- Boost - lists.preview.boost.org

Status of Disjoint Sets ?
by Brian Kuhl 15 Mar '19

15 Mar '19

Disjoint Sets is in the last few releases, has it's own test suite and github archive, but isn't mentioned on the documentation page? An oversight, or intentional? https://www.boost.org/doc/libs/1_69_0/ I see a defect that says it should move into Graph, is that part of the puzzle? https://github.com/boostorg/disjoint_sets/issues Brian

2 1

Networking TS + Beast, NEW Tutorials, Read this to learn std::net !!!
by Mike 15 Mar '19

15 Mar '19

> All in all nice intro, unfortunately it requires users to understand > templates quite well, but honestly not sure if it is avoidable, with a > gazillion customization points in ASIO. Anyone aware of a wrapper library around ASIO that doesn't offer all those customization points and maybe not the full functionality/ efficiency but is easier to use? Am 15.03.19, 01:09, Ivan Matek via Boost <boost(a)lists.boost.org> schrieb: Here are comments from somebody who "knows C++" but does not know a lot about NW and has read only the 1st <[1]https://www.boost.org/doc/libs/master/libs/beast/doc/html/beast/ using_io/asio_refresher.html> page. // get iterators representing the range of characters in the buffer auto begin = range::begin(buffer.data()); auto end = range::end(buffer.data()); Quite a weird pattern, where buffer does not have member begin, end. Maybe a sentence or two describing the rationale and complexity would be nice. I know it is O(1), but it is weird that begin/end are called on pointer. std::size_t read_lineI am tired, but is this function O(n^2)? cb += bytes_transferred; // adjust the pointer and sizeI would maybe switch to write instead of write_some to keep the code shorter, and mention write_some as another option in text. [](error_code ec, std::size_t bytes_transferred) For learning purposes I would capture something in lambda by value and note that reference capture would be buggy. using type = std::allocator<void>; Personal preference: not a fan of typedefs for relatively short types used once. All in all nice intro, unfortunately it requires users to understand templates quite well, but honestly not sure if it is avoidable, with a gazillion customization points in ASIO. _______________________________________________ Unsubscribe & other changes: [2]http://lists.boost.org/mailman/listinfo.cgi/boost References 1. https://www.boost.org/doc/libs/master/libs/beast/doc/html/beast/using_io/as… 2. http://lists.boost.org/mailman/listinfo.cgi/boost

1 0

Coordinating Boost Releases with compiler releases
by Marshall Clow 14 Mar '19

14 Mar '19

In the thread "MSVC 2019 and 1.70 beta", there has been a discussion about what to do about support for MSVC 2019, which is due to be released right about the same time that Boost 1.70. One person suggested that we delay the boost release until after the MSVC release (I think the suggestion was "May"). This general question has been discussed before (several times), since tools get released fairly often - and so does Boost. In general, I am not in favor of delaying boost releases in order to add support for 'about to be released' development tools. There are two main reasons for this: * There's always another compiler / tool about to be released. Clang 8 is imminent - currently at RC5. GCC 9.1 is scheduled to be released in early May. * Boost releases happen every four months. August is not really that far off. People who want support for tool XXXX can use "the trunk", getting it either from git or from bintray. [ Everyone knows about the tarballs at bintray, right? https://dl.bintray.com/boostorg/master/ ] Also, if we were to delay the boost release until May, then what would happen to the August release? The release process for 1.71.0 is scheduled to start at the end of June. -- Marshall

5 7

[filesystems]
by Florian Lindner 14 Mar '19

14 Mar '19

Hello, I have this innocent piece of code: namespace fs = boost::filesystem; auto path = getFilename(); // returns a string fs::create_directories(fs::path(path).parent_path()); std::ofstream ofs(path + "~"); ofs << info; ofs.close(); fs::rename(path + "~", path); which causes the exception: boost::filesystem::rename: No such file or directory: "../9f/061b4f7a5e529c964659226eedd4e5~", "../9f/061b4f7a5e529c964659226eedd4e5" However, I have no idea how that could happen. I use the rename, so that a reading process never sees an empty file, but only no file or filed with info. Is there any race involved between ofs.close() and fs:rename()? The code was executed on a distributed network filesystem (lustre). Any ideas anyone? Thanks, Florian

