October 2014 - Boost - lists.boost.org

[1.56.0] Release candidates available
by Marshall Clow 31 Oct '14

31 Oct '14

Release candidate files for 1.65.0 are available at http://boost.cowic.de/rc/ 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. This helps ensure the candidates build OK before we push them out to SourceForge. The files (and associated md5s) are: MD5 (./boost_1_57_0_rc1.7z) = 17c98dd78d6180f553fbefe5a0f57d12 MD5 (./boost_1_57_0_rc1.tar.bz2) = 1be49befbdd9a5ce9def2983ba3e7b76 MD5 (./boost_1_57_0_rc1.tar.gz) = 25f9a8ac28beeb5ab84aa98510305299 MD5 (./boost_1_57_0_rc1.zip) = 5e040e578e3f0ba879da04a1e0cd55ff Thanks! -- The release managers P.S. Release notes are here: http://www.boost.org/users/history/version_1_57_0.html

1 0

[PropertyTree][variant] Reading JSON into a recursive variant.
by Arindam Mukherjee 31 Oct '14

31 Oct '14

I have a recursive variant definition that can handle JSON data. Is there a way to read JSON data using PropertyTree into a recursive JSON? Here is the variant defn. #define BOOST_VARIANT_NO_FULL_RECURSIVE_VARIANT_SUPPORT #include <boost/variant.hpp> JSONNullType {}; boost::variant<std::string, double, bool, JSONNullType> jsonToken; typedef boost::make_recursive_variant< std::string, double, JSONNullType, std::map<std::string, boost::recursive_variant_>, std::vector<boost::recursive_variant_> >::type JSONValue; Regards, Arindam

1 0

Re: [boost] Coming up on 1.57 release date - get your fixes in
by Felix Uhl 30 Oct '14

30 Oct '14

Edward Diener wrote: > Marshall Clow wrote: > > Just this morning we received a report of Boost.PP failing on VS 2013. > > It is not failing. It is just giving a warning. I don’t know for sure, but Marshall may be referring to this ticket: https://svn.boost.org/trac/boost/ticket/10686 On VS2013, VARIADICS and MSVC_VARIADICS don’t get defined despite the fact that they should, rendering some variadic macros useless. Von: Edward Diener Gesendet: ‎Donnerstag‎, ‎30‎. ‎Oktober‎ ‎2014 ‎04‎:‎10 An: boost(a)lists.boost.org On 10/29/2014 10:50 AM, Marshall Clow wrote: > We’ll be closing the master branch tomorrow (Thursday) for the 1.57.0 release. > Changes not in by then will not go into the release. > > Also, what issues in the beta have not been addressed? > >> Visual C++ 14 CTP 4 without BOOST_MSVC_ENABLE_2014_JUN_CTP: all libraries >> built successfully (but some are not usable, mostly because of missing >> noexcept) > >> >> All projects using Boost I tried (Debug 32-bit + 64-bit) compiled >> successfully with MSVC12. For MSVC14, TypeTraits 9de6a49 is missing in >> master. > > I believe that these have been merged. > >> Interprocess is broken for MinGW and there are some issues with Intrusive in GCC. I've committed fixes for both. > > I believe that this has been merged. > > Thread fixes have been merged. > > I do not know if the patches described in http://lists.boost.org/Archives/boost/2014/10/217257.php have been merged. > >> range/detail/any_iterator.hpp is broken because it expects >> postfix_increment_proxy and writable_postfix_increment_proxy to be in >> namespace boost::detail, but they are in boost::iterators::detail now. > > This does NOT appear to have been fixed > >> boost serialization is still broken(said so: http://article.gmane.org/gmane.comp.lib.boost.user/82715 ) and to date i do know no workaround to get my library to work on fx Arch Linux without telling users to rebuild with boost 1.55 (or sending them a saved package from my system…) > > I do not know what the status of this is; but there have been no commits to range/master in October. > > Just this morning we received a report of Boost.PP failing on VS 2013. It is not failing. It is just giving a warning. _______________________________________________ Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost

2 1

[testing] Compile failures on MSVC (teeks)
by Andrey Semashev 30 Oct '14

30 Oct '14

Hi, I'm seeing weird compilation failures on Teeks (MSVC 10 and 11): http://www.boost.org/development/tests/develop/developer/output/teeks99-08c… http://www.boost.org/development/tests/develop/developer/output/teeks99-08d… The weird part is this: D:\teeks99-08\d\boost_root\boost/fusion/support/detail/result_of.hpp(32) : error C2143: syntax error : missing ',' before '...' I cannot find the boost/fusion/support/detail/result_of.hpp file in git. Am I missing something or these test machines are broken somehow?

