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

synchronous vs asynchronous...
by Ozgur Ozturk 23 Jul '09

23 Jul '09

Hi, I am kind of new to multithreaded programming and network programming, so I had some difficulty with the examples. Can you confirm an epiphany I had? - I was thinking I had to use the asynchronous functions to handle multiple clients, now I understand that you can achieve same things with synchoronous ones w/ threads and you can write equally efficient code?. (It is only a different pattern; some people find the code more manageable the async way, since they don't need to handle threads themselves.) Is this new understanding true? - If no (if there is some efficiency difference), is it significant? (Otherwise, for the initial prototype of our software, I may use still use synchronous sockets w/ threads, since I am still not very comfortable with async.) Best regards, Ozgur (Oscar) Ozturk www.DrOzturk.com Phone: +1 (908) DROZGUR i.e, +1 (908) 376-9487

3 2

ASIO async_read_until...
by Etienne Philip Pretorius 23 Jul '09

23 Jul '09

Hello list, I have a set of inherited classes as follows: client [parent of] stream [parent of] parser [parent of] grammar When I perform the following call: boost::asio::async_read_until( socket_, streambuf_, stream::parse(), boost::bind( &client::handle_read, this, boost::asio::placeholders::error, boost::asio::placeholders::bytes_transferred ) ); The xml::parser::operator() call for the match condition is a copy of the classes: parser [parent of] grammar I need to know if someone could suggest a client way for me to use (shared) pointers to the classes so that the xml::parser::operator() can keep its state changes over successive calls for async_read_until. Currently I have this (hack) for the issue, but hope there is a cleaner approach: class parser : public grammar { public: parser( ) : parser_(//stupid hack... this ) { }; virtual ~parser( ) { }; parser& parse( ) { return *this; }; template <typename T> std::pair<T, bool> operator()( T begin, T end ) { ...[state changes based on content matches]... }; private: parser* parser_;//needed for stupid hack... protected: }; When I call stream::parse() it returns the parser object, at which async_read_until makes a copy of the parser object and its base classes. When parser::operator() is called by async_read_until, I then apply all state changes via parser* parser_ to the original class. Any suggestions are welcome. Etienne.

2 2

Statechart compile error on OS X
by Daniel Dilts 23 Jul '09

23 Jul '09

I'm trying to use Boost.Statechart in an iPhone development project. When I build the Hello World example in a .mm file (Objective C++) I get an error "error: no class template named 'rebind' in 'struct Machine'". If, in the same project, I build the statemachine in a .cpp file (C++) and call a C function from a .mm file it works perfectly. Is there a compiler flag or project setting that I need to change to get the state machine to work from a .mm file? I would really like the state machine to be a member of an Objective C class.

2 3

Important: Subversion upgrade on Saturday, July 25
by Doug Gregor 23 Jul '09

23 Jul '09

We will be upgrading Boost's Subversion server and the Boost Subversion repository to Subversion 1.6.3. This upgrade will bring the new features provided by the Subversion 1.5.x and 1.6.x released to Boost, including merge tracking: http://subversion.tigris.org/svn_1.5_releasenotes.html http://subversion.tigris.org/svn_1.6_releasenotes.html The upgrade will begin on Saturday, July 25 at 8:00AM EST, and is likely to take all day. During this upgrade, the Boost Subversion repository, Trac, and mailing list archives will be inaccessible. However, the Boost mailing lists should still function normally. I will send a note to the Boost list when the upgrade is complete. Thank you for your patience. - Doug

1 0

boost 1.39 serialization via base pointer ok on gcc, but crashes on AIX V10.1
by Avi Bahra 23 Jul '09

