April 2008 - Boost-users - lists.preview.boost.org

Re: [Boost-users] Newbie question-- how do I get a Lockable?
by Henry Holmes 25 Apr '08

25 Apr '08

Thanks, Bill. I can declare a mutex but it doesn't have the lock/unlock methods described in the documentation-- the only thing that's public is the constructor/destructor. Could I have botched my installation of boost somehow? I just got it out of synaptic, thinking that that would be the least error-prone way to go. Also, in the interest of all that "teach a man to fish" stuff so I can quit bothering you, is there a document somewhere that explains what a "concept" is in boost terms? (ie, is it a design specification for developers to work off of, or is it a pure virtual class actually described in a .hpp file somewhere?) Thanks, Henry ----- Original Message ---- From: Bill Somerville <bill(a)classdesign.com> To: boost-users(a)lists.boost.org Sent: Thursday, April 24, 2008 4:02:44 PM Subject: Re: [Boost-users] Newbie question-- how do I get a Lockable? Hi Henry, Henry Holmes wrote: > Sorry if this is a dumb question-- it looks like just about every kind > of lock/mutex I could declare in the documentation wants to be passed > a Lockable object, but I can find no such thing in any of the header > files. I have the Boost Thread library installed and am able to make > and run threads, and I have all sorts of other lock and mutex stuff > defined, it just does not seem to correspond to the documentation at > all. Am I just horribly confused about what the documentation means? > Any help is greatly appreciated. Lockable is not a type, it is a concept. Concepts are requirements a type must comply with to be used by a type or function. The Lockable concept is defined here http://www.boost.org/doc/libs/1_35_0/doc/html/thread/synchronization.html#t… .. The boost thread library provides types that implement these concepts and they may be substitued where a suitable concept is required. This is a form of polymorphism you may not be aware of. The types you may want to instantiate for your requirements are probably described here http://www.boost.org/doc/libs/1_35_0/doc/html/thread/synchronization.html#t… note that they all implement at least the Locakble concept. HTH -- Bill Somerville Class Design Limited _______________________________________________ Boost-users mailing list Boost-users(a)lists.boost.org http://lists.boost.org/mailman/listinfo.cgi/boost-users ____________________________________________________________________________________ Be a better friend, newshound, and know-it-all with Yahoo! Mobile. Try it now. http://mobile.yahoo.com/;_ylt=Ahu06i62sR8HDtDypao8Wcj9tAcJ

2 1

MPL Vector item limits
by Neal Meyer 25 Apr '08

25 Apr '08

I've been searching thru the achive to try to get this to work, but haven't been able to get it to compile. I need a mpl::vector with up to 100 items. In the preprocessed headers there is a big comment that says // -- DO NOT modify by hand!. How can I regenerate these preprocessed headers with more possible items in the list. -Neal

2 1

Newbie question-- how do I get a Lockable?
by Henry Holmes 24 Apr '08

24 Apr '08

Sorry if this is a dumb question-- it looks like just about every kind of lock/mutex I could declare in the documentation wants to be passed a Lockable object, but I can find no such thing in any of the header files. I have the Boost Thread library installed and am able to make and run threads, and I have all sorts of other lock and mutex stuff defined, it just does not seem to correspond to the documentation at all. Am I just horribly confused about what the documentation means? Any help is greatly appreciated. ____________________________________________________________________________________ Be a better friend, newshound, and know-it-all with Yahoo! Mobile. Try it now. http://mobile.yahoo.com/;_ylt=Ahu06i62sR8HDtDypao8Wcj9tAcJ

