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

boost::interprocess::string
by Josmon Paul 29 Dec '11

29 Dec '11

2 1

Problem with Interpocess offset_ptr
by mike＠tesla.rcub.bg.ac.rs 29 Dec '11

29 Dec '11

Hi, I need to store in the shared memory structure for Quadtree. It is a class with 4 pointers for tree regions and vector of points: class CNode4 { ... CNode4* node0; CNode4* node1; CNode4* node2; CNode4* node3; vector<vec2> pointArray; }; I have tried with class CNode4 { ... offset_ptr<CSNode4> node0; ... }; and using as: typedef allocator<void, managed_shared_memory::segment_manager> void_allocator; const void_allocator alloc_void(segment.get_segment_manager()); CSNode4* rootNode = segment.construct<CSNode4> (_RootNode) (alloc_void); rootNode->node0 = new CSNode4(alloc_void,...); this works until I try to static_cast node0 offset_ptr<CSNode4> N1 = static_cast<CSNode4*>(rootNode->node0); I get: (MSVC 2010, boost 1.42) error C2440: 'static_cast' : cannot convert from 'boost::interprocess::offset_ptr<PointedType>' to 'CSNode4 *' what am I doing wrong? Thanks, Ivan

2 1

Build boost with Apple's LLVM 3.0
by Daniel Dilts 29 Dec '11

29 Dec '11

I am building boost on OS X 10.7.2. Using ./bootstrap.sh ./b2 sudo ./b2 install it appears to build with the GCC 4.2.1 toolchain. How would I change this so that it builds using the LLVM 3.0 toolchain?

2 2

Using the /RTC option without specifying a debug runtime will lead to linker errors
by Sleat, Ben 28 Dec '11

28 Dec '11

Hi I'm trying to track down a very nasty stack corruption bug in my release code and therefore I would like to enabled the Visual C++ RTCs option to enable runtime stack checks in my release build, however when I try to do this I get an "Incompatible build options" error from the boost libraries. Specifically I get this from boosts AutoLink.hpp: #if defined(__MSVC_RUNTIME_CHECKS) && !defined(_DEBUG) # pragma message("Using the /RTC option without specifying a debug runtime will lead to linker errors") # pragma message("Hint: go to the code generation options and switch to one of the debugging runtimes") # error "Incompatible build options" #endif It seems to be suggesting that in order to use /RTC I need to do a Debug build, buts that's exactly what I don't want to do... I need to enable Runtime Checks in the Release build! Is there any way to get this to work in a release build that uses boost? Does anyone know the details of the incompatibility? Surely /RTC doesn't require all linked in DLLs also to be compiled with /RTC? Thanks in advance. Ben This email is from Verint Systems UK Ltd and is confidential. If you have received it in error, please advise us and then delete the original and any attachments it may have contained. Verint Systems UK Ltd does not accept any responsibility for viruses. Verint Systems UK Ltd is a limited company registered in England and Wales at 241 Brooklands Road, Weybridge, Surrey KT13 0RH ,registration number 2602824. VAT registration number GB566153238.

3 5

boost::asio::buffer + std::wstring
by Luke Walker 28 Dec '11

28 Dec '11

Hi, I'm sending unicode text over a socket using boost::asio::async_write and boost::asio::buffer however only 1/4 of the string ends up at the receiver. What must I do so that all bytes are sent? Thanks!

2 6

[Python] Wrapping Visitor Pattern and Object Slicing
by Alex Olivas 27 Dec '11

27 Dec '11

Hey All, I'm trying to wrap a third party lib that uses the visitor pattern and their own smart pointers ('ref_ptr' below), which manages objects that inherit from 'Referenced'. I've put together a simple example that illustrates the problem. Code for python bindings : http://codepad.org/wq47m5CI Simple test script : http://codepad.org/Jfme3U6U Output for two cases : http://codepad.org/gbFBxy04 Case 1) 'apply' override is commented out in the python class Case 2) 'apply' override exists The problem that I'm having is that the 'apply' method ( lines 83 - 86 ) slices everything to base unless I use boost::ref(). However this only works for wrapped classes and in this lib several classes use derived visitors that are not part of the public API ( which I modeled as HelperVisitor in the above example ). Is there anyway to prevent object slicing for both wrapped and unwrapped objects in this scenario? I'm building against boost 1.47 and if anyone's interested in digging deeper the 3rd party lib is Open Scene Graph 3.0.1. Thanks, Alex.

