November 2009 - Boost-users - lists.preview.boost.org

Serializing of object with map and multimap. Questions.
by B Hart 13 Nov '09

13 Nov '09

I've been having some problems storing and loading a class with a map and a multimap. The map contains values that are simple objects of another class. My questions are below. Here is also a link to the same question posted on stackoverflow. Thanks. http://stackoverflow.com/questions/1676781/boost-serializing-of-object-cont… See below a main() and two very simple classes. Then per Boost serialization (and what is shown) my questions are: 1) Does class B need the normal overloaded stream insertion operators '<<' and '>>' to be defined? Currently in my real code it doesn't have these. 2) Does class A in the store() and load() methods need to iterate through the map and multimap containers explicitly, storing/loading their key:value pairs explicitely? e.g. something like: void A::store(const char* filename){ std::ofstream ofs(filename); boost::archive::text_oarchive oa(ofs); std::map< std::string, B >::iterator it; BMap.size(); oa << BMap.size(); for( it = BMap.begin(); it != BMap.end(); it++ ){ oa << it->first; oa << it->second; } //similar for strMultimap } I assume that I don't need to do this, but am not certain. 3) Assuming class B has only the two data members shown, does it need a default contructor included explicitely? (as opposed to the implicit default constructor) 4) Does B need to have an overide for the comparison operator '>'? I assume that it doesn't since this is a very simple class. Finally, any other comments per anything that I've failed to cover is appreciated! Example code for my above questions: //includes ommitted int main() { std::string file("test.dat"); A * pA = new A; pA->store(file.c_str()); pA->fillMaps(); //release data pA->load(file.c_str()); return 0; } //includes ommitted class A { friend class boost::serialization::access; public: std::map< std::string, B > BMap; std::multimap< std::string, std::string > strMultimap; template<class Archive> void serialize(Archive & ar, const unsigned int version) { ar & BMap; ar & strMultimap; } void store(const char* filename){ std::ofstream ofs(filename); boost::archive::text_oarchive oa(ofs); oa << this; } void load(const char* filename){ std::ifstream ifs(filename); boost::archive::text_iarchive ia(ifs); ia >> this; } void fillMaps(){ //code to allocate B objects and put them in BMap and fill strMultimap with whatever number of key:value pairs } class B { friend class boost::serialization::access; public: std::string str; unsigned int num; B::B(void) : str("a string") , num(7) { } template<class Archive> void serialize(Archive & ar, const unsigned int version) { ar & str; ar & num; } }

5 8

Building on Mac with Snow Leopard
by DeNigris Sean 13 Nov '09

13 Nov '09

Hi list, I downloaded 1.40.0 and started to build on my mac, but "./bootstrap.sh --help" gave me back "command not found" in Terminal. It turned out that the permissions for bootstrap.sh were wrong (executable bit not set), so I had to "chmod 755 bootstrap.sh" first. Just a heads-up if anyone else runs into this. Cheers! Sean DeNigris sean(a)clipperadams.com

3 4

[Serialization] Serialization of function pointers
by elizabeta petreska 12 Nov '09

12 Nov '09

Hello everybody This is my first post to the list. Can I serialize C++ function pointer ? I search around and found that it could not be done ( although I am not sure ). If it can not be done, whats the explanation about it ? Thanks

6 5

BJam scans entire disk for compilation
by Parijat Garg 12 Nov '09

12 Nov '09

Hi! I recently migrated all my C++ project compilation to BJam. While it has saved me a lot of hassles, I found that BJam takes too much time preparing for compilation. I ran an strace and found that BJam is scanning through my entire disk - perhaps looking for compilation targets. Is this a usual thing or are my compilation scripts likely to be misconfigured? Help will be much appreciated. Parijat

2 1

[Review] We're halfway through the GGL review
by Hartmut Kaiser 12 Nov '09

12 Nov '09

Hey all, We're halfway through the GGL review (see attached review announcement) it is planned to end November 15th. Now would be a good time to send in your reviews! Regards Hartmut ------------------- Meet me at BoostCon http://boostcon.com

1 0

[BGL] Schultes and Sanders Contraction Hierarchies Routing
by Stephen Woodbridge 12 Nov '09

