April 2014 - Boost - lists.preview.boost.org

Re: [boost] [serialization] proposed improvements: forward-compatibility of serialization
by Brian Wood 18 Apr '14

18 Apr '14

On Fri, Apr 18, 2014 at 10:52 AM, Robert Ramey < robertmacleranramey(a)gmail.com> wrote: > > > On Friday, April 18, 2014 6:40:32 AM UTC-7, Brian Wood wrote: >> >> niXman writes: >> >> > Hi, >> > >> > YAS(https://github.com/niXman/yas) ? >> >> The test results on your page are impressive, but there's >> not much info on what or how you tested. Please give >> some more info on that. >> > > Looks like the YAS version uses a special mem_ostream/mem_istream classes > rather than std::strstream. Since the largest part of any of these > benchmarks is spent on the i/o part this would account for a huge > difference. > > >> >> Also it looks to me like the results you list for text >> serialization are off by a factor of ten. >> > > This is common when the benchmark is compiled in debug mode. The boost > serialization library has a huge amount of extra checking in debug mode > which generally slows things down by a factor of 10. I don't know if > that's the case here though. > > He has this table: | boost | YAS | +------+---------+---------+---------+ | save | 8090 ms | 277 ms | +28000% | +------+---------+---------+---------+ | load | 8794 ms | 265 ms | +33000% | +------------------------------------+ I think the last column is off. It should be +2800% and +3300%. Personally I wouldn't use the percents and would use 28 and 33. -- Brian Ebenezer Enterprises - Heavenly code. http://webEbenezer.net

1 0

Re: [boost] [serialization] proposed improvements: forward-compatibility of serialization
by Brian Wood 18 Apr '14

18 Apr '14

niXman writes: > Hi, > > YAS(https://github.com/niXman/yas) ? The test results on your page are impressive, but there's not much info on what or how you tested. Please give some more info on that. Also it looks to me like the results you list for text serialization are off by a factor of ten. -- Brian Ebenezer Enterprises - So far G-d has helped us. http://webEbenezer.net

1 0

[phoenix] Test failures on clang
by Andrey Semashev 18 Apr '14

18 Apr '14

Hi, I'm seeing Boost.Log test failures on clang. The errors come from Boost.Phoenix/Boost.Proto: http://www.boost.org/development/tests/develop/developer/output/marshall-ma… It seems like actor assignment operator got broken recently.

2 1

Benchmarks comparing JSON libraries - including JSon Spirit
by Klaim - Joël Lamotte 17 Apr '14

17 Apr '14

http://blog.thousandeyes.com/efficiency-comparison-c-json-libraries/ I think this benchmark (with sources) might interest people who were recently planning to make an official Boost library for JSON parsing. In particular, this benchmark include JSon Spirit which is based on Boost Spirit. These benchmark might be useful for future tests.

4 5

Help with a shared_ptr problem on Android (ARM)
by Peter Dimov 17 Apr '14

17 Apr '14

https://svn.boost.org/trac/boost/ticket/9912 reports that spinlock_gcc_arm.hpp segfaults on a Nexus 7. Unfortunately, I know next to nothing about ARM or Android, so I wonder if someone can help with that.

3 4

Application : Question about Review/Review Manager
by Renato Forti 17 Apr '14

17 Apr '14

Hi All, The proposed Boost.Application is on review schedule. http://www.boost.org/community/review_schedule.html I have one question! How get a Review Manager? Anyone can help in this process? How this work? Thanks

2 1

[algorithm] Split that appends
by Sebastian Karlsson 17 Apr '14

17 Apr '14

First of all I think it's a bit unintuitive that boost::algorithm::split swaps with target, I would've assumed it appended. The fact that it in effect clears target makes it cumbersome to incrementally split into a destination, as you need a new temporary for every call to split, and then explicitly transfer everything over to your final destination. The docs do mention the behavior as a note ( which I presume is precisely because it's a bit unintuitive ). My question then naturally becomes why does split behave like this? As changing the behavior would be a breaking change, do you think it would be unreasonable to have a split_append? Kind regards, Sebastian Karlsson

5 5

[property_tree] xml_parser::read_xml() and write_xml() are limited to std::string keys
by Alexander Bogdashevsky 17 Apr '14

17 Apr '14

boost::property_tree::xml_parser and json_parser are hardcoded to work only with keys of std::string type. The following code will fail to compile: boost::property_tree::basic_ptree<boost::container::string, boost::container::string> pt; std::string filename("123.xml"); boost::property_tree:xml_parser::read_xml(filename, pt); I have prepared a patch that removes this limitation. Please consider pulling https://github.com/sanderbog/property_tree/commit/8af8b6bf3b65fa59792d849b5… Kind Regards, Alexander Bogdashevsky

