- Boost-users - lists.boost.org

Re: [Boost-users] [boost] [thread] synchronized_value: value and move semantics
by Klaim - Joël Lamotte 26 Jun '13

26 Jun '13

On Wed, Jun 26, 2013 at 12:37 PM, Ben Pope <benpope81(a)gmail.com> wrote: > > I think this should just not exist in C++11 and instead be replaced by > something like monitor<T> described by Herb Sutter [1]: I'm using several variants of Monitor in my projects. It don't work well in cases where what you really want is syncrhonization, which happen sometime when the design suggests it. In this case you need a mutex, so synchronized_value helps with that. In my opinion, If the multiple access is a bottleneck in real application, it is very easy to spot with a profiler and modify the code as you pointed. Joel Lamotte

1 0

Problem Iterating a directory
by Sarma Tangirala 26 Jun '13

26 Jun '13

Hello, I am using boost filesystem to iterate through my directory and have encountered a problem that I am unsure of how to debug. try { //for ( boostfs::directory_iterator dir(boostfs::path("/home/vtangira/Desktop/person_finder_v20130513/data/dummy1")), end_itr; for ( boostfs::directory_iterator dir(boostfs::path(paths_to_files.at(0).c_str())), end_itr; dir != end_itr; dir++ ) { std::cout << "File: " << dir->path() << std::endl; } } catch (boostfs::filesystem_error e) { cout << "Error Code: " << e.code() << endl << "Error Msg: " << e.code().message() << endl; } In this snippet, the line with the for loop that has been commented out works but the the one below it with the vector does not. The vector version represents the same string. Any advice is appreciated. Thanks -- 0 1 0 0 0 1 1 1 1

3 4

Re: [Boost-users] [Asio] Does stop() cancel handlers post()ed to io_service?
by Marat Abrarov 26 Jun '13

26 Jun '13

> However, I'm still interested whether stop() should always cancel the > queued handlers, i.e. if this is a bug (on windows). The io_service > documentation says this regarding the use of a work object > (http://alturl.com/98rvq) > > "To effect a shutdown, the application will then need to call the > io_service > object's stop() member function. This will cause the io_service run() > call > to return as soon as possible, abandoning unfinished operations and > without > permitting ready handlers to be dispatched. > Alternatively, if the application requires that all operations and > handlers > be allowed to finish normally, the work object may be explicitly > destroyed." I think (I've just inspected the Asio code) this behavior is designed in such a way for performance reasons - to not check "stopped" flag too often. Even the "stopped" flag is just an atomic value - it's Windows IOCP where OS uses specific technics for scheduling IOCP-tied threads and, may be, the author of Asio thinks (or even knows/has experience) that even atomic-value-check degrades performance or brakes specific scheduling of IOCP-tied threads. Regards, Marat Abrarov.

