January 2007 - Boost-users - lists.preview.boost.org

Mapping from type to value
by Johan Torp 04 Jan '07

04 Jan '07

What is the best practice for creating a map from a type to a value? Currently I do it like this: ===================================== struct SomeMap { template<typename T> static const std::string& LookUp(); }; template <> const std::string& SomeMap::LookUp<int>() { static const std::string s("int"); return s; } ===================================== ...but this seems far from optimal. -- System Developer TouchTable AB Tel: +46 (0)31 773 68 12 johan.torp(a)touchtable.se

2 2

Simple MPL question: Generate code from type sequence
by Johan Torp 04 Jan '07

04 Jan '07

I would like to generate one call to a templated function for each type in an MPL sequence. Which is the best way to do this? E.g.: typedef vector<float,double,long double> SomeVector; template <typename T> write_size() { std::cout << sizeof(T) << " "; } Should somehow expand to: write_size<float>(); write_size<double>(); write_size<long double>(); I have managed to do this "manually" in a fairly general way, what I'm looking for is some part of boost which replaces my ForEach struct below or some nicer way of writing the same code: ============================================== struct WriteSize { template<typename T> void Call() {write_size<T>();} }; template <class It, class End> struct ForEach { template <class Func> static void Call(Func& func) { func.Call<typename It::type>(); ForEach<typename next<It>::type, End>::Call(func); } }; template <class End> struct ForEach<End, End> { template <class Func> static void Call(Func&) {} }; // Then I can generate the calls by WriteSize ws; ForEach<begin<SomeVector;>::type, end<SomeVector;>::type>::Call(ws); ============================================== -- System Developer TouchTable AB Tel: +46 (0)31 773 68 12 johan.torp(a)touchtable.se

2 1

[BGL] How to get at a specific vertex and check out degree?
by Ferng, Andrew D 04 Jan '07

04 Jan '07

Hi all, I've been reading the docs and have gotten more and more confused. Figure I better stop and ask for directions. So far, I've been successful in using the following (basically copied the connected component example): adjacency_list<vecS, vecS, undirectedS> G; vector<int> linkPath(num_vertices(G)); int num = connected_components(G, &linkPath[0]); cout << "Total number of linkPaths: " << num << endl; So from the output of the algorithm, I examine all the different paths found and are interested in certain vertices. For example, for some component, I want to know if a certain vertex has degrees over 2. I'm using undirected so I assume out_degree will give me the total degrees? >From the docs: out_degree(vertex_descriptor u, const adjacency_list& g) and graph_traits<adjacency_list>::vertex_descriptor I think the 2nd param in out_degree is just G. But what about the first param? Is that like an iterator? how do I specify the vertex that I'm interested in? Thanks!

1 1

Re: [Boost-users] [Boost.Multi-Index] ICE w/ VC7.1
by Marc Viala 03 Jan '07

03 Jan '07

De la part de John Maddock > Marc Viala wrote: > > Before trying to recompile some codes from B.MI/test to crash my > > compiler w/ an ICE, I've just played with some compilation options in > > my IDE and the results are: > > > > In Debug mode: > > If the C/C++ options "/Zi" & "/Gm" are set: KO -> w/ ICE in > > apply_wrap.hpp > > If the C/C++ options "/ZI" & "/Gm" are set: KO -> w/ ICE in > > apply_wrap.hpp > > If the C/C++ option "/Gm" is not set and "Zi" or "ZI" is set: OK > > > > > > In Release mode: > > with or without the "/Gm" option set: OK > > > > Funny... It seems that the /Gm option is quite unstable in debug > > mode? I don't know exactly the goal of this option but our project > > template set this option by default in this mode. > > > > Joaquin or John, do you know the meaning of this compilation option > > for VC7.1? Do you know the recommended options w/ this compiler to > > use Boost libraries, more specifically B.MI? For information, up to > > now, we use "/ZI" or "/Zi" in conjunction with "/Gm" option with all > > other Boost libraries. > > With apologies I think I misread your original post: I took your post to > mean you were seeing a runtime crash, but you mean an internal compiler > error: and I did see that at one point when testing your original code. I > use the same project options as you BTW /Zi and /Gm: the /Zi enables build > and continue, /Gm enables a minimal rebuild (compile only the changes to a > source). Generally they work quite well, but if you see an internal > compiler error the trick is just to do a clean, then rebuild, and that > usually clears the problem. Of course cleaning and rebuilding a large > project can wipe out the advantages that /Zi /Gm normally bring :-( Thanks a lot this explanation... It does clarify for me the signification of these options. Now that I'm aware of /Gm option behaviour I would pay more attention. > > Strangely I was unable to reproduce the internal error this morning no > matter options I used: it seems to be very dependent on whatever VC has in > it's cache. > Yes, that's a pity... I can now go ahead with B.MI. Best regards Marc

