March 2013 - Boost-users - lists.preview.boost.org

[BGL] Having a cost function as weight map
by Maxime van Noppen 06 Sep '13

06 Sep '13

Hi, I can't find how to do this pretty basic thing : having a function instead of a field to compute the cost of edges for various path finding algorithms. I couldn't find the answer in the documentation and Google was of no help. Did I miss something obvious ? Example : ------------------- #include <boost/graph/dijkstra_shortest_paths.hpp> #include <boost/graph/graph_traits.hpp> #include <boost/graph/adjacency_list.hpp> struct my_edge { double cost() const { return cost_ * 2; } double cost_; }; struct my_node { }; int main() { typedef boost::adjacency_list< boost::vecS, boost::vecS, boost::directedS, my_node, my_edge > graph_t; typedef boost::graph_traits<graph_t>::vertex_descriptor vertex_t; graph_t graph; vertex_t v /*= ... */; using boost::weight_map; // Works : boost::dijkstra_shortest_paths(graph, v, weight_map(boost::get(&my_edge::cost_, graph))); // Doesn't compile : //boost::dijkstra_shortest_paths(graph, v, // weight_map(boost::get(&my_edge::cost, graph))); } ------------------- Thanks ! -- Maxime

4 8

Identify cycles in a undirected graph
by NsPx 20 Aug '13

20 Aug '13

Hi all, I use undirected_dfs to detect all cycles in a undirected graph. I detect a cycle with DFSVisitor.back_edge(e,g) but how can I get the list of vertices defining this cycle ? Thanks by advance

4 4

[thread] synchronized_value: value and move semantics
by Vicente J. Botet Escriba 08 Jun '13

08 Jun '13

Hi, boost::synchronized_value (not released yet [1][2][3]) is based on [4]. See below part of this paper adapted to the Boost.Thread interface. Currently boost::synchronized_value is in addition Copyable and Swappable. I was wondering if it is worth adding value and move semantics to synchronized_value. Making it EqualityComparable, LessThanComparable and Movable if the underlying type satisfy these requirements will allow to store them on a Boost.Container/C++11 container. Do you see something completely wrong with this addition? Has some of you had a need for this? Could you share the context? Best, Vicente [1] https://svn.boost.org/svn/boost/trunk/boost/thread/synchronized_value.hpp [2] https://svn.boost.org/svn/boost/trunk/libs/thread/example/synchronized_valu… [3] https://svn.boost.org/svn/boost/trunk/libs/thread/example/synchronized_pers… [4] "Enforcing Correct Mutex Usage with Synchronized Values" by Anthony Williams http://www.drdobbs.com/cpp/enforcing-correct-mutex-usage-with-synch/2252002… ------------- [section The Problem with Mutexes] The key problem with protecting shared data with a mutex is that there is no easy way to associate the mutex with the data. It is thus relatively easy to accidentally write code that fails to lock the right mutex - or even locks the wrong mutex - and the compiler will not help you. std::mutex m1; int value1; std::mutex m2; int value2; int readValue1() { boost::lock_guard<boost::mutex> lk(m1); return value1; } int readValue2() { boost::lock_guard<boost::mutex> lk(m1); // oops: wrong mutex return value2; } Moreover, managing the mutex lock also clutters the source code, making it harder to see what is really going on. The use of synchronized_value solves both these problems - the mutex is intimately tied to the value, so you cannot access it without a lock, and yet access semantics are still straightforward. For simple accesses, synchronized_value behaves like a pointer-to-T; for example: boost::synchronized_value<std::string> value3; std::string readValue3() { return *value3; } void setValue3(std::string const& newVal) { *value3=newVal; } void appendToValue3(std::string const& extra) { value3->append(extra); } Both forms of pointer dereference return a proxy object rather than a real reference, to ensure that the lock on the mutex is held across the assignment or method call, but this is transparent to the user. [endsect] [/The Problem with Mutexes] [section Beyond Simple Accesses] The pointer-like semantics work very well for simple accesses such as assignment and calls to member functions. However, sometimes you need to perform an operation that requires multiple accesses under protection of the same lock, and that's what the synchronize() method provides. By calling synchronize() you obtain an strict_lock_ptr object that holds a lock on the mutex protecting the data, and which can be used to access the protected data. The lock is held until the strict_lock_ptr object is destroyed, so you can safely perform multi-part operations. The strict_lock_ptr object also acts as a pointer-to-T, just like synchronized_value does, but this time the lock is already held. For example, the following function adds a trailing slash to a path held in a synchronized_value. The use of the strict_lock_ptr object ensures that the string hasn't changed in between the query and the update. void addTrailingSlashIfMissing(boost::synchronized_value<std::string> & path) { boost::strict_lock_ptr<std::string> u=path.synchronize(); if(u->empty() || (*u->rbegin()!='/')) { *u+='/'; } } [endsect] [/Beyond Simple Accesses] [section Operations Across Multiple Objects] Though synchronized_value works very well for protecting a single object of type T, nothing that we've seen so far solves the problem of operations that require atomic access to multiple objects unless those objects can be combined within a single structure protected by a single mutex. One way to protect access to two synchronized_value objects is to construct a strict_lock_ptr for each object and use those to access the respective protected values; for instance: synchronized_value<std::queue<MessageType> > q1,q2; void transferMessage() { strict_lock_ptr<std::queue<MessageType> > u1 = q1.synchronize(); strict_lock_ptr<std::queue<MessageType> > u2 = q2.synchronize(); if(!u1->empty()) { u2->push_back(u1->front()); u1->pop_front(); } } This works well in some scenarios, but not all - if the same two objects are updated together in different sections of code then you need to take care to ensure that the strict_lock_ptr objects are constructed in the same sequence in all cases, otherwise you have the potential for deadlock. This is just the same as when acquiring any two mutexes. In order to be able to use the dead-lock free lock algorithms we need to use instead unique_lock_ptr, which is Lockable. synchronized_value<std::queue<MessageType> > q1,q2; void transferMessage() { unique_lock_ptr<std::queue<MessageType> > u1 = q1.unique_synchronize(boost::defer_lock); unique_lock_ptr<std::queue<MessageType> > u2 = q2.unique_synchronize(boost::defer_lock); boost::lock(u1,u2); // dead-lock free algorithm if(!u1->empty()) { u2->push_back(u1->front()); u1->pop_front(); } } [endsect] [/Operations Across Multiple Objects]

