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

Re: [Boost-users] How to cast a boost::shared_ptr to void*
by Mupparthy Ravindranath 10 May '11

10 May '11

Sorry for the spam. I got the answer, how to do it. Ignore my previous mail. [quote] #include <boost/shared_ptr.hpp> #include <iostream> using namespace std; class A { public: void hello () { cout << "Hello, friends...!" << endl; } }; typedef boost::shared_ptr<A> APtr; void* callbk (void* o) { APtr a = *static_cast<APtr*>(o); a->hello(); } int main() { A* rawPtr = new A; boost::shared_ptr<A>sPtr(rawPtr); callbk((void*)&sPtr); return 0; } [/quote] warm regars, RV ________________________________ From: Mupparthy Ravindranath <rv_nath(a)yahoo.com> To: "boost-users(a)lists.boost.org" <boost-users(a)lists.boost.org> Sent: Tuesday, 10 May 2011 1:26 PM Subject: How to cast a boost::shared_ptr to void* Hi all, My requirement is to cast an object of type shared_ptr as void*, so that I can send it to a callback function that accepts a void*. I cannot change callback the function signature because it is part of an existing legacy code. In the existing code, I am sending a raw object pointer (instead of shared_ptr). Now, if I want to pass a shared_ptr, it won't compile because I cannot cast the shared ptr to void*. Can anyone suggest some workaround for this. I tried searching the mailing lists and internet, but not much luck. regards, RV

2 1

How to cast a boost::shared_ptr to void*
by Mupparthy Ravindranath 10 May '11

10 May '11

Hi all, My requirement is to cast an object of type shared_ptr as void*, so that I can send it to a callback function that accepts a void*. I cannot change callback the function signature because it is part of an existing legacy code. In the existing code, I am sending a raw object pointer (instead of shared_ptr). Now, if I want to pass a shared_ptr, it won't compile because I cannot cast the shared ptr to void*. Can anyone suggest some workaround for this. I tried searching the mailing lists and internet, but not much luck. regards, RV

4 3

boost graph: Error with compressed_sparse_row graph and strong_components
by aitor 10 May '11

10 May '11

Hi, the following example (taken from csr example in boost docs) fails when calling strong_components algorithm. The error message is: /usr/include/boost-1.46.1/boost/graph/strong_components.hpp:59: error: no matching function for call to 'get(int*&, unsigned int&)' the program works with boost 1.41.0 any help would be much appreciated :-) #include <boost/graph/compressed_sparse_row_graph.hpp> #include <string> #include <fstream> #include <boost/graph/strong_components.hpp> using namespace boost; int main() { typedef std::pair<int, int> E; E the_edges[] = { E(0, 1), E(0, 2), E(0, 3), E(1, 0), E(1, 3), E(1, 5), E(2, 0), E(2, 5), E(3, 1), E(3, 4), E(4, 1), E(5, 0), E(5, 2) }; typedef compressed_sparse_row_graph<directedS> WebGraph; WebGraph g(boost::edges_are_sorted, &the_edges[0], &the_edges[0] + sizeof(the_edges)/sizeof(E), 6); // strong components std::vector<int> v(6); int i = boost::strong_components(g, &v[0]); std::cout << i << "\n"; return 0; }

2 2

[Spirit] Alternatives and single member struct doesn't compile
by Vitaly Budovski 10 May '11

10 May '11

Hi, Please see the attached file. What am I doing wrong? qi/detail/assign_to.hpp:121: error: invalid static_cast from type ‘const test_b’ to type ‘unsigned char’ Thanks.

1 0

Re: [Boost-users] [msm] EUML and State construction
by Christophe Henry 09 May '11

09 May '11

> Thanks Christophe. One other thing I noticed. In > msm/front/euml/common.hpp, in > MSM_EUML_EVENT_INSTANCE_HELPER_ATTRIBUTE_MAP > there is an unused variable "dummyval" that causes warnings in gcc > compilers. Is naming this variable necessary to support other > compilers, or can it be unnamed? No, it's not necessary at all, it just was forgotten. Every new gcc version brings a few instances of it :( I just fixed it in the trunk (for 1.47). Thanks, Christophe

1 0

[boost::thread] using futures
by Kazz 09 May '11

09 May '11

Hi All. I am trying to use Futures, but not getting the expected result. The code below works some times, and other times it throws exception "terminate called after throwing an instance of 'boost::exception_detail::clone_impl<boost::exception_detail::error_info_injector<boost::promise_already_satisfied> >' what(): Promise already satisfied" Any help will be appreciated. Thanks Kazz. //-------------------------------------------// boost::promise<int> pt; // holds final calculated result void worker1() { // do any processing using algorithm A srand ( time ( NULL) ); int delay = rand() % 5 + 1; boost::this_thread::sleep( boost::posix_time::seconds(delay)); // result calculate pt.set_value(1); } void worker2() { // do any processing using algorithm B srand ( time ( NULL) ); int delay = rand() % 5 + 1; boost::this_thread::sleep( boost::posix_time::seconds(delay)); // result calculated pt.set_value(2); } int main() { boost::shared_future<int> ans(pt.get_future()); boost::thread th1(worker1); boost::thread th2(worker2); ans.wait(); std::cout<<ans.get()<<std::endl; std::cout<<"exiting"<<std::endl; return 0; }

2 3

String serialization
by Robert Bielik 09 May '11

09 May '11

I'd like to (XML) serialize a non-STL string, and I'd like it to serialize in the same manner as STL string (i.e. just the plain string as CDATA). I can do it like: <myString> <s>The string...</s> </myString> but I'd like: <myString>The string...</myString> the serialization implementation of std::string is a menace to decipher... TIA /Rob

1 1

[asio] IPv6 interface index
by Dylan Klomparens 09 May '11

09 May '11

Hello, I am trying to use Boost for some IPv6 and multicast network communication. I need to construct an IPv6 multicast socket that uses a specific network interface index. I was able to find the correct multicast option to set the network interface index in boost/asio/ip/detail/socket_option.hpp: explicit multicast_request(const boost::asio::ip::address_v6& multicast_address, unsigned long network_interface = 0) The problem is, I don't know how to find the correct value for the "network_interface" parameter. Is there a way to get the network_interface value using a local IPv6 address that I can provide? I looked in the documentation and examples, but couldn't find anything. -- Dylan

1 0

[msm] EUML in a header file
by Nate Knight 09 May '11

09 May '11

Christophe, One problem I've run into with the EUML frontend is the the _EVENT, _ACTION and _STATE macros define non-static instance variables. So, if I define a state machine in a header file which then gets included in multiple source files (e.g. I have two classes which use the same state machine) I get linker errors about duplicate instances. For states, I can work around this issue by doing, for example, BOOST_MSM_EUML_DECLARE_STATE((),MyStateType) static MyStateType MyState; and then using MyState in the transition table. Events and Actions don't seem to have the corresponding _DECLARE_ macros, so this workaround isn't really feasible at this point in time. Any thoughts? Thanks Nate

1 0

Usage of boost serialization binary_object
by Robert Bielik 09 May '11

09 May '11

Hi all, I'm wondering how to load a binary object from an archive ? I've managed to use binary_object to save it nicely, but I can't find any references on how to load it again :( TIA /Rob

1 1

Re: [Boost-users] [msm] EUML and State construction
by Christophe Henry 09 May '11

09 May '11

>> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ >> With gcc 4.5 and Boost 1.46.1, I get: >> >> simpletest.cpp: In function 'int main()': >> simpletest.cpp:30:75: error: no match for 'operator<<' in >> 'boost::msm::back::states_ << test_state_with_constructor(1)' > > Ok, I see what it is, I did not solve correctly my conflict between > proto and serialization. > I just committed a fix into the trunk (revision 71719), but it's not > working, I simply run into my next proto problem :( > > I'll continue working on this tomorrow. I fixed the second problem too (revision 71746) in the trunk. It should now work correctly. Thanks, Christophe

2 1

Re: [Boost-users] ASIO sockets - 2nd write fails
by bill comment 09 May '11

09 May '11

> > 'new_connection_' always points to an invalid socket because after a > connection is established you always replace it with a new one. > yes. i was suspecting this but didn't fully understand. thanks for confirming. my solution was to change my architecture to that of the ChatServer sample where a reference to my Connection object (parallel to their participant/session) is handed to a new Client object (parallel to their room) so that it will live for the duration of the client app's connection and my server can write several responses. thanks for the responses.

1 0

[forEach] Working with custom operator []
by Avishay Ben Shabtai 09 May '11

09 May '11

Hi i have a container class that defines 2 overloaded [] operators: T& operator[](const chat *); T& operator[](unsigned int); when i am using BOOST_FOREACH(T val,container) i get compilation errors but working directly with iterators is working ok, any idea?

4 3

ASIO sockets - 2nd write fails
by bill comment 09 May '11

09 May '11

I am having a problem writing a TCP socket message using boost:asio async_* I am new to using boost asio sockets. reading what I could find (this list, samples, other forums) I have added boost::asio socket use to my application. What works is: accept, receive, parse and reply to a message. After my client sends a message and receives a response it remains connected and starts an async receive. Several seconds later my server tries to send another message to the client (by calling server::write()). the code runs ok up to connection::handle_writeMessage() where error is: 2719 “The file handle supplied is not valid”. I thought I read (links below) that using strand::wrap() to queue the async_* calls and io_service::post() to post the initial function call would allow a thread that didn’t call io_service::run() to initiate the sending or a message. http://stackoverflow.com/questions/4078484/using-boost-sockets-do-i-need-on… http://stackoverflow.com/questions/4090567/better-understanding-boosts-chat… http://groups.google.com/group/boost-list/browse_thread/thread/124a643a06d6… did I interpret this incorrectly? Did I code it wrong? Thank you for your time and advice. to develop my code I started with the http3 and chat server samples. Changes include: - server::threads is a class member - server::run() returns without waiting for threads to stop so I start other threads in my program - added server::stop() to stop threads. - server::write() is called from another thread. - connection::do_writeMessage() and connection::handle_writeMessage() are from the chat server sample. A summary of my server code is below. its long, but should be familiar from the samples. Server.hpp std::vector<boost::shared_ptr<boost::thread> > threads; server::server : thread_pool_size_(2), // 1 thread for wait, read, reply. 1 for unsolicited writes. acceptor_(io_service_), new_connection_(new connection(io_service_, request_handler_)), request_handler_() { async_accept(new_connection_->socket(), boost::bind(&server::handle_accept, this, boost::asio::placeholders::error)); } void server::run() { // 1 thread for wait, read, reply. 1 for unsolicited writes. // Create a pool of threads to run all of the io_services. //std::vector<boost::shared_ptr<boost::thread> > threads; for (std::size_t i = 0; i < thread_pool_size_; ++i) { boost::shared_ptr<boost::thread> thread(new boost::thread( boost::bind(&boost::asio::io_service::run, &io_service_))); threads.push_back(thread); } } void server::stop() { io_service_.stop(); // Wait for all threads in the pool to exit. for (std::size_t i = 0; i < threads.size(); ++i) { threads[i]->join(); } } server::handle_accept (const boost::system::error_code& e) { if (!e) { // read, parse and reply new_connection_->start(); new_connection_.reset(new connection(io_service_, request_handler_)); // ready to accept another connection cout << "waiting for new accept ..." << endl; acceptor_.async_accept(new_connection_->socket(), boost::bind(&server::handle_accept, this, boost::asio::placeholders::error)); } } // public // called from another thread in my program to send an unsolicited message to my connected client void server::write(const string& msg) { io_service_.post(boost::bind(&connection::do_writeMessage, new_connection_, msg)); } - - - - - - - - - - connection::connection(boost::asio::io_service& io_service, request_handler& handler) : strand_(io_service), socket_(io_service), request_handler_(handler) { _running = true; } void connection::start() { buffer_.fill (0); socket_.async_read_some(boost::asio::buffer(buffer_), strand_.wrap( boost::bind(&connection::handle_read, shared_from_this(), boost::asio::placeholders::error, boost::asio::placeholders::bytes_transferred) ) ); } void connection::handle_read(const boost::system::error_code& e, std::size_t bytes_transferred) { if (!e) { // parse what we have so far. request_parser_.parse() is my code and it works correctly. // when successful result==true, request boost::tribool result; request_.reset(); boost::tie(result, boost::tuples::ignore) = request_parser_.parse( request_, buffer_.data(), buffer_.data() + bytes_transferred); if (result) { // we read an entire message // this is my code to process the message and put fill in the reply_ buffer. request_handler_.handle_request(request_, reply_); // send the reply boost::asio::async_write(socket_, reply_.to_buffers(), strand_.wrap( boost::bind(&connection::handle_writeResponse, shared_from_this(), boost::asio::placeholders::error) ) ); } else if (!result) { // there was an error reading the message boost::asio::async_write(socket_, reply_.to_buffers(), strand_.wrap( boost::bind(&connection::handle_writeResponse, shared_from_this(), boost::asio::placeholders::error) ) ); } else { // read some ok. need to read some more. socket_.async_read_some(boost::asio::buffer(buffer_), strand_.wrap( boost::bind(&connection::handle_read, shared_from_this(), boost::asio::placeholders::error, boost::asio::placeholders::bytes_transferred) ) ); } } else { cout << "connection::handle_read: error: " << e.value() << " " << e.message() << endl; } // If an error occurs then no new asynchronous operations are started. This // means that all shared_ptr references to the connection object will // disappear and the object will be destroyed automatically after this // handler returns. The connection class's destructor closes the socket. } void connection::handle_writeResponse(const boost::system::error_code& e) { if (!e) { // test. This message is received by my client. do_writeMessage ("hello world"); // ready to read more start(); } else { cout << "connection::handle_writeMessage: error sending to client: " << e.value() << " " << e.message() << endl; } // No new asynchronous operations are started. This means that all shared_ptr // references to the connection object will disappear and the object will be // destroyed automatically after this handler returns. The connection class's // destructor closes the socket. } - - - - - - - - - - // call to this function is posed to io_service from server::write // public void connection::do_writeMessage(string msg) { bool write_in_progress = !_messagesToWrite.empty(); _messagesToWrite.push_back(msg); // if there are any in the queue, handle_writeMessage is already sending them. if (write_in_progress) { cout << "connection::do_writeMessage: write in progress." << endl; } else { // start the async write. boost::asio::async_write(socket_, boost::asio::buffer(_messagesToWrite.front().data(), _messagesToWrite.front().length()), strand_.wrap( boost::bind(&connection::handle_writeMessage, shared_from_this()/*this*/, boost::asio::placeholders::error) ) ); } } // async write is done. if no error and more messages to write, write them, too. void connection::handle_writeMessage(const boost::system::error_code& error) { if (!error) { cout << "connection::handle_writeMessage: message written ok." << endl; // remove the message we just wrote _messagesToWrite.pop_front(); // any more? if (!_messagesToWrite.empty()) { // write the next one and call this function when that'd done. boost::asio::async_write(socket_, boost::asio::buffer(_messagesToWrite.front().data(), _messagesToWrite.front().length()), strand_.wrap( boost::bind(&connection::handle_writeMessage, shared_from_this()/*this*/, boost::asio::placeholders::error) ) ); } else { cout << "connection::handle_writeMessage: no more messages" << endl; } } else { cout << "connection::handle_writeMessage: error: " << error.value() << " " << error.message() << endl; } // let this async chain end. } - - - - - - - - - - main() { pTCPServer = new http::server3::server (argv[TCPPortNumber]); if (pTCPServer) { pTCPServer->run(); } // start thread that will initiate 2nd message … … … }

4 3

boost::function adding a lot of binary overhead
by Littlefield, Tyler 09 May '11

09 May '11

Hello all: I'm using boost::function in a few places so that I can use boost::bind to pass member pointers on objects, static member functions or just normal functions, but I'm noticing that I get a ton of extra overhead when compiling after I use boost::function in a new place. Is there a fix for this? I really like boost and some of it's features, but it's making my binaries huge. -- Take care, Ty my website: http://tds-solutions.net my blog: http://tds-solutions.net/blog skype: st8amnd127 “Programmers are in a race with the Universe to create bigger and better idiot-proof programs, while the Universe is trying to create bigger and better idiots. So far the Universe is winning.” “If Java had true garbage collection, most programs would delete themselves upon execution.”

