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

crash using boost::interprocess
by nevergone . 05 Nov '11

05 Nov '11

Hello, I'm using boost::interprocess to share config information between processes on windows platform. It random crash on custom pc. I only have a minidump dump, here is the stack output by windbg: 001df6b0 65e49d45 001df758 00302c48 7721ee1e XXXXPlugin!boost::interprocess::detail::emulation_recursive_mutex::lock+0xc 001df6f8 65e49c42 06750038 65e7b8b4 06750028 XXXXPlugin!boost::interprocess::scoped_lock<boost::interprocess::interprocess_recursive_mutex>::lock+0x15 001df720 65e494c1 06750008 00000001 6c1274eb XXXXPlugin!boost::interprocess::segment_manager<wchar_t,boost::interprocess::rbtree_best_fit<boost::interprocess::mutex_family,boost::interprocess::offset_ptr<void>,0>,boost::interprocess::iset_index>:: priv_get_lock+0x42 001df76c 65e48239 65e7b8b4 001df798 001df794 XXXXPlugin!boost::interprocess::segment_manager<wchar_t,boost::interprocess::rbtree_best_fit<boost::interprocess::mutex_family,boost::interprocess::offset_ptr<void>,0>,boost::interprocess::iset_index>:: priv_generic_find<wchar_t>+0x31 001df7bc 65e70053 001df7dc 00000001 001df8c4 XXXXPlugin!boost::interprocess::segment_manager<wchar_t,boost::interprocess::rbtree_best_fit<boost::interprocess::mutex_family,boost::interprocess::offset_ptr<void>,0>,boost::interprocess::iset_index>:: priv_find_impl<ConfigFile::_CACHE_ITEM>+0xa9 001df800 65e720c8 01ff8670 65e7b8b4 001df8cc XXXXPlugin!ConfigFile::CConfigCache::Read+0xc3 001df82c 65e5bbf5 65e7b8b4 001df8cc 00000004 XXXXPlugin!ConfigFile::CCommonConfig::Read+0x28 001dfb00 65e5e42e 020739b8 001dfb24 65e5e69a XXXXPlugin!CXXXXXMng::bFindProcessInfo+0x55 001dfb0c 65e5e69a 0000001c 001dfba0 00000113 XXXXPlugin!CXXXXXMng::OnTimer+0x5e 001dfb24 756286ef 00020448 00000113 0000001c XXXXPlugin!CXXXXXMng::StaticWndProc+0x2a 001dfb50 75628876 65e5e670 00020448 00000113 user32!InternalCallWinProc+0x23 001dfbc8 756289b5 00000000 65e5e670 00020448 user32!UserCallWinProcCheckWow+0x14b 001dfc28 75628e9c 65e5e670 00000000 001dfc70 user32!DispatchMessageWorker+0x35e 001dfc38 0137cf27 001dfc50 013eb5a8 013eb2d8 user32!DispatchMessageW+0xf 001dfc70 0134b2f5 8b56d422 013eb2d8 013eb2d8 xxxxMain!CXXXX::MessageLoop+0x97 001dfec8 67add2a7 013f11a4 00000001 00000000 xxxxMain!CXXXXApp::InitInstance+0x335 001dfecc 013f11a4 00000001 00000000 013787d8 mfc80u!AfxWinMain+0x48 [f:\dd\vctools\vc7libs\ship\atlmfc\src\mfc\winmain.cpp @ 37] 001dff6c 75b41114 7ffda000 001dffb8 7721b495 xxxxMain!__native_startup_lock 001dff78 7721b495 7ffda000 773d0f2e 00000000 kernel32!BaseThreadInitThunk+0xe 001dffb8 7721b468 01378944 7ffda000 00000000 ntdll!__RtlUserThreadStart+0x70 001dffd0 00000000 01378944 7ffda000 00000000 ntdll!_RtlUserThreadStart+0x1b EXCEPTION_RECORD: ffffffff -- (.exr 0xffffffffffffffff) ExceptionAddress: 72341bac (XXXXPlugin!boost::interprocess::detail::emulation_recursive_mutex::lock+0x0000000c) ExceptionCode: c0000005 (Access violation) ExceptionFlags: 00000000 NumberParameters: 2 Parameter[0]: 00000000 Parameter[1]: 06400030 Attempt to read from address 06400030 FOLLOWUP_IP: XXXXPlugin!boost::interprocess::detail::emulation_recursive_mutex::lock+c 65e41bac 8b4608 mov eax,dword ptr [esi+8] the code is: PCACHE_ITEM CacheItem = m_SharedMemory.find<CACHE_ITEM>(lpszName).first; Can anyone help me? thx

