- Boost - lists.boost.org

boost::any_ref and boost::any_const_ref
by Daniel Larimer 14 Feb '10

14 Feb '10

I am attempting to create a library with a generic interface run-time interface and have need for something like boost::any; however, boost::any requires heap allocation. boost::variant is not acceptable either because I do not know the full set of types in advance. Has anyone seen the equivalent of a type-checked void* or const void*? Effectively I would like to create a polymorphic method like this: void myclass::method( const_any_ref p1, const_any_ref p2, const any_ref p3 ) { const uint16_t& _p1 = p1; const double& _p2 = p2; const std::string& _p3 = p3; //If any of the types do not match it should throw an exception. alternatively the follow syntax may work if you want to avoid exceptions? boost::optional<const uint16_t&> _p1 = p1; if( !_p1 ) // cast failed? } Anyone have any nifty ideas on how to accomplish this without RTTI and no heap allocation? In place of RTTI I do have a "manual" RTTI system that I could use. get_typename<T>::str will return a unique const char* pointer. I am guessing the best I can hope for is a pair of pointers, one "void*" and one const char*. I am going to write such a class unless someone here knows of an existing solution. Dan

1 1

[vault][msm] Where has the vault gone?
by Agustín K-ballo Bergé 14 Feb '10

14 Feb '10

http://www.boostpro.com/vault/ gives a 404 - Not Found I'm trying to get the latest MSM version. Is there anywhere else I can get it? Agustín K-ballo Bergé.-

3 3

Re: [boost] [vault][msm] Where has the vault gone?
by Christophe Henry 13 Feb '10

13 Feb '10

> You can find it in sandbox > http://svn.boost.org/svn/boost/sandbox/msm/ Small precision, I use it as my "trunk" version. It is stable but the documentation still lags the code state (I am in the rewriting process). The last stable version and with up-to-date documentation (the one from the vault + a few bugfixes / performance improvements) is in: http://svn.boost.org/svn/boost/sandbox/msm/tags/version-2.0/ >For SF right link is http://sourceforge.net/projects/boost-msm/ And I didn't even advertise it :) I plan to use SF so that one could request and try some features before they make it into the official code, but I need to finish the project page first, so for the moment, you will find not much there. Christophe

1 0

Re: [boost] [serialization] Can one deserialize a derived classes viapointers to the base class using polymorphic archives in aDLL context?
by Bogdan 13 Feb '10

13 Feb '10

Robert Ramey <ramey <at> rrsd.com> writes: > >> > >> Please note BOOST_CLASS_EXPORT_IMPLEMENT is used only once per class > >> (inside polymorphic_derived2.cpp), yet the serializers are inserted > >> in the same map both from the Dll and from the executable. > >> > >> Am I doing something wrong? Has anyone been able to create a similar > >> working example? I really appreciate any feedback I could get. > > It looks like polymorphic_derived2.cpp is being included in both the > mainline as well as the DLL. > > It's not clear that polymorphic_base needs to be n the DLL and probably > shouldn't be. > > Make sure that your header includes BOOST_CLASS_EXPORT_KEY and not > BOOST_CLASS_EXPORT. > > finally, make sure that the test for this facility works. > > Robert Ramey > Thank you Robert for your quick response: I appreciate a lot the generous work you put in the library. I double checked the project I previously uploaded: I downloaded it via gmane.org, changed the name back from bin to zip and verified the following: - polymorphic_derived2.cpp file is included only in dll_polymorphic_derived2.vcproj. - there are only three projects dll_polymorphic_derived2, Library and test_dll_exported plus an explicit reference to the bjam files (I removed the latter after downloading); - there is no usage BOOST_CLASS_EXPORT; - there are exactly three declarations of BOOST_CLASS_EXPORT_KEY: one in polymorphic_base.hpp, and two in polymorphic_derved2.hpp one for polymorphic_derived1 and another for polymorphic_derived2. Similarly, there are only three usages of BOOST_CLASS_EXPORT_IMPLEMENT. There was only one question mark: test_dll_exported.vcproj depended on dll_polymorphic_derived2 both via Project Dependencies option and explicitly as an input library: to be on the safe side, I removed the explicit dependency of test_dll_exported on dll_polymorphic_derived2 from the solution. I rebuild the solution from scratch. Unfortunately I got the same exception. So I am still cornered and looking for some help. Thank you again, Bogdan