23 Jul '09

Take the following small self contained example. This compiles and runs without any problems on gcc 4.2.1 on SUSE Linux. However on AIX v10.1 compiler, you see the following on compile: "/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 "/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_oarchiv. "/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>::sa "/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_. "src/TestSerialisation2.cpp", line 86.20: 1540-0700 (I) The previous message was produced while processing "main()". vacpp.compile.c++ bin/vacpp/debug/src/TestSerialisation.o When the program is run, the following is produced by the debugger: Can't read type info for "basic_pointer_oserializer": Expecting mangled name, and didn't find one. Can't read type info for "basic_pointer_iserializer": Expecting mangled name, and didn't find one. Can't read type info for "singleton<boost__archive__detail__<unnamed>__guid_initializer<RepeatDate> >": Expecting ":", and didn't find one. Can't read type info for "singleton_wrapper<boost__archive__detail__<unnamed>__guid_initializer<RepeatDate> >": Expecting ":", and didn't find one. Can't read type info for "error_info_container": Expecting mangled name, and didn't find one. Can't read type info for "init_guid<RepeatDate>": Expecting ":", and didn't find one. it then throw an exception: archive_exception::unregistered_cast This is because the it thinks that Base/Derived class relationship has not been registered. (i.e RepeatBase/RepeatDerived in the enclosed example). It appears that: ar & boost::serialization::base_object<RepeatBase>(*this); is not registering the base/derived relationship on AIX. The actual point in the code where is throws an exception is >>>>>>>>>: // convert pointer to more derived type. if this is thrown // it means that the base/derived relationship hasn't be registered vp = serialization::void_downcast( *true_type, *this_type, static_cast<const void *>(&t) ); if(NULL == vp){ boost::serialization::throw_exception( >>>>>> archive_exception(archive_exception::unregistered_cast) ); } In the debugger variables true_type and this_type are of type boost::serialization::extended_type_info true_type: m_type_info_key unsigned int const 0x1100121b8 0x00000001 (1) m_key $string const * 0x1100121c0 0x1000b9748 -> "RepeatDerived" this_type: m_type_info_key unsigned int const 0x110012a40 0x00000001 (1) m_key $string const * 0x110012a48 0x00000000 It looks like the m_key data member of this_type is empty, I expected "RepeatBase" Has any one else hit a similar problem, where on one platform its ok and another it crashes ? Are there any workarounds I could try ? -- Best regards, Ta, Avi

2 1

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

[multi_array] Assertion when resizing empty array
by John Reid 22 Jul '09

22 Jul '09

Resizing an empty array no longer seems to work. My code worked fine with boost 1.38, now it causes an assertion with the latest boost trunk from svn. The code is: #include <boost/multi_array.hpp> int main( int argc, char * argv[] ) { boost::multi_array< double, 2 > a; a.resize( boost::extents[2][3] ); return 0; } The assertion is: multi_array_resize: /home/john/Dev/ThirdParty/boost/boost-trunk/boost/multi_array/base.hpp:446: ArrayRef boost::detail::multi_array::multi_array_impl_base<T, NumDims>::generate_array_view(boost::type<Reference>, const boost::detail::multi_array::index_gen<NumDims, NDims>&, const typename boost::detail::multi_array::associated_types<T, mpl_::size_t<NumDims> >::size_type*, const typename boost::detail::multi_array::associated_types<T, mpl_::size_t<NumDims> >::index*, const typename boost::detail::multi_array::associated_types<T, mpl_::size_t<NumDims> >::index*, TPtr) const [with ArrayRef = boost::detail::multi_array::multi_array_view<double, 2ul>, int NDims = 2, TPtr = double*, T = double, long unsigned int NumDims = 2ul]: Assertion `index_bases[n] <= start && start < index_bases[n]+index(extents[n])' failed. Aborted My compiler is: john@john-dell:~/Dev/MyProjects/Bio/hmm$ gcc -v Using built-in specs. Target: x86_64-linux-gnu Configured with: ../src/configure -v --with-pkgversion='Ubuntu 4.3.3-5ubuntu4' --with-bugurl=file:///usr/share/doc/gcc-4.3/README.Bugs --enable-languages=c,c++,fortran,objc,obj-c++ --prefix=/usr --enable-shared --with-system-zlib --libexecdir=/usr/lib --without-included-gettext --enable-threads=posix --enable-nls --with-gxx-include-dir=/usr/include/c++/4.3 --program-suffix=-4.3 --enable-clocale=gnu --enable-libstdcxx-debug --enable-objc-gc --enable-mpfr --with-tune=generic --enable-checking=release --build=x86_64-linux-gnu --host=x86_64-linux-gnu --target=x86_64-linux-gnu Thread model: posix gcc version 4.3.3 (Ubuntu 4.3.3-5ubuntu4) I updated from svn today. I think the code creates a view on the old data in order to retain it. The assertion happens because this view is empty. Regards, John.