3 3

[fusion] result_of
by Robert Ramey 08 May '11

08 May '11

I'm very interested in fusion and have been experimenting with it. There is one thing that is currently confusing me. (OK more than one thing). The documentation a typical function at .../fusion/sequence/concepts/associative_sequence.html shows valid expression has_key<K>(s) returning type result_of::has_key<S, K>::type which is OK by me. But why doesn't fusion use: #include <boost/result_of.hpp> then use boost::result_of<has_key<K>(S)>::type which would seem to me to reduce the size of fusion (and the docs) by around 50%. What am I missing here? Robert Ramey

2 1

[asio] Pending handlers not invoked in boost.asio?
by Germán Diago 08 May '11

08 May '11

Hello. I'm writing a little game which keeps four players connected. At first, my protocol does something like the following: -Server waits for 3 incoming connections. io_service::run is running in another thread (and the code I'll show you server side is executed after that). -Clients connect and send its data (name and team). When everything's done, the server continues and shows the received data from the clients. I have a loop like this on the Server side (with the great coroutine implementation from the examples): for (; num_players_connected_ !=3; ++num_players_connected_) { //Create context connection_contexts_.at(num_players_connected_) = std::make_shared<ConnectionContext>(service_); yield acceptor_.async_accept(*connection_contexts_.at(num_players_connected_)->socket(), std::ref(*this)); std::cout << "Servidor: Incoming connection" << std::endl; connection_contexts_.at(num_players_connected_)->start(); //This calls ClientContext::operator() //Start connection processing std::cout << "Servidor: Players connected: " << num_players_connected_ << std::endl; } //Here every connection is already done but must wait for data available in every context. std::cout << "Server: All " << num_players_connected_ << " players are connected" << std::endl; for (std::size_t i = 0; i < 3; i++) { //Wait for each player to have its data available server-side connection_contexts_.at(i)->playerDataReady().get(); std::cout << "Player info:\n"; std::cout << "\tPlayer name " << connection_contexts_.at(i)->playerName() << std::endl; std::cout << "\tPlayer team " << connection_contexts_.at(i)->playerTeam() << std::endl; } /////////////////////////////////// The Context for each client in the server side looks something like this: void ClientContext::operator()() { using namespace boost; const MsjDatosJugador * player_data = 0; MensajeRed msg; reenter (coro_opcall_state_) { yield asyncReadMessageFromSocketIntoQueue(std::ref(*this)); msg = getNextMessage(); std::cout << "Message has been read\n" << std::endl; if (msg.getTipo() == typeid(MsjDatosJugador)) { player_data = &(msg.getInfo<MsjDatosJugador>()); player_name_ = std::move(player_data->name_); player_team_ = player_data->team_; player_ready_ = true; } else { message_received_handler_(getNextMessage()); } } } void ClientContext::asyncReadMessageFromSocketIntoQueue( std::function<void ()> handler, const boost::system::error_code & ec, std::size_t bytes) { using namespace boost; if (!ec) { reenter (coro_read_state_) { message_size_ = std::make_shared<boost::int64_t>(); yield asio::async_read(*socket_, asio::buffer(message_size_.get(), sizeof(boost::int64_t)), std::bind(&GuinyoteClientContext::asyncReadMessageFromSocketIntoQueue, this, handler, std::placeholders::_1, std::placeholders::_2)); message_buffer_ = std::make_shared<std::vector<char> >(*message_size_); yield asio::async_read(*socket_, asio::buffer(message_buffer_->data(), *message_size_), std::bind(&GuinyoteClientContext::asyncReadMessageFromSocketIntoQueue, this, handler, std::placeholders::_1, std::placeholders::_2)); std::istringstream stream(std::string(message_buffer_->begin(), message_buffer_->end()), std::ios::binary); boost::archive::binary_iarchive archive(stream); MensajeRed msg; archive >> msg; std::cout << "GuinyoteClientContext: Pushing message into the message queue.\n"; messages_.push(msg); handler(); } } else { BOOST_THROW_EXCEPTION(std::runtime_error("GuinyoteContext: Error reading from client socket: " + ec.message())); } } My EXACT problem is that when the THIRD client connects, after invoking ClientContext::asyncReadMessageFromSocketIntoQueue once and reaching the first yield, it is never invoked again, even if I passed the handler function. Am I missing something? Must I flush the io_service handler execution in some way? Thanks in advance.

3 3

Query on proposed Boost.Move library
by Leo Carreon 07 May '11

07 May '11

Hi, I’m new to this mailing list and I haven’t figured out how to search the archives yet. At this stage, all I wanted to find out is whatever happened to the proposed Boost.Move library? Regards, Leo

2 2

Boost Install on Mac OS X using GCC
by Andrew 07 May '11

07 May '11

Hello, I'm trying to install boost on mac using the gcc compiler. I encounter these errors during installation for all libraries: "g++" -Wl,-R -Wl,"/System/Library/Frameworks/Python.framework/Versions/2.6/lib" -Wl,-R -Wl,"/System/Library/Frameworks/Python.framework/Versions/2.6/lib/python2.6/config" -o "stage/lib/libboost_prg_exec_monitor.dylib" -Wl,-h -Wl,libboost_prg_exec_monitor.dylib -shared -Wl,--start-group "/Users/andrew/boost/boost/bin.v2/libs/test/build/gcc-4.2.1/release/threading-multi/execution_monitor.o" "/Users/andrew/boost/boost/bin.v2/libs/test/build/gcc-4.2.1/release/threading-multi/debug.o" "/Users/andrew/boost/boost/bin.v2/libs/test/build/gcc-4.2.1/release/threading-multi/cpp_main.o" -Wl,-Bstatic -Wl,-Bdynamic -Wl,--end-group ...failed gcc.link.dll stage/lib/libboost_prg_exec_monitor.dylib... gcc.link.dll stage/lib/libboost_unit_test_framework.dylib ld: unknown option: -R collect2: ld returned 1 exit status I have no experience at all using boost in mac, and it is necessary that I use gcc. I am wondering if anyone could point me to a solution regarding this. Thanks, Andrew

6 7

[iterator] Problems with iterator_facade and const-qualification
by Evan Driscoll 07 May '11

07 May '11

I'm trying to use the iterator_facade and am running into trouble. It's not really caused by the iterator_facade, but I figured I'd ask anyway. This takes some explanation, so if anyone reads this, know that I *do* appreciate it. :-) TLDR version: I'm getting a compiler error about converting between pointers of iterator_facade<> instantiations with different const qualifications on the value_type. The code's attached, in addition to mostly appearing (with a few minor formatting changes) inline below. Please help me figure out what to do to fix it. What I'm doing is a bit roundabout, but here goes. One fault with C++ iterators (at least from some point of view) is that if you want to operate on different kinds of iterators you have to care about too much -- you need to know more than just the type that op* returns (which is all you care about most of the time, iterator_traits notwithstanding), but what the underlying iterator type is. This means that if you want to write code that operates on multiple kinds of iterators, it must be a template. Templates don't suit my needs at this time, so I'm trying to write a couple wrapper classes that will do dynamic dispatch to different iterator types. Think of this as writing a C++ version of Java's Iterator interface. Thus I define a base class for my iterators: template<typename ValueType, typename ReferenceType> class iterator_base : public iterator_facade<iterator_base<ValueType, ReferenceType>, ValueType, boost::forward_traversal_tag, ReferenceType> { friend class boost::iterator_core_access; protected: virtual void increment() = 0; virtual ReferenceType dereference() const = 0; virtual bool equal(iterator_base const & other) const = 0; public: virtual iterator_base* clone() const = 0; }; (I'm only worrying about forward iterators at this time. I'm also ignoring ptrdiff_t) The first three functions are straight from the iterator_facade class, and the fourth is your typical clone() method. Now, I define a template that wraps any iterator that doesn't implement my iterator_base; it just stores an instance of the backing iterator type and does the corresponding thing (I'm removing some 'typenames' in there to avoid line wraps): template<typename BackingIter> class cplusplus_iterator_adaptor : public iterator_base<iterator_traits<BackingIter>::value_type, iterator_traits<BackingIter>::reference> { private: BackingIter backing; typedef typename iterator_traits<BackingIter>::value_type ValueType; typedef typename iterator_traits<BackingIter>::reference Reference; public: cplusplus_iterator_adaptor() {} cplusplus_iterator_adaptor(cplusplus_iterator_adaptor const & other) : backing(other.backing) {} explicit cplusplus_iterator_adaptor(BackingIter iter) : backing(iter) {} cplusplus_iterator_adaptor* clone() const { return new cplusplus_iterator_adaptor(*this); } // Actual iterator functions void increment() { ++backing; } bool equal(iterator_base<ValueType, Reference> const& other) const { const cplusplus_iterator_adaptor * p = dynamic_cast<cplusplus_iterator_adaptor const*>(&other); return p && backing == p->backing; } Reference dereference() const { return *backing; } }; Okay, now we have something that we can use like cplusplus_iterator_adaptor<vector<int>::iterator> my_itr(vec.begin()); *my_itr; ++my_itr; and it will work the same as vector<int>::iterator, except that you can treat it polymorphically from an iterator_base<int> reference or pointer, thus freeing you from caring that it wraps a vector iterator instead of just caring that it gives you an int&. There's one problem with this however, which is that if I <i>actually</i> want to deal with it polymorphically, I have to use a reference or pointer; but (1) this kinda sucks and (2) lots of code is written that operates on iterators by value (like, say, the STL). To solve this problem, I implemented a wrapper class that gives value semantics to a iterator_base<> pointer. This wrapper class also implements the elements required of the iterator concept, and additionally behaves like a cloning smart pointer. template<typename ValueType, typename ReferenceType = ValueType &> class iterator_base_wrapper : public iterator_facade<iterator_base_wrapper<ValueType, ReferenceType>, ValueType, boost::forward_traversal_tag, ReferenceType> { private: iterator_base<ValueType, ReferenceType> * backing; public: iterator_base_wrapper() : backing(0) {} iterator_base_wrapper(iterator_base_wrapper const & other) : backing(other.backing->clone()) {} explicit iterator_base_wrapper(iterator_base const & base) : backing(base.clone()) {} private: void swap(iterator_base_wrapper & other) const { std::swap(backing, other.backing); } public: iterator_base_wrapper& operator=(iterator_base_wrapper const&other) { iterator_base_wrapper copy(other); swap(copy); return *this; } ~iterator_base_wrapper() { delete backing; } private: friend class boost::iterator_core_access; virtual void increment() { ++(*backing); } virtual bool equal(iterator_base_wrapper const & other) const { return backing->equal(*other.backing); } ReferenceType dereference() const { return **backing; } }; OK, goal achieved! Somewhat. It's now possible to use the iterator_base_wrapper like you would a normal iterator: int run() { const int arr[] = {0,1,2,3,4,5,6,7,8,9,10}; const int size_arr = sizeof(arr)/sizeof(arr[0]); vector<int> v(arr, arr+size_arr); cplusplus_iterator_adaptor<vector<int>::iterator> a_beg(v.begin()); cplusplus_iterator_adaptor<vector<int>::iterator> a_end(v.end()); iterator_base_wrapper<int> ii(a_beg); iterator_base_wrapper<int> ii_end(a_end); vector<int>::const_iterator vi = v.begin(); while(ii != ii_end) { assert( *vi == *ii ); ++vi; ++ii; } assert( vi == v.end() ); return 0; } (BY the way, I totally intend on writing a couple make_* functions, a la make_pair.) OK, but now, what happens if we change things to be const_iterators? I can still make the cplusplus_iterator_adaptor: cplusplus_iterator_adaptor<vector<int>::const_iterator> a_beg(v.begin()); But I *can't* make my wrapper, using either iterator_base_wrapper<const int, const int &> ii(a_beg); or iterator_base_wrapper<int, const int &> ii(a_begin); because of this error: cannot convert from 'iterator_base<int, const int&>*' to 'iterator_base<const int, const int&>*'. The reason for *this* is that the cplusplus_iterator_adaptor uses the iterator_traits<>::value_type of the underlying iterator -- vector's const_iterator -- which isn't const-qualified when it subclasses iterator_facade. This means that the iterator_facade base class in both cases are different types. What's the best way to fix this? The best I can come up with is to add like a clone_const() function that returns a const-qualified version of the value type. But that seems ugly. Is there something elegant I'm missing? Evan Driscoll

4 4

cgal python compilation with boost
by Ferdi Zoet 07 May '11

07 May '11

dear boost user, there seems to be an error when I am trying to compile cgal-python. ============================================================================ 103-99:cgal-python-0.9.4-beta1 2 FerdiZoet$ make ======================================================= Compiling of CGAL package Be patient this may take a while ======================================================= Compiling All_files_at_once.cpp In file included from All_files_at_once.cpp:30: /opt/local/include/boost/python/detail/invoke.hpp: In function ‘PyObject* boost::python::detail::invoke(boost::python::detail::invoke_tag_<false, true>, const RC&, F&, TC&, AC0&) [with RC = boost::python::detail::specify_a_return_value_policy_to_wrap_functions_returning<const CGAL::Point_3<CGAL::Epeck>&>, F = const CGAL::Point_3<CGAL::Epeck>& (CGAL::Triangle_3<CGAL::Epeck>::*)(int)const, TC = boost::python::arg_from_python<CGAL::Triangle_3<CGAL::Epeck>&>, AC0 = boost::python::arg_from_python<int>]’: /opt/local/include/boost/python/detail/caller.hpp:223: instantiated from ‘PyObject* boost::python::detail::caller_arity<2u>::impl<F, Policies, Sig>::operator()(PyObject*, PyObject*) [with F = const CGAL::Point_3<CGAL::Epeck>& (CGAL::Triangle_3<CGAL::Epeck>::*)(int)const, Policies = boost::python::default_call_policies, Sig = boost::mpl::vector3<const CGAL::Point_3<CGAL::Epeck>&, CGAL::Triangle_3<CGAL::Epeck>&, int>]’ /opt/local/include/boost/python/object/py_function.hpp:38: instantiated from ‘PyObject* boost::python::objects::caller_py_function_impl<Caller>::operator()(PyObject*, PyObject*) [with Caller = boost::python::detail::caller<const CGAL::Point_3<CGAL::Epeck>& (CGAL::Triangle_3<CGAL::Epeck>::*)(int)const, boost::python::default_call_policies, boost::mpl::vector3<const CGAL::Point_3<CGAL::Epeck>&, CGAL::Triangle_3<CGAL::Epeck>&, int> >]’ Py_global_functions_2.cpp:171: instantiated from here /opt/local/include/boost/python/detail/invoke.hpp:88: error: no match for call to ‘(const boost::python::detail::specify_a_return_value_policy_to_wrap_functions_returning<const CGAL::Point_3<CGAL::Epeck>&>) (const CGAL::Point_3<CGAL::Epeck>&)’ /opt/local/include/boost/python/detail/caller.hpp: In static member function ‘static const PyTypeObject* boost::python::detail::converter_target_type<ResultConverter>::get_pytype() [with ResultConverter = boost::python::detail::specify_a_return_value_policy_to_wrap_functions_returning<const CGAL::Point_3<CGAL::Epeck>&>]’: /opt/local/include/boost/python/detail/caller.hpp:242: instantiated from ‘static boost::python::detail::py_func_sig_info boost::python::detail::caller_arity<2u>::impl<F, Policies, Sig>::signature() [with F = const CGAL::Point_3<CGAL::Epeck>& (CGAL::Triangle_3<CGAL::Epeck>::*)(int)const, Policies = boost::python::default_call_policies, Sig = boost::mpl::vector3<const CGAL::Point_3<CGAL::Epeck>&, CGAL::Triangle_3<CGAL::Epeck>&, int>]’ /opt/local/include/boost/python/object/py_function.hpp:48: instantiated from ‘boost::python::detail::py_func_sig_info boost::python::objects::caller_py_function_impl<Caller>::signature() const [with Caller = boost::python::detail::caller<const CGAL::Point_3<CGAL::Epeck>& (CGAL::Triangle_3<CGAL::Epeck>::*)(int)const, boost::python::default_call_policies, boost::mpl::vector3<const CGAL::Point_3<CGAL::Epeck>&, CGAL::Triangle_3<CGAL::Epeck>&, int> >]’ Py_global_functions_2.cpp:171: instantiated from here /opt/local/include/boost/python/detail/caller.hpp:102: error: ‘struct boost::python::detail::specify_a_return_value_policy_to_wrap_functions_returning<const CGAL::Point_3<CGAL::Epeck>&>’ has no member named ‘get_pytype’ make[1]: *** [All_files_at_once.o] Error 1 make: *** [all] Error 2 ============================================================================ anybody an idea? thanks. kind regards, Ferdi

1 0

[filesystem] boost::filesystem::file_size fails on Solaris 11
by Red Taza 07 May '11

07 May '11

Hi, I have been trying to get filesystem::file_size() on boost_1_46_1 to work but am getting the following error: $ ./tut1 tut1.cpp boost::filesystem::file_size: Not owner: "tut1.cpp" Note: To avoid core file I modified tut1.cpp so that it catches exceptions. The permissions and ownership of tut1.cpp seem fine: -rw-r--r-- 1 redtaza staff 760 2011-05-07 10:40 tut1.cpp $ uname -a SunOS lunar 5.11 snv_151a i86pc i386 i86pc Solaris $ CC -V CC: Sun C++ 5.11 SunOS_i386 2010/08/13 usage: CC [ options ] files. Use 'CC -flags' for details $ CC -g -library=stlport4 -I/usr/local/boost_1_46_1/include tut1.cpp -o tut1 /usr/local/boost_1_46_1/lib/libboost_system.a /usr/local/boost_1_46_1/lib/libboost_filesystem.a Boost was built with the toolset sun and no other options. Also if I compile and run the following code (which I believe is similar to the code that file_size()) I get expected results: #include <iostream> #include <fcntl.h> #include <sys/types.h> #include <sys/stat.h> #include <cstdio> int main(int argc, char* argv[]) { if (argc < 2) { std::cout << "Usage: tut1 path\n"; return 1; } struct stat path_stat; if(::stat(argv[1], &path_stat)!= 0) { std::perror("stat() error: "); return -1; } if(!S_ISREG(path_stat.st_mode)) { std::cerr << argv[1] << " is not a regular file\n"; return -1; } std::cout << path_stat.st_size << '\n'; return 0; } $ CC -g -library=stlport4 -I/usr/local/boost_1_46_1/include getFilesize.cpp -o getFilesize /usr/local/boost_1_46_1/lib/libboost_system.a /usr/local/boost_1_46_1/lib/libboost_filesystem.a $ ./getFilesize tut1.cpp 760 Thanks