2 4

Is boost::this_thread::sleep_for interruption point?
by Serg Gulko 23 May '13

23 May '13

Hello! According to documentation(v 1.50.0) call boost::this_thread::sleep_for can be interrupted. For some reason this functionality not works for me. Here is code sample: //Simple thread function void Test::worker(int delay) { try { boost::this_thread::sleep_for(boost::chrono::milliseconds(delay)); cout << "Doing some work" << endl; } catch (boost::thread_interrupted const& e) { cout << "INTERRUPTED" << endl; } } //Far in the code boost::thread workerThread(boost::bind(&Test::worker, this, 1200)); boost::this_thread::sleep(boost::posix_time::milliseconds(30)); workerThread.interrupt(); boost::this_thread::sleep(boost::posix_time::milliseconds(1500)); I always getting 'Doing some work'... -- View this message in context: http://boost.2283326.n4.nabble.com/Is-boost-this-thread-sleep-for-interrupt… Sent from the Boost - Users mailing list archive at Nabble.com.

3 3

[FunctionTypes] Testing for a general callable type.
by Robert Jones 24 Apr '13

24 Apr '13

Hi All Is there anything in Boost to help me test for a general type being callable with a specific signature? I've found Boost.FuncionTypes, which seems to be confined to *builtin* callable types, but has made me aware that all the const/volatile/reference variations might make a general test more complicated than I first thought. Any pointers much appreciated. Thx, Rob.

6 11

Re: [Boost-users] [CMake] [boost][b2] Differenciate 32bit and 64 bit binaries when using CMake.
by Klaim - Joël Lamotte 16 Apr '13

16 Apr '13

Júlio Hoffimann said: > > I don't know if i understood correctly, but you can set the library > directory by hand, please refer to > /usr/share/cmake-<version>/Modules/FindBoost.cmake, in particular the > variable BOOST_LIBRARYDIR. On Thu, Oct 11, 2012 at 1:29 PM, Miller, Frank <FMiller(a)sjm.com> wrote: > A simple solution would be to install only the 64bit version of boost > into a new directory and change the environment variable to point there. > Then FindBoost will only be able to see the version you want. > Thanks, this will solve my immediate problem, but it don't solve it in general. (I will use it though, so thanks) I can't guarantee where the boost libraries are installed, that's why I use FindBoost. My current understanding is that boost don't have a specific default lib dir for different address models, which is the source of the problem as FindBoost script can't rely on a protocol to find these lib files. Am I correct? Joel Lamotte

3 8

[Boost.Coroutine] Problem compiling simple code with c++98 compiler
by Lukas Czaja 03 Apr '13