2 1

[smart_ptr] Sole ownership of shared_ptr?
by Joseph Thomson 24 Apr '08

24 Apr '08

Hi everyone, Over the past few days I have been trying to integrate Boost smart pointers into my system, which is basically a manager (ObjectManager) class which creates, destroys, and keeps track of instances of another class (Object). The manager keeps a std::map of instances of type shared_ptr. The manager has a createObject() method which returns a shared_ptr instance to the user, thus the user becomes an owner of the object, alongside the manager. My problem arises because I want objects to _always_ be destroyed when the manager method destroyObject() is called. I realise that part of the reason to use smart pointers is to prevent data from being deleted while a reference to it still exists, but I figure they are still worth using just for the safety aspect. My specific reasons for wanting the manager to be able to destroy the objects on command are: 1.) The manager may keep track of the objects, but the objects are used elsewhere in the system (until their destruction). 2.) I don't want to put the responsibility with the user (me!) to have to delete all references to the object in order for it to be deleted. Even if I pass back a weak_ptr, it is likely that the user will store a locked version of the pointer for convenience. 3.) I am using lua, which has no C-style "delete" command, and handles object deletion in its own time. Thus, even if all references to the object have been deleted in lua, the object may not actually be deleted straight away (unless you invoke deletion by calling collectgarbage(), which is a very messy solution). Anyway, I came up with a solution, but am not sure if there is any better or cleaner way. I am relatively new to smart pointers and the whole idea of virtual methods, so if I use the wrong terminology I apologise. Here goes: I have an abstract ObjectInterface class, which outlines all the methods that the user will access. I then have an Object class, which is of type ObjectInterface, and implements its abstract methods. The ObjectManager will, as before, create and store a map of shared_ptr references. However, instead of returning a shared_ptr to the user, the ObjectManager::createObject() method now creates a shared_ptr<ObjectHandle> which it returns to the user. ObjectManager doesn't store a reference to the handle, so the user is the sole owner of that. ObjectHandle (I'm not sure this is the correct use of the term "Handle"?) is also of type ObjectInterface, but in its constructor it takes a shared_ptr reference from the ObjectManager, which it stores internally as a weak_ptr (so that ObjectManager remains the sole owner of Object). It implements the ObjectInterface methods by forwarding calls from the user to the Object, but only if the weak_ptr hasn't expired. I am also storing a raw pointer to the Object in ObjectHandle, so that it doesn't have to call shared_ptr<>::lock() every time it needs to forward a call; I think this is reasonable to do since I never copy the pointer, and certainly won't delete it. Attached is a fully working example that I have tried to make as concise and clear as possible. Maybe my solution is the best one and my question seems stupid, or maybe there is a far simpler way and my solution seems stupid, but either way I am asking because I am unsure about this kind of thing in general. Thanks very much for any help in advance! Joe http://www.nabble.com/file/p16528658/smart_ptr_example.cpp smart_ptr_example.cpp -- View this message in context: http://www.nabble.com/-smart_ptr--Sole-ownership-of-shared_ptr--tp16528658p… Sent from the Boost - Users mailing list archive at Nabble.com.

2 1

Serialization of multidimmensional arrays
by Jason Gorski 24 Apr '08

24 Apr '08

Hello, We have just updated to 1.35 and it seems as if serialization of multidimensional arrays is broken. In 1.34_1, the following serialization code worked fine: int ints[2][2]; template<class Archive> void serialize(Archive & ar, const unsigned int version) { ar & BOOST_SERIALIZATION_NVP( ints ); } In 1.35_0, I receive the following error at compile time: oserializer.hpp:489: error: invalid static_cast from type ‘const int (*)[2]’ to type ‘const int*’ I have been able to get around this by looping through the outer dimension of the array and serializing all inner arrays separately, however this solution is less than ideal. If this is in fact an issue with the library, I am highly concerned. It seems to me that this issue should have been caught during regression testing. Thanks! Jason -- Jason Gorski Research Engineer Oasis Advanced Engineering, Inc. (248)373-9800 x 269

2 1

[multi_index] fast_pool_allocator not working with MI in 1.35?
by Filip Konvička 24 Apr '08