5 9

Re: [boost] [Boost-users] Networking TS + Beast, NEW Tutorials, Read this to learn std::net !!!
by Vinnie Falco 14 Mar '19

14 Mar '19

On Thu, Mar 14, 2019 at 10:50 AM james <dirtydroog(a)gmail.com> wrote: > This is excellent news. Timeouts are ASIO's biggest downfall. I'll definitely check it out. > If it's low level, and performant enough, then it paves the way for ASIO-ed connectors for things like Redis and ZeroMQ. Thanks! And we are very happy to answer questions, feel free to open an issue or come to the #beast Slack channel. Regards

1 0

Cannot serialize pointer through base without void_cast_register and register_type.
by Nicolas Guillot 14 Mar '19

14 Mar '19

Hello. *** CONTEXT *** I use your serialization library since several years on linux (gcc). Now, I'm moving to windows (_MSC_VER == 1600) (boost 1.55, but same issues on boost 1.69). For my example, I use a class A in a DLL, another class B inheriting from A in its DLL and an exe, linked to those DLLs, which serializes a B object, through its A base. _______ | A | ------- /_\ | | _______ | B | ------- I make a class serializable like that (I summurize on purpose (no #include, ...)). For A it's the same thing, but it has no base class. ----> B.h ---------------------------------------- #ifdef B_EXPORTS #define B_DLL __declspec(dllexport) #else #define B_DLL __declspec(dllimport) #endif class B_DLL B: public A { public: B(); virtual ~B(); private: friend class boost::serialization::access; template <class Archive> void serialize(Archive & ar, const unsigned int version); int i; }; BOOST_CLASS_EXPORT_KEY(B) ----> B.cpp ---------------------------------------- B::B() : i(123) {} B::~B() {} template <class Archive> void B::serialize(Archive & ar, const unsigned int) { ar & SERIALIZE_BASE(ClassA); ar & SERIALIZE(i); } #define FULL_SCOPED_SERIALIZABLE_CLASS B #include "Common/InstanciateSerializationTemplate.h" ----> InstanciateSerializationTemplate.h ---------------------------------------- This file purpose is to instanciate the template B::serialize methods for each archive type, and to register the class. Here is its contents (once again, no namesapce nor include to make it easier to read): // explicit instantiate code for all archive types expected to be use: // MUST COME BEFORE BOOST_CLASS_EXPORT_IMPLEMENT // binary: template __declspec(dllexport) void FULL_SCOPED_SERIALIZABLE_CLASS::serialize<binary_iarchive>(binary_iarchive & ar, const unsigned int version); template __declspec(dllexport) void FULL_SCOPED_SERIALIZABLE_CLASS::serialize<binary_oarchive>(binary_oarchive & ar, const unsigned int version); // XML: template __declspec(dllexport) void FULL_SCOPED_SERIALIZABLE_CLASS::serialize<xml_iarchive>(xml_iarchive & ar, const unsigned int version); template __declspec(dllexport) void FULL_SCOPED_SERIALIZABLE_CLASS::serialize<xml_oarchive>(xml_oarchive & ar, const unsigned int version); // TXT: template __declspec(dllexport) void FULL_SCOPED_SERIALIZABLE_CLASS::serialize<text_iarchive>(text_iarchive & ar, const unsigned int version); template __declspec(dllexport) void FULL_SCOPED_SERIALIZABLE_CLASS::serialize<text_oarchive>(text_oarchive & ar, const unsigned int version); BOOST_CLASS_EXPORT_IMPLEMENT(FULL_SCOPED_SERIALIZABLE_CLASS) *** MY ISSUE *** All that stuff works in linux, gcc, but not in Windows. If I write in my exe this serialization code: std::ofstream ofs; ofs.open("D:\\TestSerial.xml"); boost::archive::xml_oarchive ar(ofs); shared_ptr<B> b = shared_ptr<B>(new B()); shared_ptr<A> a = b; try { ar << SERIALIZE(a); } catch (boost::archive::archive_exception& ex) { std::cout << "Serialization to stream error, what: " << ex.what() << std::endl; } catch (...) { std::cout << "Catch ... Serialization to stream error" << std::endl; } ofs.close(); With this code, I have an archive_exception: "unregistered class - derived class not registered or exported". I have to write the serialization like that: ar.register_type(static_cast<B*>(0)); boost::serialization::void_cast_register(static_cast<B*>(0), static_cast<A*>(0)); ar << SERIALIZE(a); *** QUESTION *** At least, my question: I cannot do like that: what I would like to achieve: when a new class is made serializable like above, I expect the class to be registered automatically (probabley with BOOST_CLASS_EXPORT_IMPLEMENT in InstanciateSerializationTemplate.h). It's too painfull to write the register_type and void_cast_register each time I want to serialize. What am I missing? Thank you very much for any help or recommandations. Nico

