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

multi threaded resource repository
by Michael Marcin 14 Jul '08

14 Jul '08

I'm attempting to use boost thread's synchronization primitives to implement an efficient resource repository. The resource repository identifies immutable resources by an unique id. If the resource is cached it just returns a shared_ptr to the resource. If the resource is not yet loaded it does a potentially expensive construction on the resource stores it in the cache and then returns a shared_ptr to the resource. Resources can potentially request other resources from the repository during their construction so that case must be considered. The single-threaded version of this class is fairly straightforward but I don't have much experience in making things parallel so I'm struggling. I was hoping someone could offer suggestions on how to make an efficient multi threaded implementation. Single threaded version below: class ResourceStream; class Repository { typedef std::map< ResourceIdentifier, boost::weak_ptr<const void> > ResourceCache; public: template< typename Resource > boost::shared_ptr<const Resource> Acquire( ResourceIdentifier id ) { boost::shared_ptr<const Resource> resource; ResourceCache::iterator it = m_resourceCache.find(id); if( it != m_resourceCache.end() ) { boost::weak_ptr<const void> cachedResource = it->second; try { resource = boost::shared_static_cast<const Resource>( cachedResource.lock() ); } catch( boost::bad_weak_ptr& ) { } } if( !resource ) { resource.reset( new Resource( *this, Fetch(id) ) ); m_resourceCache.insert( ResourceCache::value_type( id, resource ) ); } return resource; } private: // magic that retrieve an unprocessed resource based on an id ResourceStream Fetch( ResourceIdentifier id ); ResourceCache m_resourceCache; }; Thanks, Michael Marcin

1 0

Re: [Boost-users] How to build 32-bit Boost libraries on a 64-bit Linux machine using bjam?
by Gehua Yang 14 Jul '08

14 Jul '08

Hi Folks, I am using Boost.Serialization and Boost.Thread in my project. Because current Boost release had problems with GCC 4.3, I had to build my own Boost libraries from a svn checkout. Fortunately, with the help from "Getting Started Guide", I accomplished the mission with ease. The commands I used are: bjam --build-dir=/tmp/build-boost --toolset=gcc --build-type="release <threading>multi <link>static <runtime-link>static" stage bjam --build-dir=/tmp/build-boost --toolset=gcc --build-type="release <threading>multi <link>static <runtime-link>static" --prefix=/software/boost-20080710 install The above produced 64-bit libraries. My question is, how I can build 32-bit Boost libraries on the same machine? Does it take command line variables, such as CXXFLAGS=-m32? Thanks. Gehua Yang

3 2

serialization of boolean values hardcoded to '1' and '0'
by Andrew McDonald 14 Jul '08

14 Jul '08

I was wondering why in the archive library the input and output text streams are hardcoded with the boolalpha bit as unset, e.g. (is >> std::noboolalpha;)? Specifically the code is in the directory: boost_1_35_0/boost/archive/impl/ In basic_text_iprimitive.ipp template<class IStream> BOOST_ARCHIVE_OR_WARCHIVE_DECL(BOOST_PP_EMPTY()) basic_text_iprimitive<IStream>::basic_text_iprimitive( IStream &is_, bool no_codecvt ) : is(is_), flags_saver(is_), precision_saver(is_), archive_locale(NULL), locale_saver(is_) { if(! no_codecvt){ archive_locale.reset( add_facet( std::locale::classic(), new codecvt_null<BOOST_DEDUCED_TYPENAME IStream::char_type> ) ); is.imbue(* archive_locale); } is >> std::noboolalpha; } and basic_text_oprimitive.ipp template<class OStream> BOOST_ARCHIVE_OR_WARCHIVE_DECL(BOOST_PP_EMPTY()) basic_text_oprimitive<OStream>::basic_text_oprimitive( OStream & os_, bool no_codecvt ) : os(os_), flags_saver(os_), precision_saver(os_), archive_locale(NULL), locale_saver(os_) { if(! no_codecvt){ archive_locale.reset( add_facet( std::locale::classic(), new codecvt_null<BOOST_DEDUCED_TYPENAME OStream::char_type> ) ); os.imbue(* archive_locale); } os << std::noboolalpha; } Andrew McDonald Software Engineer "Zufall ist nur der Ausdruck unserer Unfähigkeit, den Dingen auf Grund zu kommen" - Albert Einstein

2 2

Boost Python : Generic Function Call
by William Marié 14 Jul '08

14 Jul '08

Hi, I would like to find a way to call Python functions from C++ in a generic way. With Python you can use PyObject_CallObject( PyObject* pFun, PyObject* pyArg ) where pyArgs is a tuple, this allows the user to genericaly call Python functions assuming that you create the right tuple before calling PyObject_CallObject. With boost python i tried to do the same thing but i can't find a way to do it, using the call function or directly the () operator i always have to pass all the arguments. Is there a easy way to do that using boost python ? -- View this message in context: http://www.nabble.com/Boost-Python-%3A-Generic-Function-Call-tp18431821p184… Sent from the Boost - Users mailing list archive at Nabble.com.

