- Boost-users - lists.boost.org

Thread locks
by Larry E. Ramey 03 Feb '09

03 Feb '09

Under 1.33.1 we used to do this: boost::try_mutex::scoped_try_lock lock(m_mutex); if (lock.locked() && m_threadState == UploadMap) { I'm not sure, but I think under 1.37 I should do: boost::try_mutex::scoped_try_lock lock(m_mutex); if (lock.owns_lock() && m_threadState == UploadMap) { When I do that, I get a crash when I exit that I never used to have. Any ideas? Larry

2 2

[boost-users][lambda] crash in std::for_each
by Igor R 03 Feb '09

03 Feb '09

The following code crashes in for_each when trying to output the 1st pair. What's wrong with it? Thanks! namespace bl = boost::lambda; std::map<std::string, std::string> temp; temp["a"] = "1"; temp["b"] = "2"; const std::map<std::string, std::string> attrs(temp); typedef std::pair<std::string, std::string> string_pair; std::stringstream stream; std::for_each(attrs.begin(), attrs.end(), stream << bl::bind(&string_pair::first, bl::_1) << bl::bind(&string_pair::second, bl::_1));

4 7

[asio] Cloning a WSA socket (windows)
by Tore Halvorsen 03 Feb '09

03 Feb '09

Hi, I am trying to clone a socket from process to another, mostly for having one process per client in a network server. Some legacy stuff makes this the only design possible, so the threaded approach is simply not feasible... Anyway, when I'm trying to create a boost socket from a cloned WSA socket I get an exception. class boost::exception_detail::clone_impl< struct boost::exception_detail::error_info_injector< class boost::system::system_error > > : The parameter is incorrect This same happens for this example program, using boost 1.37 Does anyone have an idea why this does not work? #ifndef _WIN32_WINNT #define _WIN32_WINNT 0x0501 #endif #define WIN32_LEAN_AND_MEAN #define NOMINMAX #include <windows.h> #include <winsock2.h> #pragma comment(lib, "ws2_32") #include <boost/asio.hpp> namespace ip = boost::asio::ip; int main_impl(int argc, char* argv[]) { boost::asio::io_service ios; ip::tcp::socket s(ios); ip::tcp::acceptor a(ios, ip::tcp::endpoint(ip::tcp::v4(), 1777)); a.accept(s); WSAPROTOCOL_INFO info; if (WSADuplicateSocket(s.native(), GetCurrentProcessId(), &info)) { std::cerr << "WSAError: " << WSAGetLastError() << std::endl; return EXIT_FAILURE; } // pass info over a pipe to the client... SOCKET copy(WSASocket(FROM_PROTOCOL_INFO, FROM_PROTOCOL_INFO , FROM_PROTOCOL_INFO, &info, NULL, NULL)); ip::tcp::socket scopy(ios, ip::tcp::v4(), copy); return EXIT_SUCCESS; } int main(int argc, char* argv[]) { try { return main_impl(argc, argv); } catch(std::exception const& ex) { std::cerr << typeid(ex).name() << ":" << ex.what() << std::endl; return EXIT_FAILURE; } } -- Eld på åren og sol på eng gjer mannen fegen og fjåg. [Jøtul] <demo> 2009 Tore Halvorsen || +052 0553034554

2 4

[mpi] Non-default arguments for running regression tests through bjam
by Pete Bartlett 03 Feb '09

03 Feb '09

Dear Boosters, I am just going through the learning curve for Boost.MPI. My underlying target platform is MPICH2 on Windows XP/Vista. Given I am taking baby-steps I would like run the regression tests to check my install. Mpirun doesn't exist in my install - mpiexec is required instead. The docs and mpi.jam comments say "The (optional) fourth argument configures Boost.MPI for running regression tests. These parameters specify the executable used to launch jobs (default: "mpirun") followed by any necessary arguments to this to run tests and tell the program to expect the number of processors to follow (default: "-np")." I'm no jam expert, but that suggests that my user-config.jam should change to using mpi : : <find-shared-library>cxx <find-shared-library>mpi <library-path>C:/MPICH2/lib <include>C:/MPICH2/include : : mpiexec -n ; where the " : : mpiexec -n " is the new part. However this results in a slew of jam errors. I am able to run the regression tests by hacking mpi.jam to change the defaults, however I'd prefer to know the correct syntax in user-config.jam please. Thanks in advance for any help, Pete Bartlett

2 2

Boost.lambda: can I do this with a lambda function (multiplestatements?
by Joost Kraaijeveld 03 Feb '09

03 Feb '09

Hi, I want to embed some debug printing in a lambda function which is a predicate used to remove elements from a vector like this: collection.erase( std::remove_if( collection.begin(), collection.end(), boost::lambda::if_then_else_return( boost::lambda::_1 <= "s5", // if (std::cout << "deleted: " << boost::lambda::_1, true), // then (std::cout << "not deleted: " << boost::lambda::_1, false ))), // else collection.end()); This actually compiles, runs and removes all the right elements but it does not print for every element it is checking. It just prints "not deleted: deleted: s4s5s6s8" , the last string being all the elements of the collection. Is what I want possible and if so how? TIA, -- Groeten, Joost Kraaijeveld Askesis B.V. Molukkenstraat 14 6524NB Nijmegen tel: 024-3888063 / 06-51855277 fax: 024-3608416 web: www.askesis.nl

