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

Re: [Boost-users] Boost.Asio -- combining sync read with async read
by Sameer Parekh 14 Oct '09

14 Oct '09

> Right. The most simple way to accomplish this is to use > asio::streambuf as a > buffer. When you process the input data, you call consume() function > only > for the amout you've really processed, and the rest of the data > remains > unused in the streambuf - until the next time you get something from > the > server (look at the asio examples with streambuf). Yes, I'm using the streambuf, and then constructing an istream from the streambuf, and then calling getline on the istream. (The chat client/server was the primary example that I used to build my app.) If I have 2.5 lines in the buffer then the first two calls to getline will give me the first two lines, but won't the third call to getline get me the half line, and consume that half line from the streambuf? Do I need to process the streambuf in a more manual fashion rather than calling getline on the istream? Also to make sure I understand it how it works with the streambuf, that I am understanding you correctly: if I was just processing one line in the async_handler, and the first time the server sends me 3 lines, then the handler is called and will process the first line sent. Then the server sends another 3 lines, the async handler is called and the client will process the 2nd line that it received, not the 4th line, right? Thanks for your help, -Sameer

2 1

[lambda] Initializing a std::list of unsigned short points
by Ryan McConnehey 14 Oct '09

14 Oct '09

In my project I'm trying to initialize a std::list of unsigned short pointers. I've created a couple of simple hand written loops but would like to convert them to use std::for_each. I've been unsuccessful in the past getting boost::lambda to work with std container of pointers. In the lambda documentation I know that new_ptr and delete_ptr are need in creating and destroying pointers in lambda functions. I've provided the hand written loop that does work and the "attempted" conversion to std::for_each. Any help on the correct implementation for the lambda values would be appreciated. Ryan typedef std::list<unsigned short *> ml; for (ml::iterator itor = m_list.begin(); itor != m_list.end(); ++itor, startValue++) { *itor = new unsigned short(startValue); } for (ml::iterator itor = m_list.begin(); itor != m_list.end(); ++itor) { delete *itor; } std::for_each(m_list.begin(), m_list.end(), *(bl::_1) = bl::new_ptr<unsigned short>()(bl::var(startValue)++) ); std::for_each(m_list.begin(), m_list.end(), boost::lambda::delete_ptr(*(boost::lambda::_1)));

3 3

Re: [Boost-users] Boost.Asio -- combining sync read with async read
by Sameer Parekh 14 Oct '09

14 Oct '09

> > No, it will read anything available in the socket and put it in the > buffer. > It can be 1 line, 2 lines or 3.25 lines. What I mean is that you > have to be > aware of the fact that the buffer might contain more data than you > "requested" from read_until. Aha, thanks Igor. So my handler needs to be able to deal with a buffer that contains more than one line. That's not so bad, I can just run the processing code for each line until the buffer is empty, that should be fine. However, what concerns me is if the handler receives a buffer that has just 2.5 lines in it. All the messages from the server are going to be one line long. So if the server sends three messages in succession but say there is some network hiccup partway through the 3rd line, then my client will only receive 2.5 lines, and the handler will get called with the partial line? Then the handler has to cache the partial line so that when the rest of the line comes through it can paste them together again? That seems hard... thanks, -s

2 1

[asio] deadline_timer callback is not called
by Daniele Barzotti 14 Oct '09

14 Oct '09

Hi, I've yet posted a question about this topic but I not solve, so I try to ask again. PS: I used the same solution (in conjunction with asio::serial_port) in another project and, there, it works! I'm working on a library and I have only one static io_service and io_service::work like in the following code: ---------------------- START CODE ------------------------ typedef boost::shared_ptr<boost::asio::io_service::work> io_work_ptr; // Main io_service static boost::asio::io_service io_service; static boost::scoped_ptr<boost::thread> io_service_thread; static io_work_ptr p_work; static bool thread_started; void io_worker_thread(void) { #if(WIN32 && _WIN32_DCOM) struct com_init { com_init() { CoInitializeEx(NULL, COINIT_APARTMENTTHREADED); } ~com_init() { CoUninitialize(); } } initializer_object; #endif io_service.run(); }; bool CreateMainThread() { if (!thread_started) { try { // create the work object on the heap: // [Reference counting] // This object is a simple pointer because // it is assigned to each created objects // These objects store it in a shared_ptr so, // when the last object is destroyed // The io_service::work is destroyed as well p_work.reset( new boost::asio::io_service::work(io_service) ); // Instead of start the worker thread directly by // passing the io_service object, // I call the io_worker_thread function where // I can do some initialization tasks like // calling CoInitializeEx for COM io_service_thread.reset(new boost::thread(io_worker_thread)); thread_started = !thread_started; return true; } catch (boost::thread_resource_error e) { // Failed to create the new thread return false; } } else return true; } /*! * \brief Factory function for AudioPlayer objects creation */ EXTERN_C EUROAUDIOLIB_API AudioPlayerHndl CALL CreateAudioPlayer() { if (!CreateMainThread()) return NULL; // create the new object AudioPlayer* pAudioPlayer = new AudioPlayer; pAudioPlayer->io_worker_ptr = p_work; return pAudioPlayer; }; ---------------------- END CODE ------------------------ In the above code pAudioPlayer->io_worker_ptr is declared as: io_work_ptr p_work; In a class I have to use the timer, so I initialize it with the global io_service object: ---------------------- START CODE ------------------------ boost::asio::deadline_timer rx_timer_; WaveStream::WaveStream() : rx_timer_(io_service) { }; void WaveStream::ProcessRTPPacket() { // RTP Packet received ...... //Start a new timer or renew it rx_timer_.expires_from_now( boost::posix_time::milliseconds( 1000 ) ); io_service.reset(); // We managed to cancel the timer. Start new asynchronous wait. rx_timer_.async_wait( boost::bind(&WaveStream::handle_timeout, this, _1) ); .... } void WaveStream::handle_timeout(const boost::system::error_code& error) { std::cout << "BOOST TIMER RAISED!!" << endl; // A Timeout is raised: a stream is end if (!error) {...} }; ---------------------- END CODE ------------------------ But the callbak is never called!! Since ProcessRTPPacket is called frequently, I try also to move the timer on another point just for test with the same effect! Where could be the mistake? Thanks in advance! Daniele.