1 0

[math] & [random] distribution object compatibility
by Thijs van den Berg 16 Apr '14

16 Apr '14

Boost [random] has probability distribution objects for drawing random samples of those distributions. Boost [math] also has probability distributions, those provide free functions for computing properties of the distributions, like mean, pdf etc. I like the free function design of math variant e.g. it uses pdf(distribution,x) and I was wondering why this hasn’t been adopt in C++11 for <random>? To me a free function syntax like random(distribution,engine) would make sense. It would align with the fact that C++11 has added begin(container) and end(container) free functions. I think it’s a bit inconvenient to have two libs that both contain probability distributions. Especially since I’m hoping that boost/math would some day en up in the standard? One way we could integrate the two libs is by adding a random(distribution,engine) free function to the math lib which would accept both distribution objects from the random lib and the math lib. This however could stimulate users in writing less C++11 portable code. My questions are: * Why wasn’t the free function syntax adopted in C++11? Perhaps because of performance options? * Is there an effort to integrate the distributions in math and random within boost? * Are there concrete plans to add elements of math to the C++ standard? We already have <random> now, I would love more.

3 5

clang++ isn't getting the BOOST_NO_CXX11_HDR_FUNCTIONAL flag that it needs
by Chris Cooper 16 Apr '14

16 Apr '14

Trying to build & run the regression tests for boost 1.55 on my Mac using the clang++ that comes with Xcode 5.1 ("Apple LLVM version 5.1 (clang-503.0.38) (based on LLVM 3.4svn)”), it can’t build config_test.cpp because BOOST_NO_CXX11_HDR_FUNCTIONAL isn’t set, and libs/config/test/boost_no_ccx_hdr_functional.ipp definitely doesn’t build (after renaming it to .cpp of course). If I set CXXFLAGS to "-std=c++11 -stdlib=libc++” everything is fine, but I also need a build of boost that _doesn’t_ use c++11, and it’s the non-c++-11 version that is failing. I can get around this by manually setting CXXFLAGS to “-DBOOST_NO_CXX11_HDR_FUNCTIONAL” but it seems like the correct way to do this is to add a test to boost/config/compiler/clang.hpp that tests … something… and then sets BOOST_NO_CXX11_HDR_FUNCTIONAL if appropriate. 1. Would you agree that’s the right place to set BOOST_NO_CXX11_HDR_FUNCTIONAL (when appropriate)? 2. Any suggestions on the correct test to determine whether BOOST_NO_CXX11_HDR_FUNCTIONAL is needed? It looks like the _has_feature() macro is the preferred method, but I can’t figure out which feature to check for … Thanks, Chris p.s. The text of the 3 errors (when trying to compile boost_no_cxx_hdr_functional) is exceedingly lengthy, but to summarize: usr/bin/../lib/c++/v1/__functional_base_03:962:22: error: no matching function for call to '__invoke' typedef decltype(__invoke(_VSTD::declval<_Fp>())) type; usr/bin/../lib/c++/v1/__config:496:32: note: expanded from macro 'decltype' #define decltype(x) __typeof__(x) usr/bin/../lib/c++/v1/__functional_base_03:1019:14: note: in instantiation of template class 'std::__1::__invoke_return<int &, false>' requested here typename __invoke_return<type&>::type ./boost_no_cxx_hdr_functional.ipp:22:4: note: in instantiation of template class 'std::__1::reference_wrapper<int>' requested here std::ref(i); usr/bin/../lib/c++/v1/__functional_base_03:962:22: error: no matching function for call to '__invoke' typedef decltype(__invoke(_VSTD::declval<_Fp>())) type; usr/bin/../lib/c++/v1/__config:496:32: note: expanded from macro 'decltype' #define decltype(x) __typeof__(x) usr/bin/../lib/c++/v1/__functional_base_03:1019:14: note: in instantiation of template class 'std::__1::__invoke_return<const int &, false>' requested here typename __invoke_return<type&>::type ./boost_no_cxx_hdr_functional.ipp:23:4: note: in instantiation of template class 'std::__1::reference_wrapper<const int>' requested here std::cref(i); ./boost_no_cxx_hdr_functional.ipp:36:9: error: no member named 'bind' in namespace 'std'; did you mean 'find'? std::bind(f, std::placeholders::_1, 0.0f);

3 2