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

AIX Boost 1.39 serialization warning message:
by Avi Bahra 23 Jul '09

23 Jul '09

Take the following simple use of serialization: //=============================================== #include <boost/archive/text_iarchive.hpp> #include <boost/archive/text_oarchive.hpp> #include <boost/serialization/base_object.hpp> #include <boost/serialization/utility.hpp> #include <boost/serialization/vector.hpp> #include <boost/serialization/string.hpp> #include <fstream> #include <assert.h> class Defs { public: Defs(const std::string& f) : fileName_(f) {} Defs() {} bool operator==(const Defs& rhs) const { return fileName_ == rhs.fileName_;} const std::string& fileName() { return fileName_;} private: std::string fileName_; friend class boost::serialization::access; template<class Archive> void serialize(Archive & ar, const unsigned int /*version*/) { ar & fileName_; } }; int main() { Defs saveDefs("saveFile.txt"); { // Save the defs file std::ofstream ofs( saveDefs.fileName().c_str() ); boost::archive::text_oarchive oa( ofs ); oa << saveDefs; } { // Restore the defs file Defs loadDefs; std::ifstream ifs(saveDefs.fileName().c_str()); boost::archive::text_iarchive ia(ifs); ia >> loadDefs; assert( saveDefs == loadDefs ); } return 0; } //=============================================== This produces the following messages from the AIX v10.1 compiler. vacpp.compile.c++ bin/vacpp/debug/src/TestSerialisation.o "/s1a/emos_esuite/emos_data/sms/boost_1_39_0/boost/archive/detail/oserializer.hpp", line 538.5: 1540-2400 (W) "boost::serialization::STATIC_WARNING" is undefined. The delete operator will not call a destructor. "/s1a/emos_esuite/emos_data/sms/boost_1_39_0/boost/archive/detail/oserializer.hpp", line 530.13: 1540-0700 (I) The previous message was produced while processing "boost::archive::save<boost::archive::text_oarchive,Defs>(text_oarchive &, Defs &)". "/s1a/emos_esuite/emos_data/sms/boost_1_39_0/boost/archive/detail/common_oarchive.hpp", line 64.9: 1540-0700 (I) The previous message was produced while processing "boost::archive::detail::common_oarchive<boost::archive::text_oarchive>::save_override<Defs>(Defs &, int)". "/s1a/emos_esuite/emos_data/sms/boost_1_39_0/boost/archive/basic_text_oarchive.hpp", line 75.13: 1540-0700 (I) The previous message was produced while processing "boost::archive::basic_text_oarchive<boost::archive::text_oarchive>::save_override<Defs>(Defs &, int)". "/s1a/emos_esuite/emos_data/sms/boost_1_39_0/boost/archive/detail/interface_oarchive.hpp", line 64.21: 1540-0700 (I) The previous message was produced while processing "boost::archive::detail::interface_oarchive<boost::archive::text_oarchive>::operator<< <Defs>(Defs &)" Does anyone know if the boost serialisation library is still actively maintained on this compiler ? Best regards, Ta, Avi

2 1

strange serialization compilation issue - serialize function not being compiled
by G, Benjamin (NSN - IN/Bangalore) 23 Jul '09

23 Jul '09

Hi, I have a class which has got the serialize function implemented. template<class Archive>void myClassB::serialize( Archive& ar, const unsigned int) Somehow, this function is skipped and doesn't appear even in the object file created for this file which gives me undefined reference errors later when linking. I have included : #include <boost/archive/text_iarchive.hpp> #include <boost/archive/text_oarchive.hpp> #include <boost/archive/binary_oarchive.hpp> #include <boost/archive/binary_iarchive.hpp> In my cpp file. Does this sound faimiliar to anyone? Or am I going wrong somewhere? Thanks in advance - Benjamin Jacob.