2 1

Boost:interprocess 1.36.0 - Minimum size for a shared memory segment
by Federico Golfrè Andreasi 03 Feb '09

03 Feb '09

Hi ! which is the minimum number of bytes I have to allocate to construct a managed shared memory? Or how can I calculate it? I did some tests and I believe is 112 bytes for 32 bit machines and 224 bytes for 64 bit, am I right ? And for the allocators? What I want to do is set the size of the shared memory smaller than I can, knowing the dimension of the variables I want to put in; even if the code is compiled in 64 bit or 32 bit machines. Thank you, Federico

2 1

boost::tokenizer with char*
by Archie14 03 Feb '09

03 Feb '09

I was wandering if anyone can help me to understand how to use boost::tokenizer with char* buffer. tokenizer template parameters include string as default Type. I don't want to copy char* into string.

1 1

Using SFINAE to detect presence of member function - Unreliable
by Naik, Roshan 03 Feb '09

03 Feb '09

I have taken a look at anything I could find through Boost lists and Google that came up as a solution to the problem of detecting the existence of a member function in a type... using SFINAE. All of them have one problem.. they are unable to detect the existence of the member function if it was defined in a base class. They can only detect if the member function was directly defined in the class being tested. At their heart, all solutions boil down to some variation of the following template class with a template parameter that is a (non-type) pointer to member function: template<class T, void (T::*)(void) > struct Tester; Then we try to SFINAE on it by the usual sizeof trick. However the problem is that if method 'void foo(int)' is defined in a Base class. Then the type of &Derived::foo is actually void (Base::*)(int) ... and not void (Derived::*)(int) Although during argument passing the conversion of the Base member function pointer to Derived member function pointer is acceptable to C++, it will NOT do such conversion when instantiating templates. So trying to use Tester<Derived, &Derived::foo> will always fail us if 'foo' is defined in Base. So .. my question is ....does anyone have solution that works for the case when the method is defined in the base class too ? I am thinking ... if it is possible to extract the class name from the actual type of &Derived::foo, then the problem can be solved... but the type extraction mechanism must not error out on types that have absolutley no member function called foo. I know type erasure (which i prefer to call External polymorhism) can solve this issue but SFINAE is more efficient at runtime. -Roshan

3 7

[asio] How to get the name of the endpoint from the ip address of the endpoint.
by Andrew Maclean 03 Feb '09

03 Feb '09

If I have an endpoint as an ip address (IP V4 or IPv6), how can I do a DNS lookup to get the name of the endpoint? In the example below, I understand that the host_name is coming from the query. But how do I do a lookup to get the host name from the IP? This may help. If I do something like this: ip::tcp::resolver_iterator res = ip::tcp::resolver(ios).resolve(ip::tcp::resolver_query("google.com", "80")); for(;res != ip::tcp::resolver_iterator(); ++res) { std::cout << res->host_name() << ":" << res->service_name() << " "; std::cout << res->endpoint().address().to_string() << std::endl; } The the output is: google.com:80 209.85.171.100 google.com:80 74.125.67.100 google.com:80 74.125.45.100 However if I do: ip::tcp::resolver_iterator res = ip::tcp::resolver(ios).resolve(ip::tcp::resolver_query("209.85.171.100", "80")); I get: 209.85.171.100:80 209.85.171.100 so th ehosr name comes from the query. I really want: google.com:80 209.85.171.100 Thanks in advance for any help. Andrew -- ___________________________________________ Andrew J. P. Maclean Centre for Autonomous Systems The Rose Street Building J04 The University of Sydney 2006 NSW AUSTRALIA Ph: +61 2 9351 3283 Fax: +61 2 9351 7474 URL: http://www.acfr.usyd.edu.au/ ___________________________________________

2 1

Question in usage of boost::bind
by Archie14 03 Feb '09

03 Feb '09

Below is an example that fails to compile. I have a vector of pointers to a templated class and I need to use find_if. My understanding of boost::bind. is that the first argument, when binding member function, is an explicit "this" pointer, that should be A* in my case. Any help in explaining what I am doing wrong will be greatly appreciated. #include <algorithm> #include <string> #include <vector> #include <boost/bind.hpp> #include <algorithm> #include <string> #include <vector> #include <boost/bind.hpp> template<typename T> class A { T _val; public: A(T val) : _val(val){}; T Val() {return _val;} bool CompareByVal(A<T>* a, int iType, T ofval) { return ( (a != NULL) && (iType == 1) && (a->Val() == ofval) ); } }; int _tmain(int argc, _TCHAR* argv[]) { std::vector<A<int>*> a; std::find_if (a.begin(), a.end(), boost::bind(&A<int>::CompareByVal, _1, _2, 1,1)); }

3 2