1 0

[serialization] Can one deserialize a derived classes via pointers to the base class using polymorphic archives in a DLL context?
by Bogdan 13 Feb '10

13 Feb '10

Hi all, I am trying to upgrade to boost 1.42 using Visual Studio 2008 and I have difficulties using serialization with polymorphic archives in a DLL setup. I would like to encapsulate in one DLL the class serialization functionality based on a polymorphic base archive. In another place, I would like to perform the actual serialization/deserialization using a specific archive (e.g., polymorphic_text_[io]archive). The section "Special Considerations" of the documentation indicates in two paragraphs ("DLLS - Serialization and Runtime Linking" and " Plugins") that this scenario is supported by the library. However I couldn't find any example that shows how to do that. Therefore, I changed slightly the boost test that includes test_dll_exported.cpp and polymorphic_derived2.cpp as follows: 1. I moved the classes polymorphic_base and polymorphic_derived1 to the project polymorphic_derived2 This way, we have one place where those classes serialize their members (i.e., polymorphic_derived2.dll) and a separate binary where we serialize them via a derived archive (i.e., test_main() inside test_dll_exported.exe); 2. I replaced the template version of the serialization functions with the equivalent version based on boost::archive::polymorphic_[io]archives. I also removed the text-based instantiation inside the Dll. As indicated in the boost documentation, the class serialization will only have to be compiled once for all archive implementations (present or future). Therefore, the DLL should contain only the serialization functions to the base polymorphic archives. While everything builds correctly, at run time I get a C++ exception BEFORE main() in basic_serializer_map.cpp around the line 48: // if this fails, it's because it's been instantiated // in multiple modules - DLLS - a recipe for problems. // So trap this here if(!result.second){ boost::serialization::throw_exception( archive_exception( archive_exception::multiple_code_instantiation, bs->get_debug_info() ) ); } Please note BOOST_CLASS_EXPORT_IMPLEMENT is used only once per class (inside polymorphic_derived2.cpp), yet the serializers are inserted in the same map both from the Dll and from the executable. Am I doing something wrong? Has anyone been able to create a similar working example? I really appreciate any feedback I could get. Thank you, Bogdan PS For more info I attached the files describing my problem plus the .sln file (I used the configuration "Debug runtime-dynamic threading-multi")

2 1

Re: [boost] [contract] Contract Programming Library
by Andrzej Krzemienski 13 Feb '10

13 Feb '10

> I am thinking to submit a library for Contract Programming (a.k.a. > Design By Contract (TM) ) for C++. > I have drafted some of the library documentation in Boost-like format: > http://dbcpp.sourceforge.net/boost/libs/contract/doc/html/ > Comments? Hi, I would be very interested in trying such a library. I have one suggestion, which may be impossible to implement, but I think it would be worth at least looking at. The macros for defining assertions spoil function declarations. I know it is inevitable, etc., but.. read on. I know at least one library in Boost that also spoils function declarations in order to provide additional functionality: Concept Check library; they may be others too (MPL?). My suggestion is that if there are (or will be) libraries that require spoiling function declarations, they should all provide the same "spoiled" syntax. Otherwise I will be always asking "how do I declare a function in this library?". It would be very convenient if Boost provided one alternative function definition syntax that when used would enable all its libraries to work. Regards, &rzej