1 0

Invitation to connect on LinkedIn
by Ted McCormack via LinkedIn 07 May '11

07 May '11

LinkedIn ------------Ted McCormack requested to add you as a connection on LinkedIn: ------------------------------------------ Hermann, I'd like to add you to my professional network on LinkedIn. - Ted Accept invitation from Ted McCormack http://www.linkedin.com/e/9m2kgu-gne8wkqz-0/vyhGWmEqB2IKeRN9DMTel5y44rvJcs9… View invitation from Ted McCormack http://www.linkedin.com/e/9m2kgu-gne8wkqz-0/vyhGWmEqB2IKeRN9DMTel5y44rvJcs9… ------------------------------------------ DID YOU KNOW you can conduct a more credible and powerful reference check using LinkedIn? Enter the company name and years of employment or the prospective employee to find their colleagues that are also in your network. This provides you with a more balanced set of feedback to evaluate that new hire. http://www.linkedin.com/e/9m2kgu-gne8wkqz-0/rsr/inv-27/ -- (c) 2011, LinkedIn Corporation

1 0

[pool] singleton_pool crash in multithreaded mode
by Phil Bouchard 07 May '11

07 May '11

Hi, I have a strange bug here. I am using singleton_pool and it crashes at a call to singleton_pool::ordered_malloc in multithreaded mode. The funny thing is I am locking a mutex before the call and it still crashes. The test file can be found here: https://svn.boost.org/svn/boost/sandbox/block_ptr/libs/smart_ptr/example/th… And the calls to the pool (allocate & deallocate) a made from class "pool" of the following file: https://svn.boost.org/svn/boost/sandbox/block_ptr/boost/detail/block_base.h… The backtrace I get is: #0 0x0000000000407842 in boost::simple_segregated_storage<unsigned long>::try_malloc_n (start=@0x42e7ddc0, n=10, partition_size=8) at /home/vnmr1/phil/boost_1_46_1/boost/pool/simple_segregated_storage.hpp:236 #1 0x00000000004078f7 in boost::simple_segregated_storage<unsigned long>::malloc_n (this=0x614d88, n=11, partition_size=8) at /home/vnmr1/phil/boost_1_46_1/boost/pool/simple_segregated_storage.hpp:260 #2 0x000000000040c2ab in boost::pool<boost::default_user_allocator_new_delete>::ordered_malloc (this=0x614d88, n=88) at /home/vnmr1/phil/boost_1_46_1/boost/pool/pool.hpp:527 #3 0x000000000040c9b8 in boost::singleton_pool<boost::detail::sh::pool, 1u, boost::default_user_allocator_new_delete, boost::details::pool::pthread_mutex, 32u, 0u>::ordered_malloc (n=88) at /home/vnmr1/phil/boost_1_46_1/boost/pool/singleton_pool.hpp:72 #4 0x000000000040d937 in boost::detail::sh::pool::allocate (s=88) at ../../../boost/detail/block_base.hpp:150 #5 0x000000000040d9cd in boost::detail::sh::block<std::pair<int, int> >::operator new (s=88) at ../../../boost/detail/block_base.hpp:301 #6 0x000000000040d9f1 in boost::detail::sh::make_block<std::pair<int, int>, std::pair<int, int> > (t0=@0x42e7e030) at ../../../boost/block_ptr.hpp:492 #7 0x0000000000405fd3 in worker (id=1) at thread_test.cpp:17 #8 0x0000000000408b5c in boost::_bi::list1<boost::_bi::value<int> >::operator()<void (*)(int), boost::_bi::list0> (this=0x1776e660, f=@0x1776e658, a=@0x42e7e0cf) at /home/vnmr1/phil/boost_1_46_1/boost/bind/bind.hpp:253 #9 0x0000000000408b9a in boost::_bi::bind_t<void, void (*)(int), boost::_bi::list1<boost::_bi::value<int> > >::operator() (this=0x1776e658) at /home/vnmr1/phil/boost_1_46_1/boost/bind/bind_template.hpp:20 #10 0x0000000000408bb8 in boost::detail::thread_data<boost::_bi::bind_t<void, void (*)(int), boost::_bi::list1<boost::_bi::value<int> > > >::run ( this=0x1776e4d0) at /home/vnmr1/phil/boost_1_46_1/boost/thread/detail/thread.hpp:61 #11 0x00002b6910e0a914 in thread_proxy () from /home/vnmr1/phil/lib/libboost_thread.so.1.46.1 #12 0x0000003906a06367 in start_thread () from /lib64/libpthread.so.0 #13 0x0000003905ed30ad in clone () from /lib64/libc.so.6 Anybody has any idea? Thanks, -Phil

1 0

Re: [Boost-users] [iostreams] Flushing problems with :filtering_ostream
by eg 06 May '11

06 May '11

On Wed, May 4, 2011 at 2:31 PM, Bo Jensen <jensen.bo(a)gmail.com> wrote: > > > In the interest of closure, can you confirm (from the comment you made in > > another thread), that this now works for you after changing the stream > open > > call to be: > > > > std::ofstream strm("test.gz", std::ios_base::out | > std::ios_base::binary); > > > > I know that it works for me when I tried it. > > > Thanks for your interest in this, much appreciated. The problem > persists, the written file can not be decompressed, I have worked my > way around the problem, so it's not a issue for me anymore, but I > would be happy to help track it down. I think it's very strange you > can run without problems. > > To clarify, when I built and ran it with VS 2008 and boost 1.35 it worked. When I just re-tried using VS 2008 and boost 1.45 or boost 1.46.1, it seems to produce invalid .gz files. Perhaps you can create a New Ticket and enter this as a bug at: https://svn.boost.org/trac/boost/newticket

1 0

Re: [Boost-users] Boost-users Digest, Vol 2714, Issue 3
by bill comment 06 May '11

06 May '11

> > 2719 ?The file handle supplied is not valid?. > > FWIW, this usually means that the socket is already closed. > yes. by the time my connection::do_writeMessage() is called, socket_.is_open() returs false. but, why has it been closed and what the correct way to keep it open? i attempt to keep it open after writing the 1st response as follows: in connection::handle_writeResponse() call start(), which calls socket_.async_read_some() this is similar to the chat client sample. apparently that's not sufficient to keep the socket open for when i want to write to it again. fwiw, TcpView (from SysInternals) shows the socket's state as "established" after the 1st successful write and after the failed 2nd attempt. ???

1 0

MPI with threads - adress not mapped problem
by Kraus Philipp 06 May '11

06 May '11

Hello. I'm using Boost 1.46.0 with OpenMPI 1.5.3 under Gentoo x64. I have compiled OpenMPI with thread and mpithread option and Boost with MPI support. A little testprogram with: #include <boost/mpi.hpp> namespace mpi = boost::mpi; int main(int argc, char* argv[]) { mpi::environment loMPIenv(argc, argv); mpi::communicator loMPICom; return 0; } crashes alway if I run it with "mpiexec -np" (without mpiexec it runs). Sometimes the program hangs or creates the message "adress not mapped" with "Segmentation fault". I have recompiled OpenMPI without threads and mpithreads and it seems, that the errors are gone. Is this a problem with OpenMPI or Boost MPI? Phil

2 3

[boost] [xint] Formal Review Result
by Vladimir Prus 06 May '11

06 May '11

I apologize for the delay in producing the final summary and result of the Boost.Xint review, but here it is. Votes ===== YES: - Christian Henning - Steven Watanabe - Jarrad Waterloo - Edward Diener - Paul A. Bristow NO : - Mathias Gaunard - Joel Falcou - Anders Dalvander - Joachim Faulhaber - Phil Endecott - Domagoj Saric - Gordon Woodhull ("fraction of a vote") I might have missed somebody inside the 454 messages I have as result of the review, but the overall picture is clear -- the votes are split. High-level issues ================= - The claim that the library is "fast" is not necessary true. Phil has did some checking against hand-written assembler that don't look too good. - It is not clear whether the design allows for improving performance of fixed-size integers using specialization (again, see Phil's comments) - There were concerns about COW. In particular, it was very clearly stated that if COW is used, it should work fine in multi-threaded programs. - It was suggested to use ET. - Some folks suggested better separation between algorithms and data representation. Specific issues =============== - It was pointed out that "a + b + c" creates a pile of temporaries, and this can be improved without using ET. - Joachim's review raised a lot of valid points. Section 3 of his reviews in something that should be addressed, more or less entirely. Note that while we normally don't force specific coding style, some of formatting pointed out is simply unacceptable if this code is to be read by anybody. - 'Secure' flag is probably not so much of it. Future directions ================= I think that reviews by Phil and Joachim are most important in specying evolutionary directions for Boost.XInt. Chad made several comments about what he plans to do, in particular: - http://article.gmane.org/gmane.comp.lib.boost.devel/216935 Conclusion ========== I think there's no doubt that (i) this domain is important and (ii) Chad has put a lot of work into this library, producing several iterations. It was pointed out that we have many arbitrary-precision libraries skeletons all around Boost, and providing such a solid offering for review is a significant accomplishment. And it everybody learned a lot during the review. There was a couple of important meta-points raised during review -- whether review should focus on external interface only, and what is the scope of the library. Some reviewers argued that only external interface matters. However, it seems to me (as review manager), that concerns about internal design are valid, and may not be rejected offhand. After all, external interface of an integer library is more or less defined by mathematics, and internal design plays a key role in successfull evolution of the library. There was also heated discussion about the scope of the library, in particular support for fixed-width integers. I think that it's abstractly OK for this library to treat fixed-width integers as second-class citizens. However, I am having hard time determining the intended scope of the library. If I'm doing cryptography, would I really want to implement cryptography algorithms from the books with Boost.XInt, only to introduce bugs that are probably fixed in existing implementations (often, included in the OS). And if I'm really intent on doing that, I probably want raw performance, including a way to plug in optimized backends using SSE 2011, CUDA, and whatever fancy technology is there. If I'm doing general programming, then I'm most likely interested in not too big integers. They might be fixed-width 128-bit integers, or they might be something larger-than-64, but not necessary fixed. However, such integers are not the scope of the library, and they were found to have suboptimal performance. It is certainly possible to refine/define the scope, optimize everything for that scope, accomodate the proposals made during review, and obtain an excellent Boost library. However, there are some many suggestions, and so many planned redesigns and changes, that we need a second look at the library. Further, because of the global nature of the changes, it would be very hard to identify a specific subset that must be improved and then go through mini-review. For that reason, I've arrived at the the following result: Boost.Xint library is not accepted at this time. We'll be happy to have another full review of the library when the next version is ready. Thanks to Chad for the submission and for everybody who participated in the review! - Volodya -- Vladimir Prus Mentor Graphics +7 (812) 677-68-40 _______________________________________________ Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost

2 1

[Phoenix] Can we use boost::phoenix::function<Impl const&>?
by TONGARI 06 May '11

06 May '11

Greetings, The intuitive way to make a PFO having state is letting boost::phoenix::function hold the impl's reference, however, the following does not work. struct Impl { ...defines a functor having state... }; Impl impl; boost::phoenix::function<Impl const&> f(impl); f(...lazy call...); // should return an actor here, but not compiles Is this intended behavior?

1 0

Syntax Error identifier heap_memory when using Thread library from precompiled boost version 1.46.1
by Lasse Laursen 06 May '11

06 May '11

Hello boostah's, Sorry for the bad bro lingo. I'm having a spot of trouble getting the latest precompiled boost version to work with vs2010. So far I've only added the include directory, added the lib directory and then tried to include the thread.hpp file which results in the following error: Error 1 error C2061: syntax error : identifier 'heap_memory' c:\boost_1_46_1\boost\thread\win32\thread_heap_alloc.hpp 97 1 texsyn I googled this problem-o and found a response asking if the user had any code which replaced the "new" command for debugging purposes. I did have such code and proceeded to comment it out. Unfortunately, no change. Still the same error. Has anyone experienced anything similar and have an idea of how to best proceed? The bestest of all regards in the world, Lasse

1 1

[Boost.Spirit] RFC: bin_float parser
by Vitaly Budovski 06 May '11

06 May '11

Hello, Please see the attached headers and test case for a bin_float parser. Please provide comments/suggestions/criticisms. Regards, Vitaly Budovski

3 3

serialization
by grahamo 06 May '11

06 May '11

Hi, I'm working on an atrocious project whereby the persistence/serialization of various objects is done by an internal framework. Data is stored in binary format. Each piece of data is stored with an integer identifier that assicates it with the bindary data. I won't go into the details here. Anyways, this framework is very old and appears to present more problems than it solves. My goal is to convince the project manager that boost serialization is the way to go. They are against external 3rd party libraries despite the fact that boost is maintained by hundreds of developers (perhaps more). If I can provide a persuasive argument to migrating to boost, I can rid this project of a lot of issues. Here's the issues we currently face; A) The objects we persist are very much subject to change. New fields are added, moved around and deleted more or less as the wind blows. With each version of the application, various objects are persisted to disk in a file. This file *must* be compatible with the next version of the application. V1 objects and V2 objects are similar but not always identical (sometimes they are). Migration from V1 to V2 is done by massacring the the codebase with if/else statements. -- View this message in context: http://boost.2283326.n4.nabble.com/serialization-tp3498451p3498451.html Sent from the Boost - Users mailing list archive at Nabble.com.

2 1

mpi all reduce with std::vector and back_inserter
by Kraus Philipp 05 May '11

05 May '11

Hello, I have a std::vector<std::string> vector at each CPU. I would like to "reduce" all vectors on all CPUs. For example: CPU 0: vec = [a, b, c, d] CPU 1: vec = [e, f, g, h] std::vector<std::string> myreduce; mpi::all_reduce( mpi com object, vec, myreduce, #); myreduce = [a, b, c, d, e, f, g, h] How can I do this? Should I replace my # with a back_inserter of the std::vector? Thx for help Phil

2 3

Getting multi_index_container in 1.46.1 to work with VC++ 6.0
by tangleman＠secules.net 05 May '11

05 May '11

I've got a legacy MFC project that was built in VC++ 6.0. It needs to stay compatible with VC6 runtime so that I can just drop it into Windows NT4 where the original program was and have it just work. I've finally gotten MFC and VC6 runtime to play nice in VS2003. Now I'm trying to work in boost libraries so that I can use the ethernet and serial aspects of ASIO. Of course, I downloaded and am working with the latest version of boost (1.46.1), and I have it to the point where 311 errors are reported at compile time, all having to do with multi_index_container. I've looked through the compiler specifics pages of the 1.46.1 and 1.45.0 releases with respect to the multi_index and have made the workaround changes that I could (i.e. member->member_offset, boost::multi_index::multi_index_container), but I am still getting syntax errors related to the typedef of my multi_index_container. If it will help, here is my definition: struct ID{}; struct Name{}; typedef boost::multi_index::multi_index_container< PDCUTestMessage, boost::multi_index::indexed_by< boost::multi_index::ordered_unique< boost::multi_index::tag<ID>, boost::multi_index::member_offset<PDCUTestMessage, unsigned char, offsetof(PDCUTestMessage, id)>>, boost::multi_index::ordered_unique< boost::multi_index::tag<Name>, boost::multi_index::member_offset<PDCUTestMessage, string, offsetof(PDCUTestMessage, name)>> > > PDCUMessageList; The first error I'm getting is a syntax error calling out the comma on the line with member_offset PDCUTestMessage id. Then, since that doesn't compile, subsequent error messages snowball for every time I instantiate PDCUTestMessage and every time I add to the multi_index_container. Please help as soon as possible. I'm on a tight schedule, and this is the only major hurdle I see at this point. Thanks! Chris