3 4

Compiling Boost 1.47 with GCC 4.6.1 and --std=c++0x
by Travis Gockel 04 Nov '11

04 Nov '11

I'm building Boost 1.47.0 with GCC 4.6.1 and C++11 support enabled and get the following two errors building the Signals library: libs/signals/src/named_slot_map.cpp: In member function ‘void boost::signals::detail::named_slot_map::disconnect(const boost::signals::detail::stored_group&)’: libs/signals/src/named_slot_map.cpp:105:23: error: call of overloaded ‘erase(boost::signals::detail::named_slot_map::group_iterator&)’ is ambiguous libs/signals/src/named_slot_map.cpp:105:23: note: candidates are: /usr/local/include/c++/4.6.1/bits/stl_map.h:613:7: note: std::map<_Key, _Tp, _Compare, _Alloc>::iterator std::map<_Key, _Tp, _Compare, _Alloc>::erase(std::map<_Key, _Tp, _Compare, _Alloc>::const_iterator) [with _Key = boost::signals::detail::stored_group, _Tp = std::list<boost::signals::detail::connection_slot_pair>, _Compare = boost::function2<bool, boost::signals::detail::stored_group, boost::signals::detail::stored_group>, _Alloc = std::allocator<std::pair<const boost::signals::detail::stored_group, std::list<boost::signals::detail::connection_slot_pair> > >, std::map<_Key, _Tp, _Compare, _Alloc>::iterator = std::_Rb_tree_iterator<std::pair<const boost::signals::detail::stored_group, std::list<boost::signals::detail::connection_slot_pair> > >, std::map<_Key, _Tp, _Compare, _Alloc>::const_iterator = std::_Rb_tree_const_iterator<std::pair<const boost::signals::detail::stored_group, std::list<boost::signals::detail::connection_slot_pair> > >] /usr/local/include/c++/4.6.1/bits/stl_map.h:643:7: note: std::map<_Key, _Tp, _Compare, _Alloc>::size_type std::map<_Key, _Tp, _Compare, _Alloc>::erase(const key_type&) [with _Key = boost::signals::detail::stored_group, _Tp = std::list<boost::signals::detail::connection_slot_pair>, _Compare = boost::function2<bool, boost::signals::detail::stored_group, boost::signals::detail::stored_group>, _Alloc = std::allocator<std::pair<const boost::signals::detail::stored_group, std::list<boost::signals::detail::connection_slot_pair> > >, std::map<_Key, _Tp, _Compare, _Alloc>::size_type = long unsigned int, std::map<_Key, _Tp, _Compare, _Alloc>::key_type = boost::signals::detail::stored_group] libs/signals/src/named_slot_map.cpp: In member function ‘void boost::signals::detail::named_slot_map::remove_disconnected_slots()’: libs/signals/src/named_slot_map.cpp:128:35: error: call of overloaded ‘erase(std::_Rb_tree_iterator<std::pair<const boost::signals::detail::stored_group, std::list<boost::signals::detail::connection_slot_pair> > >::_Self)’ is ambiguous libs/signals/src/named_slot_map.cpp:128:35: note: candidates are: /usr/local/include/c++/4.6.1/bits/stl_map.h:613:7: note: std::map<_Key, _Tp, _Compare, _Alloc>::iterator std::map<_Key, _Tp, _Compare, _Alloc>::erase(std::map<_Key, _Tp, _Compare, _Alloc>::const_iterator) [with _Key = boost::signals::detail::stored_group, _Tp = std::list<boost::signals::detail::connection_slot_pair>, _Compare = boost::function2<bool, boost::signals::detail::stored_group, boost::signals::detail::stored_group>, _Alloc = std::allocator<std::pair<const boost::signals::detail::stored_group, std::list<boost::signals::detail::connection_slot_pair> > >, std::map<_Key, _Tp, _Compare, _Alloc>::iterator = std::_Rb_tree_iterator<std::pair<const boost::signals::detail::stored_group, std::list<boost::signals::detail::connection_slot_pair> > >, std::map<_Key, _Tp, _Compare, _Alloc>::const_iterator = std::_Rb_tree_const_iterator<std::pair<const boost::signals::detail::stored_group, std::list<boost::signals::detail::connection_slot_pair> > >] /usr/local/include/c++/4.6.1/bits/stl_map.h:643:7: note: std::map<_Key, _Tp, _Compare, _Alloc>::size_type std::map<_Key, _Tp, _Compare, _Alloc>::erase(const key_type&) [with _Key = boost::signals::detail::stored_group, _Tp = std::list<boost::signals::detail::connection_slot_pair>, _Compare = boost::function2<bool, boost::signals::detail::stored_group, boost::signals::detail::stored_group>, _Alloc = std::allocator<std::pair<const boost::signals::detail::stored_group, std::list<boost::signals::detail::connection_slot_pair> > >, std::map<_Key, _Tp, _Compare, _Alloc>::size_type = long unsigned int, std::map<_Key, _Tp, _Compare, _Alloc>::key_type = boost::signals::detail::stored_group] WHOA BUDDY! Anyway, if you look into this, it is because my STL implementation has changed to the new definition for map::erase which takes a const_iterator. As you can see, trying to erase with an iterator creates an ambiguous call between the const_iterator and key_type. I came up with two solutions: One was to explicitly cast to a const_iterator and the other was to mark stored_group::stored_group<T>(const T&) as explicit. Both work perfectly fine for me, but I like the explicit constructor solution better. I attached that change as a patch...feel free to use it if it doesn't break anything (which it probably does). -- Travis Gockel

