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

[iostreams] container_device example does not compile?
by Robert Dailey 26 Aug '09

26 Aug '09

Hello, I'm using the container_device example listed here (The full example near the bottom): http://www.boost.org/doc/libs/1_39_0/libs/iostreams/doc/tutorial/container_… I modified this class to exist in the global namespace (i.e. not in boost::iostreams::example). It does not compile under MSVC9. I get the following errors. I have NO idea what the compiler is trying to tell me. 1>c:\it\work\vfdtester\ContainerDevice.hpp(50) : error C2061: syntax error : identifier 'stream_offset' 1> c:\it\work\vfdtester\ContainerDevice.hpp(79) : see reference to class template instantiation 'container_device<Container>' being compiled 1>c:\it\work\vfdtester\ContainerDevice.hpp(50) : error C2061: syntax error : identifier 'stream_offset' 1> C:\IT\third_party\boost_1_40_0_beta1\boost/iostreams/traits.hpp(176) : see reference to class template instantiation 'container_device<Container>' being compiled 1> with 1> [ 1> Container=std::vector<boost::uint8_t> 1> ] 1> C:\IT\third_party\boost_1_40_0_beta1\boost/iostreams/traits.hpp(188) : see reference to class template instantiation 'boost::iostreams::detail::member_char_type<T>' being compiled 1> with 1> [ 1> T=container_device<std::vector<boost::uint8_t>> 1> ] 1> C:\IT\third_party\boost_1_40_0_beta1\boost/iostreams/stream.hpp(104) : see reference to class template instantiation 'boost::iostreams::char_type_of<T>' being compiled 1> with 1> [ 1> T=container_device<std::vector<boost::uint8_t>> 1> ] --------- Robert Dailey

2 2

[serialization] Binary archive headers?
by Robert Dailey 26 Aug '09

26 Aug '09

Hi, I'm looking at the streambuf filled with data after I do this: boost::asio::streambuf buf; std::ostream stream( &buf ); boost::archive::binary_oarchive archive( stream ); archive << "Hello World"; And I see that there's a lot of binary data that preceeds the "Hello World" data. Looks like custom header information that boost::serialization inserts into the stream. Is there a way to prevent any automatic header information from being inserted into the archive's stream? I'm using this serialized buffer to communicate with a device over a COM port and I can't have this extra information in my buffer. Thanks. --------- Robert Dailey

3 2

Transmitting serialized data of multiple types and variable length over a socket
by A.Agrawal＠patrick.com.au 26 Aug '09

26 Aug '09

Hello, I am new to boost (few months). I have a client - server situation where there can be multiple clients (Linux or Window) connected to one server (Linux). Clients send requests and Server sends response to the clients. Server also sends new updates which the clients will keep receiving. The request /response has following structure: struct Message { uint8_t data_name; uint16_t count_of_data_records; // uint16_t data_len; //length of serialized_data string serialized_data; //serialized_data is binary serialized data using boost serialization. }; We have 6 types of structs (identified by data_name) that are serialized into serialized_data. I understand that I have to send two buffers one (ASCII buffer) with data_name, count_of_data_records and data_len and other binary buffer. I have seen example: http://www.boost.org/doc/libs/1_39_0/doc/html/boost_asio/example/serializat… Questions: 1. I am not sure if I should use synchronous or asynchronous communication, I am inclined towards synchronous. 2. What is the best way to send this data across from a server to multiple clients and from one client to one server.I don't know the best way to implement it. Please advise what is the best way or if you know of an example. Regards, Anil Agrawal

2 1

[asio] Passing streams into async_write()?
by Robert Dailey 25 Aug '09

25 Aug '09

Hello, I have a std::istream that I pass into a function that calls async_write. Currently here is what I have: void SerialClient::DoWrite( std::istream const& data ) { boost::asio::async_write( m_port, boost::asio::buffer( data.rdbuf() ), boost::bind( &SerialClient::WriteCompleted, this, boost::asio::placeholders::error ) ); } Obviously this does not compile. There is no overload of buffer() that takes a streambuf. How can I get asio to write out my stream? --------- Robert Dailey

1 1

[serialization] Stop tracking a specific item
by Chard 25 Aug '09

25 Aug '09

I have no access to the default constructor of a type I want to serialize (and no knowledge of any other constructors). However, the type is a friend to boost::serialization::access. Rather than duplicating a chunk of the serialization code, I thought I could just reuse the guts of it by serializing the object through a pointer. However, this would require me to construct the pointer during the serialization process, which would mess up the object tracking. Is it possible to turn the tracking off for a specific serialized object, but not the whole type ? E.g. a contrived example: template ... void serialize(Archive &ar, const unsigned) { Type *p = 0; if (Archive::is_saving::value) { p = FuncThatCanGiveMeAPointerToType(); } // Want to turn off object tracking on p here ar & p; // or remove it here if (Archive::is_loading::value) { FuncThatCanAssignTypeByValue(*p); delete p; } } It's like I require a remove_object_address, as opposed to reset_object_address. Or is there an alternative approach?