1 0

Re: [Boost-users] filesystem::path threadsafe ?? (1.48) Options
by Aris Basic 27 Dec '11

27 Dec '11

Here is the sample code that crashes every time i compile project as multibyte and not unicode and ahve user wchar_t as default type #include <boost/thread.hpp> #include <boost/filesystem.hpp> int main(void) { std::string sPath("c:\\Development"); boost::thread_group tg; for(int i = 0; i < 10; i++) { tg.create_thread([&sPath](){ boost::this_thread::sleep(boost::posix_time::milliseconds(10)); boost::filesystem::path p(sPath); boost::filesystem::directory_iterator di(p), end; while(di != end) std::cout << (*di).path().string() << std::endl; }); } tg.join_all(); }

3 3

Boost.Container string operator+ problem
by dave_chp 27 Dec '11

27 Dec '11

Using boost 1.48 and vs2010 gives me an error when trying to use boost::container::string. #include <boost/container/string.hpp> int main( int, char *[] ) { using boost::container::string; string hello = "hello "; string world = "world"; string hello_world = hello + world; } 1>d:\sdk\boost_1_48_0\boost\container\string.hpp(2368): error C2248: 'boost::container::containers_detail::basic_string_base ::alloc' : cannot access protected member declared in class 'boost::container::containers_detail::basic_string_base ' 1> with 1> [ 1> A=std::allocator<char> 1> ] 1> d:\sdk\boost_1_48_0\boost\container\string.hpp(214) : see declaration of 'boost::container::containers_detail::basic_string_base ::alloc' 1> with 1> [ 1> A=std::allocator<char> 1> ] 1> d:\tmp\xxx.cpp(15) : see reference to function template instantiation 'boost::container::basic_string<CharT,Traits,A> boost::container::operator +<char,std::char_traits<char>,std::allocator<_Ty>>(const boost::container::basic_string<CharT,Traits,A> &,const boost::container::basic_string<CharT,Traits,A> &)' being compiled 1> with 1> [ 1> CharT=char, 1> Traits=std::char_traits<char>, 1> A=std::allocator<char>, 1> _Ty=char 1> ] 1> d:\tmp\xxx.cpp(20) : see reference to function template instantiation 'void hello_world_test<boost::container::string>(void)' being compiled 1> template <class CharT, class Traits, class A> inline basic_string<CharT,Traits,A> operator+(const basic_string<CharT,Traits,A>& x, const basic_string<CharT,Traits,A>& y) { ... str_t result(reserve, x.size() + y.size(), x.alloc()); ... } should this maybe be x.get_allocator() instead of x.alloc()? -- View this message in context: http://boost.2283326.n4.nabble.com/Boost-Container-string-operator-problem-… Sent from the Boost - Users mailing list archive at Nabble.com.

2 2

[smart_ptr] Inclusion of unique_ptr
by Szymon Gatner 27 Dec '11

27 Dec '11

Hi, as suggested by Dave Abrahams I am now turning here with proposition of including unique_ptr<> to smart_ptr library. Newly accepted boost.move enables portable implementation and move-aware containers in boost.container allow to take full advantage of this now standard pointer type. Attaching proposed implementation. -- Szymon Gatner The Lordz Games Studio www.thelordzgamesstudio.com

3 2

whitebox testing with boost unit test framework?
by Chris Cleeland 26 Dec '11

26 Dec '11

I'm using the boost unit test framework to implement unit tests for code. I'm wondering if there is a best-practice for doing white-box testing/internal state inspection of types under test. My hack right now is to strip out "protected" and "private" from the header, but I hold my nose as I do that. Thoughts/suggestions? -- Chris Cleeland

7 16