3 2

Spirit qi recommendations ? (avi)
by Avi Bahra 23 Jul '09

23 Jul '09

Hartmut Kaiser wrote: >> In order to reduce compile times I have decided to rewrite the >> parser/grammar >> from classic to spirit qi (spirit 2.1). My understanding is that >> faster >> compiles come from easier modularisation available under spirit qi. >> (As shown in mini_c example, in the boost tree) >> >> However I need this to be portable to as many platforms as >> possible. >> In my case minimum set is Linux,HP-UX, AIX. >> > > We never tried to compile Spirit on AIX/xlc. So any feedback is welcome. > Using gcc should be fine. > Steven Watanabe wrote: >>The trunk regression tests for spirit with xlc mostly fail >>with an internal compiler error. On AIX v10.1 boost 1.39, spirit classic gets stuck in an infinite loop, and spirit qi produces the following even on the simplest of grammars. My plans to migrate to AIX are equally stuck. "/s1a/emos_esuite/emos_data/sms/boost_1_39_0/boost/spirit/home/support/component.hpp", line 219.37: 1540-0711 (S) Too few template arguments have been specified. "/s1a/emos_esuite/emos_data/sms/boost_1_39_0/boost/spirit/home/support/modifier.hpp", line 83.59: 1540-0711 (S) Too few template arguments have been specified. "/s1a/emos_esuite/emos_data/sms/boost_1_39_0/boost/spirit/home/support/attribute_transform.hpp", line 71.17: 1540-0711 (S) Too few template arguments have been specified. >From googling around a similar error was found with boost tuple, (but must have been with a different compiler version) see: http://lists.boost.org/Archives/boost/2004/06/66550.php I have seen that a new patch for the v10.1 compiler was released in May 2009. I am not sure yet if I can access this patched compiler, If I can will let you know. Best regards, Ta, Avi

2 1

Using boost::asio::async_write to stream out data
by Alex Black 23 Jul '09

23 Jul '09

My app writes out a large amount of data to a socket, in a loop, in pieces, say 8kb at a time. I'd like to switch from using Socket::send to use asio::async_write. At a high level, what do I need to do? From my reading, I think I need to: 1. Call async_write instead of send 2. Ensure that my buffers have the correct lifetime - today since my writes are synchronous I re-use the same buffer over and over again - so instead, should I make a new buffer for each write? Whats a good way to ensure the buffer has the correct lifetime, e.g. how do I delete it once its been written out? 3. I understand I can't call async_write again before the first one has been completed (since I don't want my 8kb chunks interleaved), so should I maintain a queue of buffers waiting to go out? Then in my handle_write method should I check the queue, if at least one buffer is in the queue then I should start a new async_write? Thanks! - Alex

4 20

[coroutine] Assembler snippet for linux x86_64
by Pavel Shevaev 23 Jul '09

23 Jul '09

Hi folks, First of all I'd like to thank Giovanni P. Deretta for the brilliant idea and implementation of courotines [1]. Unfortunately, coroutines haven't been reviewed and included into the official boost repository yet but I believe it's just a matter of time. I downloaded the latest version "boost-coroutine-2009-04-30.tar.gz" from the boost vault[2] and it seems to build/run just fine on linux x86. However trying to build it on x86_64 gave me a compile time error - "This file is for x86 CPUs only" in swapcontext.cpp, since there is a #if !defined(i386) guard which produces this error. I was naive and tried to change "#if !defined(i386)" to "if !defined(i386) && !defined(__x86_64__)" in hope it would "just work" for x86_64...well, it didn't ;) I'm not that good in assembler(I have only basic knowledge), maybe someone stumbled upon the same problem and has a working patch for x86_64? Or maybe someone can give some pointers on how to make couroutines x86_64 compatible? IMHO, it shouldn't be very hard and I can try to do it myself, I just don't know where to start :) Thanks a lot in advance. [1] http://www.crystalclearsoftware.com/soc/coroutine/index.html [2] http://www.boostpro.com/vault/index.php?direction=0&order=&directory=Concur… -- Best regards, Pavel

3 7

A linking error on using lambda expression in dll . Can anyone with kindness help me?
by fmingu 23 Jul '09

23 Jul '09

I am using Dev-C++ and lambda expression (boost 1.34.1). I created a dll containing a class. And want other programs to use the memberfunctions in the class. the dll.h is: .......................... #if BUILDING_DLL # define DLLIMPORT __declspec (dllexport) #else /* Not BUILDING_DLL */ # define DLLIMPORT __declspec (dllimport) #endif /* Not BUILDING_DLL */ ............ class DLLIMPORT classA{ private: ............ public: int function1(const int& ); ............... }; the dll.cpp is: # include "dll.h" ....................... int classA::function1(const int& testnumber){ ..................... } ........................ In the classA::function1 in dll.cpp I used the lambda expression and the above dll is created OK. and the main executable using the dll is: # include "dll.h" ................................ int WINAPI WinMain (HINSTANCE hThisInstance, HINSTANCE hPrevInstance, LPSTR lpszArgument, int nFunsterStil) ....................... classA instanceA; instanceA.function1(3); ..................... the complier told me no errors but the linker told me that : [Linker error] undefined reference to `classA::Function1(int const&)' And I do not know how to solve it. Does this error occur due to the use of lambda expression? Can anyone with kindness help me?