1 0

[Boost.Process] Status request
by Damien Kick 26 Jun '13

26 Jun '13

Thought that perhaps somebody on this list would know if/where a/the most current version of Boost.Process might be found. Or has this library been abandoned? Begin forwarded message: From: Damien Kick <dkick(a)shoretel.com<mailto:dkick@shoretel.com>> Subject: Boost.Process Date: June 25, 2013 5:16:10 PM CDT To: <boris(a)highscore.de<mailto:boris@highscore.de>> I've found your version of Boost.Process at <http://www.highscore.de/cpp/process/index.html>. I noticed that the date on this page is from 2009; are you still working on this library? If not, thank you for your effort so far. If you are still working on the library, I was hoping you might have some insights and/or help for a compilation error I'm finding when trying to use Boost.Process with Boost 1.53 on Windows 7 with Visual Studio 2010 SP1. First of all, I had to add a header file to .../process/operation.hpp; i.e. I had to add #include <boost/filesystem/operation.hpp> to get the declaration of boost::filesystem::filesystem_error. But then I'm finding the following error 1>------ Build started: Project: ZERO_CHECK, Configuration: Debug Win32 ------ 2>------ Build started: Project: test-cma, Configuration: Debug Win32 ------ 2> cma.cpp 2>Z:\Documents\perforce\dkick-mac\projects\DandM\DMTarget\windows-6.1\msvc-10.0\include\boost/process/detail/pipe.hpp(129): error C2665: 'boost::system::system_error::system_error' : none of the 7 overloads could convert all the argument types 2> Z:\Documents\perforce\dkick-mac\projects\DandM\DMTarget\windows-6.1\msvc-10.0\include\boost/system/system_error.hpp(39): could be 'boost::system::system_error::system_error(int,const boost::system::error_category &,const std::string &)' 2> Z:\Documents\perforce\dkick-mac\projects\DandM\DMTarget\windows-6.1\msvc-10.0\include\boost/system/system_error.hpp(43): or 'boost::system::system_error::system_error(int,const boost::system::error_category &,const char *)' 2> while trying to match the argument list '(DWORD, overloaded-function, const char [54])' ========== Build: 1 succeeded, 1 failed, 3 up-to-date, 0 skipped ========== Which is coming from .../boost/process/detail/pipe.hpp (the bold italic line below). From the quick skim I've given some of the documentation, I think the fact that BOOST_PROCESS_WINDOWS_USE_NAMED_PIPE is not defined means that I would find some features disabled, right? How is this determined; i.e. whether or not to define BOOST_PROCESS_WINDOWS_USE_NAMED_PIPE? # if defined(BOOST_PROCESS_WINDOWS_USE_NAMED_PIPE) static unsigned int nextid = 0; std::string pipe = "\\\\.\\pipe\\boost_process_" + boost::lexical_cast<std::string>(::GetCurrentProcessId()) + "_" + boost::lexical_cast<std::string>(nextid++); hs[0] = ::CreateNamedPipeA(pipe.c_str(), PIPE_ACCESS_INBOUND | FILE_FLAG_OVERLAPPED, 0, 1, 8192, 8192, 0, &sa); if (hs[0] == INVALID_HANDLE_VALUE) boost::throw_exception(boost::system::system_error(::GetLastError(), boost::system::system_category, "boost::process::detail::pipe::pipe: CreateNamedPipe failed")); hs[1] = ::CreateFileA(pipe.c_str(), GENERIC_WRITE, 0, NULL, OPEN_EXISTING, FILE_FLAG_OVERLAPPED, NULL); if (hs[1] == INVALID_HANDLE_VALUE) boost::throw_exception(boost::system::system_error(::GetLastError(), boost::system::system_category, "boost::process::detail::pipe::pipe: CreateFile failed")); OVERLAPPED overlapped; ZeroMemory(&overlapped, sizeof(overlapped)); overlapped.hEvent = ::CreateEvent(NULL, TRUE, FALSE, NULL); if (!overlapped.hEvent) boost::throw_exception(boost::system::system_error(::GetLastError(), boost::system::system_category, "boost::process::detail::pipe::pipe: CreateEvent failed")); BOOL b = ::ConnectNamedPipe(hs[0], &overlapped); if (!b) { if (::GetLastError() == ERROR_IO_PENDING) { if (::WaitForSingleObject(overlapped.hEvent, INFINITE) == WAIT_FAILED) { ::CloseHandle(overlapped.hEvent); boost::throw_exception(boost::system::system_error(::GetLastError(), boost::system::system_category, "boost::process::detail::pipe::pipe: WaitForSingleObject failed")); } } else if (::GetLastError() != ERROR_PIPE_CONNECTED) { ::CloseHandle(overlapped.hEvent); boost::throw_exception(boost::system::system_error(::GetLastError(), boost::system::system_category, "boost::process::detail::pipe::pipe: ConnectNamedPipe failed")); } } ::CloseHandle(overlapped.hEvent); # else if (!::CreatePipe(&hs[0], &hs[1], &sa, 0)) boost::throw_exception(boost::system::system_error(::GetLastError(), boost::system::system_category, "boost::process::detail::pipe::pipe: CreatePipe failed")); # endif ________________________________ This e-mail and any attachments are confidential. If it is not intended for you, please notify the sender, and please erase and ignore the contents.

