results\boost\bin.v2\standalone\msvc\msvc-14.0\adrs-mdl-64\archt-x86\msvc-setup.bat" amd64 >nul link /NOLOGO /INCREMENTAL:NO /DLL /DEBUG /MACHINE:X64 /MANIFEST /subsystem:console /out:"D:\teeks99-09\run\results\boost\bin.v2\libs\system\build\msvc-14.0\dbg\adrs-mdl-64\thrd-mlt\boost_system-vc140-mt-gd-x64-1_67.dll" /IMPLIB:"D:\teeks99-09\run\results\boost\bin.v2\libs\system\build\msvc-14.0\dbg\adrs-mdl-64\thrd-mlt\boost_system-vc140-mt-gd-x64-1_67.lib" @"D:\teeks99-09\run\results\boost\bin.v2\libs\system\build\msvc-14.0\dbg\adrs-mdl-64\thrd-mlt\boost_system-vc140-mt-gd-x64-1_67.dll.rsp" if %ERRORLEVEL% NEQ 0 EXIT %ERRORLEVEL% Creating library D:\teeks99-09\run\results\boost\bin.v2\libs\system\build\msvc-14.0\dbg\adrs-mdl-64\thrd-mlt\boost_system-vc140-mt-gd-x64-1_67.lib and object D:\teeks99-09\run\results\boost\bin.v2\libs\system\build\msvc-14.0\dbg\adrs-mdl-64\thrd-mlt\boost_system-vc140-mt-gd-x64-1_67.exp msvc.manifest.dll D:\teeks99-09\run\results\boost\bin.v2\libs\system\build\msvc-14.0\dbg\adrs-mdl-64\thrd-mlt\boost_system-vc140-mt-gd-x64-1_67.dll if exist "D:\teeks99-09\run\results\boost\bin.v2\libs\system\build\msvc-14.0\dbg\adrs-mdl-64\thrd-mlt\boost_system-vc140-mt-gd-x64-1_67.dll.manifest" ( call "D:\teeks99-09\run\results\boost\bin.v2\standalone\msvc\msvc-14.0\adrs-mdl-64\archt-x86\msvc-setup.bat" amd64 >nul mt -nologo -manifest "D:\teeks99-09\run\results\boost\bin.v2\libs\system\build\msvc-14.0\dbg\adrs-mdl-64\thrd-mlt\boost_system-vc140-mt-gd-x64-1_67.dll.manifest" "-outputresource:D:\teeks99-09\run\results\boost\bin.v2\libs\system\build\msvc-14.0\dbg\adrs-mdl-64\thrd-mlt\boost_system-vc140-mt-gd-x64-1_67.dll;2" ) 'mt' is not recognized as an internal or external command, operable program or batch file. ...failed msvc.manifest.dll D:\teeks99-09\run\results\boost\bin.v2\libs\system\build\msvc-14.0\dbg\adrs-mdl-64\thrd-mlt\boost_system-vc140-mt-gd-x64-1_67.dll D:\teeks99-09\run\results\boost\bin.v2\libs\system\build\msvc-14.0\dbg\adrs-mdl-64\thrd-mlt\boost_system-vc140-mt-gd-x64-1_67.lib D:\teeks99-09\run\results\boost\bin.v2\libs\system\build\msvc-14.0\dbg\adrs-mdl-64\thrd-mlt\boost_system-vc140-mt-gd-x64-1_67.pdb D:\teeks99-09\run\results\boost\bin.v2\libs\system\build\msvc-14.0\dbg\adrs-mdl-64\thrd-mlt\boost_system-vc140-mt-gd-x64-1_67.pdb D:\teeks99-09\run\results\boost\bin.v2\libs\system\build\msvc-14.0\dbg\adrs-mdl-64\thrd-mlt\boost_system-vc140-mt-gd-x64-1_67.pdb D:\teeks99-09\run\results\boost\bin.v2\libs\system\build\msvc-14.0\dbg\adrs-mdl-64\thrd-mlt\boost_system-vc140-mt-gd-x64-1_67.pdb D:\teeks99-09\run\results\boost\bin.v2\libs\system\build\msvc-14.0\dbg\adrs-mdl-64\thrd-mlt\boost_system-vc140-mt-gd-x64-1_67.pdb... ...removing D:\teeks99-09\run\results\boost\bin.v2\libs\system\build\msvc-14.0\dbg\adrs-mdl-64\thrd-mlt\boost_system-vc140-mt-gd-x64-1_67.dll ...removing D:\teeks99-09\run\results\boost\bin.v2\libs\system\build\msvc-14.0\dbg\adrs-mdl-64\thrd-mlt\boost_system-vc140-mt-gd-x64-1_67.lib ...removing D:\teeks99-09\run\results\boost\bin.v2\libs\system\build\msvc-14.0\dbg\adrs-mdl-64\thrd-mlt\boost_system-vc140-mt-gd-x64-1_67.pdb I think, this problem has been discussed on the list not long ago - `mt` being not found on the system after some MSVC update: http://boost.2283326.n4.nabble.com/build-msvc-14-linker-step-failing-tc4701… Looks like MSVC needs to be reinstalled or installed on a separate machine. Will you be able to fix this?

