- Boost-users - lists.boost.org

[serialization] Skipping unexpected elements in XML archive
by Todd Greer 22 Jul '06

22 Jul '06

Hi, I'm using the Serialization library, but need to skip unrecognized data when reading in an archive. This is for forward compatability purposes; I need for older versions of my software to consume files produced by newer ones, which may have added elements. As this functionality is not currently in the library, I have created a new archive type (as I think Robert Ramey suggested at some point). It inherits from xml_iarchive_impl and overrides load_override to identify when it has just read an opening tag with an unexpected name. It then skips whatever XML it finds within the unexpected element, reads the closing tag, and returns to normal processing. I use it with the unaltered xml_oarchive, and the pair satisfy my requirements. I'm including it below in the hopes that someone will find it useful. I've tested it on MSVC 7.1 (Visual Studio .NET 2003) with a reasonably complex XML file with both nested and non-nested unexpected elements. If you'd like, I can post some tests as well (I just need to strip out some project-specific details). Please let me know if you see any places where it could be improved, or where I've completely missed something; this is my first use of the serialization library, and my understanding is far from comprehensive. Many thanks to all who've contributed to the Serialization library, Todd Greer <tgreer <at> affinegy dot com> Senior Software Developer, Affinegy LLC #include <boost/archive/xml_iarchive.hpp> #include <boost/range/end.hpp> #include <algorithm> const char whitespace[] = " \t\n\r"; //Skip whitespace, then return true if we're looking at '</'. //Leave the stream right before the '<'. template<class IStream> bool at_close_tag(IStream& is) { char ch; using boost::end; do { ch = is.get(); } while(std::find(whitespace, end(whitespace), ch) != end(whitespace)); assert(ch == '<'); ch = is.get(); bool const end_tag = ch == '/'; is.putback(ch); is.putback('<'); return end_tag; } template<class Archive> class skip_xml_iarchive_impl: public boost::archive::xml_iarchive_impl<Archive> { #if BOOST_WORKAROUND(BOOST_MSVC, <= 1300) //see xml_iarchive_impl public: #elif defined(BOOST_MSVC) friend boost::archive::detail::interface_iarchive<Archive>; protected: #else friend class boost::archive::detail::interface_iarchive<Archive>; protected: #endif skip_xml_iarchive_impl(std::istream & is, unsigned int flags = 0): xml_iarchive_impl<Archive>(is, flags) {} void skip_all_elements() { std::string ignore; if(this->This()->gimpl->parse_string(this->This()->get_is(), ignore) && ignore.npos != ignore.find_first_not_of(whitespace)) return; //We're either looking at a start tag or an end tag. while(!at_close_tag(this->This()->get_is())) {//It's a start tag. if(!this->This()->gimpl->parse_start_tag(this->This()->get_is())) boost::throw_exception(archive_exception(archive_exception::stream_error )); ++depth; std::string const name = this->This()->gimpl->rv.object_name; skip_all_elements(); load_end(name.c_str()); } } template<class T> void load_override( #ifndef BOOST_NO_FUNCTION_TEMPLATE_ORDERING const #endif boost::serialization::nvp<T> & t, int ){ if(!t.name()) return xml_iarchive_impl<Archive>::load_override(t, 0); load_start(t.name()); // don't check start tag at highest level const std::string name_found = this->This()->gimpl->rv.object_name; bool const skip = depth > 1 && 0 != (this->get_flags() & no_xml_tag_checking) && name_found != t.name(); if(skip) skip_all_elements(); else boost::archive::load(* this->This(), t.value()); load_end(skip ? name_found.c_str() : t.name()); if(skip) //We still haven't loaded t... load_override(t, 0); //...so load t. } template<class T> void load_override(T & t, BOOST_PFTO int i) { xml_iarchive_impl<Archive>::load_override(t, i); } }; class skip_xml_iarchive: public skip_xml_iarchive_impl<skip_xml_iarchive> { public: skip_xml_iarchive(std::istream & is, unsigned int flags = 0) : skip_xml_iarchive_impl<skip_xml_iarchive>(is, flags) {} }; -- Todd Greer <tgreer <at> affinegy dot com> Senior Software Developer, Affinegy LLC