2 1

Re: [Boost-users] strange serialization compilation issue - serialize function not being compiled
by G, Benjamin (NSN - IN/Bangalore) 23 Jul '09

23 Jul '09

I forgot to mention, I am using boost 1.33.1 and compiling on windriver (x86) linux. ________________________________ From: G, Benjamin (NSN - IN/Bangalore) Sent: 23 July, 2009 1:24 PM To: 'boost-users(a)lists.boost.org' Subject: strange serialization compilation issue - serialize function not being compiled Hi, I have a class which has got the serialize function implemented. template<class Archive>void myClassB::serialize( Archive& ar, const unsigned int) Somehow, this function is skipped and doesn't appear even in the object file created for this file which gives me undefined reference errors later when linking. I have included : #include <boost/archive/text_iarchive.hpp> #include <boost/archive/text_oarchive.hpp> #include <boost/archive/binary_oarchive.hpp> #include <boost/archive/binary_iarchive.hpp> In my cpp file. Does this sound faimiliar to anyone? Or am I going wrong somewhere? Thanks in advance - Benjamin Jacob.

1 0

multiple levels of bind nesting
by Zachary Turner 23 Jul '09

23 Jul '09

I have some code that implements an asynchronous worker model where a client calls some function to request an asynchronous operation and passes a callback function, whose type is determined by a template parameter. e.g. class worker { template<class Handler> void async_do_work(Handler h) { ... } }; class client { void start_work(); void work_finished(boost::system::error_code ec, size_t bytes_transferred); }; However there are many classes implementing async_do_work but which do the work in a different way. For this I use a boost::variant<A&, B&, ...> for each type and use a boost::static_visitor to delegate to the appropriate async_do_work method. This uses code such as the following: void client::start_work() { apply_visitor( bind(do_work_visitor(), _1, protect(bind(&client::work_finished, this, _1, _2))) , work_variant_); } So from async_do_work's point of view, it receives a function with 2 arguments. How async_do_work operates internally should be separated from the client. It may not even use another thread, or it may combine the efforts of multiple threads to achieve the result. So because of that, I want async_do_work to intercept the callback before it gets posted back to the original caller. To do this, I tried the following in the implementation of worker: class worker { public: template<class Handler> void async_do_work(Handler h) { //impl supports the same basic interface as worker impl->async_do_work( bind(&worker::intercept_handler<Handler>, this, h, _1, _2)); } private: template<class Handler> void intercept_handler(Handler real_handler, boost::system::error_code ec, size_t bytes_transferred) { //do whatever is necessary and then call the real handler. real_handler(ec, bytes_transferred); } }; However this doesn't work. I've also tried protect(bind in the async_do_work function and that doesn't work either. Is it supposed to be possible to nest binds arbitrarily deep or is this just a limitation? Or am I just doing it wrong? Regards

5 5

[program_options] positional options starting with a '-' sign
by Ian McCulloch 23 Jul '09

23 Jul '09

Hi, I have an app using program_options, which has various optional arguments, followed by any number of floating point numbers, which are read as positional options. A user just pointed out that it fails with an 'unknown option -0' error if any of those positional options are negative, because it tries to interpret a number like "-0.34" as an option. Is there a way to treat unregistered options as positional arguments? allow_unregistered() doesn't quite do it - it doesn't give an error anymore but neither does it treat "-0.34" as a positional option (even if it is not the first positional argument). Regards, Ian McCulloch

2 2

[iostreams] gzip filter produces files with trailing garbage
by Michael Harris 23 Jul '09

23 Jul '09

Hi, I am using boost::iostreams in a project that I am currently working on. The relevant part of my code is shown below: #include <boost/iostreams/device/file_descriptor.hpp> #include <boost/iostreams/filtering_stream.hpp> #include <boost/iostreams/filter/gzip.hpp> ... void FileDataStore::write(DataSet *ds) { filtering_ostream os; // Check if we are writing a gzipped file int iext = this->filename.find_last_of('.'); if (iext != std::string::npos && this->filename.substr(iext+1) == "gz") { os.push(gzip_compressor()); } try { os.push(file_descriptor_sink(this->filename)); } catch (std::ios_base::failure e) { throw IOException() << errmsg_info((boost::format("Error opening file '%s': %s") % this->filename % strerror(errno)).str().c_str()); } // Write my stuff to os os.flush(); os.pop(); } Basically, if the filename being used ends in .gz I want the file written through a gzip filter so it is stored gzipped. The problem is: if I unzip the resulting file using gzip I get a message like this: gzip: myfile.gz: decompression OK, trailing garbage ignored The file content is OK but this message is annoying. What am I doing wrong that causes this? Many thanks, Regards // Mike

1 0