- Boost-users - lists.boost.org

[interprocess] another newbie question: clean-up
by Wolfgang Hottgenroth 24 Jan '07

24 Jan '07

Hi, after I go really quick help here from Ion, I've another question. I managed it to implement a map<string, string> in shared memory and I'm able to put pairs into the map in one process and read them in another process. To put pair into the map I wrote: typedef boost::interprocess::allocator<char, boost::interprocess::managed_shared_memory::segment_manager> ShmemStringAllocator; typedef boost::interprocess::basic_string<char, char_traits<char>, ShmemStringAllocator> MyString; MyString key(segment.get_segment_manager()); MyString value(segment.get_segment_manager()); key = "key1"; value = "value1"; mymap->insert(std::pair<MyString, MyString>(key, value)); (For the complete code see http://www.hottis.de/boost/strings/. Thanks.) What I'm wondering is: will the memory required for key and value be freed automatically in the writing process (put.C)? When I like to erase an element from the map, do I need to do more than just calling mymap->erase(key)? And finally when I'm done and like to remove the whole map, is more required than segment.destroy<MyMap>("MyMap")? Will all the MyString's in the map go too? Furthermore: is it possible to use a map in shared memory as an associative array and write: MyString value = (*mymap)[key]; It doesn't compile for me. Thanks, Wolfgang

2 3

[fusion][mpl::math] generic math operations
by Chris Weed 24 Jan '07

24 Jan '07

Hi, I am interested in combining the fusion and mpl::math libraries, but I am not sure if they currently support what I want to do. I want to use generic math operations that can easily switch between run-time and compile-time. I think this would require writing the mpl::math functions as "Polymorphic Function Object" as described in the fusion documentation. I am not sure if there is any mpl::math documentation that describes this. It looks like the MPL currently has some math operations that are strictly meta-functions in boost/mpl/aux_/arithmetic_op.hpp. Are the any plans to expand these to be functions also? Chris

4 13

code snippet to add line number to lexical_cast exceptions
by bert hubert 24 Jan '07

24 Jan '07

Hi people, I find that I rely on lexical_cast<> a lot, it is a great thing. However, when it throws an exception, and I catch it, I lose information on which of the many lexical_casts had a problem. This is not a lexical_cast specific problem of course, it would be great to have a more general way to add line numbers to exceptions! This led me to code up this small gcc specific hack, but I wonder if there is a better way: #define wrapped_atox(x, y) ({ \ x tmp; \ try { tmp=boost::lexical_cast<x>(y); } \ catch(exception &e) { \ throw std::runtime_error(std::string("On line " + \ boost::lexical_cast<std::string>(__LINE__)+" of " +\ std::string(__PRETTY_FUNCTION__)+" in " +std::string(__FILE__) +\ ", while converting '"+string(y)+"': ") + e.what()); \ } \ tmp; \ }) Which can be used as: cout << wrapped_atox(int, argv[1]) << endl; The code above works, but is gcc specific, and well, it doesn't look right. Perhaps someone has a better idea? Otherwise, people may benefit from the snippet above, feel free to copy it! Bert -- http://www.PowerDNS.com Open source, database driven DNS Software http://netherlabs.nl Open and Closed source services

1 0

Missing libboost_random ?
by Koen Deforche 24 Jan '07

24 Jan '07

All, I'm trying to use some of the functions in the boost random library. I get linking errors on: undefined reference to `boost::random_device::random_device(std::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)' It turns out I do not find any libboost_random.so and none of the installed library has the symbol random_device. Are they missing? Are we not yet supposed to use Boost.Random ? This is the case with both the libraries provided by my distribution (both gentoo and fedora core) as well as a CVS snapshot that I compiled myself. Where are they ? Regards, koen

1 0

[mpl] compile time integer power
by Sami 23 Jan '07

23 Jan '07

Hi All, This is a newbie mpl question, and here goes: How can evaluate integer powers using mpl? Such as, template < unsigned Pow, class Real > Real pow( Real r ) { return <compile time r*r*r* ... r > ; // the expression is compile time, // not the result of course. } int main() { float k = pow<6>(3.14f); return k; // just kidding :) } An also as a side question, is there a wrapper type container (or iterator?) that holds a given constant value (or type)? Such as, the aptly named "constant_wrapper" used below with fold algorithm, using namespace boost::mpl; const unsigned SIZE = 5; typedef fold < constant_wrapper< int_<-3>, SIZE> , int_<0> , plus<_,_> >::type number; BOOST_STATIC_ASSERT( number::value == -3*SIZE ); Thanks a lot