3 2

bjam crashes when building Boost on Windows
by Adam Wright 22 Jul '06

22 Jul '06

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi all, When trying to build boost 1.33.1 this afternoon, I encountered a problem with bjam just crashing every time I invoked it when targeting Windows. I've debugged the problem as being files with dates < 1900 in directories being searched by the glob expansions (I had a rogue file from 1757 in C:\Windows\). file_dirscan in filent.c invokes the MSCRT function _tfindnext64i32, which will eventually assert if the date is too far back in time. It's a documented limitation of the function, so my suggestion would be to FAQ this "bug" for the (very few) people it's likely to bite. Thanks, adamw -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.3 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFEwlW/9mihfAI44D0RAtyDAJ46wfH9UBhsz58nNTtKTyftI0l1igCghVD1 s18ec8+vbIevjIskC40W9Rc= =i7Ap -----END PGP SIGNATURE-----

1 0

[mpl] Combining Type Sequences
by Eric Lemings 22 Jul '06

22 Jul '06

Giving the following code: #include <boost/mpl/integral_c.hpp> namespace mpl = boost::mpl; template < unsigned U, int I = 1 > struct T : mpl::integral_c< unsigned, U > { BOOST_STATIC_CONST(int, N = I); }; and two vectors of type T of varying lengths: typedef mpl::vector < T<0>, T<1>, T<2> > t1; typedef mpl::vector < T<2, 2>, T<0, 0> > t2; How would you create a joint view of t1 and t2, sorted by increasing U values, where all elements of t1 and t2 that have the same U value have their N constants added together. In the example above, the resulting type t0 would be equivalent to: typedef mpl::vector < T<0, 1>, T<1, 1>, T<2, 3> > t0; Thanks, Eric.

2 1

[iterator]counting_iterator question
by Sean Huang 22 Jul '06

22 Jul '06

I encountered some problems while using counting_iterator. Suppose we have a range called RangeType, and we define: typedef boost::counting_iterator< boost::range_value< RangeType >::type > MyCountingIterator; if I do: std::vector< int > vec( 100 ); std::vector< int >:: iterator iter = vec.begin() + ( countingIter2 - counitingIter1 ); VC8 would then complain about a conversion from __int64 to int because the difference type of MyCountingIterator (using default) is __int64. Since we treat warnings as errors. I then changed MyCountingIterator to typedef boost::counting_iterator< boost::range_value< RangeType >::type, boost::use_default, boost::range_difference< RangeType >::type > MyCountingIterator; This time I got a different error: iterator_facade.hpp(554) : error C2784: '__w64 int boost::counting_iterator<Incrementable,CategoryOrTraversal,Difference>::distance_to(const boost::counting_iterator<OtherIncrementable> &) const' : could not deduce template argument for 'const boost::counting_iterator<OtherIncrementable> &' from 'const boost::counting_iterator<Incrementable,CategoryOrTraversal,Difference>' My question is why limiting the last three template parameters to default in counting_iterator::distance_to. Thanks, Sean

2 3

cast shared_array<T> to shared_array<const T>
by Jarda Grešula 21 Jul '06

21 Jul '06

Hello, is there any reason, why the conversion below is not supported, whereas the same conversion works for shared_ptr? shared_array<int> i; shared_array<const int> ci(i); // not possible Thanks, Jarda.

1 0

newbie:BoostGraphLibrary - brandes_betweenness -> closenness
by Ioannis Filippis 21 Jul '06

21 Jul '06

Hi all, I wonder whether it is possible to calculate the closenness centrality using the brandes_betweenness_centrality. The closeness centrality of a vertex is defined by the inverse of the average length of the shortest paths to/from all the other vertices in the graph. I know that in theory this is possible and it would actually save a lot of time if someone wants to calculate both properties. Currently I am running brandes_betweenness like this brandes_betweenness_centrality(g, get(&VertexInfo::betweenness, g), get(&EdgeInfo::betweenness, g)); I prefer bundled properties as for me are more clear in concept. I wonder whether: a.I could use Util parameters like DistanceMap to calculate closenness. However, I can not really understand how I can create an iterator_property_map using bundled properties. OR b.I have to modify the brandes_betweenness_centrality to calculate also closenness. I apologise in case the answer to my problem is trivial but I am just newbie in the Boost Graph Library. Thanks for your help. Ioannis