03 Apr '13

Hi, i was checking if my code is compatible with c++98 standard and i've encounterd compile error. With -std c++11 option everything works fine. Below i've written code causing same problem. #include <boost/coroutine/coroutine.hpp> #include <boost/coroutine/attributes.hpp> typedef boost::coroutines::coroutine<void( boost::system::error_code , std::size_t)> coro_t; struct coro_test{ coro_test() {} void coro_fun(typename coro_t::caller_type& ca)const{ boost::system::error_code ec; boost::tie(ec, boost::tuples::ignore) = ca.get(); ca(); } coro_t coro; }; int main(int argc, char *argv[]) { { boost::system::error_code ec; coro_t coro2; //default construction is ok coro_test ctest; //compile error , why ??? coro_t coro( boost::bind(&coro_test::coro_fun, &ctest, _1),coro_t::arguments( ec, 7)); coro = coro_t(boost::bind(&coro_test::coro_fun, &ctest, _1),coro_t::arguments( ec, 7)); } ... /usr/local/include/boost/type_traits/is_function.hpp:50: błąd:forming pointer to reference type 'boost::rv<boost::coroutines::coroutine<void(boost::system::error_code, long unsigned int)> >&' /usr/local/include/boost/type_traits/is_function.hpp:102: błąd:'value' is not a member of 'boost::detail::is_function_impl<boost::rv<boost::coroutines::coroutine<void(boost::system::error_code, long unsigned int)> >&>' /usr/local/include/boost/type_traits/is_convertible.hpp:354: błąd:'value' is not a member of 'boost::is_function<boost::rv<boost::coroutines::coroutine<void(boost::system::error_code, long unsigned int)> >&>' /usr/local/include/boost/utility/enable_if.hpp:59: błąd:'value' is not a member of 'boost::is_convertible<boost::coroutines::coroutine<void(boost::system::error_code, long unsigned int)>&, boost::rv<boost::coroutines::coroutine<void(boost::system::error_code, long unsigned int)> >&>' I work with Qt, g++ compiler on ubuntu 12.10 Is this library bug ?

2 3

[serialization] serializing a pointer without causing object tracking
by Stefan Strasser 02 Apr '13

02 Apr '13

Hi everyone, Hi Robert, in case you'd like to answer this one, because I have a rather specific question about your library. Thank you in advance! I'd like to serialize an object and later reconstruct it from the serialized data: void save(T const &t){ ar << t; } T* load(){ T *t=new T(); //error ar >> *t; } T might not have a default constructor. so load_construct_data should be used. apparently the way to do this is to serialize through a pointer: void save(T const *t){ ar << t; } T *load(){ T *t; ar >> t; } this works, but it causes the type T to be tracked, if its object tracking is set to track_selectively. I don't want that. but I can't set it to track_never either, because if the type is serialized at another point in the program through a pointer, the type should be tracked. is there a solution to this? is there a serialization wrapper that says "don't track this type just because of this pointer"? what comes to mind as a solution is serializing as an object, and calling load/save_construct_data manually. but then there is another problem: void save(T const &t){ save_construct_data(t,version<T>::value); ar << t; } void load(T *unconstructed_t){ load_construct_data(unconstructed_t, ??? ); ar >> *t; } if the implementation level of T is object_class_info, the object version is loaded from the preamble when the object itself is loaded. there is no in between loading the version and loading the object, it's entirely handled in the non-header part of the library. so basically I have a solution when the implementation level is object_serializable (no version info, call load_construct_data manually), and when tracking is set to track_never (serialize as a pointer). but not for the default case, object_class_info with track_selectively. am I missing something? Stefan

3 6

Newbie Interprocess message_thread user
by John Davies 01 Apr '13

01 Apr '13

I'm new to Boost and new to Linux. But old to Windows. So I'm doing some initial development in Windows XP. I have two executables made in different projects. I planned on using a message_queue to communicate between the two applications. But sending a message from one app does not make it to the other. I can add and handle messages from one thread to another within the same application. SO I know I have things partially setup correctly. But now I'm at a complete loss. Should two exe files be able to communicate through Boost message_queues? Or am I misunderstanding the whole thing?

2 2

Problems using for_each
by Michael Powell 31 Mar '13

31 Mar '13

Hello, I am struggling making for_each work right. I am using Sourcery targeting ARM. AFAIK, GCC is actually 4.7.2, and I am flagging -std=c++11. I'm not really picky how we get for_each done (or constructs like it). I'd like to avoid "macro-expansion" for obvious reasons (i.e. debug), and the always-infamous verbosity of a for loop, if possible. I've got the following: vector<rgb_type*> all; rgb_type rgb; rgb_type rgb2 = rgb.clone(); for_each(all.begin(), all.end(), cout << _1->r() << endl); With errors: _1 was not declared in this scope. (C++ Problem) That or, void rgb_printer_func(rgb_type* prgb) { cout << prgb->r() << endl; } for_each(all.begin(), all.end(), rgb_printer_func); With errors: Invalid arguments Candidates are: #1 for_each(#0, #0, #1) (Semantic Error) Thank you. Regards, Michael Powell

2 4