2 1

[serialization] [archive] Bug serializing vectors consecutively
by Paul Eipper 25 Aug '09

25 Aug '09

Hi, I've hit what I believe is a bug in boost serialization, at least on binary and text archive modes. If I have two member vectors (of int, for example), and the serialize code is like this: ar & vector1; ar & vector2; I cannot deserialize it back, since boost throws "stream error" if it's a text archive, or "std::length_error" on vector::reserve if it's a binary archive. Attached is an example code. att, -- Paul Eipper

4 5

[asio] It pollutes my O-Zone!
by Robert Dailey 25 Aug '09

25 Aug '09

I noticed that <boost/asio.hpp> and <boost/asio/io_service.hpp> both include windows specific header files through some level of indirection. I know this because it is causing problems with wxWidgets depending on the order of my include directives. I was considering using a mediator/pimpl of some sort to keep the asio includes limited to a very confined translation unit, but this would complicate the usage of ASIO in general. I have a class that owns an io_service object by-value in its header. Anyone have any suggestions to reduce the amount of namespace polluting ASIO does? I really blame this on the Windows API, since ASIO is just as limited by it as I am. Thanks in advance. --------- Robert Dailey

3 3

[proto] another const issue
by Maurizio Vitale 25 Aug '09

25 Aug '09

Eric, this may or may not be related to the other problem w/ the assignment operator. In my application eventually I'll want to allow literals of builtin constants on the right hand side of the assignment, but not on the left hand side. So given: my_int my_i; // a proto terminal int i; The following should happen: my_i = 5; // OK (i,my_i) = my_i; // OK (5,my_i) = my_i; // NOT OK For that I have a grammar for the left hand side and one for the right hand side. In the attached file, the grammar for the left hand side doesn't match and I'm not sure why. What I'm doing wrong? TIA, Maurizio

2 3

Problem with compiler options ?
by Victor Perron 25 Aug '09

25 Aug '09

Hello, I, trying to use some phoenix functions inside of a big project. The problem is that, although everything runs well in my test cases, when I integrate the very same code inside of the huge project, it always fails in a segfault. Let's explain it further : I have a code in main(), directly, both for test case and program, which is the same : int main(int argc, char *argv[]) { State my_state; my_state.setPosition(new Point(1250,8907)); my_state.setName(new string("myname")); setRobotState(&my_state); if(areamanager(localhost, 888) == false) cerr << "Could not Connect to server." << endl; while(true) {}; return 0; } ... quite simple. It is not interesting, but at some point inside of the parsing, I call (within a grammar) the following line : [...] confix_p( str_p(Message::nameMark.c_str()) >> ch_p('='), strval[phoenix::bind(&Message::setName)(self.variable1,phoenix::new_<std::string>(arg1, arg2))], ch_p(';')) [...] Where str_p is a very simple parser for strings, nameMark is a string marker for the parsing, and I'm supposed to call the member function variable1::setName(new string(arg1,arg2)) ... this is basically what phoenix does. And it works, my function is called. But in only one case... First one : (on the same computer, Linux mpc182 2.6.23.1-42.fc8 #1 SMP Tue Oct 30 13:55:12 EDT 2007 i686 i686 i386 GNU/Linux ) g++ main.o -L./ -lboost_system-mt -lboost_thread-mt -lareacommunication -o demo : This works well, no segfault during parsing, member function is called. Second one: gcc -g -O3 main.o -odemo_main -lfacetracker `xml2-config --libs` `Magick-config --ldflags` `Magick-config --libs` -lmpisearch -L/usr/lib/mysql/ -lboost_system-mt -lboost_thread-mt -lareacommunication -lpthread -lstdc++ What should make it segfault ? I tried changing back gcc to g++, then remove the -O3 flag, the -g flag, the -lstdc++... I really don't know where is the problem. Any idea ? -- Victor

2 1

Problem with casting numeric array to/from PyArrayObject
by Lionel LE ROUX 25 Aug '09

25 Aug '09

Hi, To start with Boost, I wrote a (very simple) test_that_() function in C/Boost with a numeric::array& as argument. I compiled a .so file and I'm then able to seamlessly use this function from Python : that's for the successful part... Now I want to trace memory usage (and possible leaks) using Valgrind : then I don't use Python anymore and decide to stay in the C world : so I wrote a short test program in C which only includes a call to the function in the .so : /int main(void) { Py_Initialize(); import_array(); npy_intp dim[]={1000}; PyArrayObject* X = (PyArrayObject*)PyArray_SimpleNew(1, dim, PyArray_FLOAT); test_that_((numeric::array&)X); Py_DECREF(X); Py_Finalize(); return 0; }/ I can compile and build the executable file with no error but when I run it : segmentation fault when I try to use PyArray_Size within the test_that() function, which only "link" a PyArrayObject to the numeric::array argument. The manipulation of the data seems to be OK but problem arises with PyArray_Size. Well, looks like the double cast (PyArrayObject -> numeric::array -> PyArrayObject) is not valid. Could someone confirm (or not) my (very) bad coding ? Thanks.

2 1