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

some questions regarding ref
by abir basak 03 Feb '07

03 Feb '07

Hi, I want to have a few classes to be stored inside a vector by reference. I thought to use boost::ref for that purpose , instead of storing pointer. However I am facing some problem regarding the operator & . Here is a sample code which gives the problem, struct T2{ int x; T2(int x) :x(x){} }; int main(int argc,char** argv) { std::vector<boost::reference_wrapper<T2> > v; /// vector stores reference, rather than value. T2 t1 = 10; T2 t2 = 15; v.push_back(boost::ref(t1)); /// push ref to t1. v.push_back(boost::ref(t2)); ///push ref to t2. v[0].get().x = 5; ///get by ref & change. /// v[0].x = 5 ; this statement doesn't work! std::cout<<t1.x<<"\n"; ///change is reflected. return 0; } I am not sure why I have to use get for this purpose, while & operator is written for that purpose. Using MSVC 7.1 Even I face problem like this, struct T3{ T2 t2_; T3(int x) : t2_(x){} T2& operator&(){ return t2_; } }; std::vector<T3> v3; v3.push_back(T3(4)); v3.push_back(T3(5)); /// v3[0].x = 6; This doesn't work. (&v3[0]).x = 6; ///This works! Can anyone say a little detail about the problem ? -- Abir Basak, Member IEEE Software Engineer, Read Ink Technologies B. Tech, IIT Kharagpur email: abir(a)abirbasak.com homepage: www.abirbasak.com

2 1

boost::serialization throws random exceptions
by Adrian Grigore 03 Feb '07

03 Feb '07

Hi, I'm experiencing occasional problems with boost::serialization when trying to write then read two simple classes, to a file. I am using boost 1.33.1 Most of the time everything's working fine, but sometimes the serialized data seems to be corrupt and boost::archive throws an exception when I try to load it back into the program. The error occurs about about 1 in 10 times I try this, and seems to depend on the exact values of the variables being saved. Here's a call stack of the thrown application: WxNonogram.exe!_CxxThrowException(void * pExceptionObject=0x00129680, const _s__ThrowInfo * pThrowInfo=0x00ff4954) + 0x50 bytes C++ WxNonogram.exe!boost::throw_exception<boost::archive::archive_exception>(const boost::archive::archive_exception & e={...}) + 0x40 bytes C++ WxNonogram.exe!boost::archive::basic_binary_iprimitive<boost::archive::binary_iarchive,std::basic_istream<char,std::char_traits<char> > >::load_binary(void * address=0x0012d05c, unsigned int count=4) + 0xaf bytes C++ WxNonogram.exe!boost::archive::basic_binary_iprimitive<boost::archive::binary_iarchive,std::basic_istream<char,std::char_traits<char> > >::load<int>(int & t=-858993460) + 0x31 bytes C++ WxNonogram.exe!boost::archive::load_access::load_primitive<boost::archive::binary_iarchive,int>(boost::archive::binary_iarchive & ar={...}, int & t=-858993460) + 0x2d bytes C++ WxNonogram.exe!boost::archive::detail::load_non_pointer_type<boost::archive::binary_iarchive,int>::load_primitive::invoke(boost::archive::binary_iarchive & ar={...}, int & t=-858993460) + 0x2b bytes C++ WxNonogram.exe!boost::archive::detail::load_non_pointer_type<boost::archive::binary_iarchive,int>::invoke(boost::archive::binary_iarchive & ar={...}, int & t=-858993460) + 0x2b bytes C++ WxNonogram.exe!boost::archive::load<boost::archive::binary_iarchive,int>(boost::archive::binary_iarchive & ar={...}, int & t=-858993460) + 0x2b bytes C++ WxNonogram.exe!boost::archive::basic_binary_iarchive<boost::archive::binary_iarchive>::load_override<int>(int & t=-858993460, int __formal=0) + 0x38 bytes C++ WxNonogram.exe!boost::archive::binary_iarchive_impl<boost::archive::binary_iarchive>::load_override<int>(int & t=-858993460, int __formal=0) + 0x31 bytes C++ WxNonogram.exe!boost::archive::detail::interface_iarchive<boost::archive::binary_iarchive>::operator>><int>(int & t=-858993460) + 0x38 bytes C++ WxNonogram.exe!boost::archive::detail::interface_iarchive<boost::archive::binary_iarchive>::operator&<int>(int & t=-858993460) + 0x39 bytes C++ WxNonogram.exe!NoNo::TGameProgress::TProfileState::serialize<boost::archive::binary_iarchive>(boost::archive::binary_iarchive & ar={...}, const unsigned int Version=4) + 0x7a bytes C++ Please help! I'm at a loss on how to proceed. Thanks for your time, Adrian -- Adrian Grigore adrian (AT) lobstersoft.com