1 0

Re: [Boost-users] [boost] Boost threads + MSVSC + /Za + client code
by Sohail Somani 03 Jan '07

03 Jan '07

Hi Anthony, Thanks for your response. > if BOOST_DISABLE_WIN32 is > defined and > BOOST_HAS_PTHREADS is not; the implication is that if you > can't use the > Windows headers, then you can't use the Windows thread API, > so if we haven't > got pthread-win32 configured, you can't use Boost threads. This pretty much sums it up, yes. > I am not sure what we can do about this. One issue is that > currently the > thread lib requires windows.h, and therefore compiler > extensions, when the > library sources are built. I'm not sure whether it is safe to > link something > compiled with extensions on against something compiled with > extensions off --- > is the ABI the same? If it *is* safe, then we should remove > this define from > boost/config/suffix.hpp, since the boost thread headers don't > explicitly > include windows.h. I think this is correct but I think the restriction only makes sense if you are building the thread library. Atleast the 1.33.1 code appears to have been written such that headers don't require language extensions. My gut feeling is that a solution like this should work: // BOOST_BUILDING is defined whenever building a boost library #if defined(BOOST_BUILDING) && defined(BOOST_MSVC) # define BOOST_REQUIRE_MSVC_LANGUAGE_EXTENSIONS #endif #include <boost/config/requires_threads.hpp> #if defined(BOOST_REQUIRE_MSVC_LANGUAGE_EXTENSIONS) # undef BOOST_REQUIRE_MSVC_LANGUAGE_EXTENSIONS #endif > OTOH, someone put the restriction there for a reason, and I > wouldn't want to > remove it without careful thought. Of course. Thanks again, Sohail

1 0

Using serialization of a class with a wrapped shared_ptr - compil ation and run-time issues !!!
by Jon Belinfante 03 Jan '07