1 0

[config] TI compiler warning "static local variables of extern inline function are not resolved to single copy"
by Krzysztof Czainski 04 Nov '11

04 Nov '11

Hello, I use some boost libraries with a TI compiler. One of it's defects is "static local variables of extern inline function are not resolved to single copy". It emits a warning about it. My typical workaround for this is making such a function non-inline. This may require to move such a function definition out of a header file, if it isn't a template function. When I need a non-template function to be defined inside a header file, and return an instance of a static variable, i can't just make that function static (or put it into an unnamed namespace), because there will no longer be a single instance of the returned variable.Therefore I came up with the following: instead of: inline int& get_single_int() { static int i = 0; return i; } I need: template < class > // not inline int& get_single_int_impl() { static int i = 0; return i; } inline int& get_single_int() { return get_single_int_impl<void>(); } I needed to make modifications in two places in boost to work around this defect. I'll be happy to share those modifications, but first I wanted to ask if anyone can think of a better/simpler workaround. What comes into my mind, that could make this workaround more general, is replacing the line "// not inline" with some macro BOOST_INLINE_STATIC_VAR defined conditionally to 'inline' or nothing. Thoughts? Regards, Kris

2 1

[graph] Implementing turn restrictions
by Stephen Woodbridge 04 Nov '11

04 Nov '11

Hi all, I need to solve Dijkstra shortest path with turn restrictions. I am wondering if anyone has an example of how to implement this? I'm thinking that this can be done something like this: 1. a restriction(s) could be represented as: at a given node X if my ancestry is some list of of nodes, then restrict going forward to a node or list of nodes. 2. any given node can optionally have a list of restrictions attached to it 3. a visitor function would then get the list of restrictions for the visited node and compare ancestor list of each restriction to the current path ancestry and then filter forward nodes if there is an ancestry match(s). Does this sound like a reasonable approach? Does anyone have an example of this or something similar? Any thoughts on how to efficiently implement matching the path to the restriction lists? Many thanks, -Steve