3 7

[release] 1.67.0 closed for Beta
by Daniel James 07 Mar '18

07 Mar '18

The master branch is now closed for all changes except by permission for the Beta release. Also, remember to add your libraries' changes to the release notes at: <https://github.com/boostorg/website/blob/master/feed/history/boost_1_67_0.q…> You can preview the release notes at: <http://www.boost.org/users/history/in_progress.html> Thanks, -- The Boost Release Team

1 0

About authorizing CircleCI
by Vicente J. Botet Escriba 07 Mar '18

07 Mar '18

Hi, recently I heard of CircleCI (https://circleci.com/) and it seems much faster than Travis. Is someone using it? What is your feedback? If the feedback is positive, could someone authorize CircleCI to access boostorg? Best, Vicente

6 11

[gsoc18] Mentor Request for uBLAS (prefer support for tensors)
by HiImKarl 07 Mar '18

07 Mar '18

Hi, I am Karl, a financial engineering student that is very passionate about C++. I think my background in applied math and stats would make me a great candidate for writing a tensor library. I would also like to point to this header-only multi-dimensional array template library with compile time tensor rank that I've written in the past: https://github.com/HiImKarl/Tensors, which may be good groundwork. It is a little late into March, so I was wondering if there were any mentors still looking for students for this project. I have already completed the competency test. -- Sent from: http://boost.2283326.n4.nabble.com/Boost-Dev-f2600599.html

2 1

[gsoc2018] Project 1.1: Add statistics and machine learning functions
by Dattatreya Mohapatra 07 Mar '18

07 Mar '18

Hi, My name is Dattatreya Mohapatra. I am an undergrad at IIIT Delhi, majoring in CS. I am interested in working on extending the Boost.uBLAS library as a project. I have taken courses on Linear Algebra and Machine Learning in the past, so I have a decent basic understanding of the subject. I have used C++11 in several of my course projects, so I have some familiarity with advanced features like lambdas and template programming. I have gone through the earlier threads on the same, and have started working on the core-competency test (CCT). Can the mentors of this project please guide me as to how should I get the CCT reviewed when its ready? Also, is there anything else I should look at apart from the documentation which would be helpful? Thanks and regards, Dattatreya Mohapatra

2 2

VS 2017 15.6 released, _MSC_VER=1913...
by Peter Dimov 07 Mar '18

07 Mar '18

... and Boost.Config once again warns: 1>c:\boost-git\develop\boost\config\compiler\visualc.hpp(349): note: Info: Boost.Config is older than your compiler version - probably nothing bad will happen - but you may wish to look for an updated Boost version. Define BOOST_CONFIG_SUPPRESS_OUTDATED_MESSAGE to suppress this message. Maybe we should display this only on 1920 and above, as I suggested last time? :-)