2 11

Re: [Boost-users] Boost.Asio -- combining sync read with async read (Igor R)
by Sameer Parekh 14 Oct '09

14 Oct '09

> > It's ok to use both sync. and async. calls, but you should not call > boost::asio::async_read_until() and/or boost::asio::async_read() until > previous read is complete (for the same socket/buffer). Besides, pay > attention that reading "until" might read more data than you expect. Thanks Igor, I'm a bit confused here though.. how do I know if the previous read is complete? I figured when async_read_until() calls the handler, the read was complete and everything read was sitting in the buffer? Do you mean to say that if the client receives two lines, then async_read_until() will read both lines but then put only the first line in the buffer? Then I can't call async_read_until() again because there is still another pending call to the handler that will process the second line? So how do I know at the end of the handler whether or not to call async_read_until() ? thanks! -Sameer

2 1

[fusion] probably a bug in result_of::join
by jl 14 Oct '09

14 Oct '09

This is nearly a copy-paste from the doc (and agrees with intuition): #include <boost/fusion/algorithm/transformation/join.hpp> #include <boost/fusion/include/vector.hpp> #include <boost/fusion/include/join.hpp> void foo() { using namespace boost::fusion; vector<> v1, v2; result_of::join< vector<>, vector<> >::type j = join(v1, v2); } ...but it doesn't compile (msvc9, boost 1.40): c:\UserTemp\x54209\BoostTests\Any\main.cpp(20) : error C2440: 'initializing' : cannot convert from 'boost::fusion::result_of::join<LhSequence,RhSequence>::type' to 'boost::fusion::joint_view<Sequence1,Sequence2>' with [ LhSequence=const boost::fusion::vector<>, RhSequence=const boost::fusion::vector<> ] and [ Sequence1=boost::fusion::vector<>, Sequence2=boost::fusion::vector<> ] No constructor could take the source type, or constructor overload resolution was ambiguous J-L

2 3

[Variant] Sorting data out of a vector of variant
by Oliver Rudolph 14 Oct '09

14 Oct '09

Hi all! I have some class with a member which is a vector of boost.variant, i.e. class Example { public: std::vector<variant<typeA, typeB, typeC> > content; } I want to add three members to my class that are vectors of pointers to the specific content items class Example { public: std::vector<variant<typeA, typeB, typeC> > content; std::vector<typeA> *contentA; std::vector<typeB> *contentB; std::vector<typeC> *contentC; } Now the question: how can I iterate through the vector content whilst adding its items to the other 3 vectors depending on their type? I know there has to be some way using the apply_visitor functor, but I just don't see how exactly... Thanks for your help! Best regards, Oli

2 1

Re: [Boost-users] Boost.Serialization not building
by OvermindDL1 14 Oct '09

14 Oct '09

On Wed, Oct 21, 2009 at 8:53 AM, Stefan Strasser <strasser(a)uni-bremen.de> wrote: > >> >> So what do I need to do to the xml archive to get it to work like the >> others? _______________________________________________ > > use NVPs I guess > not sure if that's the cause of THIS error though I tried looking at that, but have no clue what that is (other then some lines of code referencing it I found), or how I could apply it to this situation. Should it not work the same for the xml archive as it does text and binary, which both work fine.

2 1

Boost.Asio -- combining sync read with async read
by Sameer Parekh 14 Oct '09

14 Oct '09

Hello everyone! I am writing a client for a protocol which goes something like this: The client connects to the server, and it submits "orders" -- every time it gets an order it gets back an "ack". However, the server might asynchronously send a "fill" at any time. So I want to be listening for a fill even while the client might be doing other things (listening to another server for one-way data stream. The client then chooses to submit an order or not based on the information it gets in the datastream, or it might choose to submit an order if it receives a fill.) I thought the way to do it would be to do async_read_until() and have it call my fill handler when it receives something. Then the function that submits orders would call write() and then block waiting for the ack message using read_until() However, when I implement that, the client never seems to notice that a fill message is sent. It receives the acks just fine, and it successfully reads the datafeed coming in from the other tcp connection. I thought maybe the read_until() would cancel the async_read_until() so after I received the ack message I called async_read_until() again, but that didn't work either... (The server works as it should, I tested it with telnet.) Is it possible to mix sync/async reads in this way? Can anyone tell if I am doing something wrong? Do I need to do something else to make this work? Thanks, I appreciate your help! Here is the code in question: (as an aside, is it bad form to call connect() in the constructor? Should I call connect() and setup the async_read_until() in a separate start() method?) MarketConnectionTCP::MarketConnectionTCP(boost::asio::io_service &io_service_, std::string ip_, unsigned short agentPort_, unsigned short feedPort_) : _io_service(io_service_), _agentSocket(_io_service), _feedSocket(_io_service), _timer(_io_service, boost::posix_time::seconds(1)) { boost::asio::ip::address ipAddress(boost::asio::ip::address::from_string( ip_)); tcp::endpoint agentEndpoint(ipAddress, agentPort_); Logger::stream() << "Trying to connect to agent:" << agentEndpoint << std::endl; _agentSocket.connect(agentEndpoint); Logger::stream() << "Agent connected:" << _agentSocket.remote_endpoint() << std::endl; boost::asio::ip::tcp::endpoint feedEndpoint(ipAddress, feedPort_); Logger::stream() << "Trying to connect to feed:" << feedEndpoint << std::endl; _feedSocket.connect(feedEndpoint); Logger::stream() << "Feed connected:" << _feedSocket.remote_endpoint() << std::endl; boost::asio::async_read_until(_agentSocket, _fillBuf, "\r\n", boost::bind( &MarketConnectionTCP::handleReadFill, this, boost::asio::placeholders::error)); Logger::stream() << "Listening on agent connection:" << _agentSocket.remote_endpoint() << std::endl; boost::asio::async_read_until(_feedSocket, _feedBuf, "\r\n", boost::bind( &MarketConnectionTCP::handleReadFeed, this, boost::asio::placeholders::error)); Logger::stream() << "Listening on feed connection:" << _feedSocket.remote_endpoint() << std::endl; _timer.async_wait(boost::bind(&MarketConnectionTCP::handleTimer, this, boost::asio::placeholders::error)); Logger::stream() << "Timer is waiting" << std::endl; } OrderID MarketConnectionTCP::sendOrder(const Order &order) { boost::asio::streambuf orderBuf; std::ostream orderStream(&orderBuf); orderStream << order << "\r\n"; // This blocks until we get an ack with the order id Logger::stream() << "Sending order to:" << _agentSocket.remote_endpoint() << ":" << order << std::endl; boost::asio::write(_agentSocket, orderBuf); Logger::stream() << "Waiting for ack from:" << _agentSocket.remote_endpoint() << std::endl; boost::asio::streambuf ackBuf; std::istream ackStream(&ackBuf); boost::asio::read_until(_agentSocket, ackBuf, "\r\n"); OrderID id; ackStream >> id; Logger::stream() << "Received ack from:" << _agentSocket.remote_endpoint() << ":" << id << std::endl; // listen again boost::asio::async_read_until(_agentSocket, _fillBuf, "\r\n", boost::bind(&MarketConnectionTCP::handleReadFill, this, boost::asio::placeholders::error)); Logger::stream() << "Listening on agent connection:" << _agentSocket.remote_endpoint() << std::endl; return id; } void MarketConnectionTCP::handleReadFill(const boost::system::error_code& err) { if (!err) { std::istream fillStream(&_fillBuf); Fill newFill; fillStream >> newFill; if (newFill.isValid()) { Logger::stream() << "Fill received on:" << _agentSocket.remote_endpoint() << ":" << newFill << std::endl; receiveFill(newFill); } else { Logger::stream() << "Invalid fill received on:" << _agentSocket.remote_endpoint() << std::endl; } // listen again boost::asio::async_read_until(_agentSocket, _fillBuf, "\r\n", boost::bind(&MarketConnectionTCP::handleReadFill, this, boost::asio::placeholders::error)); Logger::stream() << "Listening on agent connection:" << _agentSocket.remote_endpoint() << std::endl; } else { delete this; } }