2 3

How to build Boost libs with stlportstld versions of stlport??
by Manish Mangal 03 Feb '07

03 Feb '07

Hi How Do I build boost libs with stlportstld dll of stlport when I build the debug versions of the boost libs. I do not want to use the stlportd versions . The command that I use for building boost is: jam "-sTOOLS=vc-8_0-stlport" "-sSTLPORT_VERSION=5.0.0" "-sSTLPORT_5.0.0_PATH=C:\SourceLibrary\STLport" "-sBUILD=debug release <runtime-link>dynamic <threading>multi <stlport-iostream>on <native-wchar_t>on" install Manish Mangal Software Engineer atlas(r) search Partner for Results 206-816-8349 // phone <mailto:t-Matt.Koch@AtlasSolutions.com> Manish.Mangal(a)AtlasSolutions.com www.AtlasSolutions.com <outbind://116-000000001368D4B5F8F6D0488191444C76FD7C9507008921F4CB773B0 545AAED425DEF553BC300003A3500100000DB0C51F64BED3447AA6EC585072C4EC300000 1E3B7A70000/jenniferh/Local%20Settings/Temporary%20Internet%20Files/OLK3 /www.AtlasSolutions.com> // web

1 0

[Filesystem] Folders with spaces in the name
by Nathan Pierluissi 02 Feb '07

02 Feb '07

Greetings: I am having trouble using the /= operator to open folders with spaces in the name, ex. "My Documents" or "Program Files". Here is an example code: path full_path("c:\"); string Fname="Program Files"; full_path/=Fname; It works for any Fname (folder name) that doesn't have a space. Let me know if you need more information. Thanks in advance. -N. Pierluissi --------------------------------- Never Miss an Email Stay connected with Yahoo! Mail on your mobile. Get started!

1 0

boost::interprocess::node_allocator problem (VC8, release build)
by alexmark 02 Feb '07

02 Feb '07

I have a problem with node_allocator. The "repro" code: #include <boost/interprocess/managed_external_buffer.hpp> #include <boost/interprocess/allocators/node_allocator.hpp> #include <boost/interprocess/containers/list.hpp> //////////////// using namespace boost::interprocess; //////////////// enum { A_SHARED_MEMORY_SIZE = 4096 * 50, A_POOL_SIZE = 512, }; //////////////// static char g_static_buffer[A_SHARED_MEMORY_SIZE]; static managed_external_buffer g_sharm (create_only, g_static_buffer, A_SHARED_MEMORY_SIZE); //////////////// typedef node_allocator<int, A_POOL_SIZE, managed_external_buffer::segment_manager> int_allocator_t; typedef list<int, int_allocator_t> IntList; //////////////// struct T { IntList int_list; // T () : int_list (int_allocator_t(g_sharm.get_segment_manager())) {} }; typedef node_allocator<T, A_POOL_SIZE, managed_external_buffer::segment_manager> tis_allocator_t; typedef list<T, tis_allocator_t> TList; //////////////// int main() { TList* p = g_sharm.construct<TList>(anonymous_instance)(tis_allocator_t(g_sharm.get_segment_manager())); try { enum { N = 2 }; // Create objects then delete them for (int n = 0; n < N; n++) { // Create T-type objects for (int i = 0; i < A_POOL_SIZE; i++) { p->push_back(T()); // push new object to the list } p->clear(); // Delete all T objects } } catch (bad_alloc &) { cout << "No memory" << endl; } g_sharm.destroy_ptr(p); return 0; } The problem is the folowing. If N > 1, on _second_ "pass" (n=1) the bad_alloc exception is thrown while pushing new T object to the list. Why? I supposed that on any "pass", except the first one, no new T-type objects are actually created in shared memory. They should be taken from the pool instead. I.e., if there's enough memory on the first "pass", it must be also enough on any other one. The problem is specific for Visual Studio 2005, when producing release build (all options carry their default values). Thanks, Alex.