5 5

[gsoc18] [Boost.Intrusive] Mini version of project
by Rishija Mangla 07 Mar '18

07 Mar '18

Hello, My name is Rishija Mangla. I’m a mathematics graduate and pursuing my MCA (2nd year) from University of Delhi, India. I want to contribute in enhancing Boost.Intrusive library and have implemented a mini library for the same ( as asked for competency test ). Library includes: - Generic segment tree ( support range queries where operation is given by the user over any general class ) - Expression templates for lazy evaluation - Written test programs to check the accuracy ( using Boost.test unit test framework ) - Test cases are generated randomly Link to library implemented https://github.com/Rishija/boost_competency_test <https://github.com/Rishija/boost_competency_test> I plan to implement more complex data structures that have been mentioned in the list. I have already implemented prefix tree, both dynamic and via filesystem ( will provide link to the project in case you want to have a look ). So I’ll be pretty comfortable in developing other similar data structures and algorithms for them. Thanks

1 0

[Boost.Yap] Review results
by Louis Dionne 06 Mar '18

06 Mar '18

Dear Boost, I am pleased to announce that the Yap library is CONDITIONALLY ACCEPTED into Boost. I counted the following formal reviews: - Alex Hagen-Zanker: REJECT / CONDITIONAL ACCEPT - Barrett Adair: ACCEPT - Brook Milligan: CONDITIONAL ACCEPT - John Fletcher: ACCEPT - Paul Fultz II: ACCEPT - Steven Watanabe: ACCEPT I'd like to thank everyone who participated to the review for their feedback, and also Zach Laine for his submission. The following is a summary of the main issues that were raised during the review: - Contextual evaluation of terminals The concern was that there was no way of evaluating an expression tree with a custom context other than to first transform it with a custom transform, and then evaluate it. This, however, requires encoding the recursion into the transform even when we only want to modify terminals. After a lot of discussion, everybody seemed to agree that the following would be desirable (paraphrasing Steven): `transform_evaluate` would behave exactly like `transform`, except that the default behavior for nodes that are not handled by the context is to evaluate the operators instead of building a new expression. Zach fixed that in https://github.com/tzlaine/yap/issues/63 by introducing a new transform called `evaluation()` which evaluates terminals, and the ability to pass multiple transforms `t1`, ..., `tN` to `transform`. The first matching transform is executed, and `transform` will copy-and-recurse if no transforms match. `transform_evaluate(expr, tform)` can then be implemented as `yap::transform(expr, tform, yap::evaluation())`. - Auto-unwrapping of terminals in transform and automatic application of terminal transforms before unwrapping There was a concern that automatically unwrapping terminals in transforms and applying terminal transforms before unwrapping would result in counter intuitive behavior. After some discussion, it was determined that automatic application of transforms before unwrapping could indeed lead to surprising results and was removed. However, unwrapping of terminals is necessary to keep the matching of terminals reasonable, so it was kept. I am comfortable with that decision, as the unwrapping of terminals, which leads to a very intuitive way of matching terminals, was one of the features that sprung to me as being a great usability boost when I first looked at the library. - Positional placeholders should stay or go? Some of the reviewers felt like it was not desirable to provide positional placeholders as part of Yap, because those aren't the primary purpose of the library and C++14 obsoletes them with lambdas. Other reviewers felt like they were in fact very useful and shouldn't be removed. Since no clear consensus could be reached, no acceptance condition is attached to this issue. However, I personally favor the removal of placeholders from the core API on the basis of minimality. I would like to encourage Zach to check whether their removal would result in API simplifications; if so, I would encourage (but not require) their removal, at Zach's discretion. - Documentation The documentation is not sufficiently extensive, and takes too much for granted from the reader. This was raised by a couple of reviewers and it has been my largest problem with the library so far. More on this below. - The eval_xxx customization points. Nobody liked them, and they are already gone. - BOOST_YAP_CONVERSION_OPERATOR_TEMPLATE Nobody liked it, and it is already gone. - Short-circuiting of && and || in evaluate Steven