2 1

Boost compatibility with Old Compilers
by chin 14 Oct '09

14 Oct '09

HI, We are planning to use Boost version 1.39 on Windows. In docs i found the following: ---------------------------------------------------------------------------------------------------- Compilers Tested Boost's primary test compilers are: o Linux: o GCC 4.3.3 on Ubuntu Linux. o Windows: o Visual C++ 7.1 SP1, 8.0 SP1 and 9.0 SP1 on Windows XP. Boost's additional test compilers include: o Linux: o Intel 9.0 on Red Hat Enterprise Linux. o Intel 10.0 on Red Hat Enterprise Linux. o Intel 10.1 on 64-bit Linux Redhat 5.1 Server. o Intel 10.1 on Suse Linux on 64 bit Itanium. o Intel 11.0 on Red Hat Enterprise Linux. o Intel 11.1 on Red Hat Enterprise Linux. o GCC 3.4.3, GCC 4.0.1, GCC 4.2.4, GCC 4.3.3 and GCC 4.4.0 on Red Hat Enterprise Linux. o GCC 4.3.3 and GCC 4.4.0 with C++0x extensions on Red Hat Enterprise Linux. o GCC 4.1.1, 4.2.1 on 64-bit Red Hat Enterprise Linux. o GCC 4.1.2 on Suse Linux on 64 bit Itanium. o GCC 4.1.2 on 64-bit Redhat Server 5.1. o GCC Open64 4.2.2 on Red Hat Enterprise Linux. o GCC 4.3.4 on Debian unstable. o QLogic PathScale(TM) Compiler Suite: Version 3.1 on Red Hat Enterprise Linux. o GCC version 4.2.0 (PathScale 3.2 driver) on 64-bit Red Hat Enterprise Linux. o Sun 5.9 on Red Hat Enterprise Linux. o Windows: o Visual C++ 7.1 on XP. o Visual C++ 9.0 on XP. o Visual C++ 9.0 on Vista. o Visual C++ 9.0 on Vista 64-bit. o Visual C++ 9.0, using STLport 5.2, on XP and Windows Mobile 5.0. o Visual C++ 10.0 beta. o Borland 5.9.3, 6.1.0, 6.1.3. o Borland C++ Builder 2007 and 2009. o Intel C++ 11.1, with a Visual C++ 9.0 backend, on Vista 32-bit. o Intel C++ 11.1, with a Visual C++ 9.0 backend, on Vista 64-bit. o GCC 4.3.3 and 4.4.0, on Mingw with C++0x features. o Solaris: o Sun C++ 5.7, 5.8, 5.9 on Solaris 5.10. o GCC 3.4.6 on Solaris 5.10. ---------------------------------------------------------------------------------------------------- My concern is: We are supposed to use Boost both for Linux and Solaris, but the compilers being used are slightly old. On Debian Linux we have GCC 3.3 and on Solaris we have Sun WorkShop 6 update 2 C++ 5.3. Is Boost compatible with these versions? Does this list mean that Boost is not compatible with older versions of Compiler? -- View this message in context: http://www.nabble.com/Boost-compatibility-with-Old-Compilers-tp25885255p258… Sent from the Boost - Users mailing list archive at Nabble.com.

2 1

TCP boost server, but non-boost client
by G, Benjamin (NSN - IN/Bangalore) 14 Oct '09

14 Oct '09

Hi, I have a server which provides configuration to different clients. This server uses boost's serialization to serialize the C++ configuration objects. Some of my clients don't want to use boost for the (de-)serialization work. Will this cause any issues? Can any client deserialize the data serialized by a boost-using entity? What are the things that I need to take care of? I did read somewhere that boost serialization should be portable (I wonder if it means the same). All help appreciated. Regards, - Ben

2 1

[asio] How to know if the io_service is runing
by Daniele Barzotti 14 Oct '09

14 Oct '09

In my library I have this code (yet used in another project) where pwork is assigned to a shared_ptr: -------------------------------------------------------------- typedef boost::shared_ptr<boost::asio::io_service::work> io_work_ptr; // Main io_service static boost::asio::io_service io_service; static boost::scoped_ptr<boost::thread> io_service_thread; static boost::asio::io_service::work* p_work; static bool thread_started; bool CreateMainThread() { if (!thread_started) { try { // create the work object on the heap p_work = new boost::asio::io_service::work(io_service); // run the IO service as a separate thread: io_service_thread.reset ( new boost::thread ( boost::bind ( &boost::asio::io_service::run, &io_service ) ) ); thread_started = !thread_started; return true; } catch (boost::thread_resource_error e) { // Failed to create the new thread return false; } } } -------------------------------------------------------------- WaveStream::WaveStream() : rx_timer_(io_service) { }; WaveStream::Test() { //Start a new timer or renew it rx_timer_.expires_from_now( boost::posix_time::milliseconds(500) ); // We managed to cancel the timer. Start new asynchronous wait. rx_timer_.async_wait( boost::bind(&WaveStream::handle_timeout, this, _1) ); } void WaveStream::handle_timeout(const boost::system::error_code& error) { .... } -------------------------------------------------------------- The timer callback is never called. Is there a way to know if the io_service is runing? Daniele.