3 6

boost::unit_test::framework needs fini_unit_test_suite
by pwm 23 Jan '07

23 Jan '07

I have just run into an issue with using boost unit test. To initialize a test one must implement a function init_unit_test_suite. There needs to be an analogous fini_unit_test_suite so the test suite can properly release any resources acquired by the init method. Phil

2 1

[pool] object_pool::destroy_all?
by Michael Marcin 23 Jan '07

23 Jan '07

I started using object_pool recently to solve a problem. I have lots of objects being created and destroyed frequently and stored in a collection. Every so often an event happens which triggers destroying all the objects. Essentially I have: boost::object_pool<Object> m_object_pool; std::vector<Object*> m_objects; Then I have a function void DestroyAllObjects() { std::vector<Object*>::iterator it, iend = m_objects.end(); for( it = m_objects.begin(); it != iend; ++it ) { m_object_pool.destroy(*it); } m_objects.clear(); } It seems like there should be a better way of doing this... i.e. void DestroyAllObjects() { m_object_pool.destroy_all(); m_objects.clear(); } Also it might be nice to iterate over all the objects in the pool (in which case I wouldn't even need the m_objects vector. Perhaps there is a better way to solve the problem? Thanks, Michael Marcin

1 0

Re: [Boost-users] [general question] on threading anddoublecheckedlocking pattern
by Sohail Somani 23 Jan '07

23 Jan '07

> -----Original Message----- > From: boost-users-bounces(a)lists.boost.org > [mailto:boost-users-bounces@lists.boost.org] On Behalf Of > Jason Olenhouse > > Meyers and Alexandrescu have commented on this as well: > http://www.aristeia.com/Papers/DDJ_Jul_Aug_2004_revised.pdf A very good article I might add.

1 0

[Boost 1.33.1].[64-bit]
by Prashant Thakre 23 Jan '07

23 Jan '07

Hi, I am trying to get 64-bit build of Boost 1.33.1 on following platforms: 1.Solaris 10 - Sun Studio 11 - Sparc. URL's listed below helped me to build some if not all. Patch -- http://blogs.sun.com/sga/entry/boost_1_33_1 Instruction for passing the relevant options to bjam -- http://blogs.sun.com/sga/category/Boost 2.HP-UX 11.23 - aCC A.06.02 - Itanium. How do I pass "+DD64" option to the compiler and the linker for 64-bit build. 3.AIX 5.3 - xlC/C++ 8.0 - PowerPC. Correct way to pass "-q64" option to the compiler and the linker. I do not really want to use the environment variable OBJECT_MODE. -- regards, Prashant Thakre

3 7

gdb and Invalid Address specified to RtlFreeHeap
by gelbeiche 23 Jan '07

23 Jan '07

Hi, I have an C++ application that uses the Oracle OCI libraries(oci.dll), which itself has references to msvcr71.dll. I compiled the application with gcc 3.4.2 under MinGW and for runtime usage I downloaded/installed the desired msvcr71.dll. From "installed.ini": [components] runtime=mingw-runtime-3.11.tar.gz w32api=w32api-3.8.tar.gz ... The program crashed at exit with the following error message in a dialog window: Microsoft Visual C++ Runtime Library Runtime Error ! abnormal program termination Because I want debug the error I installed gdb 5.2.1. When I run the program in gdb I get the following message: Program received signal SIGTRAP. Trace/breakpoint trap. ... warning: HEAP[...] warning: Invalid Address specified to RtlFreeHeap(230000, 246950) In the backtrace I see that it happens in OCIInitialize. So I suppose it has something to do with the oci.dll and msvcr71.dll. But I'm able to run the application outside the debugger and beside the crash at exit it is working. How can I debug my application ? My OS is Windows2000. Thomas -- http://www.randspringer.de

2 1