1 0

BGL serialization
by Camilo Melani 21 Jul '06

21 Jul '06

Hello, I'd like to save and restore a adjacency_list graph in a XML file, Does anyone has some example code or can help me? I can serialize std::map and others classes, but when I try to serialize the graph I get compiler errors. Thank for the help!! Camilo Some part of the code: #include <boost/config.hpp> #include <boost/graph/adjacency_list.hpp> #include <boost/graph/adj_list_serialize.hpp> #include <boost/dynamic_bitset.hpp> #include <boost/archive/xml_iarchive.hpp> #include <boost/serialization/map.hpp> typedef adjacency_list< vecS, vecS, bidirectionalS, no_property, no_property > Graph; template<class Archive> void serialize(Archive & ar, const unsigned int version) { ar & BOOST_SERIALIZATION_NVP(graph); } Error!!! C:\Boost\include\boost-1_35\boost/archive/basic_xml_oarchive.hpp(86) : error C2664: 'boost::mpl::assertion_failed' : cannot convert parameter 1 from 'boost::mpl::failed ************boost::serialization::is_wrapper<T>::* ***********' to 'boost::mpl::assert<false>::type' with [ T=boost::no_property ] No constructor could take the source type, or constructor overload resolution was ambiguous __________________________________________________ Preguntá. Respondé. Descubrí. Todo lo que querías saber, y lo que ni imaginabas, está en Yahoo! Respuestas (Beta). ¡Probalo ya! http://www.yahoo.com.ar/respuestas

1 0

Re: [Boost-users] [Serialization] Problem serializing via baseclassptr
by Guy Létourneau 21 Jul '06