2 1

Coming up on 1.57 release date - get your fixes in
by Marshall Clow 30 Oct '14

30 Oct '14

We’ll be closing the master branch tomorrow (Thursday) for the 1.57.0 release. Changes not in by then will not go into the release. Also, what issues in the beta have not been addressed? > Visual C++ 14 CTP 4 without BOOST_MSVC_ENABLE_2014_JUN_CTP: all libraries > built successfully (but some are not usable, mostly because of missing > noexcept) > > All projects using Boost I tried (Debug 32-bit + 64-bit) compiled > successfully with MSVC12. For MSVC14, TypeTraits 9de6a49 is missing in > master. I believe that these have been merged. > Interprocess is broken for MinGW and there are some issues with Intrusive in GCC. I've committed fixes for both. I believe that this has been merged. Thread fixes have been merged. I do not know if the patches described in http://lists.boost.org/Archives/boost/2014/10/217257.php have been merged. > range/detail/any_iterator.hpp is broken because it expects > postfix_increment_proxy and writable_postfix_increment_proxy to be in > namespace boost::detail, but they are in boost::iterators::detail now. This does NOT appear to have been fixed > boost serialization is still broken(said so: http://article.gmane.org/gmane.comp.lib.boost.user/82715 ) and to date i do know no workaround to get my library to work on fx Arch Linux without telling users to rebuild with boost 1.55 (or sending them a saved package from my system…) I do not know what the status of this is; but there have been no commits to range/master in October. Just this morning we received a report of Boost.PP failing on VS 2013. Have I missed any of the beta issues? — Marshall

7 11

[fusion] Broken compilation on MSVC (was: [testing] Compile failures on MSVC (teeks))
by Andrey Semashev 30 Oct '14

30 Oct '14

On Thu, Oct 30, 2014 at 1:28 PM, Andrey Semashev <andrey.semashev(a)gmail.com> wrote: > Hi, > > I'm seeing weird compilation failures on Teeks (MSVC 10 and 11): > > http://www.boost.org/development/tests/develop/developer/output/teeks99-08c… > > http://www.boost.org/development/tests/develop/developer/output/teeks99-08d… > > The weird part is this: > > D:\teeks99-08\d\boost_root\boost/fusion/support/detail/result_of.hpp(32) > : error C2143: syntax error : missing ',' before '...' > > I cannot find the boost/fusion/support/detail/result_of.hpp file in > git. Am I missing something or these test machines are broken somehow? Oh, sorry, I was looking at the master local branch, and it fails on develop. Anyway, the header is there and it is broken for compilers not supporting variadic templates.

1 0

Re: [boost] 1.57.0 beta 1 is now available
by Felix Uhl 30 Oct '14

30 Oct '14

Edward Diener wrote: >John Doré wrote: >> Unfortunately this gives us the warning: >> >> warning C4003: not enough actual parameters for macro 'BOOST_PP_EXPAND_I' > > It is an expected warning from Boost PP, which can be ignored. The VC++ > preprocessor is not a C++ standard conforming preprocessor so it gives > warnings when none are necessary. There is virtually nothing that can be > done against such warnings except to attempt to turn them off completely > within Boost PP code. I will look into doing this but often it is > impossible to do since VC++ often does not honor where the preprocessor > warning actually takes place when an effort is made to eliminate the > warning at that place. Does that mean the regular >#pragma warning(disable : 4003) doesn’t work for pre-processor related warnings?

1 0

1.57.0 beta 1 is now available
by Marshall Clow 30 Oct '14

30 Oct '14

Boost release 1.57.0 beta 1 is now available from SourceForge See http://sourceforge.net/projects/boost/files/boost-binaries/1.57.0.beta.1/ This release contains no new libraries. For details of what's in the release, see http://www.boost.org/users/history/version_1_57_0.html. Note that the links to files on this web page are for the final release - use the SourceForge link above to get the beta files. Please download the beta, give it a try, and report any problems you encounter. Thanks, -- The Boost Release Team P.S. There are a couple of known problems with this beta: * There are problems building with MSVC 8.0 * There are some problems with range/detail/any_iterator.hpp. These will be fixed before the release.

9 10

Re: [boost] Is there interest in a library for string-convertible enums?
by Felix Uhl 30 Oct '14

30 Oct '14

Artyom Tokmakov wrote: > Have you considered X-macro trick? Yes, and I heavily dislike the syntax. I played around with it a bit and it is absolutely not feasible for this use case. My goal is a simple, intuitive and easy to use syntax. With xmacros you’d have to define an enum in a separate file, Which only makes sense for really, really large enumerations.

1 0

Re: [boost] Is there interest in a library for string-convertible enums?
by Felix Uhl 29 Oct '14

29 Oct '14

Artyom Tokmakov wrote: >In addition to various useful suggestions on the list, I would ask for >implementation to support (almost) any large number of enum elements >(enumerators) within enum declaration.: >I have ad-hoc implementation of my own which uses Boost.Proprocessor's >sequence to define enum elements. Then I can generate whatever >parser/printer/other code I want (I find that very useful). >There's a disadvantage, however: obvious limitation of max 256 >elements (at least, by default) coming from Boost.Preprocessor. And that’s where the problem lies. To enable large numbers of enum values, I’d have to find a solution without using Boost.PP. Without C++ introducing argument number based macro overloading, that is not possible, short of defining my own macro sequences for more than 256 elements, which I would like to avoid. I am still thankful that you asked for this, because it uncovered a huge implementation problem which induces enormous compilation overhead (1 minute for 48 enum values on my laptop). I will do my best to make the library comprehend large enums, but I can’t guarantee anything larger than 256. --- Felix Uhl

2 1