3 6

How to change a local variable in the if_then algorithm
by Meryl Silverburgh 02 Feb '07

02 Feb '07

Hi, I have a sorted int vector 'orgWidth'. I would like to save the save into a new int vector 'newWidth' if the value of the current element is 10% bigger than the previous one. For example, orgWidth int vector : 300 366 729 769 781 1297 newWidth int vector: 300 366 729 1297 I know how to add the value to the newWidth vector using the if_then with boost. But my question is how can I update the value of the 'previousWidth' as I use for_each() algorithm. int previousWidth = -1; vector<int> orgWidth; for_each(orgWidth.begin(), orgWidth.end(), if_then( (_1 > (previousWidth * 1.1)) , bind (&vector<int>::push_back,boost::ref(newWidth), _1 ))); Thanks for any help.

2 1

installing boost at work
by liamv7 02 Feb '07

02 Feb '07

Hi, I want to use boost at work. What's the best way to make this available to our developers? The options I see: 1. Put the include files and built libs/dlls in version control. I don't see much point to this and binaries take up a lot a space and it's annoying when upgrading. 2. Zip up the include files and dlls and have developers unzip in appropriate location. 3. Let developers access boost over network. To slow. How do you guys manage this? Cheers, Liam.

2 1

Serialization and const
by Anders Sundman 02 Feb '07

02 Feb '07

Hi! I'm having some problems grasping how the serialization library works with respect to const/non-const data. Ex: Consider the following simple class that I would like to serialize: struct Simple { Simple() : member(0) {} Simple(int m) : member(m) {} int member; }; BOOST_CLASS_EXPORT(Simple); BOOST_SERIALIZATION_SPLIT_FREE(Simple); //BOOST_CLASS_TRACKING(Simple, boost::serialization::track_never); namespace boost { namespace serialization { template<class Archive> void save(Archive & ar, const Simple & t, unsigned int version) { ar << t.member; } template<class Archive> void load(Archive & ar, Simple & t, unsigned int version) { ar >> t.member; } } } But when I do the serialization, strange things happen. Ex: const Simple s1(7); //Simple s1(7); boost::archive::text_oarchive oa(ss, f); oa << s1; The non-const alternative above doesn't work unless I do BOOST_CLASS_TRACKING(Simple, boost::serialization::track_never); And, Simple * const s1 = new Simple(7); //Simple * s1 = new Simple(7); boost::archive::text_oarchive oa(ss, f); oa << s1; The non-const alternative above doesn't work even if I do BOOST_CLASS_TRACKING(Simple, boost::serialization::track_never); I'm obviously missing something here, please help me out. Best regards, Anders Sundman

2 2

Re: [Boost-users] installing boost at work
by Nat Goodspeed 02 Feb '07

02 Feb '07

________________________________________ From: boost-users-bounces(a)lists.boost.org [mailto:boost-users-bounces@lists.boost.org] On Behalf Of Sohail Somani Sent: Friday, February 02, 2007 10:40 AM To: boost-users(a)lists.boost.org; boost-users(a)lists.boost.org Subject: Re: [Boost-users] installing boost at work -----Original Message----- From: boost-users-bounces(a)lists.boost.org on behalf of Nat Goodspeed [Nat] we checked the Boost sources & includes into a subtree in our version-control system. we created platform-specific project files and build them along with our own source code. ------------- I do the same thing but I'm curious why you wouldn't just use bjam to build boost? [Nat] That was the first thing we tried, but it failed spectacularly, with voluminous and mysterious output. After a few hours of fruitless twiddling, punctuated by exclamations at the opacity of bjam config files, command line syntax and documentation in general, our frantic combing of the Boost documentation turned up the interesting fact that it also endorses the idea of building Boost libraries with your own toolchain. We grabbed that ball and ran with it, and have never looked back. My colleagues and I have used a lot of computer languages and are not afraid of learning new syntax. But concerning bjam, I'd better stop now lest this message devolve into outright flame. ;-) I like the *idea* of a truly cross-platform build tool. But traffic on this list suggests that, rather than making life easier, bjam is a major stumbling block for a significant number of would-be Boost adopters. To be fair, there's room to hope that the build system v2 is significantly easier to use. None of us here have any desire to investigate, though. And the build-problem messages from other people keep coming in. [Sohail] One thing I do is also check in the binaries per platform/compiler/variant and the build scripts are smart enough to know where to look. The logic behind this was that there is no point rebuilding as 90% of the developers would use the same platform and that the builds would just build our code. [Nat] I've worked with homebrew build systems in which dependency analysis can fetch unmodified executable/library modules even for your own code. It's a good optimization, though it can be a bit finicky to get it right. And you do have to establish the convention that in case of a mysterious crash, do a full rebuild and try again. That said, as I noted earlier, most of the time Boost modules are only built the FIRST time a given developer fetches them from the version-control system. The cost quickly amortizes away.