2 1

[filesystem] error compiling win win7/mingw
by Gustavo Villada 04 Nov '11

04 Nov '11

Hi Guys, I'm trying to compile boost.filesystem on a win7 with mingw. MinGW is installed by default. I've already compile bjam, b2 and successful boostrap.bat I got this error, suggestions? C:\boost_1_47_0>bjam toolset=gcc --with-filesystem stage Component configuration: - chrono : not building - date_time : not building - exception : not building - filesystem : building - graph : not building - graph_parallel : not building - iostreams : not building - math : not building - mpi : not building - program_options : not building - python : not building - random : not building - regex : not building - serialization : not building - signals : not building - system : not building - test : not building - thread : not building - wave : not building ...patience... ...found 525 targets... ...updating 8 targets... gcc.compile.c++ bin.v2\libs\filesystem\build\gcc-mingw-4.5.2\debug\link-static\threading-multi\v3\src\operations.o In file included from c:\mingw\bin\../lib/gcc/mingw32/4.5.2/../../../../include/sys/utime.h:1:0, from c:\mingw\bin\../lib/gcc/mingw32/4.5.2/../../../../include/sys/utime.h:1, from c:\mingw\bin\../lib/gcc/mingw32/4.5.2/../../../../include/sys/utime.h:1, from c:\mingw\bin\../lib/gcc/mingw32/4.5.2/../../../../include/sys/utime.h:1, from c:\mingw\bin\../lib/gcc/mingw32/4.5.2/../../../../include/sys/utime.h:1, from c:\mingw\bin\../lib/gcc/mingw32/4.5.2/../../../../include/sys/utime.h:1, from c:\mingw\bin\../lib/gcc/mingw32/4.5.2/../../../../include/sys/utime.h:1, from c:\mingw\bin\../lib/gcc/mingw32/4.5.2/../../../../include/sys/utime.h:1, from libs\filesystem\v3\src\operations.cpp:117: c:\mingw\bin\../lib/gcc/mingw32/4.5.2/../../../../include/sys/utime.h:1:23: error: #include nested too deeply libs\filesystem\v3\src\operations.cpp:655:26: warning: '<unnamed>::create_symbolic_link_api' defined but not used "g++" -ftemplate-depth-128 -O0 -fno-inline -Wall -g -mthreads -DBOOST_ALL_NO_LIB=1 -DBOOST_SYSTEM_STATIC_LINK=1 -I"." -c -o "bin.v2\libs\filesystem\build\gcc-mingw-4.5.2\debug\link-static\threading-multi\v3\src\operations.o" "libs\filesystem\v3\src\operations.cpp" ...failed gcc.compile.c++ bin.v2\libs\filesystem\build\gcc-mingw-4.5.2\debug\link-static\threading-multi\v3\src\operations.o... ...skipped <pbin.v2\libs\filesystem\build\gcc-mingw-4.5.2\debug\link-static\threading-multi>libboost_filesystem-mgw45-mt-d-1_47.a(clean) for lack of <pbin.v2\libs\filesystem\build\gcc-mingw-4.5.2\debug\link-static\threading-multi>v3\src\operations.o... ...skipped <pbin.v2\libs\filesystem\build\gcc-mingw-4.5.2\debug\link-static\threading-multi>libboost_filesystem-mgw45-mt-d-1_47.a for lack of <pbin.v2\libs\filesystem\build\gcc-mingw-4.5.2\debug\link-static\threading-multi>v3\src\operations.o... ...skipped <pstage\lib>libboost_filesystem-mgw45-mt-d-1_47.a for lack of <pbin.v2\libs\filesystem\build\gcc-mingw-4.5.2\debug\link-static\threading-multi>libboost_filesystem-mgw45-mt-d-1_47.a... gcc.compile.c++ bin.v2\libs\filesystem\build\gcc-mingw-4.5.2\release\link-static\threading-multi\v3\src\operations.o In file included from c:\mingw\bin\../lib/gcc/mingw32/4.5.2/../../../../include/sys/utime.h:1:0, from c:\mingw\bin\../lib/gcc/mingw32/4.5.2/../../../../include/sys/utime.h:1, from c:\mingw\bin\../lib/gcc/mingw32/4.5.2/../../../../include/sys/utime.h:1, from c:\mingw\bin\../lib/gcc/mingw32/4.5.2/../../../../include/sys/utime.h:1, from c:\mingw\bin\../lib/gcc/mingw32/4.5.2/../../../../include/sys/utime.h:1, from c:\mingw\bin\../lib/gcc/mingw32/4.5.2/../../../../include/sys/utime.h:1, from c:\mingw\bin\../lib/gcc/mingw32/4.5.2/../../../../include/sys/utime.h:1, from c:\mingw\bin\../lib/gcc/mingw32/4.5.2/../../../../include/sys/utime.h:1, from libs\filesystem\v3\src\operations.cpp:117: c:\mingw\bin\../lib/gcc/mingw32/4.5.2/../../../../include/sys/utime.h:1:23: error: #include nested too deeply libs\filesystem\v3\src\operations.cpp:655:26: warning: '<unnamed>::create_symbolic_link_api' defined but not used "g++" -ftemplate-depth-128 -O3 -finline-functions -Wno-inline -Wall -mthreads -DBOOST_ALL_NO_LIB=1 -DBOOST_SYSTEM_STATIC_LINK=1 -DNDEBUG -I"." -c -o "bin.v2\libs\filesystem\build\gcc-mingw-4.5.2\release\link-static\threading-multi\v3\src\operations.o" "libs\filesystem\v3\src\operations.cpp" ...failed gcc.compile.c++ bin.v2\libs\filesystem\build\gcc-mingw-4.5.2\release\link-static\threading-multi\v3\src\operations.o... ...skipped <pbin.v2\libs\filesystem\build\gcc-mingw-4.5.2\release\link-static\threading-multi>libboost_filesystem-mgw45-mt-1_47.a(clean) for lack of <pbin.v2\libs\filesystem\build\gcc-mingw-4.5.2\release\link-static\threading-multi>v3\src\operations.o... ...skipped <pbin.v2\libs\filesystem\build\gcc-mingw-4.5.2\release\link-static\threading-multi>libboost_filesystem-mgw45-mt-1_47.a for lack of <pbin.v2\libs\filesystem\build\gcc-mingw-4.5.2\release\link-static\threading-multi>v3\src\operations.o... ...skipped <pstage\lib>libboost_filesystem-mgw45-mt-1_47.a for lack of <pbin.v2\libs\filesystem\build\gcc-mingw-4.5.2\release\link-static\threading-multi>libboost_filesystem-mgw45-mt-1_47.a... ...failed updating 2 targets... ...skipped 6 targets... C:\boost_1_47_0>