03 Jan '07

Hi I am at present introducing myself to Boost Serialization which from my first impressions of is very impressive to use and I like what I see. I have an ongoing set of classes to serialiise and find most of the problems I have found can be resolved from the Boost documentation. However one problem I have is that a huge bulk of my classes used boost::smart_ptrs each of which seem to add an extra level of complexity and at present I can not see my own resolution to. I have one simple case of a smart string class - as shown below which causes me a problem. This source code is to create a smart string class which uses a wrapped shared_ptr to a std::string - the idea behind this Smart String class is to reduce memory sizes for huge quantity of repeated strings in an application. -Please see source code. This class works fine apart the serialization which either does not compile or crashes with assertion errors. Please see below source code. As this code stands with line in the serialization block ar & m_stringPtr; I get a long compilation error - c:\Boost\boost_1_33_1\boost\serialization\shared_ptr.hpp(235) : error C2027: use of undefined type 'boost::STATIC_ASSERTION_FAILURE<x>' - with it complatining with the line oa << a; which is in the main code block. However if I change to code ar & * m_stringPtr; The code compiles correctly and it serializes the SmartString - however it run-time throws in trying to de-serialise at ia >> b; Any ideas here where I am going wrong - I would be most greatful. Thanks /********************************************/ #include <set> #include <vector> #include <string> #include <boost/shared_ptr.hpp> #include <boost/scoped_ptr.hpp> #include <boost/archive/text_oarchive.hpp> #include <boost/archive/text_iarchive.hpp> #include <boost/serialization/shared_ptr.hpp> BOOST_SERIALIZATION_SHARED_PTR(std::string) /** Smart String Class Basically holds a String only by a ptr and the pimpl method. Idea is to only ever store a shared_ptr and let boost do a reference count on its usage. Idea to of course save time/speed of string storage. */ class SmartString { public: /** overriden stream operator */ friend std::ostream& operator<< (std::ostream &os, const SmartString &cSmartString); SmartString() { m_stringPtr.reset(); } SmartString( const std::string& s) { boost::shared_ptr< std::string > tempPtr( new std::string( s) ); m_stringPtr = tempPtr; } SmartString( const char* s) { boost::shared_ptr< std::string > tempPtr( new std::string( s) ); m_stringPtr = tempPtr; } SmartString( const SmartString& other) { m_stringPtr = other.m_stringPtr; } SmartString& operator=(const SmartString& other) { m_stringPtr = other.m_stringPtr; return *this; } inline SmartString& operator=( const std::string& otherStr) { boost::shared_ptr< std::string > tempPtr( new std::string( otherStr) ); m_stringPtr = tempPtr; return *this; } bool operator() ( const SmartString& t1, const SmartString& t2) const; inline bool operator<( const SmartString& otherStr) { const std::string other= *otherStr.m_stringPtr; return *m_stringPtr < other; } inline bool operator<=( const SmartString& otherStr) { const std::string other= *otherStr.m_stringPtr; return *m_stringPtr <= other; } inline bool operator>( const SmartString& otherStr) { const std::string other= *otherStr.m_stringPtr; return *m_stringPtr > other; } inline bool operator>=( const SmartString& otherStr) { const std::string other= *otherStr.m_stringPtr; return *m_stringPtr >= other; } inline bool operator==( const SmartString& otherStr ) { return *m_stringPtr == *otherStr.m_stringPtr; } inline bool operator==( const SmartString& otherStr ) const { return *m_stringPtr == *otherStr.m_stringPtr; } inline bool operator==( const std::string& otherStr ) { return *m_stringPtr == otherStr; } inline bool operator==( const char* c ) { int x = ::strcmp( (*m_stringPtr).c_str(), c); return x? false:true; } inline const std::string& getString( ) const { return *m_stringPtr; } inline bool operator!=( const SmartString& otherStr) { const std::string other= *otherStr.m_stringPtr; return *m_stringPtr != other; } inline bool operator!= ( const std::string& otherStr) { return *m_stringPtr != otherStr; } inline const char& operator[] (const int index) const { if ( (index >=0) && ( index <= static_cast<int>((*m_stringPtr).size() ))) return (*m_stringPtr)[index]; else { std::string message( "Out of range index on " + (*m_stringPtr) ); throw std::out_of_range(message); } } /** overriden + operation to retunn addition of two strings */ friend SmartString operator+(const SmartString& s1,const SmartString &s2); virtual ~SmartString(){} friend class boost::serialization::access; protected: template <class Archive> void serialize ( Archive& ar, const unsigned long file_version) { ar & m_stringPtr; } boost::shared_ptr<std::string> m_stringPtr; }; std::ostream& operator<< (std::ostream &os, const SmartString& cSmartString) { const std::string str( cSmartString.getString() ); os << str.c_str(); return os; } SmartString operator+(const SmartString& s1,const SmartString &s2) { std::string str = s1.getString(); str += s2.getString(); return SmartString ( str.c_str() ); } int main( int, const char*[] ) { const SmartString a("ABC"); const char* fileName = "smartStringFile"; { std::ofstream ofs(fileName); boost::archive::text_oarchive oa( ofs ); oa << a; } SmartString b; { std::ifstream ifs( fileName, std::ios::binary ); boost::archive::text_iarchive ia( ifs ); ia >> b; } } /********************************************/ ********************************************************** Mitsubishi UFJ Securities International plc ("MUSI") is registered in England, company number 1698498 and registered office at 6 Broadgate, London EC2M 2AA. MUSI is part of the Mitsubishi UFJ Financial Group and is authorised and regulated by The Financial Services Authority. This message is intended solely for the individual addressee named above. The information contained in this e-mail is confidential and may be legally privileged. If you are not the intended recipient please delete in its entirety. Messages sent via this medium may be subject to delays, non-delivery and unauthorised alteration. The information contained herein or attached hereto has been obtained from sources we believe to be reliable but we do not represent that it is accurate or complete. Any reference to past performance should not be taken as an indication of future performance. The information contained herein or attached hereto is not to be construed as an offer or solicitation to buy or sell any security, instrument or investment. MUSI or any affiliated company, may have an interest, position, or effect transactions, in any investment mentioned herein. Any opinions or recommendations expressed herein are solely those of the author or analyst and are subject to change without notice.