1 0

Re: [Boost-users] installing boost at work
by Nat Goodspeed 02 Feb '07

02 Feb '07

> -----Original Message----- > From: boost-users-bounces(a)lists.boost.org [mailto:boost-users- > bounces(a)lists.boost.org] On Behalf Of liamv7 > Sent: Friday, February 02, 2007 5:04 AM > To: Boost-users(a)lists.boost.org > Subject: [Boost-users] installing boost at work > > Hi, I want to use boost at work. What's the best way to make this > available to our developers? The options I see: > 1. Put the include files and built libs/dlls in version control. I > don't see much point to this and binaries take up a lot a space and > it's annoying when upgrading. > 2. Zip up the include files and dlls and have developers unzip in > appropriate location. > 3. Let developers access boost over network. To slow. > > How do you guys manage this? [Nat] Probably not what you want to hear, but we checked the Boost sources & includes into a subtree in our version-control system. For the few Boost libraries we use that require building, we created platform-specific project files and build them along with our own source code. Naturally they don't require rebuilding as often as our stuff, but it makes it more uniform for a new developer to get started. This also allows us to use the VCS (Subversion in our case) to manage the problem of merging local patches with new Boost releases.

2 1

[multi_index] no non-const reference even for simple sorted container
by Markus Werle 02 Feb '07

02 Feb '07

> Hi! I tried to use multi_index_container as a drop in replacement for std::map and hoped that I could obtain non-const references to the values therein. Although returning a non-const reference might not be a problem for a multi_index_container with only 1 sorting criterium the lib and the compiler insist on const &. Workarounds beyond modify(...) available? Markus #include <boost/tuple/tuple.hpp> #include <boost/multi_index_container.hpp> #include <boost/multi_index/sequenced_index.hpp> #include <boost/multi_index/key_extractors.hpp> #include <boost/multi_index/ordered_index.hpp> #include <string> namespace bm = ::boost::multi_index; // http://lists.boost.org/boost-users/2005/12/16088.php template<typename Tuple,int N> struct tuple_member_extractor { typedef typename boost::tuples::element<N,Tuple>::type result_type; const result_type& operator()(const Tuple& t) const { return boost::tuples::get<N>(t); } }; typedef boost::tuple<std::string, std::string> element; typedef bm::multi_index_container< element, bm::indexed_by<bm::ordered_unique <tuple_member_extractor<element, 0> > > > multi_t; int main() { multi_t test; test.insert(boost::make_tuple("a", "aa")); // will not compile element & ref = test.find("a"); std::string & s = ref.get<1>(); }

2 1

[bind] how to get a funtor of a function since the function's template argument is a functor
by hongleij＠126.com 02 Feb '07

02 Feb '07

// how to get the functor of function caller below,which could be used in a boost::thread or other condition #include <boost/bind.hpp> #include <iostream> class BufferMoniter { public: void maitain() { std::cout <<" BufferMoniter" << std::endl; } }; class LocalPeer { public: template <class T> void caller(unsigned int interval,T t) { //do something t(); } void fork() { BufferMoniter *pBuffer=new BufferMoniter( ); caller(500,boost::bind(&BufferMoniter::maitain,pBuffer)); //I want a functor of it //boost::thread thrd( ); //insert here //thre.join(); } }; //best wishes