1 0

Serialize abstract base class
by Fabian Weiss 04 Nov '11

04 Nov '11

I try to serialize objects wich are derived of an abstract base class. class Modul() { Modul(); int a; virtual void function() = 0; private: friend class boost::serialization::access; template <typename Archive> void serialize(Archive &ar, const unsigned int version) { ar & boost::serialization::make_nvp("a", a); }; }; BOOST_SERIALIZATION_ASSUME_ABSTRACT(Modul) BOOST_CLASS_TRACKING(Modul, track_never) class ExampleModul : public Modul { ExampleModul(){}; int b; void function(); private: friend class boost::serialization::access; template <typename Archive> void serialize(Archive &ar, const unsigned int version) { ar & boost::serialization::make_nvp("Modul", boost::serialization::base_object<Modul>(*this)); ar & boost::serialization::make_nvp("b", b); } }; For serializing the classes I use a boost::archive::xml_oarchive and register the classes: boost::archive::xml_oarchive archfile; archfile.register_type<Modul>(); archfile.register_type<ExampleModul>(); I can compile serializing and run the code - works. But when it comes to deserializing I get a compile error: ..instantiated from here /usr/include/boost/serialization/access.hpp:132:9: error: cannot allocate an object of abstract type 'Modul' note: because the following virtual functions are pure within 'Modul': virtual void Modul::function() If I do not register <Modul> in the deserializing code, then it compiles but I get runtime error of course: terminate called after throwing an instance of 'boost::archive::archive_exception' I found this example but I dont understand what they are doing and where the classes are registered: http://www.boost.org/doc/libs/1_47_0/libs/serialization/example/demo.cpp How can I serialize my abstract base class?

2 3

[type_traits] an is_iterator, not is_pointer, trait
by Belcourt, K. Noel 03 Nov '11

03 Nov '11

Hi, I'm looking to use a trait in enable_if that can distinguish between a raw pointer and any other valid iterator type, essentially the difference between iterator traits and the raw pointer specialization. template <class Iterator> struct iterator_traits; template <class T> struct iterator_traits<T*>; I was hoping to find a boost::is_iterator type trait which would return false_type for is_iterator<int*> (any raw pointer) and true type for any other valid iterator. This would allow me to group iterators via is_pointer and is_iterator. I was just wondering if this already exists somewhere? Thanks. -- Noel Belcourt