2 2

Spirit qi recommendations ?
by Avi Bahra 22 Jul '09

22 Jul '09

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. So my fall-back is boost 1.38. So how much has boost spirit qi changed between 1.38 and present ? How are defects reported against spirit qi, integrated back to past releases, or is there any recommended patch type mechanism, that supports redistribution of my local boost tree. Or is simply better to use spirit independently from boost? Are there any issues/guidelines that people have found in transitioning from classic to Qi. In my case I made heavy use of closures. I have not seen closures in any of spirit qi examples, Assuming there is equivalent functionality to closures ? I would interested in any recommendations before starting on this project. Best regards, Ta, Avi

3 2

weak_ptr and thread safety
by Pavel Shevaev 22 Jul '09

22 Jul '09

Guys, could you please clarify whether the following usage of weak_ptr is thread safe: //thread A: shared_ptr<Foo> p = wp.lock(); //thread B: p.reset(); I believe it's not thread safe as it looks very similar to the example 3 from the shared_ptr documentation: // thread A p = p3; // reads p3, writes p // thread B p3.reset(); // writes p3; undefined, simultaneous read/write Am I right in my assumption? And if yes, could you please give any recommendations for the following use case. I have a networking thread which owns a set of net session objects stored as shared_ptrs. I also have another thread where some entities have weak_ptrs members pointing to net session objects from networking thread. From time to time net sessions are erased and I think when at the very this moment entity object from another thread tries to lock() the net session and access it seg. fault happens. Thanks in advance. -- Best regards, Pavel

4 5

Re: [Boost-users] [Fusion] Understanding Fusion
by Cromwell Enage 22 Jul '09

22 Jul '09

--- On Sat, 7/18/09, Robert Jones wrote: > Hi All > > I'm very much a Fusion newbie, just getting may head around it, so a > really basic question! > > From the intro page, when we write > > #include <boost/fusion/sequence.hpp> > #include <boost/fusion/include/sequence.hpp> > > vector<int, char, std::string> stuff > for_each(stuff, print_xml()); > > > am I correct in believing that there is no loop involved in the > for_each, and that this resolves, at compile time, to a linear > list of function calls? Yes. IIRC, it resolves to the equivalent of: print_xml()(int()); print_xml()(char()); print_xml()(std::string()); HTH, Cromwell D. Enage

1 0

Creating a fixed size managed_mapped_file
by Eli Zakashansky 22 Jul '09

22 Jul '09

Hello, I would like to create a managed_mapped_file which should contain a certain number of elements[n] of a certain size [k], therefore the required size which I would like to use is n*k (size of element * number of elements). But when the file is created [new bip::managed_mapped_file (bip::create_only, fileName, n*k);] besides the elements which I would like to placed in this file, a segment manager<http://www.boost.org/doc/libs/1_39_0/doc/html/interprocess/architecture.htm…> is also placed there and occupies an uncertain space [http://www.boost.org/doc/libs/1_39_0/doc/html/interprocess/architecture.htm…] I would like to know how much memory (size of file) shall I require in the C-tor so that my file will have enough space for n elements of size k, and be as compact as possible. Thanks in advance, Eli.

1 0

Invoking 'brent_find_minima' for a member function of a class?
by s seelenluft 22 Jul '09

22 Jul '09

Hi, I found this example: http://net.pku.edu.cn/~webg/src/paradise/thirdparty/installed/include/boost… works fine but I have trouble figuring out how to adapt this if the function to minimize is a member function of a class (and invoked in another member function). Any example of how to call brent_find_minima (or similar functions) from a member function and minimizing a member function would be very welcome. I have to the advice to wrap the function to be minimized into a static wrapper function but could not find an example how to do this from within a member function (I have it working when calling from main). Thanks. Markus

1 1

Iterator to wrap around ends of the container. Does it exist?
by PB 22 Jul '09

22 Jul '09

Hi, Just wondered if there was anything in Boost so that when an iterator is incremented, if it's at the last element it goes to the begin() of the container. And for decrement if at the beginning it would refer to one before end(). The sequence would have to be finite, so just selecting a different logical begin and end really. I'm looking to solve this using some kind of new iterator (iterator adaptor) rather than a change of container. If there anything in Boost for this? Thanks, Pete