1 0

Re: [Boost-users] [Asio] Does stop() cancel handlers post()ed to io_service?
by Marat Abrarov 26 Jun '13

26 Jun '13

> The program below gives different results on Windows (Visual Studio > 2008) and Linux. While stop() does cancel the handlers on Linux, the same > is not true for Windows. I also tried to destroy the io_service object > instead of calling stop, but this does not cancel the handlers on either > system, even though the documentation seems to state it should. It would be more accurate to say "on Windows asio::io_service::stop() does not cancel already queued (!) handlers". See example below. ~~~~~~~~~~~~~~~~~~~~~~~ #if defined(WIN32) #include <tchar.h> #endif #include <cstdlib> #include <cstddef> #include <iostream> #include <boost/ref.hpp> #include <boost/asio.hpp> #include <boost/bind.hpp> #include <boost/thread.hpp> #include <boost/function.hpp> namespace { typedef boost::function<void (boost::asio::io_service&)> work_factory; void run_test(work_factory wf) { { boost::asio::io_service io_service; { wf(io_service); std::cout << "io_service work created" << std::endl; { std::cout << "work thread starting..." << std::endl; boost::thread work_thread(boost::bind( &boost::asio::io_service::run, boost::ref(io_service))); boost::this_thread::sleep(boost::posix_time::millisec(300)); io_service.stop(); std::cout << "io_service signalled to stop" << std::endl; work_thread.join(); std::cout << "work thread stopped" << std::endl; } } std::cout << "destroying io_service..." << std::endl; } std::cout << "io_service destroyed" << std::endl; } void test_function1() { static int counter = 0; std::cout << "handler #" << counter++ << std::endl; boost::this_thread::sleep(boost::posix_time::millisec(100)); } void post_test_function1(boost::asio::io_service& io_service) { for (std::size_t i = 0; i != 10; ++i) { io_service.post(&test_function1); } } void post_test_function2(boost::asio::io_service&); void test_function2(boost::asio::io_service& io_service) { static int counter = 0; std::cout << "handler #" << counter++ << std::endl; boost::this_thread::sleep(boost::posix_time::millisec(100)); post_test_function2(io_service); } void post_test_function2(boost::asio::io_service& io_service) { io_service.post(boost::bind(test_function2, boost::ref(io_service))); } } // anonymous namespace #if defined(WIN32) int _tmain(int /*argc*/, _TCHAR* /*argv*/[]) #else int main(int /*argc*/, char* /*argv*/[]) #endif { std::cout << "Test 1" << std::endl; run_test(post_test_function1); std::cout << std::endl << "Test 2" << std::endl; run_test(post_test_function2); return EXIT_SUCCESS; } ~~~~~~~~~~~~~~~~~~~~~~~ Regards, Marat Abrarov.

2 1

Declaring tricky templated variables such as disjoint_sets.
by Jeremy Murphy 25 Jun '13

25 Jun '13

Hi, this is a somewhat general C++ question, but I'm specifically concerned about a Boost class in this case. How do I declare a boost::disjoint_sets<Rank, Parent, FindCompress> variable without some kind of automatic template deduction? Usually I see such classes declared taking their template parameters from within a function such as: template <typename Rank, typename Parent> void foo(Rank rank, Parent parent) { disjoint_sets<Rank, Parent> dset; } but what if I want to declare dset at file scope? I've been trying to manually construct the Rank and Parent types, but it's slightly doing my head in and the iterator_property_map example seems to be stale. Is there a general C++ technique or a Boost-specific one that can help me out here? Or is it just a matter of laboriously building up the typedefs? Thanks, cheers. Jeremy

2 1

[Asio] Does stop() cancel handlers post()ed to io_service?
by Martin Gernhard 25 Jun '13