2 1

Making a range from a single element?
by Mathias Gaunard 14 Jul '08

14 Jul '08

Is there an utility within Boost.Range (or Range_ex) that allows making a range of Ts from a single T? It seems quite useful and practical. Basically a std::make_pair(&foo, &foo+1) works but requires foo to be an lvalue.

2 1

Unexpected warning relating to boost::function, and size_t / unsigned
by Matthew Bourdua 14 Jul '08

14 Jul '08

Hello, I'm receiving a compiler warning (MSVC8.0) that does not make sense to me, and I wonder if this might be a boost related issue: #include <vector> #include <functional> #include <boost/function.hpp> int main() { std::vector< std::size_t > sizes; std::vector< unsigned > data; boost::function< bool (unsigned, unsigned) > fn = std::greater< unsigned >(); fn( data.front(), 10 ); // NOTE: data is empty - will cause runtime error } The warning I get is: warning C4267: 'argument' : conversion from 'size_t' to 'unsigned int', possible loss of data referring to the last line in main(). This warning goes away if I do either of the following: - Comment out the std::vector< std::size_t > sizes - this one baffles me. - Use a 'naked' std::greater< unsigned >() instead of a boost::function< bool (unsigned, unsigned) >() object. I wonder if size_t is somehow redefined in boost in a way that might be causing me some trouble? Any help would be appreciated, thanks. Matt

3 3

Re: [Boost-users] Unexpected warning relating to boost::function, and size_t / unsigned
by Anders Dalvander 14 Jul '08

14 Jul '08

Hi Matt, Disable the /Wp64 switch. It is deprecated, as it gave false-positives and false-negatives, for explanation see: http://blogs.msdn.com/vcblog/archive/2007/08/10/the-future-of-the-c-languag… Regards, Anders Dalvander > Hello, > > I'm receiving a compiler warning (MSVC8.0) that does not make sense to > me, and I wonder if this might be a boost related issue: > > int main() > { > std::vector< std::size_t > sizes; > std::vector< unsigned > data; > > boost::function< bool (unsigned, unsigned) > fn = std::greater< > unsigned >(); > fn( data.front(), 10 ); // NOTE: data is empty - will cause runtime > error > } > > The warning I get is: > warning C4267: 'argument' : conversion from 'size_t' to 'unsigned int', > possible loss of data > referring to the last line in main(). > > Matt

1 0

About max in boost
by Peng Yu 14 Jul '08

14 Jul '08

Hi, I'm wondering if there is a max function boost, which accept arbitrary number of arguments. For example, max(a, b) max(a, b, c) max(a, b, ...., z) Thanks, Peng

2 1

Accessing vertices by their name
by Anthony Labarre 13 Jul '08

13 Jul '08

X-No-Archive: yes Hi all, I'm interested in adding edges to a graph whose vertices are labelled, and I would like to do this using the vertex NAMES instead of their indices (or "descriptors"). I know that add_edge uses vertex_descriptor, so I probably won't be able to achieve what I want directly, but I thought property maps would allow to easily recover the vertex_descriptor corresponding to a given label. Unfortunately, although I have spent a lot of time looking for that in the documentation, I didn't succeed in achieving my goal. My graph is declared as follows: typedef adjacency_list< vecS, vecS, undirectedS, property< vertex_name_t, std::string > > Graph; Graph G(0); I hoped that boost::property_map<Graph, vertex_name_t>::type vertex_names = get(vertex_name, G); would be exactly what I needed, but it seems that vertex_names can only be accessed using vertex descriptors and returns strings, which is exactly the opposite of what I need. If anyone has ideas, I'd love to hear them. Thank you so much, Anthony -- http://homepages.ulb.ac.be/~alabarre/

1 0

[BGL] kolmogorov_max_flow: can't compile even the example
by Niko Vuokko 13 Jul '08

13 Jul '08

I'm trying to compile the example for kolmogorov_max_flow in Graph library, but it gives me an error (more precisely 1023K worth of errors...). Tried compiling on Cygwin, GCC 3.4.4, g++ -c -I/path/to/boost -O2 example.cpp. I don't really know how to look for the problem, I tried compiling also with gSTLfilt (filters STL errors), but couldn't find anything. Any ideas?

3 5