21 Jul '06

Any updates on my problem Robert? Thanks, Guy > >Robert, > >I've created a simple test as you suggested. The problem is that the test >does not even build (in oserializer.hpp, there are 2 conditions for this to >happen but none seems to match my test..). Here's what the compiler gives >me: > >C:/Boost/include/boost-1_33_1/boost/archive/detail/oserializer.hpp: In >function `void boost::archive::save(Archive&, T&) [with Archive = >boost::archive::text_oarchive, T = A]': >C:/Boost/include/boost-1_33_1/boost/archive/basic_text_oarchive.hpp:78: >instantiated from `void >boost::archive::basic_text_oarchive<Archive>::save_override(T&, int) [with >T = A, Archive = boost::archive::text_oarchive]' >1_33_1/boost/archive/detail/interface_oarchive.hpp:78: instantiated from >`Archive& >boost::archive::detail::interface_oarchive<Archive>::operator<<(T&) [with T >= A, Archive = boost::archive::text_oarchive]' >../main.cpp:90: instantiated from here >C:/Boost/include/boost-1_33_1/boost/archive/detail/oserializer.hpp:567: >error: incomplete type `boost::STATIC_ASSERTION_FAILURE<false>' does not >have member `value' >#include <iostream> >#include <fstream> >#include <boost/archive/text_oarchive.hpp> >#include <boost/archive/text_iarchive.hpp> >#include <boost/serialization/export.hpp> > >class A >{ > public: > > A() > { > i = 1; > }; > > virtual ~A() > { > }; > > virtual void Dummy() = 0; > > private: > > int i; >}; >BOOST_IS_ABSTRACT(A) > >class B : public A >{ > public: > > B() > { > }; > > virtual ~B() > { > }; > > void Dummy() > { > }; > > void SetValue( std::string& value ) > { > Data = value; > } > > private: > > friend > class boost::serialization::access; > > template <class Archive> > void serialize( Archive& ar, const unsigned int version ) > { > ar & boost::serialization::base_object<A>( *this ); > ar& Data; > std::cout << "Serialization of B" << std::endl; > std::cout.flush(); > } > > std::string Data; >}; >BOOST_CLASS_EXPORT(B); >int main() >{ > // Create a B object > B* Child = new B; > > // Set some data > std::string Data = "test"; > Child->SetValue( Data ); > > // Assign it to a A ptr > A* Parent = Child; > > // create and open a character archive for output > std::ofstream ofs( "filename" ); > > // save data to archive > { > try > { > boost::archive::text_oarchive oa( ofs ); > // write class instance to archive > oa << *Parent; > } > catch( boost::archive::archive_exception& e ) > { > std::cout << "Save failed! : " << e.what(); > std::cout.flush(); > > return false; > } > catch( ... ) > { > std::cerr << "Unknown Exception " << std::endl; > > return false; > } > > // archive and stream closed when destructors are called > } > > delete Child; >} > >The example is very simple. Again, I have problems only when serializing >via a base class ptr (serializing object "directly" works fine). > >Thanks again for your help, > >Guy >Hmm - looks like it should work to me. > >I would try the following: > >a) make a tiny test - a main which which does nothing but tests the >serialization. >This you can ship to us in necessary. >b) For now - remove all the BOOST_CLASS_TRACKING macros. >It shouldn't make a difference but let's start simple. >c) You'll have to add BOOST_CLASS_EXPORT for SomeData >d) double check that everything is in the same namespace. >e) try removing the friend class boost::serialization::access; from >the base class. NonVolatileData::serialize(..) should never be called >by the serialization library in this case - so if this traps during >compile, >it will show a chain of "compile time calls". By looking at this chain >one can see where things went wrong. Unfortunately this requires >looking at the header source code - oh well. >f) double check that the test corresponding to your example builds >and runs in your environment. If that passes, but yours does not, try >modifying the test bit by bit by commenting out the code that is there >and replacing it with your own code. If you finish this and the test >still works - then just remove all the commented code and you'll be >done. > >Robert Ramey Guy Létourneau wrote: > Robert, > I've tried to include your changes but it doesn't solve the problem. > I had to add the following line: BOOST_IS_ABSTRACT(NonVolatileData) > for the program to compile. I agree with you, to have an abstract > class is what is needed in that case as the NonVolatileData class is > not intended to do something useful except for allowing my data > storage layer to be generic. However, abstract or not, I still have > the same problem: when serializing my SomeData object through a > pointer to the NonVolatileData, only the data of the parent class > gets serialized. Anything else you want me to try? > > Thanks, > Guy > >> Try making the base class polymorphic - that it define at least >> one abstract function. See my changes below. This will have >> the beneficial effect of inhibiting anyone from creating an instance >> of your base class directly. >> Robert Ramey > >> Guy Létourneau wrote: >>> Hello, >>> >>> I need to develop a generic data storage layer. Its job is to handle >>> the serialization of all objects that need to be persistent in my >>> system and handle management of archives, backups and restores and >>> so forth. The most important point is that it must stay generic, >>> i.e. have no knowledge of the type of objects being passed by other >>> modules in the system. Because of this, I define a base class >>> NonVolatileData from which all data classes in the system will >>> derive: >>> >>> /////////////////////////////////////////////////////////////////////// >>> // Base class >>> class NonVolatileData >>> { >>> friend >>> class boost::serialization::access; >>> >>> public: >>> >>> NonVolatileData(){ i = 1; }; >>> virtual ~NonVolatileData() = 0; // Note - pure virtual function >>> make class abstract- >>> >>> >>> protected: >>> >>> virtual void Dummy() =0; // Note - pure virtual function make >>> class abstract- >> ; >>> >>> int i; >>> template <class Archive> >>> void serialize( Archive& ar, const unsigned int version ) >>> { >>> ar & i; >>> } >>> }; >>> BOOST_CLASS_TRACKING(NonVolatileData, >>> boost::serialization::track_never) >>> >>> >>> class SomeData: public NonVolatileData >>> { >>> friend >>> class boost::serialization::access; >>> >>> public: >>> >>> SomeData () >>> { >>> }; >>> >>> ~SomeData () >>> { >>> }; >>> >>> void SetValue( std::string& Address ) >>> { >>> this->Address = Address; >>> } >>> >>> void Dummy() >>> { >>> } >>> >>> private: >>> >>> template <class Archive> >>> void serialize( Archive& ar, const unsigned int version ) >>> { >>> ar & >>> >>> boost::serialization::base_object<NonVolatileData>(*this); >>> ar& Address; } >>> >>> std::string Address; >>> }; >>> BOOST_CLASS_TRACKING(SomeData, boost::serialization::track_never) >>> ///////////////////////////////////////////////////////////////// >>> >>> My data storage interface will take a ptr to the base class i.e. >>> Save(NonVolatileData* data) so the serialization will be performed >>> through a ptr to the base class. I've read the documentation >>> multiple times and I'm still unable to make that work correctly. >>> When a do a save, only the base class is serialized even though the >>> base class is polymorphic. I've tried using the >>> BOOST_CLASS_EXPORT_GUID macro the force a registration of my derived >>> class without success. I've also checked the example from demo.cpp. >>> It uses the direct call to register_type() but in my case, I don't >>> want to change the code everytime a new class is defined in the >>> system so I went with option 2 (the macro). There's probably a >>> subtle detail I do not do correctly (polite way of saying I'm >>> missing something :-)). >>> >>> Thanks for your help, >>> >>> Guy Letourneau =