24 Apr '08

Hi, trying to use fast_pool_allocator in Boost 1.35 fails with MSVC 9.0; it worked well with MSVC 8.0. The following code fragment shows the error: #include <boost/multi_index_container.hpp> #include <boost/multi_index/indexed_by.hpp> #include <boost/multi_index/sequenced_index.hpp> #include <boost/multi_index/identity.hpp> #include <boost/pool/pool_alloc.hpp> using namespace boost; using namespace boost::multi_index; typedef multi_index_container<int, indexed_by<sequenced<> >, fast_pool_allocator<int> > IntSet; IntSet s; Below you can see the compiler error message. Am I doing something wrong? (From what I've seen digging through the code, it seems to me that the allocator gets rebound from int to void, which does not work for fast_pool_allocator.) Thanks, Filip C:\lib\Boost-1_35_0\include\boost-1_35\boost/pool/pool_alloc.hpp(130) : error C2182: 'reference' : illegal use of type 'void' C:\lib\Boost-1_35_0\include\boost-1_35\boost/detail/allocator_utilities.hpp(149) : see reference to class template instantiation 'boost::fast_pool_allocator<T,UserAllocator,Mutex,NextSize>' being compiled with [ T=void, UserAllocator=boost::default_user_allocator_new_delete, Mutex=boost::details::pool::default_mutex, NextSize=32 ] C:\lib\Boost-1_35_0\include\boost-1_35\boost/detail/allocator_utilities.hpp(158) : see reference to class template instantiation 'boost::detail::allocator::rebinder<Allocator>::result<Type>' being compiled with [ Allocator=boost::fast_pool_allocator<void,boost::default_user_allocator_new_delete,boost::details::pool::default_mutex,32>, Type=boost::multi_index::detail::sequenced_index_node_impl<boost::fast_pool_allocator<void,boost::default_user_allocator_new_delete,boost::details::pool::default_mutex,32>> ] C:\lib\Boost-1_35_0\include\boost-1_35\boost/mpl/eval_if.hpp(63) : see reference to class template instantiation 'boost::detail::allocator::compliant_allocator_rebind_to<Allocator,Type>' being compiled with [ Allocator=boost::fast_pool_allocator<void,boost::default_user_allocator_new_delete,boost::details::pool::default_mutex,32>, Type=boost::multi_index::detail::sequenced_index_node_impl<boost::fast_pool_allocator<void,boost::default_user_allocator_new_delete,boost::details::pool::default_mutex,32>> ] C:\lib\Boost-1_35_0\include\boost-1_35\boost/detail/allocator_utilities.hpp(170) : see reference to class template instantiation 'boost::mpl::eval_if_c<C,F1,F2>' being compiled with [ C=false, F1=boost::detail::allocator::partial_std_allocator_rebind_to<boost::fast_pool_allocator<void,boost::default_user_allocator_new_delete,boost::details::pool::default_mutex,32>,boost::multi_index::detail::sequenced_index_node_impl<boost::fast_pool_allocator<void,boost::default_user_allocator_new_delete,boost::details::pool::default_mutex,32>>>, F2=boost::detail::allocator::compliant_allocator_rebind_to<boost::fast_pool_allocator<void,boost::default_user_allocator_new_delete,boost::details::pool::default_mutex,32>,boost::multi_index::detail::sequenced_index_node_impl<boost::fast_pool_allocator<void,boost::default_user_allocator_new_delete,boost::details::pool::default_mutex,32>>> ] C:\lib\Boost-1_35_0\include\boost-1_35\boost/multi_index/detail/seq_index_node.hpp(36) : see reference to class template instantiation 'boost::detail::allocator::rebind_to<Allocator,Type>' being compiled with [ Allocator=boost::fast_pool_allocator<void,boost::default_user_allocator_new_delete,boost::details::pool::default_mutex,32>, Type=boost::multi_index::detail::sequenced_index_node_impl<boost::fast_pool_allocator<void,boost::default_user_allocator_new_delete,boost::details::pool::default_mutex,32>> ] C:\lib\Boost-1_35_0\include\boost-1_35\boost/multi_index/detail/seq_index_node.hpp(148) : see reference to class template instantiation 'boost::multi_index::detail::sequenced_index_node_impl<Allocator>' being compiled with [ Allocator=boost::fast_pool_allocator<void,boost::default_user_allocator_new_delete,boost::details::pool::default_mutex,32> ] C:\lib\Boost-1_35_0\include\boost-1_35\boost/multi_index/detail/seq_index_node.hpp(162) : see reference to class template instantiation 'boost::multi_index::detail::sequenced_index_node_trampoline<Super>' being compiled with [ Super=boost::multi_index::detail::index_node_base<int,boost::fast_pool_allocator<int>> ] C:\lib\Boost-1_35_0\include\boost-1_35\boost/multi_index/sequenced_index.hpp(95) : see reference to class template instantiation 'boost::multi_index::detail::sequenced_index_node<Super>' being compiled with [ Super=boost::multi_index::detail::index_node_base<int,boost::fast_pool_allocator<int>> ] C:\lib\Boost-1_35_0\include\boost-1_35\boost/multi_index_container.hpp(86) : see reference to class template instantiation 'boost::multi_index::detail::sequenced_index<SuperMeta,TagList>' being compiled with [ SuperMeta=boost::multi_index::detail::nth_layer<1,int,boost::multi_index::indexed_by<boost::multi_index::sequenced<>>,boost::fast_pool_allocator<int>>, TagList=boost::mpl::vector0<boost::mpl::na> ] .\relink.cpp(97) : see reference to class template instantiation 'boost::multi_index::multi_index_container<Value,IndexSpecifierList,Allocator>' being compiled with [ Value=int, IndexSpecifierList=boost::multi_index::indexed_by<boost::multi_index::sequenced<>>, Allocator=boost::fast_pool_allocator<int> ]

