May 2009 - Boost-users - lists.preview.boost.org

[uBLAS] Initialize a matrix by T*?
by Marco Guazzone 10 May '09

10 May '09

Hi, I'm new to uBLAS. I wonder if there is an efficient way to initialize a ublas::matrix<T> by a T*. Somthing like a copy(matrix<T>& m, T* p, size_t nrows, size_t ncols). For instance: --- [code_snip] --- double* pd = new double[10*20]; //... ublas::matrix<double> m; copy(m, pd, 10, 20); --- [/code_snip] --- Obviously I can create a 10x20 empty matrix (matrix<T> m(10,20)) and then scan the T* assigning each element one at time. But is there a more efficient way? Thank you very much in advance!! -- Marco

1 0

boost::asio and boost::statecharts newbie question
by Dave Mazzoni 10 May '09

10 May '09

Hello all - I am using the boost::asio library to do serial I/O with simple Linux serial port(s) like /dev/ttyS0. I have found out how this works, and it is a very impressive library (I will offer a small example program so that you might include it for other users of ASIO serial ports once I've got it cleaned up). However, in addition to using boost::asio, I am using the boost::statechart library as the controller (FSM) for a project. I have managed to get the serial port working with the boost::asio library. Now, my problem is that I have an io_service defined and initialized in my statechart FSM. Is there some way to cast or otherwise use the io_service that I have therein into a serial_port? In other words, I would rather not have to put the serial port handling code in the FSM (I'd rather 'refer' to it using the io_service if it is possible). Thanks for your help! Dave

3 2

boost::thread build with 64-bit mingw gcc
by Serge Skorokhodov 09 May '09

09 May '09

Hi, I tried to build boost 1.39 with 64-bit gcc 4.4 from www.equation.com and the thread library failed to compile. As it is stated here (http://sourceforge.net/forum/message.php?msg_id=5570690) the problem is (partly at least) lack of boost::thread support for 64-bit windows gcc. Will coming 1.40 version address this problem? TIA -- The fish that is singing in Ucayali river...

1 0

mpl::find_if / placeholder question
by vasyl_y 09 May '09

09 May '09

Hi, I started learn mpl and here we go... I have an mpl::vector of compund types and i want to find an element in vector based on property of compound type and cannot find the way to do it, any suggestions? Here is an example how i'm trying to do it: #include <boost/mpl/vector.hpp> #include <boost/mpl/find_if.hpp> #include <boost/type_traits.hpp> namespace mpl = boost::mpl; template<typename T, int id> struct Type2IdLink { typedef T value_type; typedef mpl::int_<id> type_id; }; typedef Type2IdLink<char, 1> v1; typedef Type2IdLink<int, 2> v2; typedef Type2IdLink<unsigned, 3> v3; typedef mpl::vector<v1, v2, v3> types; // This work typedef mpl::find_if<types, boost::is_same<mpl::_1, v2> >::type iter; // This does not work: // value_type is not a member of mpl_::_1 typedef mpl::find_if<types, boost::is_same<mpl::_1::value_type, v2> >::type iter; BOOST_MPL_ASSERT_RELATION( iter::pos::value, == , 1); Using BOOST version 1.38, GCC4.3. Thank You -- View this message in context: http://www.nabble.com/mpl%3A%3Afind_if---placeholder-question-tp23463044p23… Sent from the Boost - Users mailing list archive at Nabble.com.

2 1

boost - asio - advise
by Tim Pynegar 09 May '09

09 May '09

Hi, I'm trying to use asio for a udp client / server arrangment. I started by calling run on the io_service object on a separate thread and the using the async_read_from and the async_send_to. I was calling these from the main application thread. I was getting a lot of these errors that was causing an exception on the io_service run method. First-chance exception at 0x7c812afb in Testbed.exe: Microsoft C++ exception: boost::system::system_error at memory location 0x0012e29c. An invalid argument was supplied This happens after calling the async_send_to method. I've added in a send buffer and started to poll the io_service object instead of running it on it's own thread which has cut down the freqency this is happening however on occations it still does and i'd like to have it running on it's own thread. The question is really should i put in a mutex lock in my sendto function that call's the async_send_to that is unlocked when i recieve the handlesentto callback that's called when the async_send_to is finished. If i do this doesn't this basically make it a blocking function call in the same way as if i call the regular send_to function. There are other options i've thought of but they all seem to result in it basically becoming blocking. I've looked through the examples and there is a lot of different potential choices to make and was looking to see if anyone could advise on an arrangement that would allow for using multiple threads but block as little as possible. Thanks, Tim.