2 2

[statechart] forward_event()
by Commander Pirx 14 Oct '09

14 Oct '09

Hi, I have a state [A] with three orthogonal substates [X,Y,Z]. The states [A] and [X] provide custom reactions for an event [E]. If processing event [E] the state machine calls first the custom reaction in [A]. This method returns with forward_event() ; After this the custom reaction in [A] is called again. And then custom reaction in [X] is called. This is a little cumbersome. Why is the custom reaction in [A] called twice? And why is the outermost state called first? It would be something easier to work vice versa. First the more "specialized" inner states could do there work and then the "generalized" outer states can do the everything else. Pirx!

3 14

problem with enable_if_c and gcc4
by Lewis Hyatt 14 Oct '09

14 Oct '09

Hello Everyone- The following code compiles fine with g++ 4.1.2 but does not compile with 4.4.1, I get an error message saying the prototype does not match any declared in the class. This happens both with Boost 1.40.0 and 1.33.1, on Debian (x86 64-bit). Does anyone know what's up with this, is there some workaround? Thanks very much. #include <boost/utility/enable_if.hpp> #include <boost/type_traits/is_base_of.hpp> #include <boost/type_traits/is_same.hpp> template<typename T> struct A { struct B {}; template<typename U> typename boost::enable_if_c< boost::is_base_of<B, U>::value && !boost::is_same<B,U>::value> ::type f(U&); }; template<typename T> template<typename U> typename boost::enable_if_c< boost::is_base_of<typename A<T>::B, U>::value && !boost::is_same<typename A<T>::B, U>::value> ::type A<T>::f(U&) {} int main() { } t.cpp:20: error: prototype for 'typename boost::enable_if_c<(boost::is_base_of::value && (! boost::is_same::value)), void>::type A<T>::f(U&)' does not match any in class 'A<T>' t.cpp:12: error: candidate is: template<class T> template<class U> typename boost::enable_if_c<(boost::is_base_of::value && (! boost::is_same::value)), void>::type A::f(U&) -Lewis

1 1

[boost][filesystem] re-assign a recursive_directory_iterator
by Sascha Ochsenknecht 14 Oct '09

14 Oct '09

Hi, should it be possible to re-assign a filesystem::recursive_directory_iterator? I tried something like the following (could strip out my code and build a small test case if needed but this would take some time): fs::path aPath = ...; // assign to some valid path const fs::recursive_directory_iterator initial(path); const fs::recursive_directory_iterator end(); fs::recursive_directory_iterator current(path); // now increment current and do something with it (traverse directories etc). // ... // Later I want to go back to initial position with: current = initial; // but this causes a crash later when traversing again, seems that the (current == end) check gives wrong results and I dereference invalid current // I tried also the following: current = fs::recursive_directory_iterator(path); // this doesn't crash but (current == end) seems to give also wrong results I don't want to create 'current' as a temporary, so constructing a new 'current' when needed. In my case it is a class member. Anyhow, I can implement a workaround for me. Is it a bug or is it not wanted to re-assign the iterator? If it is not wanted to re-assign, maybe the assignment operator should be disabled? Thanks in advance, Sascha

2 1

[log] Status & getting started?
by Allen Cronce 13 Oct '09

13 Oct '09

Hi all, Currently we're using the John Torjo boost log implementation. Nothing seems to be happening with that library, so I'm exploring moving to a different logging solution. The Andrey Semashev log candidate seems to have some steam behind it. It got reviewed in February, an "RC3" version was deposited in the vault in August, and it looks like there's recent activity in its svn repository. But it's hard to get a clear picture as to its status. Can someone (Andrey?) give me some idea as to where this library is in the acceptance process? Assuming that this library is moving forward, what's the best way to get started with it? One advantage of the Torjo library was that it was header only. Looks like the Semashev version requires building. I have some reluctant experience with building boost as a whole (because I seem to end up doing it once a year or so). But I have no experience with building individual pieces like this. We're still using boost 1.36. I can see a post from Boris Schaeling indicating that the log library from February seems to work with this version. That's good news. But we're moving to 1.40 shortly. Does it make sense to take the 1.40 sources, graft in the log candidate sources in the expected places, then turn a build of the whole shebang? If not, can someone provide pointers regarding building individual libraries? Thanks in advance for information regarding the logging library's status, and any suggestions getting started with it. Best, -- Allen Cronce

1 0

New library for RPC-calls using only boost and C++0x
by Siegfried Kettlitz 13 Oct '09

