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

Re: [Boost-users] Linking with dynamic boost thread library
by John Wilkinson 09 Apr '09

09 Apr '09

> If the program compiles and runs the same only based on whether boost*.lib > is there, what function does the boost*.dll serve The boost*.lib files, as I think someone said earlier, are the import libraries for the boost*.dll run-time libraries. These are what you link to, in order to use boost*.dll at run time. The DLL files do not need to be in the link path, only in the executable path when your program is run. The boost*.lib files need to be in the link path when you build your program. John

2 2

[lambda] how to get rid of the reference argument in lambda functor?
by Pobinger Gerhard Stefan 09 Apr '09

09 Apr '09

Hi *, I was desparately trying to compile something like this (VS2008), always getting the error that it cannot convert from int* to int*&. :boost::shared_ptr<int> c; ::boost::lambda::var(c) =::boost::lambda::constructor <::boost::shared_ptr<int>>() (::boost::lambda::new_ptr<int>()() ) [This is "short form" for the conventional code ::boost::shared_ptr<int>c=::boost::shared_ptr<int>(new int());] (1) The reason is that all lambda functors take arguments by reference. In this case, make_const in front of shared_ptr did not solve the problem, it causes immediate evaluation of the new_ptr functor... Is there any way to rewrite this (1) short line of code in boost-lambda? /Gerhard Pobinger

2 1

Re: [Boost-users] Linking with dynamic boost thread library
by Steven Watanabe 09 Apr '09

09 Apr '09

AMDG Green, Jason M NSWCDL, W33 wrote: > I have a directory in my project call Boost. I added the directory to the > library files list. > In this directory I have the libboost*.lib files and the boost*.dll files. > I am able to get it working with no problem if I only have the libboost*.lib > files in there. If I rename them, and add the dll into the directory, it > will not build. That is where it will give me the error saying that it is > looking for libboost*.lib > Whoa. I'm confused. Is the linker complaining about libboost*.lib or boost*.lib? If you define BOOST_ALL_DYN_LINK or BOOST_<library name>_DYN_LINK the compiler will look for the import libraries for the dlls. > Green, Jason M NSWCDL, W33 wrote: > >> the compile warns me that it "cannot open file >> 'boost_regex-vc80-mt-gd-1_38.lib'" >> >> In Christ, Steven Watanabe

2 1

[interprocess] warning C4251: need dll-interface
by Brent Arias 09 Apr '09

09 Apr '09

I've defined a class to use within shared memory, and I've decorated it with a Windows _declspec(export) to expose it from a DLL. When I compiled with VC 2008, I got this warning: Warning C4251: class 'boost::interprocess::vector<T,A>' needs to have dll-interface to be used by clients of 'MyClass' I searched boost user archives for any discussion of this, and in the context of boost.signals I've seen someone comment to simply disable the warning with: #pragma warning (disable: 4251) Std::list<OtherClass*> talkers; #pragma warning (default: 4151) Does this also apply to boost.interprocess? Is it really safe to ignore this warning and continue to use the interprocess code from within a DLL? Brent Arias

4 3

Concept DefaultConstructible vs type traits has_trivial_constructor
by Hicham Mouline 09 Apr '09

09 Apr '09

Hi, What is the difference between the 2? For a type T, can it be DefaultConstructible and has_trivial_constructor<T> is false? Rds,

2 1

Re: [Boost-users] boost build interface
by Rudolf Leitgeb 09 Apr '09

09 Apr '09

FWIW this doesn't have to be a windows only GUI. We could use the QT widget set and make this portable to most platforms supported by boost. Note, that Qt is LGPL now, so you could even include this into commercial distributions if needed. Am 08.04.2009 um 16:10 schrieb Green, Jason M NSWCDL, W33: > Not that I am volunteering, but it isn't that hard to come up with a > simple > dialog based mfc project that would set up the commands to properly > run the > boost build.

2 2

[interprocess] placement new allocator for vector
by Brent Arias 09 Apr '09

09 Apr '09

I'd like to have a user type that contains a vector and yet is still binary serializable, *AND* that can be used entirely inside or outside of shared memory. My idea is to give the vector a placement new allocator, it goes like this: class MyBinarySerialType { struct SomeItem { char first[10]; char second[10]; }; SomeItem m_Items[16]; //The placement new allocator points with an offset_ptr<SomeItem> //to the location &m_Items[0] boost::interprocess::vector<SomeItem, /*placement new allocator*/> m_vItems; public: MyBinarySerialType(const /*placement new allocator*/ &alloc); } I'd like to know if interprocess allocators can already do this for me (via template params, etc) or if I need to write my own allocator. If this is done correctly, I think it means I can "new" MyBinarySerialType from the regular heap and it will work entirely outside of shared memory. Likewise if I use segment.construct<MyBinarySerialType>()(...), then the whole thing should be created inside of shared memory. Either way, the vector will allocate strictly to a piece of memory inside the size of the class. Brent Arias

2 1

[date_time] Time duration parsing
by Konstantin Litvinenko 09 Apr '09

09 Apr '09

Hi All! I have a string "14:34" and want to parse it into boost::posix_time::time_duration. How can I do that?