brought up the fact that `yap::evaluate` would not short-circuit the evaluation of && and ||. This issue was mooted when customization points were removed. Also, the new `evaluation()` transform, which could suffer from the same short-circuiting problems, doesn't because it was implemented properly. - Support for Boost.Build is missing. - Hana vs not Hana Full disclosure: I am the author of Boost.Hana. A couple of reviewers raised concerns with Hana being part of the public interface. The main concern seemed to be about MSVC not being able to compile Hana, and hence Yap being unusable on MSVC. While I agree that MSVC is a big player, I believe that should not dictate what all Boost libraries do. Some libraries are meant to bridge gaps with older compilers, some libraries are cutting edge. Yap is the latter, and that's fine. The MSVC folks are working hard on their C++ conformance, and they will eventually get there. Also, it is likely that Hana wouldn't be the barrier anyway: Yap relies a lot on SFINAE and MSVC is known to be poor at it. Many of the features that one would likely want to use when using Yap are also not too well supported by MSVC, so users would probably hit limitations there anyway. Plus, as soon as MSVC is close enough, I'll start working on Hana's side to support it. Taking my review manager hat off for a second, I'd also like to point out that Yap is exactly the kind of use case that Hana was built for. Just like MPL was (and is) used in many libraries' implementation and APIs, I don't see a problem with using Hana in Yap's API, since that is the right tool for the job. That being said, if Zach wishes to explore using Hana's concepts (like hana::Sequence) to conceptify his API, I encourage him to do so. This may make it possible to use any tuple-like sequence for the children of an expression, thereby reducing the coupling between the two libraries. If Zach wants to explore using Fusion instead of Hana, that is also an option that would make MSVC support achievable. - A single macro for symmetric binary operators It was brought up that there's no macro to define both `arg <op> expr` and `expr <op> arg` at once. I believe this was initially done to avoid potential conflicts with the member variants, which already provide `expr <op> arg`. I believe it should still be possible to do it, since, as Steven said, most of the time a user will need symmetric behavior anyway. - External dependencies checked-into the repository The library includes internal copies of some of its dependencies, which is unusual for a Boost library. Zach said it wasn't his intent to put the library into Boost that way if it were accepted. To see all the issues that were raised during the review, you can search for tickets labeled with `Boost Review` on GitHub: https://github.com/tzlaine/yap/issues?q=label%3A%22Boost+Review%22 Based on the concerns above, the following lists the conditions for acceptance: - Documentation In my opinion, this was the biggest concern to be addressed. A lot of progress was made during and after the review, but I still think the documentation can be improved and made more complete. I especially believe that integrating more examples inside the "tutorial" (not just at the end) and introducing new concepts more gradually would help readers that are not already expression template gurus. It is very difficult to set a clear action item for "improving the documentation". For this reason, Brook Milligan and I will work with Zach to define what it means for the documentation to meet this acceptance condition. Brook was kind enough to agree to this when I asked him in a private communication. - A single macro for symmetric binary operators The library must provide a way to define symmetric binary operators in one go. Exactly how this can be achieved and whether we should be able to SFINAE-constrain the arguments is left to Zach's discretion. - External dependencies The external dependencies must not be packaged with the library itself. Another mechanism should be used; I leave that mechanism unspecified, but it should be possible to test the library using the usual Boost.Build workflow (e.g. an ExternalPackage download that would only work in CMake is not acceptable). - Boost.Build Yap needs to support Boost.Build like all the other Boost libraries. I will post an update to this list when the conditions of the review are met. If they wish to do so, reviewers can then verify that the conditions are met in a satisfactory fashion and the library will automatically be accepted unless someone objects. I would like to thank everyone who participated in the review, and congratulations to Zach for his library. Regards, Louis Dionne

5 7