1 1

Re: [boost] MSVC 2019 and 1.70 beta
by Peter Dimov 14 Mar '19

14 Mar '19

> Tom Kent wrote: > > > Build support for VS 2019 RC (msvc-14.2) is on develop in boost build. > > Do we want to merge that to master for the release? > > I think we do. 1.70 and 2019 will release at about the same time. In fact looking at the calendar, 2019 will release (Apr 2) just before 1.70 is closed for release (Apr 3.)

3 3

boost-install-ing zlib (Was: Re: [Release] Boost 1.70.0 Beta 1 Release Candidate 2)
by Peter Dimov 13 Mar '19

13 Mar '19

> I'll need to try and see what happens with <source> in user-config. > Probably nothing good, because one would expect neither <build-name> nor > <tag> to be specified in such a case. If I remove the check for ZLIB_SOURCE, and just do alias boost_zlib : /zlib//zlib ; libraries-to-install += boost_zlib ; unconditionally in libs/iostreams/build/Jamfile, this is what happens: - when `using zlib : : <source>C:/Projects/zlib-1.2.11 ;` is in user-config, the default `b2 install` fails because `address-model=32,64 variant=debug,release` conflict as all try to build `libz.lib`. When address-model=32 variant=debug is given, `libz.lib` is installed in C:\Boost\lib, and the CMake configuration for boost_zlib correctly points to it. - when using a prebuilt `using zlib : : <include>C:/Projects/include <search>C:/Projects/lib <name>zlib-8 ;`, the CMake configuration file for boost_zlib points to "${_BOOST_LIBDIR}/zlib-8", which is not correct.

2 2

(no subject)
by Marshall Clow 13 Mar '19

13 Mar '19

Boost release 1.70.0 beta 1 is now available at: <https://dl.bintray.com/boostorg/beta/1.70.0.beta1/source/> The SHA256 checksums are as follows: dbf89645e509acb2e53eac3c633834f1dec80bd8622e6e652a9741dcf0807f17 ./boost_1_70_0_b1.7z 957b984b58656a3ec00661ecad7ef82e3a094bad36e37b9e6624843237854679 ./boost_1_70_0_b1.tar.bz2 76d149c1fb9e8a5ceb3bc1e5587edc40fa79a43e8b990d877c2210f0382b9421 ./boost_1_70_0_b1.tar.gz 71e990b2c26880521fd98459b184ee172ce1c46dcf554b6eaab7776b8b6369a6 ./boost_1_70_0_b1.zip For details of what's in the release, see < https://www.boost.org/users/history/version_1_70_0.html>. Please download the beta, give it a try, and report any problems you encounter. Thanks, -- The Boost Release Team

2 3

[rational] permission to merge cmake change?
by James E. King III 13 Mar '19

13 Mar '19

Before 1.70.0 beta started, some cmake changes from Mike Dev were merged into rational. There was a typo in one of the cmake options (BOOST_RATIONA_INCLUDE_TESTS). This does not affect code nor the boost build system. Would it be okay to merge this now or should I wait? https://github.com/boostorg/rational/pull/35 Thanks. - Jim

1 1