1 0

[serialization] Export question
by Tom Aylesworth 09 Apr '09

09 Apr '09

I just want to make sure that I understand the use of the Serialization library's EXPORT macros. On the "Serializable Concepts" page of the documentation, it states: -- The macro BOOST_CLASS_EXPORT_GUID associates a string literal with a class. ... This permits each class to be in a separate header file along with its string identifier. -- However, on the "Special Considerations" page, I find: -- Note that the implementation of this functionality requires that the BOOST_CLASS_EXPORT macro appear after and the inclusion of any archive class headers for which code is to be instantiated. ... Note that including BOOST_CLASS_EXPORT in the "a.hpp" header itself as one would do with other serialization traits will make it difficult or impossible to follow the rule above regarding inclusion of archive headers before BOOST_CLASS_EXPORT is invoked. This is different than other serialization traits which would normally be included in same file as the class declaration. -- The associated example shows the EXPORT macro following the #include of the header defining the class to be exported. I would certainly much rather have the EXPORT macros appear in the header with the class that defines them. I take it that means I have to guarantee that the file actually responsible for operating on an archive includes my headers with the class definitions and EXPORT macros *after* including the appropriate archive headers (e.g., "boost/archive/xml_oarchive.hpp"). Is that correct? Are there any other gotchas I should be aware of if I place the EXPORT macros in the headers with the classes to be exported? Thanks, Thomas Aylesworth Senior Software Engineer Reality Mobile

3 7

boost build interface
by Igor 09 Apr '09

09 Apr '09

I have a suggestion: Why not make a simple dialog window with several check buttons and few radio buttons where you will choose what kind of library you want to build(static/dynamic, release/debug), one edit line in which you enter the location where you want your libraries to be built, and one push button that will execute the build process? It would make the building process of the boost libraries 100 times easier.

4 3

[interprocess] named mutex clean up
by Chard 09 Apr '09

09 Apr '09

Does calling named_mutex::remove() have consistent cross-platform behaviour? The reason I ask is that I am trying to use a named mutex for an "I'm the only process" check. That is, the process takes a shared lock on the named mutex (at start up), then, at points within the program, it attempts to get an exclusive lock in order to perform the check/actions. When finished, the mutex is returned to a shared lock. The problem is, if a process dies in some unforeseen manner - and does not get to call named_mutex::remove() - the named_mutex is left orphaned in the system (in a shared lock) such that other processes can never get an exclusive lock. On the Windows platform, I've found I can call named_mutex::remove() at start up, which seems to clean up the named_mutex if it is orphaned, or just do nothing if another, active, process has a lock - exactly the behaviour I require. However, is this platform independent, and a suitable technique?

4 6

Boost.Random - random_device error LNK2019 ()
by Valen 09 Apr '09

09 Apr '09

Hello there! Windows XP, Boost 1.38, VC++ 7.1 / Mingw 3.4.2 sample code: //----------------------------------------------------- #include <boost/random/uniform_int.hpp> #include <boost/random/variate_generator.hpp> #include <boost/nondet_random.hpp> int main() { boost::random_device m; typedef boost::uniform_int<> distribution_type; distribution_type generator2(m()); return 0; } //----------------------------------------------------- VC++ Output: error LNK2019: unresolved external symbol "public: __thiscall boost::random_device::~random_device(void)" (??1random_device@boost@@QAE@XZ) referenced in function "protected: void __thiscall CGraphDoc::OnBtnNewnode(void)" (?OnBtnNewnode@CGraphDoc@@IAEXXZ) error LNK2019: unresolved external symbol "public: unsigned int __thiscall boost::random_device::operator()(void)" (??Rrandom_device@boost@@QAEIXZ) referenced in function "protected: void __thiscall CGraphDoc::OnBtnNewnode(void)" (?OnBtnNewnode@CGraphDoc@@IAEXXZ) error LNK2019: unresolved external symbol "public: __thiscall boost::random_device::random_device(class std::basic_string<char,struct std::char_traits<char>,class std::allocator<char> > const &)" (??0random_device@boost@@QAE@ABV?$basic_string@DU?$char_traits@D@std@@V?$allocator@D@2@@std@@@Z) referenced in function "protected: void __thiscall CGraphDoc::OnBtnNewnode(void)" (?OnBtnNewnode@CGraphDoc@@IAEXXZ) error LNK2001: unresolved external symbol "private: static char const * const boost::random_device::default_token" (?default_token@random_device@boost@@0QBDB) fatal error LNK1120: 4 unresolved externals MinGW gave same output I checked the sources and find no implementation of ~random_device and etc... Can anyone help please? -- View this message in context: http://www.nabble.com/Boost.Random---random_device-error-LNK2019-%28%29-tp2… Sent from the Boost - Users mailing list archive at Nabble.com.