2 1

[variant] Custom swap() Not Called
by Andreas Grosam 05 May '11

05 May '11

Hi, I'm using a boost::variant whose types have a custom swap() defined. When calling swap() on the variant variable, the custom swap() of the current type will not be called. The "swapable" class has a member function swap(), and there is also a free function swap() defined in the same namespace (see below). To illustrate and test the behavior I defined a special "swapable" class: namespace test { // This class is for testing purpose only! class swapable { public: swapable() : swapped_(0) {} swapable(const swapable& other) : swapped_(0) { } ~swapable() { } int swapped() const { return swapped_; } void swap(swapable& other) { ++swapped_; ++other.swapped_; } swapable& operator= (const swapable& other) { return *this; } private: int swapped_; }; void swap(swapable& lhv, swapable& rhv) { lhv.swap(rhv); } } The calling context looks as follows: void test() { typedef boost::variant< test::swapable > variant_t; variant_t v0( (test::swapable()) ); variant_t v1( (test::swapable()) ); swap(v0, v1); bool swapCalled = boost::get<test::swapable>(v0).swapped() == 1 and boost::get<test::swapable>(v1).swapped() == 1; std::cout << "boost::variant swap test: " << (swapCalled ? "PASSED":"FAILED" ) << std::endl; } So, assuming Koenig lookup works (which in fact does work for the compilers I'm using) calling swap() for the varaint variable should finally call the custom swap(), but it doesn't. Looking deeper into the sources, the culprit seem to be a workaround declaration in file move.hpp (please see below). Note: I'm using gcc, clang and apple's versions of clang. For these compilers __GNUC__ is defined as 4, and in the current config setup BOOST_WORKAROUND(__GNUC__, BOOST_TESTED_AT(2)) evaluates to true. So, when the workaround has been selected, it seems this effectively prevents Koenig lookup which then uses a std::swap like mechanism. Well, now since the compilers I'm using do not have issues with argument-dependent name lookup, how can I fix this - respectively, what is the preferred way to make this work (config setting, maybe)? Note: for testing purposes, I defined BOOST_STRICT_CONFIG 1 (which disables all workarounds) - and this test ran fine. As a side note: I only figured that there is an issue with custom swap() through making performance test - as they ran as slow as a copy. As far as I can see, the boost::variant test suite does not cover such cases. The types which the variant shall hold are copy-expensive - but they are very fast when swap is used (or rvalue reference is enabled some times for boost::variant). Thanks in advance for tips! Regards Andreas ////////////////////////////////////////////////////////////////////////// // function template move_swap // // Swaps using Koenig lookup but falls back to move-swap for primitive // types and on non-conforming compilers. // #if defined(BOOST_NO_ARGUMENT_DEPENDENT_LOOKUP) \ || BOOST_WORKAROUND(__GNUC__, BOOST_TESTED_AT(2)) // [Indicate that move_swap by overload is disabled...] #define BOOST_NO_MOVE_SWAP_BY_OVERLOAD // [...and provide straight swap-by-move implementation:] template <typename T> inline void move_swap(T& lhs, T& rhs) { T tmp( boost::detail::variant::move(lhs) ); lhs = boost::detail::variant::move(rhs); rhs = boost::detail::variant::move(tmp); } #else// !workaround namespace detail { namespace move_swap { template <typename T> inline void swap(T& lhs, T& rhs) { T tmp( boost::detail::variant::move(lhs) ); lhs = boost::detail::variant::move(rhs); rhs = boost::detail::variant::move(tmp); } }} // namespace detail::move_swap

1 0

Need help with compiling boost::function and ublas::matrix
by archie14 05 May '11

05 May '11

I am getting compiler error using boost::function and boost::numeric::ublas as function return value. I am sure I am not declaring something right, but I can't quite catch where (probably the "const" member function specifier). Any help will be greatly appreciated. Code to reproduce the problem is shown below. If #include <vector> #include <boost/function.hpp> #include <boost/numeric/ublas/matrix.hpp> using namespace boost::numeric::ublas; class Foo { public: int bar() const { return 1; } const matrix<double>& stdev() const { return _stdev; } private: matrix<double> _stdev; }; typedef std::vector<Foo> container; class invoker { public: invoker (container* cont, boost::function<int (Foo*) > t) : f(t), _cont(cont) { for (container::iterator it = _cont->begin(); it != _cont->end(); ++it) { Foo& foo = *it; int i = f(&foo); } } // compiler generates error if following code // is uncomennted together with the call in _tmain // invoker (container* cont, // boost::function<const matrix<double>& (Foo* const) > fmatrix) // : _fmatrix(fmatrix), _cont(cont) // { // for (container::iterator it = _cont->begin(); // it != _cont->end(); ++it) // { // Foo& foo = *it; // const matrix<double>& stddev = _fmatrix(&foo); // } // } private: boost::function<int (Foo*) > f; boost::function<const matrix<double>& (Foo* const) > _fmatrix; container* _cont; }; int _tmain(int argc, _TCHAR* argv[]) { container lst (5); std::fill(lst.begin(), lst.begin() + 5, Foo()); invoker invbar(&lst, &Foo::bar); // uncomment this line together with method in "invoker" to generate error // invoker invstdev(lst, &Foo::stdev); return 0; }

1 0

Adjacency List vertex access
by Carlo jardinier 05 May '11

05 May '11

I try to use the adjacency list from boost. I can fill the list with vertices and edges, but somehow I cannot access a specific vertex or edge. For example this is not working (see test.cpp): ** struct EdgeProperties { bool bConnectedBond; EdgeProperties(){bConnectedBond=false;} }; struct VertexProperties{ int index; int spin; bool bOnCluster; VertexProperties(): spin(){ bOnCluster = false;} ~VertexProperties(){ } }; typedef adjacency_list< vecS, vecS, bidirectionalS, VertexProperties, EdgeProperties > Graph; typedef graph_traits< Graph >::vertex_iterator Vit; Graph g(9); unsigned int i = 3; Vit vi; vi = vertex(i,g); ** I want vi to bi an iterator that points on the 3. vertex of the graph g. I attached the error message from g++ when compiling. Can anyone help?

4 7

[Signals] Is connection::disconnect() "const-correct"?
by Greg Hickman 05 May '11

05 May '11

For a connection 'c', if c.connected() == true prior to calling c.disconnect(); after which c.connected() == false, how can connection::disconnect() be considered a "const" member function since it clearly changes the observable state of a connection?

3 3

dereferencing type-punned pointer will break strict-aliasing rules
by Hossein Haeri 05 May '11

05 May '11

Dear all, [Boost 1.46, GCC 4.5.1, MinGW32, WinXP SP3] I use a number of Boost libraries that I see emit the following warning in GCC under release build: "dereferencing type-punned pointer will break strict-aliasing rules" I have traced the code (of SmartPtr and Function specifically) and it's always with an operations like reinterpret_cast<T*>(...)->~T(). Now, I'm wondering how safe this reinterpret_cast is? How serious should I take this warning here? (I do already know how to turn it off or the strict aliasing check.) And, more than anything else: Is this warning suggesting an improvement in my own code? TIA, --Hossein

1 0

Specifying an outgoing interface for IPv6 multicast
by Jay Katti 04 May '11

04 May '11

Folks, Has anyone figured out how to specify the outgoing interface for IPv6 multicast using the boost asio library socket options on Linux ? See snippets below... What value does one specify for "unsigned int ipv6_interface" in the network_interface constructor ? Tried specifying the interface index (1..3 on my host) but that did not work. Thanks, Jay .. >From multicast.hpp * Implements the IPPROTO_IP/IP_MULTICAST_IF socket option. * * @par Examples * Setting the option: * @code * boost::asio::ip::udp::socket socket(io_service); * ... * boost::asio::ip::address_v4 local_interface = * boost::asio::ip::address_v4::from_string("1.2.3.4"); * boost::asio::ip::multicast::outbound_interface option(local_interface); * socket.set_option(option); * @endcode * * @par Concepts: * SettableSocketOption. */ #if defined(GENERATING_DOCUMENTATION) typedef implementation_defined outbound_interface; #else typedef boost::asio::ip::detail::socket_option::network_interface< IPPROTO_IP, IP_MULTICAST_IF, IPPROTO_IPV6, IPV6_MULTICAST_IF> outbound_interface; #endif ========================================================== From: socket_options.hpp ... // Construct with IPv6 interface. explicit network_interface(unsigned int ipv6_interface) { ipv4_value_.s_addr = boost::asio::detail::socket_ops::host_to_network_long( boost::asio::ip::address_v4::any().to_ulong()); ipv6_value_ = ipv6_interface; }

1 0

Re: [Boost-users] [msm] EUML and State construction
by Christophe Henry 04 May '11

04 May '11

> I'm trying to construct some euml states with data at the time of the state > machine construction. The below boils down what I'm trying to do and > fails to compile. Should I expect this mechanism to work with the euml > frontend ? Yes you should but there is a bug, thanks for reporting it. <snip code> > ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ > With gcc 4.5 and Boost 1.46.1, I get: > > simpletest.cpp: In function 'int main()': > simpletest.cpp:30:75: error: no match for 'operator<<' in > 'boost::msm::back::states_ << test_state_with_constructor(1)' Ok, I see what it is, I did not solve correctly my conflict between proto and serialization. I just committed a fix into the trunk (revision 71719), but it's not working, I simply run into my next proto problem :( I'll continue working on this tomorrow. Thanks, Christophe

1 0

[unordered_map] Trying to understand some errors
by Craig Longman 04 May '11

04 May '11

I'm using VS v9, just in case that matters in this case. I'm : using namespace std; using namespace boost; unordered_map< string, uint32_t > xrf1; unordered_map< string, uint32_t >::const_iterator& iter = xrf1.find( "hello" ); This is something I've used many times with the other containers (well, std::map at least), but here I'm getting a conversion error. This is resolved by either removing the reference marker, or removing the const qualifier. Can someone explain this please? I'd always thought that applying const was easily done, and if so, then a reference should be sufficient, right? Unless the const cast is marked as explicit somehow? And as a side(ish) question, is there any real need to worry about have an object versus reference to something like an iterator? I've always tried to use references where I can, but I wonder if I should "just say no" and use regular objects from now on? Thanks, -- Craig Longman If it's not *signed* by me, then please don't assume it *is* from me.

3 2

[BGL] Extending Floyd Warshall (was: All-pairs SP algorithms -> Determine the path)
by Nicholas Mario Wardhana 04 May '11

04 May '11

On 16 April 2011 02:42, Jeremiah Willcock <jewillco(a)osl.iu.edu> wrote: > Hi Jeremiah, My apology for digging up an old post. I'd like to ask how Boost's Floyd Warshall can extended to provide further functionality. I cannot find any notion of visitor for Floyd Warshall, unlike Dijkstra and A*, and I am a bit confused about the types of the arguments that have to be passed to the function. > One thing you can do in BGL is to compute the paths using the distances; > look for edges where the source and target distances from a particular > starting node differ by the edge's weight, and that will be in the SSSP tree > for that starting node. You can also change the distances to <distance, > predecessor> pairs (just by changing the types being passed into the > algorithm); Will that be a variable of type std::vector<std::vector<std::pair<float, boost::vertex_descriptor>>> ? I tried this and the code failed to compile. > you would still compare just the distances, and the weight of an > edge would be a pair <actual_weight, source>. Will that be a std::pair<float, vertex_descriptor> variable? Additionally, is it also possible to know the number of nodes in a shortest path between two vertices in the graph by using Boost's Floyd-Warshall implementation? > > -- Jeremiah Willcock Thank you. Best regards, Nicholas

2 3

[asio] Joining a group on a specific interface not documented
by Dylan Klomparens 04 May '11

04 May '11

I discovered today that the boost::asio::ip::multicast::join_group option has a constructor to specify a local address to use when joining a multicast group. Unfortunately, this constructor is not documented. How can I contribute to the boost documentation or contact the author to request an update in the documentation?

2 1

Force boost::asio::buffer to copy by value.
by Slav 04 May '11

04 May '11

Hello. I use boost::asio::buffer to send a message using void Send(const std::string& messageData) { socket.async_write(boost::asio::buffer(messageData), ...); } And encounter "string iterator not dereferencable" runtime error somewhere within io_service' thread. When I create objects' variable to store message data for the buffer: void Send(const std::string& messageData) { this->tempStorage = messageData; socket.async_write(boost::asio::buffer(this->tempStorage), ...); } the error is never occured. std::string (to which messageData is referenced to) is freed almost right after Send() calling - does boost::asio::buffer stores just a reference to object? If so, how can I force it to store the data by value?

2 2

Re: [Boost-users] [unordered_map] Trying to understand some errors
by Craig Longman 04 May '11

04 May '11

On 03/05/2011 4:59 PM, Nat Goodspeed wrote: > > unordered_map< string, uint32_t> xrf1; > > unordered_map< string, uint32_t>::const_iterator& iter = xrf1.find( "hello" ); > > > > This code makes me shudder. Capturing the value returned from a function with a reference variable asserts > that the function is itself returning a reference, not a value - and always will, for the life of the codebase. > > This is far more general than iterators. > > My rule of thumb is: return by value, store returned data by value. Perhaps a bit dramatic, but you're quite right. =) And it is something I'm (normally) aware and careful of. Returning values is getting more and more acceptable as the optimizations get better and better. However, references do have their place, even in return values. I'm actually honestly not sure where I picked that up with iterators, but it worked and I guess it just kinda stuck. I try and use references over copies where appropriate, and as it seemed to work. I didn't think it through properly. > To answer your question, because xrf1 isn't const, its find() returns a non-const iterator. You're trying > to capture a reference to a const_iterator, which isn't what you have in hand. But, the confusion is if I replace unordered_map with std::map, it compiles fine. I guess that means in std::map a const_iterator is just that, a "const iterator", but in unordered_map, the const_iterator is a completely different type, not just a const version of the normal iterator? I was under the impression that const could and would always be implicitly applied if needed, just not the other way round (obviously). Time to dig through some header files I guess. Thanks, -- Craig Longman If it's not *signed* by me, then please don't assume it *is* from me.

2 1

[Spirit] Generating from struct containing single std::vector member
by Vitaly Budovski 04 May '11

04 May '11

Hi, When generating from a struct containing a single std::vector member, +parser or *parser results in an error. If, however, it is followed by an eps, it succeeds. Is this expected behaviour? Please see attached code demonstrating issue.

2 1

Thread Safety of shared_ptr
by Kevin Frey 04 May '11

04 May '11

Hello, I wish to clarify a point on the thread-safety of shared_ptr. I have read the boost documentation on shared_ptr concerning Thread-Safety and amongst other things it gives this example (see http://www.boost.org/doc/libs/1_46_1/libs/smart_ptr/shared_ptr.htm for the full set of examples): shared_ptr<int> p(new int(42)); //--- Example 1 --- // thread A shared_ptr<int> p2(p); // reads p // thread B shared_ptr<int> p3(p); // OK, multiple reads are safe <snip...> //--- Example 4 --- // thread A p3 = p2; // reads p2, writes p3 // thread B // p2 goes out of scope: undefined, the destructor is considered a "write access" Given the nature of how the examples are presented, I have nonetheless presumed the examples are largely independent of one another, as opposed to each example being dependent on the results of the previous example. The specific question I have concerns the implications of the phrase "p2 goes out of scope". I have a situation where I have a shared_ptr< Session > which is initially created by thread A. Many such objects in fact. A separate thread B cleans up periodically (the exact scenario is a .NET finaliser interfacing to native C++ code). It is therefore entirely possible for the following two actions to happen simultaneously: - Thread A is instantiating a new copy of shared_ptr< Session >, copied from an existing shared_ptr< Session >. - Thread B is destroying a shared_ptr< Session >. The above shared_ptrs are all referencing the same reference-counting object. I do not actually know which shared_ptr will be the one that actually destroys the underlying Session object (which is the whole point of me using shared_ptr as my cleanup ends up being non-deterministic with respect to sequence). Example 4 is very important to me in terms of me understanding exactly what limitation they are describing. My possible interpretations are: 1. p2 goes out of scope under any circumstance. Doesn't matter if the shared_count is 1 or > 1 prior to the object going out of scope. 2. p2 goes out of scope, resulting in destruction of the underlying object. In my interpretation of the example, when p2 goes out of scope, it will not result in destruction of the object, because p2 is a copy of p, and p is [presumably] still in scope. The outcome should merely be a decrementing of the reference count. So I'm also at a bit of a loss to understand why an incrementing of a reference count is not considered a "write access" but a potential simple decrementing of a reference count (occurring via the shared_ptr destructor) is considered a "write access". In my example I know I will never have thread A attempting to increment the ref count whilst thread B is potentially destroying the object simultaneously. When I get to my final copy of my object (ie. unique( ) is true), thread A is guaranteed to not be accessing or copying the shared_ptr when thread B commences the destruction of the final copy. Will shared_ptr behave correctly or not in my situation? Because if I take the boost example on simple face value, that I cannot have one shared_ptr going out of scope (and decrementing the ref count) whilst another thread is incrementing the ref count, then I'm feeling at a loss why boost would go to the trouble of a lock-free implementation in the first-place, since I would have thought my example is a pretty common (almost canonical) of why you would want thread-safety for shared_ptr. Now I'll be the first to admit that the boost guys are a lot smarter than me, so I'm hoping my understanding of the capability is simply wrong. (The problem for me is that if shared_ptr does not provide the thread-safety requirements I need, then I must achieve that thread-safety external to shared_ptr with some kind of mutex, which puts me in a conundrum as to where such a mutex would be held. The obvious place would be my Session object since it provides the perfect granularity, but then I have the situation of having an object, which might potentially get deleted [when the last shared_ptr is destroyed], having a mutex referenced within it for the purpose of locking, which is clearly wrong.) Any help will be greatly appreciated. Thanks Kevin

7 15

distributed breadth first search doesn't work properly?
by Mattia Lambertini 04 May '11

04 May '11

Hi, i am new to boost and i'm using the PBGL to write a software that compute some graph analysis. I've installed boost 1.44.0 and openmpi 1.5 I tried to run the breadth first search example located in $BOOST_ROOT/libs/graph_parallel/example with the graph $BOOST_ROOT/libs/graph/test/weighted_graph.gr All went well since i figure out that the result is different if i run the algorithm with a different number of processors. Only with one processor the values are corrected. here's what happens: mpirun -np 1 ./breadth_first_search.out graph g { subgraph cluster_0 { label="p0"; n0[label=0]; n1[label=2]; n2[label=1]; n3[label=2]; n4[label=1]; } n0 -- n2; n1 -- n1; n1 -- n3; n1 -- n4; n2 -- n1; n2 -- n3; n3 -- n4; n4 -- n0; n4 -- n1; } --------------------- mpirun -np 2 ./breadth_first_search.out graph g { subgraph cluster_0 { label="p0"; n0[label=0]; n1[label=1]; n2[label=1]; } n0 -- n2; n1 -- n1; n1 -- n3; n1 -- n4; n2 -- n1; n2 -- n3; subgraph cluster_1 { label="p1"; n3[label=1]; n4[label=0]; } n3 -- n4; n4 -- n0; n4 -- n1; } Is that correct? From the doc i can't figure out why.. -- Cordiali saluti, Mattia Lambertini

2 11

[PBGL] Results of distributed BFS on multiple processors vs. on a single processor
by Cedric Laczny 04 May '11

04 May '11

Hi, it recently came to my mind again that the distributed BFS in the PBGL seems to produce different results when applied to multiple processors as to when applied to a single processor (s. http://lists.boost.org/boost- users/2010/12/64783.php) If this is true, I really would like to know what the reasoning is behind this and it should be noted in the documentation of course to help prevent surprises as in the discussion above. Unfortunately, at that time, no definite answer was given and this question puzzled me again now... Any ideas on that? Best, Cedric

3 6

How to get current state from state machine?
by jusva2010 04 May '11

04 May '11

Hi All, How can I get the current state of the state machine? Currently i am getting in this way.. i would like to know if there any better way statemachine --------------- struct StateMachine:sc::state_machine<StateMachine, Initial> { void setState(string s){ state = s; } string getState() { return state; } private: string state; } //event struct dummyEvent:sc::event<dummyEvent> { } //state struct Initial:sc::simple_state<Initial, StateMachine> { typedef mpl::list<sc::custom_reactions<dummyEvent>> reactions; sc::result react(const dummyEvent&); } In react function of the 'Initial' state sc::result Initial::react(const dummyEvent&) { context <StateMachine>().setState("Initial"); } In my main main() { StateMachine machine; machine.initiate() cout << "Current State "<< machine.getState() << endl; // return me the current state } Thanks in adavance justin ----- justin -- View this message in context: http://boost.2283326.n4.nabble.com/How-to-get-current-state-from-state-mach… Sent from the Boost - Users mailing list archive at Nabble.com.