2 1

Re: [Boost-users] Serialization of dynamic data
by Pranckevicius, Tadas 02 Feb '07

02 Feb '07

I have just noticed, shouldn't it be ~Dyn() { delete [] data; } -----Original Message----- From: Anders Sundman [mailto:anders.sundman@optonova.se] Sent: 01 February 2007 14:55 To: boost-users(a)lists.boost.org Subject: [Boost-users] Serialization of dynamic data Hi! About the boost::serialization library, what is the best way to handle dynamic data in classes? Any idéas, suggestions or caveats? I didn't manage to find anything about this in the docs. Consider the following sample class: struct Dyn { Dyn(int s) : data(0), size(s) { data = new char[size]; } ~Dyn() { delete data; } char* data; int size; private: friend class boost::serialization::access; template< class Archive > void serialize( Archive &ar, unsigned int ver ) { // -> ar & data; What to do here? ar & size; } }; What to do with the data pointer? Is there any way to determine (in the serialize function) if a serialization or "de-serialization" is taking place? And have different behaviours - either allocating a new char[x] or copying data to the archive? Is there any way to store dynamic data in an archive? Best regards, Anders Sundman _______________________________________________ Boost-users mailing list Boost-users(a)lists.boost.org http://lists.boost.org/mailman/listinfo.cgi/boost-users

2 1

[threads] thread_group::create_thread() locks?
by Manuel Jung 02 Feb '07

02 Feb '07

Hi, My program uses a thread group for handling threads. these are joined, when the program ist killed, which should happen not so often, because its used to be a daemon. Most time it works just as expected. but sometimes after running the program for some minutes and after creating a bunch of threads the function thread_group::create_thread() locks. Is this possible? i think it shoulnt lokc, but it does. Im sorry, i cannot show you code cause its closed source and its a liddle bit too big and also works most of the time. I hope a answer to this question will allready help fixing. Thanks for reading. Cheers Manuel Jung

2 1

help with filesystem
by Polder, Matthew J 01 Feb '07

01 Feb '07

Hi, I started using filesystem, regex, and bind today. It's been interesting. I'm using Visual C++ 8.0 on an XP machine that has network filesystems mounted to it, some using letters (H:\, Y:\, etc.), some accessable via two slashes (//linden/temp/, etc). I'd like to peruse these using boost::filesystem. The following works: std::string t = "h:\\temp"; boost::filesystem::path p(t, boost::filesystem::native); but I can't get filesystem to work with the string "h:/temp"; std::string s = p.string(); // s = "h:/temp" bool b = boost::filesystem::portable_posix_name(s); //returns false b = boost::filesystem::windows_name(s); //returns false b = boost::filesystem::portable_name(s); //returns false b = boost::filesystem::portable_directory_name(s); //returns false b = boost::filesystem::portbale_file_name(s); //returns false b = boost::filesystem::native(s); //returns false Also, I'd like to do the following std::string q = "//linden/temp/"; boost::filesystem::path p2(q); Any help is appreciated. thanks, matthew

3 3

help with boost::bind and boost::regex
by Polder, Matthew J 01 Feb '07

01 Feb '07

I am trying to do the following: std::vector< std::string > filenames; //list of filenames retrieved using filesystem boost::regex r(".*txt$"); std::vector< std::string >::const_iterator iter = std::find_if( filenames.begin(), filenames.end(), boost::bind(boost::regex_match, _1, r) ); //Doesn't compile. I'm trying to find strings that match (or don't match) a particular regular expression. Help in properly using boost::bind and negating it would be helpful. thanks, matthew

2 1

Build only static library with -fPIC option.
by Bo Peng 01 Feb '07

01 Feb '07

Dear list, I am building boost cvs. For my specific application, I need to build only static libraries with -fPIC option. I tried command > bjam -j4 --prefix=/usr "-sBUILD=release <runtime-link>static <cxxflags>-fPIC" --with-serialization --with-iostreams install But I still get .a and .so libraries under /usr/lib. I used -n option (an advise I get from one of my previous posts) and the building commands are like: "g++" -ftemplate-depth-128 -O3 -finline-functions -Wno-inline -Wall -DBOOST_ALL_NO_LIB=1 -DNDEBUG -I"." -c -o "bin.v2/libs/serialization/build/gcc-3.4.6/release/link-static/binary_wiarchive.o" "libs/serialization/src/binary_wiarchive.cpp" so no -fPIC is passed. I tried bjam -h, but get even more confused because I am told.. -n Don't actually execute the updating actions. I guess cvs/bjam handles these options totally differently than boost 1.33.1, but how can I get the latest information? Many thanks in advance. Bo