13 Oct '09

Hello boost users and developers, i'm currently writing a library for making remote procedure calls to C++ objects. The library is based on the variadic templates extension coming with C++0x (already present in GCC 4.3) in conjunction with the boost library (threads, serialization, asio). The aim is to have a library, that is easy to integrate and doesn't require additional tools. Writing it, so far, has thaught me a lot about C++, templates, variadic templates, threading, networking etc.. I post this message, because i think that this library is quite handy in situations, where a developer already uses boost libraries and wants to add some networking functionality to his/her project and the usage of variadic templates instead of additional tools is quite new. One example would be remote monitoring an application by adding a "std::string get_status()"-function to some object. Implemented as example is a chat-system with a server and clients interacting by remote calls. This example is tested with a x64-linux and win32 using the portable boost-text-serialization. Multiple clients can connect to the server via the TCP/IP-acceptor (and disconnect without crashing the system). Implemented is a console-client for text input and display and a morse client, that interacts via a digital output device (relais card), which is connected via a serial port. The serial port and the digital output driver are both also connected via the rpc-system and can be physically located on separate computers. At this point, i have the basic bits together and a working rpc-library, that needs to be cleaned, refined and extended to match the practical needs. This means, that some programs and libraries have to be written, which use the rpc-library and show its bugs and limitations. The downside of it is, that the interface of the rpc-library may change one more time until it can be considered stable. Therefore it would not be good to test it widely in libraries posing similar requirements, but to test it with a variety of possible applications and provide the implementations as examples and (regression-/unit-)tests. - One library for Measurement/Automation is already in development. - The chat-server/client exists. - A transaction based game server is under consideration. - Some distributed parallel computing application with dynamic joining/leaving of clients. Can you give me some more ideas of different applications, which are easy to implement but yet useful? What would you like to have implemented and accessible by an rpc-library? One speed measurement using local (shared memory) channels resulted in 20000 calls/s and 60000 messages/s on a 1.8GHz Opteron. Do you know how this compares to other implementations on similar hardware? The TCP/IP-channel implementation using ASIO resulted in only about 100 synchronous calls/second regardless of the payload and almost no CPU load, which suggests that ASIO somewhere "sleeps" on linux systems for a few milliseconds. This would be fixed in some future implementation. Using asynchronous calls (1000 open calls) again leads to about 10000 calls/s. I'll have a closer look on what can be merged with "channels" from the boost vault[1], but from the first look it seems that "channels" targets a much more general approach. After the first tries, i realized, that my implemention has to be very specific (e.g. templates only where necessary). So, the design approaches might be too different. After the refinement, cleanup, some testing and documenting, i'd like to publish the rpc-library (and libraries based on it) under GPL- or boost-license. Then you'll hear from me again. In future i'll probably work 2-8 hours per week on the library and use it in some projects, but there's plenty of work left to do in various regions. This will leave space for other developers who might want to use and modify it to fit their needs. Some of the harder parts that would need implementation are redundant parallel connections and advanced routing, authentication of nodes, signing of messages, encryption of payload, access restriction to objects. Please let me know if * you are interested in this library in some way * have comments * know very similar libraries or * are interested in serialization using variadic function templates. Regards, Siegfried Kettlitz --- A short introduction on the library: * "nodes" are the central parts. They contain a factory for objects, store objects and forward "messages" to objects or other nodes in the network and contain a "scheduler" for executing function calls. The nodes are connected by "channels", which forward messages between the two connected nodes. Currently there is a TCP/IPv4 implementation using ASIO and direct forwarding. * Nodes connected by channels form a network, create and monitor virtual connections between each other. The network map is used to route messages between connected nodes. * An object has to supply a template for a class name (only if it can be constructed by the factory) and a template for registering the callable functions (also for abstract base classes which can't be constructed). * A function call is executed asynchronously (wrapper for synchronous call exists). The call to node.call(...) returns a "link_caller"-object. The function to be executed itself can optionally have "link_callee" as the first argument (which is invisible for the caller). The "link_caller"/"link_callee" is used for synchronization (wait for call to begin/finish execution), message forwarding (push/pop strings at caller/callee) and transfer/extraction of the return value. * Objects of class hierarchies with abstract base classes are possible and handled transparently. * Objects can be constructed by the factory of a node or an existing (externally constructed) object can be registered by the node. * Objects don't necessarily need to know (be modified, inherit something etc.) that they are part of the RPC-system and some of their functions are called remotely. * Objects can be aware that they are constructed by the factory and can, on construction be informed about their identity and get a reference to the node to make remote calls themselves. * An object is identified by a "reference" consisting of the "node_id" and the "object_id". * Parameters are passed (serialized, deserialized) by value. All function parameters must have the possibility to be serialized. Modifying referenced parameters won't affect the parameter at the caller. Pointers obviously won't work between nodes not sharing the same memory-address-space. * Called functions can use "link_callee" to make remote calls themselves, regardless of the object. * Priorities are assigned to calls and affect message forwarding and scheduling. * Calls to objects can be transformed into a boost::function using node.bind(...). * If a called function throws, the exception is forwarded as rpc::exception or a more specialized exception class. The exception is thrown at the caller, when waiting for interaction with the callee (e.g. wait for return value). * Template functions as targets are possible when specifying the template parameters of the function pointer when registering the function. * There is no need to use macros at all, although macros could be used to simplify some things. ---------- // Example for creating and calling an object. // Create a service. boost::asio::io_service node_io_service; // Create a node. rpc::node node_s(node_io_service, config); // Register the builder in order to build the server object later. node_s.register_builder<devicelib::chat_server_impl>(); // Register the interface of the client to be able to make calls to it. node_s.register_interface<devicelib::chat_client>(); // Create the server object. rpc::reference cs( node_s.get_id(), node_s.allocate_object_template<devicelib::chat_server_impl>() ); // Store the server in name server. // rpc::reference( node_s.get_id(), rpc::oid_simple_object_name_server ) -> A reference to the object_name_server. // rpc::sched_default() -> Default scheduling parameters. // &rpc::simple_object_name_server::store, cs -> function to call. // cs, "chat_server_object" -> function parameters. node_s.wrap_call( rpc::reference( node_s.get_id(), rpc::oid_simple_object_name_server ), rpc::sched_default(), &rpc::simple_object_name_server::store, cs, "chat_server_object" ); ---------- // Example for abstract base class. namespace devicelib { class chat_server : public chat_types { public: virtual void receive( std::string xml_message ) = 0; virtual user_id_t connect_user( rpc::reference a_client, std::string xml_user_info ) = 0; virtual std::string query_user( user_id_t id ) = 0; virtual void disconnect_user( user_id_t user ) = 0; virtual void join_room( user_id_t user, room_id_t room ) = 0; virtual std::list<user_id_t> list_users( ) = 0; virtual std::list<room_id_t> list_rooms( ) = 0; }; } --------------- // Example for registering the interface of the server object. // The template is declared in namespace rpc, therefore we need to define it there. namespace rpc { // For the base object. template<> void register_proxy_functions<devicelib::chat_server>( function_call_proxy<devicelib::chat_server>& ar_proxy ) { // Register the functions of the abstract base object using unique strings. ar_proxy.register_interface( &devicelib::chat_server::connect_user, "&chat_server::connect_user" ); ar_proxy.register_interface( &devicelib::chat_server::query_user, "&chat_server::query_user" ); ar_proxy.register_interface( &devicelib::chat_server::disconnect_user, "&chat_server::disconnect_user" ); ar_proxy.register_interface( &devicelib::chat_server::join_room, "&chat_server::join_room" ); ar_proxy.register_interface( &devicelib::chat_server::list_users, "&chat_server::list_users" ); ar_proxy.register_interface( &devicelib::chat_server::list_rooms, "&chat_server::list_rooms" ); } // For each implementation. template<> void register_proxy_functions<devicelib::chat_server_impl>( function_call_proxy<devicelib::chat_server_impl>& ar_proxy ) { // Register the functions of the base object. register_proxy_functions<devicelib::chat_server>( (function_call_proxy<devicelib::chat_server>&) ar_proxy ); // Make the functions of this object available. ar_proxy.register_interface( &devicelib::chat_server_impl::connect_user, &devicelib::chat_server::connect_user ); ar_proxy.register_interface( &devicelib::chat_server_impl::disconnect_user, &devicelib::chat_server::disconnect_user ); ar_proxy.register_interface( &devicelib::chat_server_impl::join_room, &devicelib::chat_server::join_room ); ar_proxy.register_interface( &devicelib::chat_server_impl::list_users, &devicelib::chat_server::list_users ); ar_proxy.register_interface( &devicelib::chat_server_impl::list_rooms, &devicelib::chat_server::list_rooms ); } // The implementation can be built, the base object not. template<> std::string class_builder_name<devicelib::chat_server_impl>() { // Give it an unique name here. return "chat_server_impl"; } } ----------------- 1: http://www.boostpro.com/vault/index.php?&direction=0&order=&directory=Distr…