2 1

Interest in Remote Procedure Call Library?
by Daniel Larimer 13 Feb '10

13 Feb '10

I have been playing with some code and recently developed a very efficient, extensible, and easy to use API for performing remote procedure calls (with multiple protocols). I am writing to gauge interest and get feedback on the user API. The native "protocol" is a custom combined tcp/udp protocol using multi-cast for discovery, but it would be very easy to add a shared memory solution, a dbus, XML/RPC, CORBA, etc back end to the user api. Example: class SomeClass { public: int add( int a, int b ) { return a + b ); void sub( int a, int b, int* result ) { if(result) *result = a -b; }; void inout( int& a ) { a += 5; } }; META_INTERFACE( SomeClass, METHOD(add) METHOD(sub) METHOD(inout) ) // server SomeClass myclass; Server myserver( &myclass, "my.named.service" ); // client RemoteInterface<SomeClass> ri( "my.named.service" ); int result = ri.add(1,4); assert( result == 5 ) ri.sub( 5, 1, &result ); assert( result == 4 ); ri.inout(result); assert( result == 9 ); API can support any type that is serializable. It also supports the signal/slot paradigm. SomeClass could have any number of base-classes both virtual and non-virtual. With respect to performance, it can perform 15,000+ synchronous (invoke, wait for return) operations per-second on localhost/linux. Asynchronous operations are capable of 500,000+ invokes/sec. The library doubles as a built-in reflection system allowing run time inspection member functions, parameter types, inheritance, etc. I am working to convince my employer (small company of 25) that contributing this library would bring more benefits than "costs" to the company. The reasons I think it is a good idea include: 1) Expose the company to more people (marketing) 2) We can give our customers an "open" API for communicating with our products 2) Offload some of the development / maintenance / support to the open source community 3) I like open source and have personal reasons for wanting it free and open. So I need to get some more support to convince my employer that this will be a good move. Active interest would be a good sign. Other ideas on the positive aspects of contributing vs keeping it internal would also help me make my case. Thanks, Dan

7 33

WARNING: BoostBook build change..
by Rene Rivera 13 Feb '10

13 Feb '10

In response to the whole, using Boost logo outside of real Boost Libraries, I changed the BoostBook XSL slightly to: * Default to not including a navbar. * Default to not including the Boost logo in the navbar, when one expressly asks to include the navbar. I changed the Boost Jamroot to include the two options one needs to get back the standard Boost navbar+logo: <xsl:param>nav.layout=horizontal <xsl:param>boost.image=Boost The "boost.image" param is new and lets one specify: none (for no logo image), Boost (for the boost.png image), or anything else (for your own custom image. So.. WARNING: You may have to change your doc builds if you previously assumed that the navbar or logo would show up outside of Boost to include the above options (or some version thereof). -- -- Grafik - Don't Assume Anything -- Redshift Software, Inc. - http://redshift-software.com -- rrivera/acm.org (msn) - grafik/redshift-software.com -- 102708583/icq - grafikrobot/aim,yahoo,skype,efnet,gmail

3 2

bcp utility documentation
by Jeroen Habraken 13 Feb '10

13 Feb '10

Hi, I just noticed that http://www.boost.org/doc/libs/1_42_0/tools/bcp/bcp.html which is linked to from the main boost.org page under documentation leads to a 404. The page used to exist as http://www.boost.org/doc/libs/1_41_0/tools/bcp/bcp.html. Regards, Jeroen Habraken

2 2

[property_tree] Proposal for Boost.Property_Tree put function
by Andrew Hundt 12 Feb '10

12 Feb '10

I have been using the boost::property_tree in one of our products, and have become frustrated by an inconsistency between two of the convenience functions. The basic_ptree get function allows the specification of a separate delimiter when specifying the path, such as: pt.get<float>('/', "p.a.t.h/t.o/v.a.l.u.e"); However, there is no similar way to call put to place the same value as follows: pt.put('/', "p.a.t.h/t.o/v.a.l.u.e", 3.14f); I would like propose the addition of functions to allow consistency between the use of get and put functions. The basic use of these functions is described in http://www.boost.org/doc/libs/1_42_0/doc/html/boost_propertytree/accessing.… Are there any thoughts or comments? Regards, Andrew Hundt

1 0