2 1

Using filter_iterator in a for loop
by David Ward 03 Jan '07

03 Jan '07

I want to use a boost::filter_iterator to iterate through a std::vector using a for-style loop. I haven't been able to find any examples of the proper way to do this in the Boost documentation (or how to do this for any other Boost iterator, for that matter). >From searching this newsgroup, I found that it is possible to do: for (boost::filter_iterator< ... > iter(myVector.begin(), myVector.end()); iter != boost::filter_iterator< ... >(myVector.end()); ++iter ) { ... } However, iter is quite aware of when it has reached the end of the vector, since that is the second parameter in its constructor. Is it really necessary to create a second filter_iterator, set it to the end of the vector, and then test iter against it? Can't I somehow test iter alone to see if it has reached the end of the vector? If I use std::for_each, it needs both a beginning and ending iterator as parameters, leading to the same situation as above. I've also looked at the Boost Range library and BOOST_FOREACH (described at http://www.boost.org/regression-logs/cs-win32_metacomm/doc/html/foreach.html). However, from my understanding, I don't think either of these apply to or work with Boost iterators. Whatever method is appropriate, I would ask that someone please exapnd the Boost iterator documentation to include an example of using Boost iterators in a for-style loop. I feel that this is a very basic and common usage of iterators which was not addressed. Thank you. -- David Ward Communications and Networking Division Information Technology and Telecommunications Laboratory Georgia Tech Research Institute

2 1

Re: [Boost-users] [Boost.Multi-Index] ICE w/ VC7.1
by Marc Viala 03 Jan '07

03 Jan '07

Hello Joaquin & John, > > > So I gave up at that point. Adding your example programs to the > > same VC7.1 IDE project did compile so I don't think it's a > > setup issue. > > That's fine, your help has been invaluable. It definitely > looks like a local problem, I hope Marc can provide me with > more info wrt his particular environment. Thank you very much, Thanks for your help John. Before trying to recompile some codes from B.MI/test to crash my compiler w/ an ICE, I've just played with some compilation options in my IDE and the results are: In Debug mode: If the C/C++ options "/Zi" & "/Gm" are set: KO -> w/ ICE in apply_wrap.hpp If the C/C++ options "/ZI" & "/Gm" are set: KO -> w/ ICE in apply_wrap.hpp If the C/C++ option "/Gm" is not set and "Zi" or "ZI" is set: OK In Release mode: with or without the "/Gm" option set: OK Funny... It seems that the /Gm option is quite unstable in debug mode? I don't know exactly the goal of this option but our project template set this option by default in this mode. Joaquin or John, do you know the meaning of this compilation option for VC7.1? Do you know the recommended options w/ this compiler to use Boost libraries, more specifically B.MI? For information, up to now, we use "/ZI" or "/Zi" in conjunction with "/Gm" option with all other Boost libraries. Thanks for your help. The code snipped during my compilation test is the following: #include <boost/config.hpp> #include <boost/multi_index_container.hpp> #include <boost/multi_index/composite_key.hpp> #include <boost/multi_index/ordered_index.hpp> #include <boost/multi_index/member.hpp> #include <boost/multi_index/key_extractors.hpp> // STL #include <string> using namespace boost::multi_index ; struct Item { const std::string& getCls() const { return _cls ; } const std::string& getId() const { return _id ; } std::string _cls ; std::string _id ; int _i ; int _j ; } ; void main() { typedef const_mem_fun< Item , const std::string& , &Item::getCls > key_cls ; typedef const_mem_fun< Item , const std::string& , &Item::getId > key_id ; typedef composite_key<Item,key_cls,key_id> ckey_clsid ; typedef multi_index_container< Item, indexed_by< ordered_non_unique<key_id> , ordered_non_unique<key_cls> , ordered_unique<ckey_clsid> > > ObjectContainer ; ObjectContainer c ; }

2 1

Re: [Boost-users] Boost threads + MSVSC + /Za + client code
by Sohail Somani 03 Jan '07

03 Jan '07

Hi, Any takers? The workaround I've found for the problem below is to #define _MSC_EXTENSIONS before including any boost thread headers. It /appears/ that this is only necessary when separately compiling boost threads. Sorry for the cross-post. > With Visual C++, I can't seem to determine why if I simply include > boost/threads/once.hpp, I need to enable language extensions for that > translation unit. I'm going to dig into this further today but I would > appreciate if someone has figured out a solution short of allowing > language extensions. I can understand requiring it for building boost > threads (as windows.h seems to have some extended syntax) but I'm not > sure of the logic behind requiring client code to use it as well. Thanks! Sohail

2 1

[threads] Compiling boost.threads on Linux
by Hugo Vinicius Alves Soares 03 Jan '07

03 Jan '07

Hey everybody, I'm a newbie on Boost Libraies, I'm starting with the thread library. I'm using Ubuntu and a install it using apt-get, the headers are in /usr/include/boost/ but when I use g++ to compile my test codes an error message is returned: undefined reference to methods of the library classes. I know this errors are link errors, what option I must pass to g++? Thanks for everyone that answer my ask. Hugo Vinícius __________________________________________________ Fale com seus amigos de graça com o novo Yahoo! Messenger http://br.messenger.yahoo.com/

2 1