3 2

Asio in Windows/Linux
by Kenny Stone 09 May '09

09 May '09

I'm trying to use asio in linux and windows, and I have to use a very specific way of starting my IO Service to get it to work in windows xp: void start() { tcp::endpoint endpoint = *iterator_; socket_.async_connect(endpoint, boost::bind(&MessageHandler::handle_connect, this, boost::asio::placeholders::error, ++iterator_ ) ); boost::thread t( boost::bind( &boost::asio::io_service::run, &(socket_.get_io_service() ))); t.detach(); //threadPool_.schedule(boost::bind( &boost::asio::io_service::run, &(socket_.get_io_service()) ) ); } The commented out line is what I wanted to use (threadpool), as thread detach is far from an ideal situation. To work in windows, I have to call io_service.run() in this function and thread detach; it doesn't seem to work any other way. Everything pretty much works in linux. Is there a pattern that I should be using here for cross platform usage of the library? Any ideas why the threadpool solution doesn't work in windows? ( http://threadpool.sourceforge.net/) Thanks, -- Kenny Stone Connamara Systems, LLC

1 0

[BGL] Auto Connect vertexes by predicate?
by arm2arm 09 May '09

09 May '09

Hello, Are there ways to generate Graph with edges by the user defined function? Should I generate Vertex relations by hand then put it in to Graph? I would like to feed graph by points 3d and each point is linked to its closest neighbor by distance. Then as a next step I would like to dig in the graph the sub-graphs with the given linking length. My Data structure is following: _____________________ class CCoord { public: CCoord(){}; CCoord(int id_,MyFloat x_,MyFloat y_,MyFloat z_, MyFloat w_=0): id(id_), x(x_), y(y_), z(z_), w(w_){ if(w==0.0) w=sqrt(x*x+y*y+z*z); }; MyFloat x, y, z; int id; MyFloat w; } typedef std::vector<CCoord> typeVecData; typedef adjacency_list < vecS, vecS, undirectedS, no_property, property < edge_weight_t, MyFloat > > Graph; typedef graph_traits < Graph >::edge_descriptor Edge; typedef graph_traits < Graph >::vertex_descriptor Vertex; typedef std::pair<int, int> E; typeVecData data(num_nodes); srand(0); generate(data.begin(),data.end(),gencoord); typedef std::pair<int, int> Edge; Graph G(5); add_edge(0, 2, G);//how to apply relation function? ________________________________________________ What is the correst way to implement this algorithm using BGL? Thank you beforehand. Arman -- View this message in context: http://www.nabble.com/-BGL--Auto-Connect-vertexes-by-predicate--tp23451992p… Sent from the Boost - Users mailing list archive at Nabble.com.

4 6

Re: [Boost-users] [asio] sync receive_from with timeout
by Michael Dehmlow 09 May '09