2 3

Re: [Boost-users] [Gil] TIFF file with mixed 8/16 bit channel depths
by Christian Henning 03 Nov '11

03 Nov '11

Corinna, this has to wait until tomorrow. Hope that's ok with you. But I can see from the source code that the input image should be cymk16161688_image_t and not just cymk8_image_t. Christian

1 0

[Boost.Endian] Review Results
by joel falcou 03 Nov '11

03 Nov '11

Hi everyone, after some delay, here is the result of the review of Beman Dawes' Boost.Endian library. First, I want to thanks all participants of the review that lead to very deep and interesting discussions on the matter. 5 votes were cast : 2 unconditionally in favor, 3 conditionally in favor after changes made and there is a mini-review. The verdict is that Boost.Endian is **CONDITIONNALY ACCEPTED** Here is what needs to be done to get the library ready for a mini-review: * Common use case scenarios should be developed. * Example programs should be developed for the common use case scenarios. * Documentation should illuminate the differences between endian integer/float type and endian conversion approaches to the common use case scenarios, and provide guidelines for choosing the most appropriate approach in user's applications. * Conversion functions supplying results via return should be provided. * Platform specific performance enhancements such as use of compiler intrinsics or relaxed alignment requirements should be supported. * Endian integer (and floating) types should be implemented via the conversion functions. If that can't be done efficiently, consideration should be given to expanding the conversion function signatures to resolve the inefficiencies. * Benchmarks that measure performance should be provided. It should be possible to compare platform specific performance enhancements against portable base implementations, and to compare endian integer approaches against endian conversion approaches for the common use case scenarios. * Float (32-bits) and double (64-bits) should be supported. IEEE 754 is the primary use case. * Support for user defined types (UDTs) is desirable, and should be provided where there would be no conflict with the other concerns. * There is some concern that endian integer/float arithmetic operations might used inadvertently or inappropriately. The impact of adding an endian_buffer class without arithmetic operations should be investigated. * Stream insertion and extraction of the endian integer/float types should be documented and included in the test coverage. * Binary I/O support that was investigated during development of the Endian library should be put up for mini-review for inclusion in the Boost I/O library. From this point, I will let Beman prepare Endian for the mini-review and I already volunteer to manage the mini-review so informations are not scattered around. Regards

1 0

[fusion] Adapted struct I/O
by Mateusz Łoskot 03 Nov '11

03 Nov '11

Hi, I'm still taking early steps with Boost.Fusion. I'm trying to adapt a bunch of struct and template struct for easy I/O operations. I don't understand one thing from the docs. Here we go; 1) Given the example [1] of template struct adoption: namespace demo { template<typename Name, typename Age> struct employee { Name name; Age age; }; } // Any instantiated demo::employee is now a Fusion sequence BOOST_FUSION_ADAPT_TPL_STRUCT( (Name)(Age), (demo::employee) (Name)(Age), (Name, name) (Age, age)) [1] http://www.boost.org/doc/libs/1_47_0/libs/fusion/doc/html/fusion/adapted/ad… 2) The included comment says "demo::employee is now a Fusion sequence". 3) Jumping to I/O and out docs, reading that [2] "The I/O operators: << and >> work generically on all Fusion sequences. " [2] http://www.boost.org/doc/libs/1_47_0/libs/fusion/doc/html/fusion/sequence/o… 4) Means, I should be able to stream Fusion-adopted demo::employee #include <iostream> #include <string> #include <boost/fusion/adapted/struct/adapt_struct.hpp> #include <boost/fusion/include/adapt_struct.hpp> #include <boost/fusion/sequence/io.hpp> #include <boost/fusion/include/io.hpp> // code from above example goes here int main() { demo::employee<std::string, int> e; std::cout << e << std::endl; } 5) Given what I have learned above, I expect the cout << e to work: $ g++ -I/home/mloskot/dev/boost/_svn/trunk boost_fusion.cpp boost_fusion.cpp: In function ‘int main()’: boost_fusion.cpp:28:18: error: no match for ‘operator<<’ in ‘std::cout << e’ ... I assume complete error is not necessary as my question is of general nature. Is my expectation valid? Best regards, -- Mateusz Loskot, http://mateusz.loskot.net Charter Member of OSGeo, http://osgeo.org Member of ACCU, http://accu.org

3 5