8 12

[iterators] advance and distance with custom random-access iterators
by Sebastian Weber 13 Oct '09

13 Oct '09

Hi! Is there any progress in the bug #2539 ?? It seems that the patch published there which is against a rather antiquated boost version is the only solution to the problem of iterator_adaptor/facade constructs which are random_access-tagged in order to make use of true random access operations in distance/advance operations. This is critical in performance relevant code... So is there any progress in incorporating some fixes in the public trunk? I understand that the problem is a misconception of the stl there, but an official solution would be great here... Sebastian

1 0

[BGL] saving all-pair shortest paths
by Shyamendra Solanki 13 Oct '09

13 Oct '09

Hi, I understand that both the floyd-warshall and johnson algorithms populate the DistanceMatrix for shortest distances between any pair of vertices. Could you give some hint how to get the actual shortest paths? Thanks, Shyam

2 5

Multi Index with BOOST_AUTO question.
by Surya Kiran Gullapalli 13 Oct '09

13 Oct '09

Hello all,I'm a newbie to multi-index. I ran into a strange problem with VS2008. When using BOOST_AUTO with multi-index I'm getting compilation errors (C2248) . Please see the following code. *<snip>* * * #include <iostream> #include <string> using namespace std ; #include <boost/multi_index_container.hpp> #include <boost/multi_index/mem_fun.hpp> #include <boost/multi_index/ordered_index.hpp> #include <boost/multi_index/identity.hpp> #include <boost/multi_index/member.hpp> #include <boost/foreach.hpp> #include <boost/typeof/typeof.hpp> using namespace boost ; using namespace boost::multi_index ; class Test { friend bool operator< (const Test& p_lhs, const Test& p_rhs) { return p_lhs._i < p_rhs._i ; } friend ostream& operator << (ostream& os, const Test& p_test) { os << "Integer: " << p_test._i << "\tString: " << p_test._s ; return os ; } private: int _i ; string _s ; public: Test (int i, string s): _i(i), _s(s) { } string getString (void) const { return _s ; } } ; struct byNumber { } ; struct byString { } ; int main (void) { typedef multi_index_container <Test, indexed_by < ordered_unique<tag<byNumber>, identity <Test> > , ordered_non_unique < tag<byString>, BOOST_MULTI_INDEX_CONST_MEM_FUN(Test, string, getString) > > > Container ; typedef Container::index<byNumber>::type TestByNumber ; typedef Container::index<byString>::type TestByString ; Container c ; c.insert (Test(5, "Surya")) ; c.insert (Test(15, "Kiran")) ; //TestByNumber& tests = c.get<byNumber>() ; BOOST_AUTO(tests, c.get<byNumber>()) ; BOOST_FOREACH (Test t, tests) cout << t << endl ; return 0 ; } *</snip>* * * Compilation for this case fails with C2248. If I replace the BOOST_AUTO with the previous line's statement, compilation goes well. This happens with VS2008 only. with VS2005, there's no issue with BOOST_AUTO. Am I missing something here? Thanks, Surya

2 6

C# LINQ is cool. What about boost.qlang?
by Germán Diago 13 Oct '09

13 Oct '09

Hello. I've been reading a bit about C#'s linq and it's very cool in my opinion. I would like to see something similar in c++, and boost is the ideal place I think. As I would like to see something similar, I started a prototype adapted to c++. For now I've been playing with the basics: -from, where and select operators. -it just can be used (for now) with from, where and select, all in a row, but this will change. -it should work with any range: containers, boost.range, etc. -it's just a c++0x version. It works great with lambdas (gcc svn has lambdas). -Can use any boost.fusion sequence as the object type to query. (Although the plan is to be able to give data providers to use to query any type of source. The prototype works in the following way. For example (C++0x) : vector<int> numbers = {1, 2, 3, 4, 8, 5, 9 , 24, 19, 15, 12 } auto query = from(numbers).where([](int i) { return i < 15 && i > 10}). select(fields::full_object); This will construct an object with the query (without evaluating it). The original Range CANNOT be modified. If you want to modify , copy the result to a new container (in the future call query.to_vector(), query.to_list() to return a list or vector of objects. This will evaluate everything: for (auto it = query.begin(); it != query.end(); ++it) cout << *it << endl; You can traverse the range as many times as you like, and If you have, for example: struct Doctor { string name_; string surname_; string specializedin_; }; BOOST_FUSION_ADAPT_STRUCT(Doctor...) you can do, for example: vector<Doctor> doctors = {....}; auto query = from(doctors).where([](const Doctor & d) { return d.surname_ == "Smith"; }). select(fields::_1); The plan to improve readability is to use fusion associative sequence to do: auto query = from(doctors).where([](const Doctor & d) { return d.surname_ == "Smith"; }). select(Doctorfields::name, Doctorfields::specializedin); The query will return a tuple<string, string> in this case (the one before this would return tuple<string>. Queries with multiple fields don't work for now. A first implementation with the ability to select the full object or a field number (not several, but planned) is attached. Test it if you want and tell me what you think. The plan would be the same as the .NET platform: to be able to query XML, text files, objects, databases... giving extension points to add data providers. Another thing that doesn't work right now is to nest a query inside another query, but I'll work on it. Feedback is welcome.