2 1

Passing reference to boost::bind object
by david.weber＠l-3com.com 08 Apr '09

08 Apr '09

So, we are implementing boost::signals/slots, and in several cases, we have a connect() function to allow us to hook up an arbitrary slot to the signal, without exposing the details of the signal for cleanliness issues. We would like to implement the signals::trackable interface, and have the objects which are passing in the bound slot manage their connections in this manner. However, we are passing them as boost::function objects, which (as documented) does not work properly with boost::function objects, only boost::bind objects. In all of the documentation that I've found, about assigning the results of a boost::bind call were to function objects. Is there an alternative available to not destroy the trackable functionality that we're trying to use? Thanks --dw

3 3

help building boost for iphone
by Kenny Riddile 08 Apr '09

08 Apr '09

I'm fairly inexperienced with boost.build and I'm having an issue building boost for the iphone. My user-config.jam (I put this in the same dir as the bjam executable right?) looks like this: using darwin : 4.0.1~iphone : /Developer/Platforms/iPhoneOS.platform/Developer/usr/bin/arm-apple-darwin9-g++-4.0.1 -arch arm : <striper> : <architecture>arm <target-os>iphone <macosx-version>iphone-2.2 ; using darwin : 4.0.1~iphonesim : /Developer/Platforms/iPhoneSimulator.platform/Developer/usr/bin/g++ : <striper> : <architecture>x86 <target-os>iphone <macosx-version>iphonesim-2.2 ; and I'm invoking bjam like this: ./bjam debug release toolset=darwin --layout=system link=static architecture=arm target-os=iphone macosx-version=iphone-2.2 My error is that it can't find g++ becuase it isn't looking in the right place: g++-4.0: installation problem, cannot exec '/usr/bin/arm-apple-darwin9-g++-4.0.1': No such file or directory

2 2

[graph] hash_set
by liquid 08 Apr '09

08 Apr '09

How to use hash_setS? If I define BOOST_HAS_HASH the compiler will generate "error C2977: 'stdext::hash_set' : too many template arguments". This is caused by using "BOOST_STD_EXTENSION_NAMESPACE::hash_set<Key,Eq,Hash,Alloc>" in definition of "template <class Key, class Eq, class Hash, class Alloc> struct container_traits< BOOST_STD_EXTENSION_NAMESPACE::hash_set<Key,Eq,Hash,Alloc> > { typedef set_tag category; typedef stable_tag iterator_stability; // is this right? };" Class hash_set is defined in STL as follows: template<class _Kty, class _Tr = hash_compare<_Kty, _STD less<_Kty> >, class _Alloc = _STD allocator<_Kty> > class hash_set : public _Hash<_Hset_traits<_Kty, _Tr, _Alloc, false> > {...}; which has only 3 params. Regards liquid

2 2

[serialization] appending elements
by Agustín K-ballo Bergé 08 Apr '09

08 Apr '09

Is it possible to append serialized elements to an existing archive? I need to keep a set of records on a file, and periodically add new entries to it (kind of like a database table). Reading the whole set, adding the new records, and serializing the set back is not an option since I expect the set to go pretty big. I have tried to keep several archives at the same stream (with no_header and no_tracking flags), but then I don't know how to read them. Reading them as if they were a single archive does not work, should it? Agustín K-ballo Bergé.-

4 5

[serialization] Test for end of archive
by Tom Aylesworth 08 Apr '09

08 Apr '09

Is there a way to test for the end of an archive, similar to testing for end of a stream? I have an indeterminate number of objects serialized to an archive and would like to loop until the end of the archive has been reached without reading past the end of the archive and getting a stream exception. Thanks! Thomas Aylesworth Senior Software Engineer Reality Mobile

1 0

Determining thread existence
by Jeremy Rottman 08 Apr '09

08 Apr '09

I am trying to setup my application so that it runs as a daemon. One aspect that I need to run a sanity check on is to determine if there are currently any parent threads running. Currently this is what I am testing with, when all threads are finish running, I then join all the threads to terminate them. Doing so should give me a thread group size of 0. However, this is now the case. Currently after I join the threads, the thread group size is still 0. Here is my current test code that I am working with. Any help with this is greatly appreciated. int main(int argc, char *argv[]) { InitConfig mconfig; if(mconfig.setConfig()){ boost::thread_group Tg; cout << "Thread Size Before: " << Tg.size() << endl; // DISPLAYS 0 for(int i = 1; i < 25; i++){ dmParse parse; Tg.create_thread(boost::bind(&dmParse::onInit,parse,i)); } cout << "Thread Size: " << Tg.size() << endl; // DISPLAYS 24 Tg.join_all(); cout << "Thread Count After:" << Tg.size() << endl; // DISPLAYS 24 }