2 2

Variance calcs in the stats library and in Excel
by Archie14 22 Jul '09

22 Jul '09

In the docs (file:///C:/Libraries/boost/boost_1_39_0/doc/html/accumulators/user_s_guide.htm l#accumulators.user_s_guide.the_statistical_accumulators_library.variance) there is an example of calculating variance of values 1,2,3,4,5 accumulator_set<int, stats<tag::variance(lazy)> > acc1; acc1(1); acc1(2); acc1(3); acc1(4); acc1(5); BOOST_CHECK_EQUAL(5u, count(acc1)); BOOST_CHECK_CLOSE(3., mean(acc1), 1e-5); BOOST_CHECK_CLOSE(11., moment<2>(acc1), 1e-5); BOOST_CHECK_CLOSE(2., variance(acc1), 1e-5); When I am doing the same in Excel the mean is indeed 3, which matches the example, but variance is 2.5 and not 2. I am using AVERAGE and VAR formulas in Excel for mean and variance respectively Why there is a mismatch between variance values??

1 1

Re: [Boost-users] where is the pdf documentation?
by Tan, Tom (Shanghai) 22 Jul '09

22 Jul '09

http://boost.cowic.de/rc/pdf/ > ??? I see mention of pdf documentation, but couldn't find it online. > Can someone give the link to it?

1 0

[math] finding c such that w1/c+...+wn/c < math::tools::max_value<T>()
by er 21 Jul '09

21 Jul '09

Hi All, A little while back the following code was suggested to find c such that (x/c)+(y/c) <= C = math::tools::max_value<T>() template<class T>// static T scale_to_max( const T& x, const T& y ) { double max = (std::numeric_limits<T>::max)(); if(x > max - y) { double c = x/max + y/max; while(!boost::math::isinf(x/c + y/c)) { c = boost::math::float_prior(c); } while(boost::math::isinf(x/c + y/c)) { c = boost::math::float_next(c); } return(c); } else { return(static_cast<T>(1)); } } I'm now trying to generalize this i.e. find c such that w1/c+...+wn/c <= C: 1) split w1,..,wn into maximal size, sum less than C, subgroups. 2) apply scale_to_max iteratively over adjacent groups. This does not always suceed (see example at the bottom). Any suggestion would be appreciated. *.hpp template<typename T> class scale_sum_to_max{ public: typedef void result_type; typedef T argument_type; scale_sum_to_max(T cum_sum):cum_sum_(cum_sum),c_(static_cast<T>(1)){} scale_sum_to_max():cum_sum_(static_cast<T>(0)),c_(static_cast<T>(1)){} void operator()(T x){ if( math::isinf(scaled_sum()+(x/c_)) ){ T s = scale_to_max( scaled_sum(), x/c_ ); cum_sum_ /= s; c_ *= s; BOOST_ASSERT(!math::isinf(cum_sum_+x/c_)); } cum_sum_ += x/c_; } T scaling_factor()const{ return c_; } T scaled_sum()const{ return cum_sum_; } private: T cum_sum_; T c_; }; template<typename T> class finite_sums{ typedef std::vector<T> cont_t; public: typedef void result_type; typedef T argument_type; finite_sums() { coll_.push_back(static_cast<T>(0)); } void operator()(T x){ if(math::isinf(sum()+x)){ coll_.push_back(x); }else{ coll_.back() += x; } } const cont_t& collected()const{ return coll_; } T scaling_factor()const{ return for_each( begin(coll_), end(coll_), scale_sum_to_max<T>() ).scaling_factor(); } private: T sum()const{ return coll_.back(); } cont_t coll_; }; }} *.cpp typedef double value_t; typedef std::vector<value_t> vec_t; const unsigned n = 1e5; //n = 1e3 OK value_t w = math::tools::max_value<value_t>(); w /= 2; vec_t vec_w(n,w); value_t scaling_factor = for_each( begin(vec_w), end(vec_w), math_limit::finite_sums<value_t>() ).scaling_factor(); //assertion failed: (!math::isinf(cum_sum_+x/c_))

2 2