09 May '09

Ha this is exactly the problem I ran into not two days ago I agree it would be nice if there was a timeout for the library: I used to following: //Oh throw a sleep in there: while(!timeout.elapsed()) { boost::system::error_code ec; boost::asio::socket_base::non_blocking_io command(true); socket.io_control(command); bytesRead = _socket->read_some(boost::asio::buffer(data, dataBufferSize), ec); if(error&&boost::asio::error::would_block!=error) { //there is some other error the client has disconnected; } else { //I must have got some data or theres no data yet check bytesRead for that } boost::this_thread::sleep( boost::posix_time::milliseconds(100)); } No extra threads necessary. Stephan Menzel-3 wrote: > > Is the subject Jehova enough for you? ;-) > > Great. > > Well, I haven't raised the subject so far but I'd certainly like to do so > now. For years now I am not only a user but also an admirer of boost asio. > Something that has always puzzled me however is, why a timeout > functionality which is something very common and essential for sync IO > just doesn't seem to be there. > > I think there are cases where syncronous IO is a better way to go and no > async approach is sensible. In my case this is a collection of static > functions that do UDP communication and therefore have to be sync because > I just want them to be able to return and throw exceptions and be able to > be called from anyone at any time. > > As always when I stumble across the asio/timeout problem I start googling > around. When people raise the topic here they are usually pointed to an > answer Chris Kohlhoff gave to someone asking that: > > http://lists.boost.org/Archives/boost/2007/04/120339.php > > ... which is good for tcp and works alright even though in my opinion it > leaves the question why it has to be so compilicated to do something so > basic. IMHO sync IO is all but pointless without a timeout. > > Anyway, I started to turn this into a UDP version and ended up with > something very similar: > > inline void set_result(boost::optional<boost::system::error_code>* a, > boost::system::error_code b) { > a->reset(b); > }; > > template <typename MutableBufferSequence> > void udp_read_with_timeout(boost::asio::ip::udp::socket& sock, int tmo, > boost::asio::ip::udp::endpoint &endpoint, const MutableBufferSequence& > buffers) { > > boost::optional<boost::system::error_code> timer_result; > boost::asio::deadline_timer timer(sock.io_service()); > timer.expires_from_now(boost::posix_time::seconds(tmo)); > timer.async_wait(boost::bind(set_result, &timer_result, _1)); > > boost::optional<boost::system::error_code> read_result; > sock.async_receive_from(buffers, endpoint, boost::bind(set_result, > &read_result, _1)); > > sock.io_service().reset(); > while (sock.io_service().run_one()) { > if (read_result) > timer.cancel(); > else if (timer_result) > sock.cancel(); > } > > if (*read_result) > throw boost::system::system_error(*read_result); > }; > > > So now I started to go ahead with that but I had to notice two things. > First, the performance impact is massive to a point that forces me to > reconsider. Second, when I use that in a multithreaded environment the > whole thing seems to block in some threads randomly. I have made test > program where I have multiple threads using that method, just sending an > UDP request to a server and waiting for an answer using this function. In > one thread everything is slow but continously running. In more than one > thread it blocks for up to 5 or 6 seconds every few (10-20) runs for no > visible reason. In my scenario all those threads share a socket protected > by a mutex so if one of them blocks all will be halted. > > So I come to think this is not a good way to handle the problem. Does > anyone have some thoughts on that? May this be a bug in asio (1.38) or > perhaps in my function? Or is there a better way to do sync UDP IO with a > timeout functionality? > > Cheers... > > Stephan > > > > _______________________________________________ > Boost-users mailing list > Boost-users(a)lists.boost.org > http://lists.boost.org/mailman/listinfo.cgi/boost-users > > -- View this message in context: http://www.nabble.com/-asio--sync-receive_from-with-timeout-tp22326399p2233… Sent from the Boost - Users mailing list archive at Nabble.com.

3 3

is this possible with tuples in c++0x ?
by Germán Diago 09 May '09

09 May '09

template <typename... Types> class tuple ... { template <typename N> auto operator[](std::size_t n = N) -> constexpr decltype(get<N>(*this)) { return get<N>(*this); } }; If this is possible, you could do (and it would be way more natural than get<3>(t)): tuple<int, float, float, string, mytype> t; t[3] = "hello"; But I feel this is not legal because of parameter std::size_t :(.

2 1

more than one template with automated registration
by tang10＠llnl.gov 08 May '09

08 May '09

Is there a macro similar to BOOST_AUTO_TEST_CASE_TEMPLATE for this: template< typename Value, typename Field > void testFunction() { ... }; Thanks, Nu Ai.

1 0