25 Jun '13

Hello, I want to cancel all pending handlers of an io_service. Is io_service::stop() supposed to cancel handlers that have been added using post()? The documentation for stop() is somewhat ambiguous on this matter, only stating that run() "should return as soon as possible". Alternatively, should the destructor of io_service cancel the handlers? I am trying to build a "worker thread" using Boost.Asio (1.53) and Boost.Thread as outlined at http://think-async.com/Asio/Recipes. I have already looked at the documentation and searched the bug tracker, the mailing list archives, Google and Stack Overflow but only found an unanswered message from almost two years ago (http://lists.boost.org/boost-users/2011/09/70596.php) as well as another message that describes different behavior of stop() than what I am experiencing (http://thread.gmane.org/gmane.comp.lib.boost.user/73762) The program below gives different results on Windows (Visual Studio 2008) and Linux. While stop() does cancel the handlers on Linux, the same is not true for Windows. I also tried to destroy the io_service object instead of calling stop, but this does not cancel the handlers on either system, even though the documentation seems to state it should. Result on Windows for stop() and ~io_service and on Linux for ~io_service: Start 0 1 2 Stop 3 4 5 6 7 8 9 Result on Linux for stop(): Start 0 1 2 Stop For now I am using the workaround of querying a boost::atomic<bool> flag from every handler and setting it to false after calling stop(). Is there a better way to achieve this, maybe by implementing a custom IO object / service? It feels like overkill for something seemingly simple. Regards, Martin #include <boost/asio.hpp> #include <boost/bind.hpp> #include <boost/function.hpp> #include <boost/scoped_ptr.hpp> #include <boost/thread.hpp> #include <iostream> using namespace boost; using namespace boost::asio; void testFunction() { static int counter = 0; std::cout << counter++ << std::endl; this_thread::sleep( posix_time::millisec( 100 ) ); } int main() { shared_ptr<io_service> m_ioService(new io_service()); scoped_ptr<io_service::work> m_work(new io_service::work(*m_ioService)); scoped_ptr<thread> m_thread; for ( unsigned int i = 0; i != 10; ++i ) { m_ioService->post( &testFunction ); } std::cout << "Start" << std::endl; m_thread.reset( new thread( bind( &io_service::run, m_ioService ) ) ); this_thread::sleep( posix_time::millisec( 300 ) ); std::cout << "Stop" << std::endl; m_work.reset(); m_ioService->stop(); m_thread->join(); m_thread.reset(); m_ioService->reset(); return 0; }

1 0

[asio] synchronous read from sslstream with unknown length
by "Torsten Müller" 25 Jun '13

25 Jun '13

1 0

[asio] How to deal with no-connect and/or read-timeout
by Michael Powell 24 Jun '13

24 Jun '13

Hello, I posted previously, but this is a bit of a follow-on thought. Did some searching, and SO has this note: http://stackoverflow.com/questions/13126776/asioread-with-timeout Pardon me while I stare in disbelief. Seriously? Okay, I'd like to use Boost.Asio to facilitate things. Is this the way (one proposed way) of handling read-with-timeout? Our scenario goes similar to this I think: * Client may (or may not) connect, depending on the integration scenario. The main service thread should tolerate either, with plausibly the socket connection/session operating in another thread "asynchronously". * When client connects, the session logic is pretty straightforward: read if/when there is something to read, and pass it through a message framing unwrapper. Completed message gets deserialized and published into a thread-safe queue. * Events may be generated from the main service thread and subsequently published to the client: effectively a write operation, after passing a serialized message through a message framing wrapper. Doable? Feasible, possibly with some sort of a ping-ponging-back-and-forth type of session handling callback mechanism? I am cautiously optimistic that this has more to do with connection/session callback/state-machine type things, than it does Asio? Regards, Michael

2 2

[asio] Grasping the proto-typical socket server
by Michael Powell 24 Jun '13

24 Jun '13

Hello, I've dealt with sockets, and socket servers, in the past, so am familiar with them. Abstracting out a prototypical server details, drawing from the plethora of examples, this is what I can determine. I see three primary abstractions, the server itself, with endpoint binding, socket connection acceptor, so far so good. Then there's a connection, which is a sort of io_service to socket adapter. Followed closely by a session concern, where reads and writes are actually performed. Studying the examples, however, it seems to me that by design io_service wants to be *the* first class citizen that runs, with asynchronous callbacks being the mode of choice. How might we go about mixing this code, presumably in the main thread (?), with any worker service-oriented thread that might be going on. For instance, I start a thread and wait for it to finish and rejoin for a main service loop. Socket connection sessions are secondary to this service behavior taking place IMO. If a client connects, great, we'll interact with it. If not, that should be transparent to the application. Possibly session is a second thread worker with a thread-local-scope io_service, that we pub/sub messages to the main service loop. Thoughts on this type of approach? Much appreciated... Regards, Michael

2 1