1 0

BOOST_USE_WINDOWS_H and atomic ops in shared_ptr
by Joseph Lunderville 04 May '11

04 May '11

I've been trying to purge windows.h includes from a project that uses Boost, but I see compile errors when I don't define BOOST_USE_WINDOWS_H (the included prototypes don't match prototypes defined elsewhere). This is building Boost 1.42.0 in VS2010. Is this a known issue? Is it known to be fixed in a more current version? Am I doing something wrong? I worked around the problem by creating an alternate set of #defines for the atomic ops, and including <intrin.h> instead (controlled via a new BOOST_USE_INTRIN_H macro). I can supply a patch if this sounds like a reasonable thing to do. -Joe ________________________________ This message, including any attachments, may contain privileged and/or confidential information. Any distribution or use of this email by anyone other than the intended recipient(s) is strictly prohibited. If you are not the intended recipient, please notify the sender immediately and delete all copies. Thank you.

1 0

weird Link error with XCode 4
by Kumar, Akhilesh 03 May '11

03 May '11

Hi, I am having weird link error during the release build details give below, I was having some problem with the Debug build also but after playing with "InlineMethodHidden and SymbolhiddenbyDefault" flag I was able to build in Debug config, but this trick does not work in Release config. I addition to boost library I am also using JsonSpirit library. Though I have manually copied 10.5 SDK to support 10.5 system but using 10.6 SDK also result in similar looking link error. Boost and Json Spirit was compiled with 3.2 version of Xcode. This code compile and run perfectly fine in Xcode 3.2. Any pointer will be appreciated. Link Error ************** ld: bad codegen, pointer diff in void (anonymous namespace)::throw_error<boost::spirit::classic::multi_pass<std::istream_iterator<wchar_t, wchar_t, std::char_traits<wchar_t>, int>, boost::spirit::classic::multi_pass_policies::input_iterator, boost::spirit::classic::multi_pass_policies::ref_counted, boost::spirit::classic::multi_pass_policies::buf_id_check, boost::spirit::classic::multi_pass_policies::std_deque> >(boost::spirit::classic::multi_pass<std::istream_iterator<wchar_t, wchar_t, std::char_traits<wchar_t>, int>, boost::spirit::classic::multi_pass_policies::input_iterator, boost::spirit::classic::multi_pass_policies::ref_counted, boost::spirit::classic::multi_pass_policies::buf_id_check, boost::spirit::classic::multi_pass_policies::std_deque>, std::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)to global weak symbol typeinfo for std::basic_string<char, std::char_traits<char>, std::allocator<char> >for architecture i386 Command /Developer/usr/bin/g++-4.2 failed with exit code 1 GenerateDSYMFile build/Release/libSariManager.dylib.dSYM build/Release/libSariManager.dylib cd /Users/Shared/P4/code/client /Developer/usr/bin/dsymutil /Users/Shared/P4/code/client/build/Release/libSariManager.dylib -o /Users/Shared/P4/code/client/build/Release/libSariManager.dylib.dSYM error: unable to open executable '/Users/Shared/P4/code/client/build/Release/libSariManager.dylib' Akhilesh Kumar

1 0

[msm] EUML and State construction
by Nate Knight 03 May '11

03 May '11

I'm trying to construct some euml states with data at the time of the state machine construction. The below boils down what I'm trying to do and fails to compile. Should I expect this mechanism to work with the euml frontend ? ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ #include <boost/msm/back/state_machine.hpp> #include <boost/msm/front/euml/euml.hpp> namespace msm = boost::msm; using namespace msm::front::euml; BOOST_MSM_EUML_EVENT(test_event) BOOST_MSM_EUML_DECLARE_STATE((),test_state_t) struct test_state_with_constructor : public test_state_t { test_state_with_constructor() {} test_state_with_constructor(int d) : m_d(d) {} int m_d; }; test_state_with_constructor const test_state; BOOST_MSM_EUML_TRANSITION_TABLE(( test_state + test_event == test_state ), test_table) BOOST_MSM_EUML_DECLARE_STATE_MACHINE (( test_table, init_ << test_state ), test_table_) typedef msm::back::state_machine<test_table_> test_machine_t; int main() { test_machine_t tm( msm::back::states_ << test_state_with_constructor(1) ); } ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ With gcc 4.5 and Boost 1.46.1, I get: simpletest.cpp: In function 'int main()': simpletest.cpp:30:75: error: no match for 'operator<<' in 'boost::msm::back::states_ << test_state_with_constructor(1)' Thanks Nate

1 0

[serialization] pointer to non-constructable object
by Noah Roberts 03 May '11

03 May '11

The problem boils down to something similar to: struct base { virtual ~base() {} base(x_type & obj); }; struct derived { derived() : base(member_x) {} private: x_type member_x; }; ...serialize(...,base &,....) {} ...serialize(..., derived & d, ....) { ar & base_object<base>(d); ar & member_x; } boost::shared_ptr<base> ptr(new derived); ar & ptr; The problem is that base can't be constructed without the reference. It also makes no sense to override the construction stuff because it's simply not how to get that reference. There's no way to construct a base that is sensible outside of during the construction of its derived (it's a funky design I admit, but necessary). Unfortunately the serialization library insists on being able to construct a base, even though I never save one except via derived objects through a pointer to base. I noticed a check for "is_abstract" when debugging a different issue I had (that was totally my fault). So I tried adding a bogus virtual function to the interface and overriding it in derived. This fixes the problem. There must be a way to do so without such trickery though? Can this abstract check be overridden by me?

1 0

[program_options] How to extract option value?
by Krzysztof 03 May '11

03 May '11

In order to get the option value I have this code: options_description opd; string in; opd.add_options() ("help","Display help") ("input_file,I",value<string>(),"Input file with group division"); variables_map vm; store(parse_command_line(argc,argv,opd),vm); notify(vm); if(!vm["help"].empty()) cout<<opd<<'\n'; else in = vm["input_file"].as<string>(); However the last line issues this exception: terminate called after throwing an instance of 'boost::exception_detail::clone_impl<boost::exception_detail::error_info_injector<boost::bad_any_cast> >' what(): boost::bad_any_cast: failed conversion using boost::any_cast What do I wrong? -- Regards Krzysztof J.

1 0

[spirit] parsing keyword or identifier
by Brian Budge 03 May '11

03 May '11

HI all - I have what I think is a simple problem: I'm trying to parse IES files, and there is a line that can have three options: TILT=NONE -or- TILT=INCLUDE -or- TILT=<filename> The way I'm trying to parse this line is: // in the case of TILT=INCLUDE or TILT=<filename> int lampToLuminaireGeom; int numPairsWithMulFactors = 0; std::string tiltFileName; BOOST_AUTO(keywordSym, (qi::lit("NONE") | qi::lit("INCLUDE"))); BOOST_AUTO(fileName, *char_ - keywordSym); BOOST_AUTO(initial, ( "TILT" >> char_('=') >> ( "NONE" | ( "INCLUDE" >> int_[ref(lampToLuminaireGeom) = _1] >> int_[ref(numPairsWithMulFactors) = _1] ) | fileName[ref(tiltFileName)] ) ) ); if(!phrase_parse(first, last, initial, space)) { return IES_INVALID; } The code works with NONE or INCLUDE, but not with files. As far as I know, there are no real restrictions on the filenames. At this point, the code posted above can't even get simple file names, but really I think what is required is that I grab anything non-whitespace after the '=' until it reaches an EOL. Can anyone provide some insight about what I may be doing wrong? Thanks, Brian

2 2

[wave] Custom I/O handling (include paths, etc.)?
by Anteru 03 May '11

03 May '11

Hi, I'm trying to adopt Wave as my custom preprocessor for a project. So far, things work fine, but there is one problem with Wave: I need to override all I/O functions, in particular, reading from files and opening files. Reading is easily changed by the input policy, but I couldn't find a way to register an include handler hook. There's found_include_directive, but nothing that allows me to get in-between wave and the filesystem. In my case, I on-demand construct some data which I want to be able to #include (i.e. #include <generated>) and some files are stored in resources/compressed files (#include <zip/file.h>). My include handler can open them, but how can I tell Wave to ask my include handler whether it can find a file instead of using boost::filesystem? Cheers, Anteru

2 4

[Review] Boost.Convert library has ended
by Edward Diener 03 May '11

03 May '11

The review period for the Boost.Convert library has officially ended. I will continue to read posts about the library while I make my review decision, so please do not stop discussing the library as appropriate because the review period has ended. I have received one personal request for a late review, and I am willing to accept any other requests for a late review to be made through Wednesday and posted to the mailing list, but I will not consider any other late review in my decision unless someone requests it of me personally. I know this is a fine line between an official review and just comments about what others have said before, but I need to make my decision based on some finite set of reviews and their subsequent discussion. I would like to thank everyone who has officially reviewed Boost.Convert as well as all who have made subsequent comments. The more input received, the easier it becomes for me as a reviewer to make a final decision. I would also like to thank the implementer of the Boost.Convert library, Vladimir Batov, for his willingness to address issues, answer criticism, and engage in discussion about his library. My review decision should be made and posted to the mailing list no later than by the end of Sunday, May 8.

1 0

Pass data to states, in boost state chart
by jusva2010 03 May '11

03 May '11

Hi All, Is there any way I can pass data to state event handler? Any pointers really appreciated. Thanks in advance Justin -- View this message in context: http://boost.2283326.n4.nabble.com/Pass-data-to-states-in-boost-state-chart… Sent from the Boost - Users mailing list archive at Nabble.com.

2 5

Adapt rule's attribute to my own Boost::Spirit::Qi
by Egor Popov 03 May '11

03 May '11

Hello, everyone. There is a problem of attribute transfomation: I need to transfom rule's generated attribute to my own structure. I tried to parse mathematical formula and store it in std::vector in reverse Polish notation. Formula have form: math = as_string(*digit) >> (alpha) >> +(char_('+' | '-') >> math); rule<Iterator, std::string, space_type> math; Question #1: what type is it? I think that fusion::vector<std::string, char, std::vector<fusion::vector<char, std::string> > > but it seems it is not so. Question #2: how can I convert this structure, to mine (std::vector<std::string>)? -- View this message in context: http://boost.2283326.n4.nabble.com/Adapt-rule-s-attribute-to-my-own-Boost-S… Sent from the Boost - Users mailing list archive at Nabble.com.

1 0

[Multi-index] : code fail to compile with VS2010 and boost 1.46.1. Proposed workaround
by Mathieu Peyréga 03 May '11

03 May '11

Hello, I had some code using boost 1.46.1 Multi-index library that used to compile and work very well with VS2008. I recently turned to VS2010 and some code is not compiling anymore. The problem is located in the boost_1_46_1\boost\multi_index\composite_key.hpp file and the compiler complains about not being able to distinguish between ambiguous overloads... I had to modify the composite_key.hpp file and have following changes : line 1001 : //return operator()(x,make_tuple(cref(y))); return operator()(x,make_tuple(boost::cref(y))); line 1036 : //return operator()(make_tuple(cref(x)),y); return operator()(make_tuple(boost::cref(x)),y); In order to explicitely add the boost namespace for cref. With those changes, it do compile and works very well. Can those changes in composite_key.hpp break something else ? Is there another workaround I would have missed ? I'm suspecting some conflict with new TR1 things i'm not very familiar with... Best regards, Mathieu Peyréga

3 7

[Spirit][Date_Time] Parsing of boost date_time (segfault)
by Vitaly Budovski 03 May '11

03 May '11

Hi, I had a go at building a binary parser to output a date/time object. Unfortunately, I ran into some serious issues. One of my grammars causes the application to segfault when compiled with gcc, but not in clang. I'm not sure if this is a bug in gcc, boost, or simply a programming error. Please see attached file. Is there a better way to achieve what I need (final output should be ptime object)? All suggestions welcome :) Thanks, Vitaly

3 3

[Serialization - 1.46.1] : Suggestion
by Mathieu Peyréga 03 May '11

03 May '11

Hello, Here is a small feature suggestion : I've been modifying the nvp.hpp file and added after line 142 the following macro definition : // Added by Mathieu Peyréga #define BOOST_SERIALIZATION_BASE_OBJECT_TEMPLATE_NVP(name,templateparameter) \ boost::serialization::make_nvp( \ BOOST_PP_STRINGIZE(name)BOOST_PP_STRINGIZE(templateparameter), \ boost::serialization::base_object<name<templateparameter> >(*this) \ ) // Which enables "easier" serialization for case like the following where the is multiple heritage and one of the base class is a template. Do you think this is the good way to go ? Best regards, Mathieu Peyréga /*************************************************************************************************** Short example of the BOOST_SERIALIZATION_BASE_OBJECT_TEMPLATE_NVP macro ***************************************************************************************************/ class C_Battery : public C_PhysicalElement, public C_GenericDatabaseObject<C_Battery> { public: C_Battery(void); C_Battery(C_Battery const& obj); C_Battery& operator=(C_Battery const& obj); virtual ~C_Battery(void); Boolean operator==(C_Battery const& other) const; Float64 getTensionPerElement(void) const; Float64 getCapacityPerElement(void) const; Float64 getFaradicEfficiency(void) const; Float64 getUnloadDeepness(void) const; Float64 getVolumicInstallationCost(void) const; Float64 getNumberOfCycles(void) const; Float64 getMonthlySelfLosses(void) const; void setTensionPerElement(Float64 const& value); void setCapacityPerElement(Float64 const& value); void setFaradicEfficiency(Float64 const& value); void setUnloadDeepness(Float64 const& value); void setVolumicInstallationCost(Float64 const& value); void setNumberOfCycles(Float64 const& value); void setMonthlySelfLosses(Float64 const& value); void loadFromCSVDatabase(C_CSVFileParser const& parser, Int32 recordIndex); protected: friend class C_GenericDatabaseObject<C_Battery>; Float64 m_TensionPerElement; // Tension par élément en Volts Float64 m_CapacityPerElement; // Capacitée par élément en A*h Float64 m_FaradicEfficiency; // Rendement faradique Float64 m_UnloadDeepness; // Profondeur de décharge Float64 m_VolumicInstallationCost; // Cout d'installation en euro/dm3 Float64 m_NumberOfCycles; // Nombre de cycles à la profondeur autorisée Float64 m_MonthlySelfLosses; // Auto décharge mensuelle friend class boost::serialization::access; template<class Archive> void serialize(Archive & ar, const unsigned int version) { ar & BOOST_SERIALIZATION_BASE_OBJECT_NVP(C_PhysicalElement); ar & BOOST_SERIALIZATION_BASE_OBJECT_TEMPLATE_NVP(C_GenericDatabaseObject,C_Battery); ar & BOOST_SERIALIZATION_NVP(m_TensionPerElement); ar & BOOST_SERIALIZATION_NVP(m_CapacityPerElement); ar & BOOST_SERIALIZATION_NVP(m_FaradicEfficiency); ar & BOOST_SERIALIZATION_NVP(m_UnloadDeepness); ar & BOOST_SERIALIZATION_NVP(m_VolumicInstallationCost); ar & BOOST_SERIALIZATION_NVP(m_NumberOfCycles); ar & BOOST_SERIALIZATION_NVP(m_MonthlySelfLosses); } }; BOOST_CLASS_VERSION(C_Battery, 1)

1 0

Re: [Boost-users] Thread Safety of shared_ptr
by Kevin Frey 03 May '11

03 May '11

> shared_ptr< Session >* is not a shared ptr, it is a RAW ptr to shared_ptr< Session>. Yes, I know this. The pointer you call the RAW pointer is owned only by one object. It doesn't make the implementation wrong. My issue concerns the addref/decref behaviour that occurs simultaneously within two different threads.