12 Nov '09

Hi all, Has anyone looked at supporting Drs Schultes and Sanders Contraction Hierarchies Routing using Boost Graph? See here for more info on this: http://algo2.iti.kit.edu/english/routeplanning.php This requires lots of preprocessing of the graph, but offers 10-100 times faster performance on shortest path solutions. If Andrew, Jeremiah or others have time to look at some of the papers, I would be interested in your thoughts on supporting this or even knowing that BGL might be the wrong tool to implement this with. Thanks for any feedback you might be able to offer. Best regards, -Stephen Woodbridge

4 5

Exceptions and throwing across module boundaries
by Josh Faust 12 Nov '09

12 Nov '09

One of the problems with exceptions that we've run into recently is throwing across module boundaries -- we hit a case where an exception thrown in one module cannot be caught as anything but a std::exception in a module compiled with different settings (same compiler -- gcc 4.2). This is of course item #62 in "C++ Coding Standards". Since we'd prefer to use exceptions rather than return codes as our error reporting mechanism, I'm currently investigating possible solutions to this. I was hoping to find something applicable boost, but it looks like boost doesn't differentiate between throwing in a header vs. throwing in a module -- for example, Boost.Thread has a few throw statements in cpp files (both in posix and win32). Is this just an issue that hasn't been enough of a problem to address? The approach I am considering is wrapping module interfaces with a translation layer which turns an exception to an error code, which is re-translated back to an exception in the header used by client code. Has anyone done anything similar? Is there something in boost like this already? (I was hoping to find something in Boost.Exception, but it does not appear to solve this situation). Thanks Josh

2 3

[Fusion] Views of Associative Sequences
by James Knight 11 Nov '09

11 Nov '09

I'm trying to find a solution to the following problem. If I have an associative sequence such as a fusion::set and I make a filter view of it, the view is not a model of an associative sequence and does not support the has_key and at_key functions. I can use fusion::as_set on the view to get an associative sequence again, but this doesn't really have the effect I want as it results in a set holding copies of the original data whereas the view referenced the original data. Example below. Would it be possible to create a view which supported the associative sequence interface? Any suggested alternative approaches? Thanks Nate ~~~~~~~~~~~~~~~~~ #include <boost/fusion/include/set.hpp> #include <boost/fusion/include/intrinsic.hpp> #include <boost/fusion/include/filter_view.hpp> #include <string> #include <boost/type_traits/is_arithmetic.hpp> #include <boost/mpl/placeholders.hpp> namespace fusion = boost::fusion; using boost::mpl::_; int main() { typedef fusion::set<int,double,std::string> set_t; set_t s(1,2.0,"hello"); fusion::filter_view<set_t,boost::is_arithmetic<_>> filtered_view(s); fusion::front(filtered_view) = 4; //modifies s fusion::result_of::as_set<fusion::filter_view<set_t,boost::is_arithmetic<_> > >::type filtered_set(filtered_view); fusion::at_key<int>(filtered_set) = 16; // modifies the copied data in filtered_set. }

5 11

naming delayed variables string in boost lamdba
by Emanuele Rocci 11 Nov '09

11 Nov '09

Hi AllI am getting familiar with boost::lambda.Today I learnt about delayed variable and usage of boost::constant in lambda expression.Do you know why when I try to declare delayed variables for string like I do belowthe compiler gives me each time an error?I tried the following constant_type<const char*>::type _msg_a(constant("my msg"));constant_type<const std::string>::type _msg_b(constant("my msg"));constant_type<std::string>::type _msg_c(constant("my msg")); If I use constant( "my msg" ) inside a boost lambda expression everythingis fine but I cannot declare correctly a delayed variable.Do you know how to fix it or where I can get some details?Thanks in advanceMn _________________________________________________________________ Facile, veloce, sicuro: scarica Internet Explorer 8 per MSN http://www.microsoft.com/italy/windows/internet-explorer/msn.aspx

4 5

[string-algo] split() input
by Igor R 11 Nov '09

11 Nov '09

The following is split() function declaration: template<typename SequenceSequenceT, typename RangeT, typename PredicateT> SequenceSequenceT & split(SequenceSequenceT &Result, RangeT &Input, PredicateT Pred, token_compress_mode_type eCompress = token_compress_off); Why the 2nd parameter (Input) is declared as a non-const reference? Is it changed by the function? The problem is that it doesn't compile (gcc 4.2) when Input is a temporary.

1 0