1 0

Re: [Boost-users] [serialization] Skipping unexpected elements in XMLarchive
by Todd Greer 21 Jul '06

21 Jul '06

Peter wrote: > The additional need I'm trying to solve is that the XML elements can be > reordered, i.e. come in a different order than the serialization lib > expects them. You could store whatever XML elements you skip. Then, every time you get a request to read something, first check the stored XML. You would need to have a stack of storage locations to correspond to how deep you are in the XML hierarchy and discard the XML elements on the top of the stack when you parse an end tag. I think that would work, as long as you don't mind having to buffer some XML. Another approach would be to store pointers to the variables that are passed into the serialization library, rather than storing the XML. Then, when reading an XML open tag, look in your set of stored pointers for a match and deserialize into it. I'm not sure whether or not this could be made to work with the current Serialization library. Basically, to support reordering, you need to buffer one side or the other. Buffering the XML would probably work better with the current Serialization library, but would be less efficient. HTH, Todd Greer <tgreer <at> affinegy dot com> Senior Software Developer, Affinegy LLC

1 0

Re: [Boost-users] [python], [regex] Segmentation fault when deallocating match_results in a python extension module
by bernhard.maeder＠zkb.ch 21 Jul '06

21 Jul '06

Arrr, I've now compiled the example on different machines (Windows, MacOS) and it seems I'm only getting into troubles on my target system (Debian)... I've used ldd to get the library dependencies of my extension modules; the difference between building with gcc3.3 and gcc3.4 is with the libstdc++, which is used in version 5 and 6, respectively. But since python doesn't use this library, this should make no difference at all. Is there anything else I could possibly try? Thanks Bernhard "John Maddock" <john(a)johnmaddock.co.uk> Gesendet von: boost-users-bounces(a)lists.boost.org 19.07.2006 13:53 Bitte antworten an boost-users An: <boost-users(a)lists.boost.org> Kopie: Thema: Re: [Boost-users] [python], [regex] Segmentation fault when deallocating match_results in a python extension module bernhard.maeder(a)zkb.ch wrote: >> Hello everybody >> The program aborts when exiting test_func(), during deallocation of >> match_results. The error occurs somewhere in _M_deallocate of >> stl_vector.h. >> >> Can anybody help me with this problem? What could possibly go wrong >> here? In theory not much :-( >> One thing I noted: python is compiled using GCC3.3 on our system, >> while we are using GCC3.4, is this a problem? (But, trying the same >> on a system with GCC 4 for both, python and our app, also fails, so >> it seems like the problem is to be sought elsewhere). It does sound like the std::vector is allocating with one one allocator and deallocating with another. But I know nothing about python extension modules and what can go wrong there. John. _______________________________________________ Boost-users mailing list Boost-users(a)lists.boost.org http://lists.boost.org/mailman/listinfo.cgi/boost-users

2 1