2 2

Re: [Boost-users] Applying Boost license to non boost projects (e r)
by Robert Marion 24 Apr '08

24 Apr '08

> i have written some primitives to be used with Boost.Accumulators, and > i'm considering to make them available. these files are mostly *.hpp and > depend on boost headers. under what license is it preferable that I > distribute them under? Hi, That depends on your intention. The BSD style and MIT style licenses are permissive, attribution licenses. The Boost is also a permissive license. GPL and LGPL are constrained so that anyone using your code will have to make theirs open source (depending on how it is used, in the later case). If you are writing *for* Boost, the responsible thing to do is use the Boost license. If not, and you are looking for a permissive attribution license, then I suggest either the BSD or MIT style because they are commonly known. There are many other licenses, such as Apache, CDDL, etc. If you want to guarantee your code remains open source, then GPL or LGPL are a good choice (but many persons/companies will choose not to use your code if it falls under one of those licenses). Yours, Robert Marion http://www.palamida.com

3 2

Re: [Boost-users] [iterators] NULL pointer initialization and comparison using iterator_facade
by Lang Stefan 24 Apr '08

24 Apr '08

> > > > Is it possible to construct metafunctions that allow a > function argument > > of NULL, but not any other integer value? Or at least cause > the compiler > > to call out whenever a non-NULL argument is provided? > > > > Cheers, > > Stefan > > struct convertible_from_null { > struct null_pointer_impl_t; > typedef null_pointer_impl_t > (convertible_from_null::*null_pointer_t)(); > convertible_from_null(null_pointer_t* = 0) {} > }; > > convertible_from_null test = 0; > > //convertible_from_null fail = 1; > > int main() {} > > In Christ, > Steven Watanabe > Thanks a lot, this did it! I also managed to adapt this construct to provide comparison operators for comparisons with NULL values, so all my issues have been resolved. :) So now, apart from direct pointer assignments and delete statements (which I know need to be fixed anyway) my array iterator mostly behaves like a pointer. Fortunately, thanks to Steven's suggestion, the compiler does now show up all the problematic statements. Cheers, Stefan

1 0

How to build and use boost 1.3.3.1 in visual studio 2008
by Ie Siong Fat 24 Apr '08

24 Apr '08

How to build and use boost 1.3.3.1 in visual studio 2008, I can't build boost. Can someone explain me ? -- Regards, Ie Siong Fat

1 0

Re: [Boost-users] [serialization]1.35problemwith explicitinstantiation
by Michael Seibt 23 Apr '08

23 Apr '08

Robert Ramey wrote: > Peter Dimov wrote: > > Robert Ramey: > >> Peter Dimov wrote: > > > >> I believe that making a *.cpp file with the macros > >> BOOST_CLASS_EXPORT(T) would be exactly equivalent to the above. > > > > Well, it depends on where the .cpp file is. If it's part of the static > > library, it will not be included by the linker because there are no > > references to it. But it is true that > > What I meant was that if you want to make sure the stuff in the > static library is linked in, just make a small *.cpp file as part > of your main program which includes the BOOST_CLASS_EXPORT > macros. > > #include "a.hpp" > #include "b.hpp" > ... > > BOOST_CLASS_EXPORT(A) > BOOST_CLASS_EXPORT(B) This *.cpp file won't be such little! If those headers contained templates each user of the static library had to compile the template instances again. (I have a multitude of executables and dynamic libraries that use the same template code.) If you have access to the object files linked into this static library you should link them directly into every using project instead of the static library (as suggested in earlier post). MSVC 2005 and 2008 have a linker option that does this trick if the static library is part of the same solution: "Use Library Dependency Inputs" at the tab "Linker | General". (I searched the options several times until I recognized that this option could be relevant...)

1 0