1 0

[bind] how to get the type after bind
by hongleij＠126.com 01 Feb '07

01 Feb '07

//my question is how to write the type ??? below to make it run . #include <boost/bind.hpp> #include <iostream> using boost::bind; void print() { std::cout <<" print" << std::endl; } template <class T> void func(T t) { std::cout <<"func1(T t) : " ; t(); } int main() { func(bind(&print)); //ok bind ( &func<???>, bind(&print) ) ; //don't know how to write it }

2 2

Re: [Boost-users] [bind] how to get the type after bind
by Nat Goodspeed 01 Feb '07

01 Feb '07

________________________________________ From: boost-users-bounces(a)lists.boost.org [mailto:boost-users-bounces@lists.boost.org] On Behalf Of hongleij(a)126.com Sent: Thursday, February 01, 2007 9:00 AM To: boost-users(a)lists.bo... Subject: [Boost-users] [bind] how to get the type after bind //my question is how to write the type ??? below to make it run . #include <boost/bind.hpp> #include <iostream> using boost::bind; void print() { std::cout <<" print" << std::endl; } template <class T> void func(T t) { std::cout <<"func1(T t) : " ; t(); } int main() { func(bind(&print)); //ok bind ( &func<???>, bind(&print) ) ; //don't know how to write it } [Nat] This doesn't directly answer your question; the type produced by bind() is obscure. But would the following suit your need? bind ( &func< boost::function<void ()> >, bind(&print) ) ;

1 0

[boost-thread] why not writ e "sleep" in a better
by hongleij＠126.com 01 Feb '07

01 Feb '07

hi: it really a trouble thing for me to use boost::thread::sleep() : boost::xtime helloworldtime; boost::xtime_get(&helloworldtime, boost::TIME_UTC); helloworldtime.sec += 6; boost::thread::sleep(helloworldtime); i think "boost::thread::sleep(1500) //millisecond " may easy my programming;if have such interface; now i use : void thread_sleep(unsigned int interval) //millisecond { boost::xtime xt; boost::xtime_get(&xt, boost::TIME_UTC); int sec=interval/1000; int msec=interval-1000*sec; xt.nsec += 1000*1000*msec; xt.sec+= sec; boost::thread::sleep(xt); } best wishes;-)

2 1

mapped_file_sink
by Alf Kraus 01 Feb '07

01 Feb '07

does anyone have an example of using *mapped_file_sink *to create and write into a memory mapped file?

1 0

mapped_file_sink
by Alf Kraus 01 Feb '07

01 Feb '07

I'm trying to create memory mapped files using mapped_file_sink: mapped_file_params mfParams( fileName ); mfParams.new_file_size = filesize; mapped_file_sink mfSink( mfParams ); trying to use any of its funtions like: const char *ptStart = mfSink.data(); the compiler tells me that I have no access to them for they are private members since inherited from mapped_file I know thats a piece of cake but just cant solve it. Cheers Alf

1 0

problem with boost/iterator/zip_iterator.hpp
by Tim Axelrod 01 Feb '07

01 Feb '07

Hello, If more than one source file includes boost/iterator/zip_iterator.hpp, the loader complains about a multiply defined function: tsa@mops2.lsstcorp.org:~/VW/TSA> g++ -o nothing nothingMain.cc nothing.cc /tmp/ccT5Bn0N.o: In function `boost::detail::tuple_impl_specific::tuple_equal(boost::tuples::null_type, boost::tuples::null_type)': nothing.cc:(.text+0x0): multiple definition of `boost::detail::tuple_impl_specific::tuple_equal(boost::tuples::null_type, boost::tuples::null_type)' /tmp/ccH8Z9eR.o:nothingMain.cc:(.text+0x0): first defined here collect2: ld returned 1 exit status I'm using boost_1_33_1. The source files are attached. Regards, Tim Axelrod

2 1