1 0

Re: [Boost-users] Thread Safety of shared_ptr
by 乔志强 03 May '11

03 May '11

shared_ptr< Session >* sp_Session = new shared_ptr< Session >( new Session ); shared_ptr< Session >* is not a shared ptr, it is a RAW ptr to shared_ptr< Session>. So you should: 1. use shared_ptr of shared_ptr: shared_ptr<shared_ptr< Session > > sp_Session ( new shared_ptr< Session >( new Session )) ; 2. or use shared_ptr directly: shared_ptr< Session > sp_Session( new Session ); I think you should use 2. > ------------------------------ > > Message: 7 > Date: Tue, 3 May 2011 10:46:39 +1000 > From: "Kevin Frey" <kfrey(a)hardcat.com> > To: <boost-users(a)lists.boost.org> > Subject: Re: [Boost-users] Thread Safety of shared_ptr > Message-ID: > <95C19A2ED65BF54CB79DF1F9D6A488E0013E3351(a)hcmail1.MEL01.hardcat.com.a > u> > > Content-Type: text/plain; charset="us-ascii" > > > Actually, I am not sure with this statement. Destructor is only called > if shared count reaches zero. I don't think that this can happen, that a > dtor is called in one thread and the other thread is still able to > increment the shared count even from the weak_ptr. I remember reading > weak_ptr's source and saw that it is also protected again such a case. > > > > I have taken the example at face value, but in hindsight the examples > could be a lot better because there is no context to describe the > circumstances of how the shared_ptrs are instantiated. > > > > For example, is p allocated globally, and p2 & p3 allocated at the > beginning of their respective threads? If so, example 4 becomes > meaningless because Thread B cannot force p2 out of scope since it would > only go out of scope once thread A exits, since thread A controls the > scope of p2. If thread B were to force p2 "out of scope" then p2 would > have to be a shared_ptr< >*, dynamically allocated., and subsequently > deleted by Thread B. But it is not represented that way. > > > > Unfortunately none of this helps solve my original question, which given > the direction this thread is proceeding, I will re-state in pseudo-code > that approximates the *actual* situation occurring in my program: > > > > Earlier on: > > > > shared_ptr< Session >* sp_Session = new shared_ptr< Session >( new > Session ); > > > > Dependent_Class* dep = new Dependent_Class( shared_ptr< Session >& > shared_ptr_session ); > > In Dependent_Object::Dependent_Object( ) > > this->sp_Session = new shared_ptr< Session >( shared_ptr_session ); > > Now, simultaneously: > > > > Thread A: > > Dependent_Class* dep2 = new Dependent_Class( shared_ptr< Session>& > shared_ptr_session ); > > In Dependent_Object::Dependent_Object( ) > > this->sp_Session = new shared_ptr< Session >( shared_ptr_session ); > > > > Thread B: > > In Dependent_Class::~Dependent_Class( ): // the cleanup of dep > > delete this->sp_Session; > > > > So, we have: > > > > 1. A shared object (pointed to by a shared_ptr) that is presently > in existence with a reference count of at least 2. > > 2. One thread A creating another shared_ptr to the shared object, > attempting to increment the reference count. > > 3. Another thread B simultaneously destroying a *different* > shared_ptr which is connected to the same underlying shared object, > decrementing the reference count. > > > > Note that the shared object is *not being destroyed* here, just the > reference counts are being manipulated. > > > > Will shared_ptr handle this situation correctly? > > > > Thanks > > > > Kevin > > >

1 0

boost as shared library - linux
by Krzysztof 02 May '11

02 May '11

When I try to use libboost_program_options.so in my program it gives me a message: error while loading shared libraries: libboost_program_options.so.1.46.1: cannot open shared object file: No such file or directory I tried to do the command "ldconfig -n" and substitute to LD_LIBRARY_PATH the dir where it is located - the message remains the same. The boost library directory is in my home folder. How to overcome this? -- Regards Krzysztof J.

1 0

[Test] Compiling libraries for Windows Mobile 6 ARMV4I
by Paul Heil 02 May '11

02 May '11

I would like to be able to use the Boost.Test library for Windows Mobile 6 ARMV4I projects compiled by Visual Studio 2008. I can successfully compile the Boost.Signals, Boost.Thread, Boost.System, and Boost.Random libraries; but Boost.Test uses several functions that aren't available under Windows Mobile even with STLPort 5.2.1. Specifically, these are strerror() and getenv(). The actual compile errors I receive are: /test/impl/cpp_main.ipp(38) : error C2039: 'getenv' : is not a member of '`global namespace'' /test/impl/cpp_main.ipp(38) : error C2873: 'getenv' : symbol cannot be used in a using-declaration /test/impl/cpp_main.ipp(38) : error C2039: 'strerror' : is not a member of '`global namespace'' /test/impl/cpp_main.ipp(38) : error C2873: 'strerror' : symbol cannot be used in a using-declaration /test/impl/execution_monitor.ipp(1306) : error C2039: 'strerror' : is not a member of 'stlpd_std' /test/impl/execution_monitor.ipp(1306) : error C3861: 'strerror': identifier not found /test/impl/cpp_main.ipp(106) : error C2039: 'getenv' : is not a member of 'stlp_std' /test/impl/cpp_main.ipp(106) : error C3861: 'getenv': identifier not found I have seen past postings by other users that have suggested they have supplemented STLPort with these functions (particularly posts relating to VeetoFTC), but I have yet to see an actual example of the code changes necessary to do that. Can anybody suggest the least intrusive way to add this functionality so that I can compile the rest of Boost? Thanks, PaulH

2 2

[phoenix] conditional expression?
by alfC 02 May '11

02 May '11

Hi, Does Phoenix support conditional expression? I know if supports if_ *statements* but it seems that there's is no conditional (expression*. The question what is is the equivalent for "(x>2.)?(1.):(-1.)" in general. For example, I tryed with if_[].else_[] hoping that the statement is used as result but with no luck: #include<boost/phoenix/phoenix.hpp> #include<iostream> using namespace boost::phoenix; using namespace boost::phoenix::arg_names; int main(){ std::clog << (arg1*arg1)(2.) << std::endl; //returns correctly 4 std::clog << (if_(arg1>2.)[val(1.)].else_[val(-1.)])(2.) << std::endl; //syntax error return 0; } The syntax error points that the expression fails to evaluate as a number. Any ideas? -- Thanks, Alfredo usr/include/boost/proto/operators.hpp:79:9: error: forming reference to void if.cpp:8:43: error: no match for ‘operator<<’ in ‘std::clog << boost::phoenix::if_ [with Cond = boost::phoenix::actor<boost::proto::exprns_::basic_expr<boost::proto::tag::greater, boost::proto::argsns_::list2<boost::phoenix::actor<boost ... etc, etc, 100 lines

3 3

[iostreams] Flushing problems with :filtering_ostream
by Bo Jensen 02 May '11

02 May '11

I had some older code which used to work before, but seem to have flushing problems now. The problem is I can not make boost::iostreams::filtering_ostream flush properly. Here is a snippet of my code, which is not compilable in the current form, since I just pulled it out of my own code. Here it goes : /* Helper class for writing gzipped files */ class gzip_sink { public: typedef char char_type; typedef boost::iostreams::sink_tag category; Log &Logger_; /* Logger */ boost::iostreams::filtering_ostream sink_; /* Sink used to push filters */ SlmOfstream outifstr_; /* Write stream to be pushed into sink */ std::wstring name_; /* Store name of attached file */ /* Constructor */ gzip_sink(const wchar_t *filename,Log &logger) SlmExcStartConstructor : Logger_(logger), name_(filename) { /* Open for write */ SlmOpenWriteFile(Logger_, name_.c_str(), outifstr_); /* Push gzip compressor */ sink_.push(boost::iostreams::gzip_compressor()); } SlmExcStopConstructor gzip_sink(const gzip_sink &rhs) SlmExcStartConstructor : Logger_(rhs.Logger_), name_(rhs.name_) { /* Open for write */ SlmOpenWriteFile(Logger_, name_.c_str(), outifstr_); /* Push gzip compressor */ sink_.push(boost::iostreams::gzip_compressor()); } SlmExcStopConstructor void Close() { sink_.reset(); } ~gzip_sink() { sink_.reset(); } /* Write function */ std::streamsize write( const char * s, std::streamsize n ) { /* Push write stream */ /* THIS IS THE HACK !!! WE NEED TO PUSH AND POP EACH TIME OTHERWISE SOME FLUSH IS NOT DONE!! */ sink_.push(outifstr_); sink_.write(s,n); sink_.pop(); return n; } }; I have to push and pop the stream on to the sink each time I write, which should only be done once in the constructor. It is clearly some flushing issue, but I have tried flushing in all places and no luck, any help ? If you think it looks a bit odd, it's because I use it together with boost::iostreams::code_converter<gzip_sink> to be able to use gzip with wide streams.

2 5

[Review] Boost.Convert library, last day
by Edward Diener 02 May '11

02 May '11

This is a reminder that formal review of the Boost.Convert library by Vladimir Batov ends today, May 2. There have only been a few reviews of the library so the library welcomes reviews. If you are one of the people who have been dissatisfied by the perceived limitations of Boost lexical_cast, or have discussed such limitations on Boost mailing lists in the past, please look at the library and give a review of it, as its intent is to offer a greatly enhanced version of lexical_cast as a new library. *************** * Its Purpose * *************** The library builds on the past boost::lexical_cast experience, then takes those conversion-related ideas further to suit better today's applications and programing needs. It still offers simple, minimal interface, familiar conversion behavior and additionally provides: * throwing and non-throwing conversion-failure behavior; * support for the default/fallback value to be returned when conversion fails; * two types of the conversion-failure check - basic/simple and better/safe; * formatting support based on the standard std::streams and std::stream-based manipulators (like std::hex, std::scientific, etc.); * support for different locales; * support for boost::range-compliant char and wchar_t-based containers; * no DefaultConstructibility requirement for the Target/Destination type; * consistent framework to uniformly incorporate any type-to-type conversions, extensibility and additional room to grow. With its present support for string-to-type and type-to-string conversions it is an essential tool for applications making extensive use of configuration or MS-Windows-Registry-style files or having to process/prepare considerable amounts of data in, say, XML, etc. More so, it is easily extendable to accommodate, specialize and uniformly deploy new user-defined type-to-type conversions. ******************* * Where to get it * ******************* The Boost Vault is accesible again after being down for most of the week, so you can get the library from the Boost Vault at http://www.boostpro.com/vault/index.php?action=downloadfile&filename=boost-… You can also get the library, using Subversion, from https://svn.boost.org/svn/boost/sandbox/convert. The HTML documentation is part of the distribution and can be found at libs/convert/index.html in the distribution above. ******************** * Writing a review * ******************** The reviews and all comments should be submitted to the developers list, and the email should have "[convert] Review" at the beginning of the subject line to make sure it's not missed. Please explicitly state in your review whether the library should be accepted. The general review checklist: - What is your evaluation of the design? - What is your evaluation of the implementation? - What is your evaluation of the documentation? - What is your evaluation of the potential usefulness of the library? - Did you try to use the library? With what compiler? Did you have any problems? - How much effort did you put into your evaluation? A glance? A quick reading? In-depth study? - Are you knowledgeable about the problem domain? And finally, every review should answer this question: - Do you think the library should be accepted as a Boost library? Be sure to say this explicitly so that your other comments don't obscure your overall opinion. Edward Diener, Review Manager

1 0

Re: [Boost-users] [Multi-Index] Categorised Container
by Hossein Haeri 02 May '11

02 May '11

Hi Joaquín, > multi_index_conainer< > T, > ordered_non_unique< > composite_key< > T, > member<T,category_t,&T::category>, > member<T,second_type,&T::second> > > > >, > ordered_non_unique<...> > > my_container; > > Now, if you do an equal_range on a category elements will indeed be > sorted by the second member. Right. Here is a better formulated question then: Let's say c1 is an object of type category_t. When I apply equal_range for c1, will I get my desired range in log(n) time (where n == my_container.size())? Or, will I get it in O(1) for the elements are already stored in a category-by-category manner? TIA, --Hossein

2 1

Fwd: Re: [Review] Boost.Convert library, last 3 days
by Edward Diener 01 May '11

01 May '11

I am forwarding this message sent directly to me as the review manager of the Boost Convert library review. I have removed the forwarding information to protect privacy. Hi Edward, I am by no means an expert in library design, so perhaps I have a few comments from the user-only perpective: > - What is your evaluation of the design? I think the design is appropiate in terms of purely extending lexical_cast in terms of fallback values and optional exceptions. However, I was expecting more from a library that extends on lexical_cast, for example in terms of one of the goals of the library, which is compact syntax. The feature I was expecting was conversion based on default casting. In the library it seems that the source type can be omitted (just as in lexical_cast). I would be a desired feature that also destination types could be omitted. This could be possible using cast operator such that Destination d = auto_convert::from(s); // or convert<>::from(s) // default template parameter interpreted as potential auto conversion based on destination type. Instead of the redundant Destination d = convert<Destination>::from(s); I think the requested feature (unless it is not possible, in that case I am sorry about my request) is to be expected from the library. > - What is your evaluation of the implementation? I have no comments on it since I wouldn't be able to judge on implementation used. > - What is your evaluation of the documentation? The documentation is understandable and appropriate. > - What is your evaluation of the potential usefulness of the > library? I would use the library just based in considering the natural and explicit syntax introduced. Even more if the requested feature above is introduced. > - Did you try to use the library? With what compiler? Did you > have any problems? no > - How much effort did you put into your evaluation? A glance? A > quick reading? In-depth study? A glance > - Are you knowledgeable about the problem domain? no. Only from the point of view convenience. > - Do you think the library should be accepted as a Boost library? yes, I think it should be accepted. I hope this quick review is useful to make a decision. Although the validity and weight of my vote can be objected because I am not really an active player of the Boost community. Thank you, Alfredo Correa

1 0

[Spirit] Semantic actions (binary parser) - user error or bug?
by Vitaly Budovski 01 May '11

01 May '11

Hi, I've come across some behaviour with semantic actions that seems strange. I'm not sure if this is intended or if it is indeed a bug. Please see the attached test case.