2 2

Re: [Boost-users] 1.40 and pch - gcc problems?
by Vial, Florent 13 Oct '09

13 Oct '09

Hi everyone, I downloaded and built bjam-3.17 under linux x86_64 (RHEL 4.6). I then built boost 1.40.0 using following command line: ../boost-jam-3.1.17/bin.linuxx86_64/bjam --build-dir=/tmp/build-boost-1_40_0 stage --build-type=complete --layout=tagged My user-config.jam has just been modified so that the toolset used is gcc4.2.3. I got following warnings in the math library: gcc.compile.c++ /tmp/build-boost-1_40_0/boost/bin.v2/libs/math/build/gcc-4.2.3/release/l ink-static/assoc_laguerrel.o libs/math/build/../src/tr1/assoc_laguerrel.cpp:6:21: warning: /tmp/build-boost-1_40_0/boost/bin.v2/libs/math/build/gcc-4.2.3/release/l ink-static/../src/tr1/pch.hpp.gch: created and used with different settings of -fpic gcc.compile.c++ /tmp/build-boost-1_40_0/boost/bin.v2/libs/math/build/gcc-4.2.3/release/l ink-static/assoc_legendrel.o libs/math/build/../src/tr1/assoc_legendrel.cpp:6:21: warning: /tmp/build-boost-1_40_0/boost/bin.v2/libs/math/build/gcc-4.2.3/release/l ink-static/../src/tr1/pch.hpp.gch: created and used with different settings of -fpic gcc.compile.c++ /tmp/build-boost-1_40_0/boost/bin.v2/libs/math/build/gcc-4.2.3/release/l ink-static/betal.o [snip lots of them] Then the serialization library reports following warnings: gcc.compile.c++ /tmp/build-boost-1_40_0/boost/bin.v2/libs/serialization/build/gcc-4.2.3/ release/link-static/xml_grammar.o In file included from ./boost/archive/impl/basic_xml_grammar.hpp:59, from libs/serialization/src/xml_grammar.cpp:16: ./boost/spirit/core/non_terminal/rule.hpp:18:4: warning: #warning "This header is deprecated. Please use: boost/spirit/include/classic_rule.hpp" In file included from libs/serialization/src/basic_xml_grammar.ipp:20, from libs/serialization/src/xml_grammar.cpp:63: ./boost/spirit/core/composite/operators.hpp:18:4: warning: #warning "This header is deprecated. Please use: boost/spirit/include/classic_operators.hpp" In file included from libs/serialization/src/basic_xml_grammar.ipp:21, from libs/serialization/src/xml_grammar.cpp:63: ./boost/spirit/core/composite/actions.hpp:18:4: warning: #warning "This header is deprecated. Please use: boost/spirit/include/classic_actions.hpp" In file included from libs/serialization/src/basic_xml_grammar.ipp:22, from libs/serialization/src/xml_grammar.cpp:63: ./boost/spirit/core/primitives/numerics.hpp:18:4: warning: #warning "This header is deprecated. Please use: boost/spirit/include/classic_numerics.hpp" gcc.compile.c++ /tmp/build-boost-1_40_0/boost/bin.v2/libs/serialization/build/gcc-4.2.3/ release/link-static/xml_iarchive.o In file included from ./boost/archive/impl/basic_xml_grammar.hpp:59, from ./boost/archive/impl/xml_iarchive_impl.ipp:41, from libs/serialization/src/xml_iarchive.cpp:32: ./boost/spirit/core/non_terminal/rule.hpp:18:4: warning: #warning "This header is deprecated. Please use: boost/spirit/include/classic_rule.hpp" However, the build seems to go through. Then i do following: ../boost-jam-3.1.17/bin.linuxx86_64/bjam --build-dir=/tmp/build-boost-1_40_0 --build-type=complete --layout=tagged stage I then get following messages for the math library: gcc.compile.c++ /tmp/build-boost-1_40_0/boost/bin.v2/libs/math/build/gcc-4.2.3/debug/lin k-static/runtime-link-static/assoc_laguerre.o libs/math/build/../src/tr1/assoc_laguerre.cpp:6:21: warning: /tmp/build-boost-1_40_0/boost/bin.v2/libs/math/build/gcc-4.2.3/debug/lin k-static/runtime-link-static/../src/tr1/pch.hpp.gch: not used because `_REENTRANT' not defined gcc.compile.c++ /tmp/build-boost-1_40_0/boost/bin.v2/libs/math/build/gcc-4.2.3/debug/lin k-static/runtime-link-static/assoc_legendre.o libs/math/build/../src/tr1/assoc_legendre.cpp:6:21: warning: /tmp/build-boost-1_40_0/boost/bin.v2/libs/math/build/gcc-4.2.3/debug/lin k-static/runtime-link-static/../src/tr1/pch.hpp.gch: not used because `_REENTRANT' not defined gcc.compile.c++ /tmp/build-boost-1_40_0/boost/bin.v2/libs/math/build/gcc-4.2.3/debug/lin k-static/runtime-link-static/beta.o libs/math/build/../src/tr1/beta.cpp:6:21: warning: /tmp/build-boost-1_40_0/boost/bin.v2/libs/math/build/gcc-4.2.3/debug/lin k-static/runtime-link-static/../src/tr1/pch.hpp.gch: not used because `_REENTRANT' not defined [snip lots of them] I then type as root: ../boost-jam-3.1.17/bin.linuxx86_64/bjam install --prefix=/usr/local/boost1400-gcc423 And get following installation: [snip] common.copy /usr/local/boost1400-gcc423/lib/libboost_iostreams.a common.copy /usr/local/boost1400-gcc423/lib/libboost_math_tr1.a common.copy /usr/local/boost1400-gcc423/lib/libboost_math_tr1f.a common.copy /usr/local/boost1400-gcc423/lib/libboost_math_tr1l.a ...on 7500th target... common.copy /usr/local/boost1400-gcc423/lib/libboost_math_c99.a common.copy /usr/local/boost1400-gcc423/lib/libboost_math_c99f.a common.copy /usr/local/boost1400-gcc423/lib/libboost_math_c99l.a common.copy /usr/local/boost1400-gcc423/lib/libboost_program_options.a common.copy /usr/local/boost1400-gcc423/lib/libboost_python.a common.copy /usr/local/boost1400-gcc423/lib/libboost_serialization.a common.copy /usr/local/boost1400-gcc423/lib/libboost_wserialization.a common.copy /usr/local/boost1400-gcc423/lib/libboost_signals.a common.copy /usr/local/boost1400-gcc423/lib/libboost_prg_exec_monitor.a common.copy /usr/local/boost1400-gcc423/lib/libboost_unit_test_framework.a common.copy /usr/local/boost1400-gcc423/lib/libboost_thread.a common.copy /usr/local/boost1400-gcc423/lib/libboost_wave.a ...updated 7511 targets... However, for each library, i only have 2 files installed + 1 symbolic link: -rw-r--r-- 1 root root 148826 Oct 13 17:22 libboost_date_time.a lrwxrwxrwx 1 root root 28 Oct 13 17:22 libboost_date_time.so -> libboost_date_time.so.1.40.0 -rwxr-xr-x 1 root root 74552 Oct 13 17:22 libboost_date_time.so.1.40.0 Any idea what i might have done wrong ? (compiling under Win32/MSVC90 was no problem at all surprisingly) Thanks in advance, Florent ---------- Forwarded message ---------- From: "John Maddock" <j...(a)johnmaddock.co.uk> Date: 4 sep, 10:25 Subject: 1.40 and pch - gcc problems? To: BOOST Archives >I just got1.40and tried building. I kept having some pieces fail > with a complete build, sometimes I had gcc that ran until resources > exausted, sometimes the compiler would segfault. Eventaully I tracked > it down to the math library. After noticing the many errors from the > gch (things like : error _REENTERANT not defined or used diffrent > -fpic). I disabledpchfor the math library (actually the boost > project and the libs/math/test) and I did not have any more issues. Can you be more specific about the error messages? A: bjam --with-math will build just the math lib, if you could post the complete output that would be most helpful. There's something strange going on here though as it builds just fine locally here (ubuntu Linux), and on our regression test machines (various Linux flavours). Thanks, John. JOIN US FOR THE JOURNEY. View Our New Digital Cameras: www.arridigital.com Arnold & Richter Cine Technik GmbH & Co. Betriebs KG Sitz: Munchen - Registergericht: Amtsgericht Munchen - Handelsregisternummer: HRA 57918 Personlich haftender Gesellschafter: Arnold & Richter Cine Technik GmbH Sitz: Munchen - Registergericht: Amtsgericht Munchen - Handelsregisternummer: HRB 54477 Geschaftsfuhrer: Franz Kraus; Dr. Martin Prillmann