4 5

Boost deserialization of a derived class problem
by adrien.barral＠robopec.com 30 Apr '11

30 Apr '11

Hello, I have got a problem serializing data via TCP/IP. My application is highly inspired of the asio example on serialization. I have a server (my class "Recepteur"), and a client (my class "Logger"). I try to serialize two kine of data : - Loggable - HumanReadable. Loggable is a public base of HumanReadable, later I will have other Loggable class to serialize (OneDPlotable etc...). Look at my two serializable classes : [code] // file Loggable.hpp #ifndef LOGGABLE_HPP #define LOGGABLE_HPP #include #include #include #include #include class Loggable { protected : friend class boost::serialization::access; template void serialize(Archive& _archive, const unsigned int _version) { std::cout

1 0

[Multi-Index] Categorised Container
by Hossein Haeri 30 Apr '11

30 Apr '11

Dear all, I have an application where something we call a "categorised container" with colleagues comes very hand. We do currently have a working version and I am looking to see whether it's possible to do some refactoring on that using Multi-Index. Here is an exemplified explanation of what constitutes a decision-making point for me. I would be pleased if anyone here could tell me whether this facility is available under Multi-Index: multi_index_container< T hashed_non_unique<...>, ordered_non_unique<...>> my_container; Now, is there anyway I can get all elements of my_container which have a certain hash and sorted according to the second index? In our own local terminology, we call that retrieving all the elements of a given CATEGORY in a sorted manner. (Sorry I'm not still fluent in the MultiIndex syntax to make that more formal with actual code.) TIA,--Hossein

3 2

[Review] Boost.Convert library, last 3 days
by Edward Diener 30 Apr '11

30 Apr '11

This is a reminder that formal review of the Boost.Convert library by Vladimir Batov has 3 more days and ends on May 2. There have only been a small number of reviews and this is the last weekend to review the library. I am hoping that programmers who may have been busy during the week in their work will find time this weekend for a review. The review period ends on Monday at midnight, so please get your review in before the review period ends. If you intend to do a review of the library but feel you would be late in doing so, please contact me about it. *************** * Its Purpose * *************** The library builds on the past boost::lexical_cast experience, then takes those conversion-related ideas further to suit better today's applications and programing needs. It still offers simple, minimal interface, familiar conversion behavior and additionally provides: * throwing and non-throwing conversion-failure behavior; * support for the default/fallback value to be returned when conversion fails; * two types of the conversion-failure check - basic/simple and better/safe; * formatting support based on the standard std::streams and std::stream-based manipulators (like std::hex, std::scientific, etc.); * support for different locales; * support for boost::range-compliant char and wchar_t-based containers; * no DefaultConstructibility requirement for the Target/Destination type; * consistent framework to uniformly incorporate any type-to-type conversions, extensibility and additional room to grow. With its present support for string-to-type and type-to-string conversions it is an essential tool for applications making extensive use of configuration or MS-Windows-Registry-style files or having to process/prepare considerable amounts of data in, say, XML, etc. More so, it is easily extendable to accommodate, specialize and uniformly deploy new user-defined type-to-type conversions. ******************* * Where to get it * ******************* You can get the library, using Subversion, from https://svn.boost.org/svn/boost/sandbox/convert. The HTML documentation is part of the distribution and can be found at libs/convert/index.html in the distribution above. ******************** * Writing a review * ******************** The reviews and all comments should be submitted to the developers list, and the email should have "[convert] Review" at the beginning of the subject line to make sure it's not missed. Please explicitly state in your review whether the library should be accepted. The general review checklist: - What is your evaluation of the design? - What is your evaluation of the implementation? - What is your evaluation of the documentation? - What is your evaluation of the potential usefulness of the library? - Did you try to use the library? With what compiler? Did you have any problems? - How much effort did you put into your evaluation? A glance? A quick reading? In-depth study? - Are you knowledgeable about the problem domain? And finally, every review should answer this question: - Do you think the library should be accepted as a Boost library? Be sure to say this explicitly so that your other comments don't obscure your overall opinion. Edward Diener, Review Manager

1 0

[units][phoenix] static physical constants don't play well with phoenix?
by alfC 30 Apr '11

30 Apr '11

Hi, This post is just to point out a deficiency of either boost.phoenix or boost.units and see if there are workarounds. Phoenix 3 now plays well with Boos.units quantities out of the box but for some reason it does not play well with boost.units static physical constants. Like in the following example. This is because the constant 'hbar' is not a normal quantity type but a boost::units::constant< ... hbar_t > type. do you think physical constants could be defined as normal quantities in the future so it can play well with phoenix? #include<iostream> #include<boost/units/systems/si/io.hpp> #include<boost/units/systems/si/codata_constants.hpp> #include<boost/phoenix/phoenix.hpp> int main(){ using namespace boost::units; using namespace boost::phoenix; using namespace boost::phoenix::arg_names; quantity<si::action> const hbarvar = boost::units::si::constants::codata::hbar; std::clog << (arg1 * hbarvar)(5.) << std::endl; // works std::clog << (arg1 * boost::units::si::constants::codata::hbar)(5.) << std::endl; //doesn't work return 0; }

1 0

Re: [Boost-users] [Multi-Index] Categorised Container
by Hossein Haeri 29 Apr '11

29 Apr '11

Hi Marsh, multi_index_container < T hashed_non_unique<...>, ordered_non_unique<...> > my_container; > So are you expecting hash collisions and doing something useful with > them? This is interesting, can you explain more about the problem and > why this is useful? Sure. Take T = pair<TimeStamp, shared_ptr<B> > where B is some base class for a hierarchy in my application. B has a pure virtual member function called id() which returns the same string for any instance of a class D, where D is derived from B. I will need to have all the instances the dynamic type of the second element of which is of D, but sorted according to TimeStamp's operator <. My idea was to get the first index in my_container to be this id() member function and the second index to be operator < (const TimeStamp&, const TimeStamp&). Is that doable in Multi-Index? > I think you're on the right track by calling them 'categories' and > suggest you not try to overload it with the hash value used to index the > collection (even if they happen to be the same). Sorry but I'm not sure I can follow your English here. Would you mind paraphrasing it if the issue is still important? > If you didn't want to store redundant data in T, you could define a key > extractor which runs your preferred hash function to calculate the > category on the fly. This much redundancy is not a matter for me. But, I'd rather not calculate the category on-the-fly for this container is going to be heavily accessed (and also populated/garbage-collected) for every D which derives from B. Does that make it easier to answer my questions? TIA, --Hossein

1 0

ning Boost Regression Tests
by Ted Byers 29 Apr '11

29 Apr '11

I have been using boost for quite a while (for years I used boost without running any tests other than those I wrote myself), and am now looking at running the regression tests. I found the statement: "To run every library's regression tests, run bjam from Boost's /status directory." What I haven't found is how to view the results of that testing on my own machine. I don't know if that is a catastrophic failure of the test script, or if I just didn't invoke bjam correctly. I ran bjam (in cygwin's bash, on Windows 7), without any arguments, in boost's status directory, and it executed a lot of tests. Obviously, that produced a lot more output than can be copied here, but the last few lines of output (after many many hundreds of lines of output) were: 1 [main] bjam 9280 fhandler_dev_zero::fixup_mmap_after_fork: requested 0x772B0000 != 0x0 mem alloc base 0x772B0000, state 0x1000, size 8192, Win32 error 487 263 [main] bjam 9280 C:\cygwin\home\Ted\boost_1_46_1\bjam.exe: *** fatal error - recreate_mmaps_after_fork_failed Stack trace: Frame Function Args 0028E5D0 6102796B (0028E5D0, 00000000, 00000000, 74C98458) 0028E8C0 6102796B (6117EC60, 00008000, 00000000, 61180977) 0028F8F0 61004F1B (61180DAC, 00000000, 61180DA6, 0041A000) 0028F930 610061B1 (00000060, 02000000, 6116A724, 6116A720) 0028F960 610064C0 (00000000, 00000002, 00000000, 75053480) 0028FA20 6106FC15 (61000000, 00000001, 0028FD24, 00000001) 0028FA40 77169960 (6106F960, 61000000, 00000001, 0028FD24) 0028FB34 7716D8C9 (0028FD24, 7EFDD000, 7EFDE000, 7723206C) 0028FCB0 771837EE (0028FD24, 77130000, 77AB63F9, 00000000) 0028FD00 77182187 (0028FD24, 77130000, 00000000, 00000000) 0028FD10 77169E89 (0028FD24, 77130000, 00000000, 0001002F) End of stack trace 0 [main] bjam 8628 fork: child -1 - died waiting for longjmp before initialization, retry 0, exit code 0x100, errno 11 vfork: Resource temporarily unavailable Obviously the last test executed failed, but I found no summary of the results, nothing to indicate whether or not most of the built library was built correctly. Did the regression test script abort without producing the summary, or do I need to pass some argument to tell it to display a summary (I had in mind the kind of output you get when building gcc, and you execute 'make check' before executing 'make install', as one doesn't want to install a library or anything else that didn't get built correctly)? Thanks Ted

2 2

Boost in an MFC app on WinNT4
by tangleman＠secules.net 29 Apr '11

29 Apr '11

I am trying to modify a current MFC application running on Windows NT4 to use boost libraries for ethernet communcications (originally, it used CommX for serial) and general increased performance, effieciency, and clarity of code. I started out modifying it in VS2010, but I found that I was unable to produce a valid NT4 app with VS2010. I googled the issue and found that VS2005 was the last version able to create an NT4 app. I got everything to compile in VS2005, but found that I had to change certain #define statements in stdafx.h in order to get the app to run on NT4. However, even after doing that, when I execute the program on NT4, it immediately quits with no error. It doesn't show anything. I even checked the Task Manager processes while executing it. It is like it never executed at all. I got desperate and tried compiling the app in VS2003, but I got so many errors that I gave that up. My last attempt was VC6, but again, I got 262 errors and started wondering whether it was even possible for boost to run in VB6 or on an NT4 machine. This is a very tight-schedule project, so any (prompt) help would be very appreciated! Thank you!

3 2

[boost::fusion] map or set uniqueness guarantees
by Ovanes Markarian 29 Apr '11

29 Apr '11

Hello *, I am wondering if I should expect some uniqueness guarantees on contained element from fusion set or map containers? In this example: struct print_value { template<class T> void operator()(T& v)const { cout << typeid(T).name() << " has value: " << v << endl; } }; boost::fusion::set<int, int, int> s; boost::fusion::for_each(s, print_value()); I get 3 lines - one for each int I put in. I would expect only one line. What are my expectations for at_key? Which int would it return? Which type is going to be returned after I inserted the same type again? May be the better question would be, is the new type inserted as the head element or at the end of the underlying list? Many thanks for the anserws? With Kind Regards, Ovanes

1 0

Asio handle_read this=0x0 in http server3 example
by Matthijs Möhlmann 29 Apr '11

29 Apr '11

Hello, When compiling example http/server3 then I get a segmentation fault. GDB gives the following trace: #0 0x000000000040e2a2 in http::server3::request_parser::parse<char*> ( this=0x2078, req=..., begin=0x48 <Address 0x48 out of bounds>, end=0x273 <Address 0x273 out of bounds>) at request_parser.hpp:42 #1 0x000000000040743e in http::server3::connection::handle_read (this=0x0, e=..., bytes_transferred=555) at connection.cpp:48 As you see the 'this=0x0' in #1 is not correct I think. I think this has to do with the internals in Asio but I'm not sure. I'm using g++ 4.6 in combination with boost trunk. I'm not able to reproduce this with boost 1.46.1. Regards, Matthijs Möhlmann

1 0

is there a function can get two file path relative path?
by k 29 Apr '11

29 Apr '11

the boost filesystem library provide relative_ path() but it only give relate root directory. or is it other function can do it?

2 1

[Boost][enable_if] SFINAE and decltype
by Germán Diago 28 Apr '11

28 Apr '11

Hi all, I'm having a hard time trying to figure out how to make a code like this to improve error messages through static_assert. I want that when the function instantiation fails, the other function's static assert fails. I don't know how to do it. I began with something like this: template <class Range, class T> Range find(Range && haystack, const T & elem) { static_assert(std::is_convertible<decltype (elem == haystack.front())>::value, "elem and haystack must be equality comparable"); while (!haystack.empty() && !(haystack.front() == elem)) haystack.popFront(); return haystack; } But that didn't work since when decltype is applied to an invalid expression, SFINAE doesn't trigger in the function body. So now I'm trying an alternative like this, but I don't know how to express it: template <class Range, class T> Range find(Range && haystack, const T & elem, decltype (elem == haystack.front()) * = 0) { while (!haystack.empty() && !(haystack.front() == elem)) haystack.popFront(); return haystack; } template <class Range, class T> Range find(Range && haystack, const T & elem, !decltype (elem == haystack.front()) * = 0) { static_assert(..., "haystack.front() and elem must be equality comparable"); return haystack; } I can't figure out how to make it work. I also tried with boost::enable_if but the problem is that a bool expression doesn't have a ::value member. Could anyone suggest me THE SIMPLEST and most readable way to make this work? Thanks in advance. P.S.: By the way, I also tried structs and partial specialization, but no luck.

2 2

Read/write with gzip
by Anders Knudby 28 Apr '11

28 Apr '11

Hello, I previously managed to write data to a gzipped file using the following code structure (it's an example of course, the "memblock" is actually filled with real data and the real filename is specified elsewhere). #include <boost/iostreams/filtering_streambuf.hpp> #include <boost/iostreams/filtering_stream.hpp> #include <boost/iostreams/copy.hpp> #include <boost/iostreams/filter/gzip.hpp> #include <boost/iostreams/device/file_descriptor.hpp> #include <boost/filesystem.hpp> namespace io = boost::iostreams; int size = 1000; char* memblock = new char [size]; io::filtering_ostream out; out.push(io::gzip_compressor()); out.push(io::file_descriptor_sink("c:/outfile.gz")); out.write (memblock, size); That works fine. I would now like to read these data again. I imagined I could do something like the following (using the same #include statements etc. as above): int size = 1000; char* memblock = new char [size]; io::filtering_istream in; in.push(io::gzip_decompressor()); in.push(io::file_descriptor_source("c:/outfile.gz")); in.read (memblock, size); That complies and runs, but it doesn't give me the real data back. What am I doing wrong? Anders

4 7

boost build problem?...
by Krzysztof 28 Apr '11

28 Apr '11

I wrote a few lines of code just for start with a library of boost (program_options). Before I did it I had built boost library as prescribed and *.a and *.so files were created in my /home/krzysztof/Documents/Projects/boost/lib directory. However, compilation fails because of undefined references of some external symbols which belongs to program_options library. Could anyone help me with it? Below, there is the code and g++ output. (Linux 64b, Qt Creator and Code::Blocks used as IDE) #include "boost/program_options.hpp" using namespace boost::program_options; using namespace std; int main(int argc, char *argv[]) { options_description opd; opd.add_options() ("help","Display help") ("input_file,if",value<string>(),"Specifies input file") ("fet_file",value<string>(),"definitions file") ("output_file,of",value<string>(),"file after edition"); } g++ -o fetxedit main.o -L/usr/lib -L/home/krzysztof/Documents/Projects/boost/lib -lQtCore -lpthread main.o: In function `main': /home/krzysztof/Documents/Projects/fetxedit-build-desktop/../fetxedit/main.cpp:8: undefined reference to `boost::program_options::options_description::m_default_line_length' /home/krzysztof/Documents/Projects/fetxedit-build-desktop/../fetxedit/main.cpp:8: undefined reference to `boost::program_options::options_description::m_default_line_length' /home/krzysztof/Documents/Projects/fetxedit-build-desktop/../fetxedit/main.cpp:8: undefined reference to `boost::program_options::options_description::options_description(unsigned int, unsigned int)' /home/krzysztof/Documents/Projects/fetxedit-build-desktop/../fetxedit/main.cpp:14: undefined reference to `boost::program_options::options_description::add_options()' /home/krzysztof/Documents/Projects/fetxedit-build-desktop/../fetxedit/main.cpp:14: undefined reference to `boost::program_options::options_description_easy_init::operator()(char const*, char const*)' ... and so on -- Regards Krzysztof Jasieński

2 1

boost::bind
by simone pilozzi 28 Apr '11

28 Apr '11

Dear Users, I am using bind to calculate the transform of an initial vector in its element by element power (to a given exponent). I used this syntax std::transform(values.begin(), values.end(), values.begin(),boost::bind(pow,_1,exponent)); but the compiler complains for with error C2780 : expects 2 arguments 3 provided... Do you have any solution/explanation? Thanks in adavnce!

2 4

[iostream] gzip filter buggy on windows
by Bo Jensen 28 Apr '11

28 Apr '11

In my quest of making gzip work for my application on windows, I found the following bug : int main(int argc, char *argv[]) { std::ofstream strm("test.gz"); boost::iostreams::filtering_ostream sink; /* Push gzip compressor */ sink.push(boost::iostreams::gzip_compressor()); /* Push file writer */ sink.push(strm); #if 0 /* This does not work */ sink <<" OBJ N"<< std::endl; #else /* This works */ sink <<" OBJ N"<< std::endl; #endif } The first string results in a invalid gzip file i.e invalid compressed data--length error. I am aware this might be a bug in zib lib. It works on without any problems on linux. Does anyone use gzip on windows on a regular basis ?

2 4

Request for conditional index feature in boost multi-index
by Gokulakannan Somasundaram 28 Apr '11

28 Apr '11

Hi, First of all thanks for the great library, which had reduced our development time by a lot. In boost multi-index, we currently have almost all the indexing concepts integrated. Hence i am requesting for a new feature called conditional indexes, where in an index/view inside multi-index will only index a portion of the entries inside the entire multi-index(Say something that satisfies a functor). This is in resemblance to the partial indexes in Postgresql. I feel that would provide elegant solutions to lot of issues without forming round about solutions. For example i have a requirement where in there are two sets of data with the same search criteria, but different eviction criteria Say i have two sets of data with same search condition, but one set needs a MRU(Most Recently Used) list of 100 and the other set requires a MRU of 200. Say the entry is like this class Student { int student_no; char sex; std::string address; }; The search criteria is student_no, but for sex='m', we need MRU of 200 and for sex='f', we need a MRU of 100. Now i have a solution where in i introduce a new ordered index to maintain ordering. For example the IndexSpecifierList will be something like this typedef multi_index_container< Student, indexed_by< ordered_unique< member<Student, int, &Student::student_no> >, ordered_unique< composite_key< member<Student, char, &Student::sex>, member<Student, int, &Student::sex_specific_student_counter> > > > > student_set Here i have to introduce a new counter called sex_specific_student_counter, and during the eviction time, i need a logn search. Insertion also involves a extra logn complexity. Now instead if i have two sequenced indexes (like mentioned in the MRU example) in the place of second ordered index, but both with a condition on indexing only a few items, this would have looked more elegant and with better performance. Thanks in advance, Gokul.