1 0

[Property map] 1.40.0 breaks boost.graph python bindings
by John Reid 13 Oct '09

13 Oct '09

I'm using gcc 4.2.4 on Ubuntu. The boost.graph python bindings (http://www.osl.iu.edu/~dgregor/bgl-python/) work fine with boost 1.39. When I try with boost 1.40 I get many errors such as the following: ./boost/graph/python/resizable_property_map.hpp:101: error: ‘class boost::vector_property_map< boost::default_color_type, boost::graph::python::basic_index_map< boost::graph::python::basic_descriptor<boost::detail::edge_desc_impl<boost::bidirectional_tag, void *>, boost::bidirectionalS>, boost::adj_list_edge_property_map< boost::bidirectional_tag, long unsigned int, const long unsigned int &, void *, const boost::property<boost::edge_index_t, long unsigned int, boost::no_property>, boost ::edge_index_t>>>’ has no member named ‘get_store’ Does anyone know what changed between 1.39 and 1.40 to break this? The Readme.txt that comes with boost 1.40 suggests that nothing was updated in the property map library. BTW it doesn't compile with gcc 4.3 either but then neither does it compile with boost 1.39 (albeit for different reasons). I can provide all of the error messages if needed. John.

2 11

Re: [Boost-users] [Boost.MPL]is it possible to generate types
by Tan, Tom (Shanghai) 13 Oct '09

13 Oct '09

Thanks, it works like a charm. Tan, Tom (Shanghai) wrote: > I have a templated class > > template<class U, class V, class W> > class S > { > //... implementations > }; > > and some stock type implementations for type U, V and W: > > typedef boost::mpl::vector<U0, U1> u_types; > typedef boost::mpl::vector<V0, V1, V2, V3, V4> u_types; > typedef boost::mpl::vector<W0, W1, W2, W3, W4> w_types; > > I want to test class S with all possible combinations of the template > arguments, > > typedef boost::mpl::vector< > S<U0,V0,W0>, > S<U0,V0,W1>, > // ... > S<U1,V4,W4>, > > s_types; > > like this: > boost::mpl::for_each<s_types>(test_func). > > The only problem is there are 2*5*5 = 50 combinations that I do not > wish to type in one by one. > > Is there a way to generate all the combinations with boost::mpl or > Boost.Preprocessor? >> Look at BOOST_PP_SEQ_FOR_EACH_PRODUCT In Christ, Steven Watanabe ***************************************

1 0

Re: [Boost-users] Boost-users Digest, Vol 2134, Issue 4
by Tan, Tom (Shanghai) 13 Oct '09

13 Oct '09

Thanks, it works like a charm. Tan, Tom (Shanghai) wrote: > I have a templated class > > template<class U, class V, class W> > class S > { > //... implementations > }; > > and some stock type implementations for type U, V and W: > > typedef boost::mpl::vector<U0, U1> u_types; > typedef boost::mpl::vector<V0, V1, V2, V3, V4> u_types; > typedef boost::mpl::vector<W0, W1, W2, W3, W4> w_types; > > I want to test class S with all possible combinations of the template > arguments, > > typedef boost::mpl::vector< > S<U0,V0,W0>, > S<U0,V0,W1>, > // ... > S<U1,V4,W4>, > > s_types; > > like this: > boost::mpl::for_each<s_types>(test_func). > > The only problem is there are 2*5*5 = 50 combinations that I do not > wish to type in one by one. > > Is there a way to generate all the combinations with boost::mpl or > Boost.Preprocessor? >> Look at BOOST_PP_SEQ_FOR_EACH_PRODUCT In Christ, Steven Watanabe ***************************************

1 0