3 3

stop thread before socket destructor
by Gianni Ambrosio 28 Apr '11

28 Apr '11

Hi All, I modified a boost example to implement a tcp server that's able to accept clients on a thread. Unfortunately I get the following message: "The I/O operation has been aborted because of either a thread exit or an application request". I realized that the problem is related to the fact the server destructor is called before the handle_accept occurs. Now, how could I stop the thread in the server destructor so that the handle_accept is not called anymore? Here is the test code I used: typedef boost::shared_ptr<boost::asio::ip::tcp::socket> tcp_socket_ptr; class stream_handler { public: stream_handler() : acceptor_(io_service_) { boost::asio::ip::tcp::endpoint endpoint = boost::asio::ip::tcp::endpoint(boost::asio::ip::tcp::v4(), 32123); acceptor_.open(endpoint.protocol()); acceptor_.set_option(boost::asio::ip::tcp::acceptor::reuse_address(true)); acceptor_.bind(endpoint); acceptor_.listen(); new_socket = tcp_socket_ptr(new boost::asio::ip::tcp::socket(io_service_)); acceptor_.async_accept(*new_socket, boost::bind(&stream_handler::handle_accept, this, boost::asio::placeholders::error)); thread_ = boost::thread(boost::bind(&boost::asio::io_service::run, &io_service_)); } ~stream_handler() { acceptor_.close(); } void handle_accept(const boost::system::error_code& err) { if (err) { std::cout << err.message() << std::endl; } else { new_socket = tcp_socket_ptr(new boost::asio::ip::tcp::socket(io_service_)); acceptor_.async_accept(*new_socket, boost::bind(&stream_handler::handle_accept, this, boost::asio::placeholders::error)); } } private: io_service io_service_; tcp::acceptor acceptor_; boost::thread thread_; tcp_socket_ptr new_socket; }; int _tmain(int argc, _TCHAR* argv[]) { stream_handler sh; return 0; } Regards Gianni

2 2

mpi skeleton/content does not work with stl complex
by Tobias Stollenwerk 27 Apr '11

27 Apr '11

Hello all, It seems that sending stl complex data via the skeleton/content mechanism does not work. Here is a simple code that shows th problem: #include <boost/mpi.hpp> #include <boost/serialization/complex.hpp> #include <boost/serialization/vector.hpp> #include <iostream> namespace mpi = boost::mpi; using namespace std; int main(int argc, char* argv[]) { mpi::environment env(argc, argv); mpi::communicator world; int NN=world.size(); int myid=world.rank(); if (myid == 0) { int N = 10; vector <complex<double> > l(N); for (int p=1; p!=NN; p++) { int taskid=1; world.send(p, 0, taskid); world.send(p, 1, mpi::skeleton(l)); } mpi::content c = mpi::get_content(l); for (int n = 0; n!=l.size() ; n++) { l[n]=complex<double>(1.0,1.0); } for (int p=1; p!=NN; p++) { world.send(p, 1, c); } } else { vector <complex<double> > l; mpi::content c; world.recv(0, 1, mpi::skeleton(l)); c = mpi::get_content(l); world.recv(0, 1, c); for (int n=0; n!=l.size(); n++) { cout << l[n] << " "; } cout << endl; } } The entries of the vector l are not (1.0,1.0) but they seem to be uninitialized. This only occurs if one uses complex data types AND the skeleton/content mechanism. Does anyone know, if this is a build in problem, or if I am doing something wrong? Thanks in advance, Tobias

1 0

questions on the BFS example of Parallel BGL
by Da Zheng 27 Apr '11

27 Apr '11

Hello, I'm testing PBGL on my cluster. I have 4 nodes and each node has a dual-core processor. I changed the BFS example a little bit. The code generates a graph with RMAT, and it does BFS 10 times, starting from different nodes. I first run the program in a sequential mode, and the graph has 1,000,000 vertices. Then I run it in the cluster. The result is shown below, but I don't see much speedup. The running time is reduced from 14 seconds to 9.5 seconds. Can the result be right? I see very good speedup in the PBGL document. Another problem is that BFS starting from some vertices only needs very little time. I can imagine that it traverses only very few vertices. Is there any way I can count how many vertices are visited from the specific root vertex? My code is also shown below. Thanks, Da [zhengda@amdahl36 pbgl]$ ./bfs ~/bfs.out 1000000 generate graph takes 162.336 seconds to generate graph start BFS random v: 1395111516 takes 14.3496 seconds for BFS search random v: 488302563 takes 0.029389 seconds for BFS search random v: 72509101 takes 14.1103 seconds for BFS search random v: 1835953408 takes 0.028544 seconds for BFS search random v: 951813172 takes 14.7907 seconds for BFS search random v: 1708481694 takes 14.2254 seconds for BFS search random v: 305355333 takes 14.1587 seconds for BFS search random v: 1627009882 takes 0.02749 seconds for BFS search random v: 1769559585 takes 0.027495 seconds for BFS search random v: 101184463 takes 0.02756 seconds for BFS search [zhengda@amdahl36 pbgl]$ mpirun -np 8 --hostfile ~/.mpi_hostfile ./bfs ~/bfs.out 1000000 generate graph generate graph generate graph generate graph generate graph generate graph generate graph generate graph takes 136.623 seconds to generate graph start BFS random v: 1395111516 takes 9.6362 seconds for BFS search random v: 488302563 takes 0.017129 seconds for BFS search random v: 72509101 takes 9.49864 seconds for BFS search random v: 1835953408 takes 0.031777 seconds for BFS search random v: 951813172 takes 9.53063 seconds for BFS search random v: 1708481694 takes 9.45427 seconds for BFS search random v: 305355333 takes 9.58886 seconds for BFS search random v: 1627009882 takes 0.018723 seconds for BFS search random v: 1769559585 takes 0.020844 seconds for BFS search random v: 101184463 takes 0.016004 seconds for BFS search boost::minstd_rand gen; std::cout << "generate graph" << std::endl; struct timeval start_time; struct timeval end_time; gettimeofday (&start_time, NULL); Graph g(RMATGen(gen, nvertices, 16 * nvertices, 0.57, 0.19, 0.19, 0.05), RMATGen(), nvertices); gettimeofday (&end_time, NULL); if (process_id(process_group(g)) == 0) { std::cout << "takes " << (end_time.tv_sec - start_time.tv_sec) + ((double) end_time.tv_usec - start_time.tv_usec) / 1000000 << " seconds to generate graph" << std::endl; std::cout << "start BFS" << std::endl; } for (int i = 0; i < 10; i++) { int rv = gen(); if (process_id(process_group(g)) == 0) { std::cout << "random v: " << rv << std::endl; } graph_traits<Graph>::vertex_descriptor start = vertex(rv % nvertices, g); // Compute BFS levels from vertex 0 property_map<Graph, vertex_distance_t>::type distance = get(vertex_distance, g); put(distance, start, 0); gettimeofday (&start_time, NULL); breadth_first_search (g, start, visitor(make_bfs_visitor(record_distances(distance, on_tree_edge())))); gettimeofday (&end_time, NULL); if (process_id(process_group(g)) == 0) { std::cout << "takes " << (end_time.tv_sec - start_time.tv_sec) + ((double) end_time.tv_usec - start_time.tv_usec) / 1000000 << " seconds for BFS search" << std::endl; } }

1 0

converting bison to boost::spirit
by Littlefield, Tyler 27 Apr '11

27 Apr '11

Hello all: I have a server I am trying to rewrite, but it uses bison to generate the parser. I know near to nothing about either of these, and the examples kind of confuse me. Is there a way that I can somehow convert the bison to boost::spirit for parsing? Also: I'm kind of curious how one would go about making a parser with boost::spirit. So I define my tokens, (lets say I want to make a calculator as it's pretty simple). Then I would need to generate a tree, correct? My first thought was to just use the [] operator to pass in a function to receive the tokens, then push it to a vector, but that seems like it would be really slow when it comes to parsing the tokens. I also know of an abstract syntax tree, but I'm kind of confused how that works and how spirit would work with it. The calculator overloaded a lot of operators, but I sort of got lost down the line in terms of what was supposed to be happening. Anyone willing to explain this would be greatly appreciated. -- Thanks, Ty

2 1

[units][uBlas] Problems using matrix operations with dimension constraints
by Ryan 27 Apr '11

27 Apr '11

I'm currently trying to use the uBlas library with the units library and I'm running into some problems. The matrix and vector classes I've instantiated with unit constraints don't seem to play well with uBlas methods. Has anyone found ways of using the two libraries together? I've included some lines of what I'm trying to do and would really appreciate some help. Ryan using namespace boost::numeric::ublas; using namespace boost::units; //Works fine. This is expected though since dimensionless quantity can accept //regular built in types. bounded_matrix<quantity<si::dimensionless, 3, 3> rotate = identity_matrix<double>(3); //How do I initialize a vector of dimension lengths with something like the zero_matrix? bounded_vector<quantity<si::length>, 3> diff = zero_matrix<double>(3) * si::meter; //??? //Lets pretend the diff variable has values in it. This line doesn't compile. //The compiler is saying it can't deduce the template argument. //What can I do to help the compiler out so I can get the correct result? bounded_vector<quantity<si::length>, 3> result = prod(rotate, diff);

2 3

[Thread] Solution to conflict with MFC?
by Christian Larsen 27 Apr '11

27 Apr '11

Hello, I know this has been asked before, but I could not find a good solution. I'm experiencing problems with an application using both MFC and Boost.Thread. Specifically the assertion ASSERT(AfxGetModuleState() != AfxGetAppModuleState()); in AfxCoreInitModule() fails. I link Boost.Thread statically into a DLL in my project. From previous posts to this list and elsewhere, e.g. http://lists.boost.org/boost-users/2009/04/46929.php, I see that the issue is caused by setting _pRawDllMain in tss_pe.cpp. Commenting this line does "solve" the problem, but I would rather not have to hack the Boost source code this way. It seems there is a patch somewhere for this problem (referred to in http://lists.boost.org/boost-users/2009/04/47015.php) but it appears it never made it into the source. Is there any chance of this issue being solved in a better way? Best regards, Christian Larsen

4 5

boost build problem?...
by Krzysztof 27 Apr '11

27 Apr '11

I wrote a few lines of code just for start with a library of boost (program_options). Before I did it I had built boost library as prescribed and *.a and *.so files were created in my /home/krzysztof/Documents/Projects/boost/lib directory. However, compilation fails because of undefined references of some external symbols which belongs to program_options library. Could anyone help me with it? Below, there is the code and g++ output. (Linux 64b, Qt Creator and Code::Blocks used as IDE) #include "boost/program_options.hpp" using namespace boost::program_options; using namespace std; int main(int argc, char *argv[]) { options_description opd; opd.add_options() ("help","Display help") ("input_file,if",value<string>(),"Specifies input file") ("fet_file",value<string>(),"definitions file") ("output_file,of",value<string>(),"file after edition"); } g++ -o fetxedit main.o -L/usr/lib -L/home/krzysztof/Documents/Projects/boost/lib -lQtCore -lpthread main.o: In function `main': /home/krzysztof/Documents/Projects/fetxedit-build-desktop/../fetxedit/main.cpp:8: undefined reference to `boost::program_options::options_description::m_default_line_length' /home/krzysztof/Documents/Projects/fetxedit-build-desktop/../fetxedit/main.cpp:8: undefined reference to `boost::program_options::options_description::m_default_line_length' /home/krzysztof/Documents/Projects/fetxedit-build-desktop/../fetxedit/main.cpp:8: undefined reference to `boost::program_options::options_description::options_description(unsigned int, unsigned int)' /home/krzysztof/Documents/Projects/fetxedit-build-desktop/../fetxedit/main.cpp:14: undefined reference to `boost::program_options::options_description::add_options()' /home/krzysztof/Documents/Projects/fetxedit-build-desktop/../fetxedit/main.cpp:14: undefined reference to `boost::program_options::options_description_easy_init::operator()(char const*, char const*)' ... and so on -- Regards Krzysztof Jasieński

2 1

Re: [Boost-users] Boost-users Digest, Vol 2650, Issue 4
by Robert Lupton the Good 27 Apr '11

27 Apr '11

> Sorry for the slow response. I simply defined the following: > > typedef boost::gil::pixel<float, boost::gil::gray_layout_t> gray_float_pixel_t; > typedef boost::gil::image<gray_float_pixel_t, false> gray_float_image_t; > typedef gray_float_image_t::view_t gray_float_view_t; > typedef gray_float_image_t::const_view_t gray_float_const_view_t; > > As Dr. Bourdev advised, I do not use with these typedefs any of the color conversion operations offered by GIL. Apparently those rely on the scaling properties of the float wrapper type. > > So far, all I have done with these typedefs is ... > > 1. construct views using interleaved_view and pixel data pointed to by [const] float*, as illustrated near the beginning of the GIL tutorial > 2. create x- and y- iterators for the views to apply row- and column-oriented algorithms? Sorry for _my_ slow response (I was in Japan). I don't care about colour conversions, but I think that these types will convert to integral types with scaling unless you add more boilerplate. I added the following -- but it may not all have been needed; this was quite a while ago and I was new to gil. I also have some remaining conversion problems where I have to rewrite expressions to make them compile, but this may be something else. Anyway, I'd really like to see native gil support for a complete set of NON-scaling types. namespace boost { namespace gil { /* * Define a type that's a pure float, without scaling into [0, 1] */ typedef float bits32f_noscale; GIL_DEFINE_BASE_TYPEDEFS(32f_noscale, gray) GIL_DEFINE_ALL_TYPEDEFS_INTERNAL(32f_noscale, dev2n, devicen_t<2>, devicen_layout_t<2>) template<> struct channel_multiplier<bits32f_noscale> : public std::binary_function<bits32f_noscale,bits32f_noscale,bits32f_noscale> { bits32f_noscale operator()(bits32f_noscale a, bits32f_noscale b) const { return a*b; } }; template <typename DstChannelV> struct channel_converter<bits32f_noscale, DstChannelV> : public std::unary_function<bits32f_noscale,DstChannelV> { DstChannelV operator()(bits32f_noscale x) const { return DstChannelV(x + 0.5f); } }; template <typename SrcChannelV> struct channel_converter<SrcChannelV,bits32f_noscale> : public std::unary_function<SrcChannelV,bits32f_noscale> { bits32f_noscale operator()(SrcChannelV x) const { return bits32f_noscale(x); } }; // // Totally specialised templates to resolve ambiguities // #define CONVERT_NOOP(T1, T2) \ template <> \ struct channel_converter<T1, T2> : public std::unary_function<T1, T2> { \ T2 operator()(T1 x) const { return static_cast<T2>(x); } \ }; \ \ template <> \ struct channel_converter<T2, T1> : public std::unary_function<T2, T1> { \ T1 operator()(T2 x) const { return static_cast<T1>(x); } \ } CONVERT_NOOP(unsigned char, short); CONVERT_NOOP(unsigned char, unsigned short); CONVERT_NOOP(unsigned char, int); CONVERT_NOOP(unsigned short, short); CONVERT_NOOP(unsigned short, int); CONVERT_NOOP(short, int); #undef CONVERT_NOOP

3 2

[Units] Extension for units of data (and data rate)
by David, Robert (NSN - HU/Budapest) 27 Apr '11

27 Apr '11

Hello, I have created an extension to Boost.Units to support data sizes (bit / byte / kilobit / etc.) and data rates (bit per sec / Byte per sec / kbps / etc.), trying to mimic the logic behind the SI, GCS systems, so it might be added to the library after some review and fixes. Please see the attachment for the code (8 kB zip). Before that, a problem has to be fixed in my implementation: If I define the unit byte_per_second using make_scaled_unit and bit_per_second, I get a mile-long template error about not being able to figure out the prefix when I try to convert a quantity using that unit to string. I did a workaround for this problem using decltype(), but that requires including data_size.hpp and time.hpp to data_rate.hpp (although I don't know if it is a problem), and also, it only works on compilers supporting c++0x. You can see this workaround starting at line 40 in boostex/units/systems/data/data_rate.hpp. If I comment back line 41, and comment out line 44-45, the following will not compile: quantity<byte_per_sec_unit> speed(5.0 * byte_per_sec_unit); std::cout << speed << std::endl; however, auto speed = 5.0 * byte_per_sec_unit; std::cout << speed << std::endl; does (although only on compilers supporting c++0x auto), but prints out "second^-1 byte" instead of "byte/second" - effectively it does not call the name_string specialization for byte_per_sec in boostex/units/systems/data/io.hpp. All of this works correctly with the decltype() workaround. So my question is: can I define byte_per_sec (kilobyte_per_sec, megabyte_per_sec, etc.) in a c++03-compatible way? Br, Robert

1 0

Boost::serialization problem when use archive in a loop
by mikesam 27 Apr '11

27 Apr '11

I have a problem with serialization, and I do not know how to solve it. Time to serialize the class a lot of growing when I do it in a loop. The first iteration takes about 40 ms, for example, but any other takes about 400 ms. class NODE { public: NODE(){} NODE* add(NODE* node) { children.push_back(node); return node; } ~NODE() { for(vector<NODE*>::const_iterator it = children.begin(); it != children.end(); ++it) { delete *it; } } vector<NODE*> children; private: friend class boost::serialization::access; template <typename Archive> void serialize(Archive &ar, const unsigned int version) { ar & children; } }; void save(NODE* main) { std::ofstream stream("test", std::ios::binary); boost::archive::binary_oarchive oa(stream, boost::archive::no_header); oa << main; } int main() { boost::timer t; NODE* main(new NODE()); for (int i=0; i<2000;i++) { NODE* subnode = main->add( new NODE()); for (int k=0; k<20; k++) subnode->add( new NODE()); } for (int i=0; i<4; i++) { t.restart(); save(main); cout<<"serialize: "<<t.elapsed()<<endl; } delete main; return 0; } -- View this message in context: http://boost.2283326.n4.nabble.com/Boost-serialization-problem-when-use-arc… Sent from the Boost - Users mailing list archive at Nabble.com.

2 1