3 8

a vector of subgraphs
by fabricio 08 Apr '09

08 Apr '09

Hello, I'm creating a program that will require a vector, of dynamic size, of subgraphs. so let's suppose I have something like this: typedef subgraph< adjacency_list > Graph; Graph G; what I want is to create a: vector< Graph& > subG; but the usage of & here is illegal. Do you guys have any suggestion on how to do it? regards,

4 8

Re: [Boost-users] [serialization] Export question
by Tom Aylesworth 08 Apr '09

08 Apr '09

I think I just answered my own question regarding whether I can use the EXPORT macro in the header that declares the class being exported. Since the EXPORT macro instantiates a static guid_initializer for the class, including it in multiple source files creates multiple static objects and results in link errors. So at this point, I wonder, why bother with the EXPORT macro at all? Given that the class has to be registered with the archive, why not just use register_type when the archive is created? When I thought the EXPORT macro allowed me to register the class in the same header where the class is created, it seemed very useful. Now I guess I don't see any advantage of using EXPORT. Am I missing something? Thomas Aylesworth Senior Software Engineer Reality Mobile

3 2

[mpl] newbie question about if_c et al
by Jonathan Leonard 08 Apr '09

08 Apr '09

Hi, I'm trying to make a class which is conditionally synchronized. What I have is like the following: template<bool SYNCH=true> class c { public: void f() { typename boost::mpl::if_c<SYNCH, ScopedCriticalSection<>, void*>::type scs(cs); // SCS == RAII wrapper around CS // do other stuff } private: typename boost::mpl::if_c<SYNCH, CriticalSection<>, void*>::type cs; }; This is not entirely optimal though as it requires 3 MOV instructions to create the scs void* copy when SYNCH is false (and I'd rather not lean on compiler optimizations even if they will help in this case). [Something like NullType, VoidType, or OptionType (a la F#) would be ideal but I can't seem to locate one (and I'm not sure the type system would allow such a thing-it doesn't seem to like a simple 'void']. Any better idiom or alternative approach anyone can offer for this purpose would be greatly appreciated. Many thanks, Jonathan

3 7

Re: [Boost-users] Linking with dynamic boost thread library
by Steven Watanabe 08 Apr '09

08 Apr '09

AMDG Green, Jason M NSWCDL, W33 wrote: > I'm having a similar problem with my project. I'd like to use the dlls but > no matter what, the compile warns me that it "cannot open file > 'boost_regex-vc80-mt-gd-1_38.lib'" > You /are/ linking to the dll then. > I'm trying to find where I can set the project (in vis studio 2005) to use > the dynamic link libraries as opposed to the static. Based on previous > entries in this thread, I assume that is the first step that I need to take. > Have you added the location of the libraries to the library paths? (Tools->Options->Projects and Solutions->VC++ Directories, I think) > Can someone tell me where to set this? > In Christ, Steven Watanabe

2 1

Re: [Boost-users] Linking with dynamic boost thread library
by Ovanes Markarian 08 Apr '09

08 Apr '09

Project Properties -> C/C++ -> Code Generation: Runtime Library: Multi-Threaded Debug DLL (/MDd) or Multi-Threaded DLL (/MD) for release version Good Luck, Ovanes On Wed, Apr 8, 2009 at 6:40 PM, Green, Jason M NSWCDL, W33 < jason.m.green3(a)navy.mil> wrote: > I’m having a similar problem with my project. I’d like to use the dlls > but no matter what, the compile warns me that it “cannot open file > ‘boost_regex-vc80-mt-gd-1_38.lib’” > > > > I’m trying to find where I can set the project (in vis studio 2005) to use > the dynamic link libraries as opposed to the static. Based on previous > entries in this thread, I assume that is the first step that I need to take. > > > > Can someone tell me where to set this? > > Jason > > > > *From:* boost-users-bounces(a)lists.boost.org [mailto: > boost-users-bounces(a)lists.boost.org] *On Behalf Of *Ovanes Markarian > *Sent:* Wednesday, April 08, 2009 11:01 AM > *To:* boost-users(a)lists.boost.org > *Subject:* Re: [Boost-users] Linking with dynamic boost thread library > > > > What I mean is the reversed order. You can not link the static runtime and > dynamic boost_thread or any other lib. > > On Wed, Apr 8, 2009 at 4:50 PM, Steven Watanabe <watanabesj(a)gmail.com> > wrote: > > AMDG > > > > Igor MI wrote: > > But now I wonder about what Ovanes said. Why did I linked successfully with > static thread library when my project is set to multi threaded debug dll? > > > > > You can link using the dll runtime and the static boost_thread. > > > > In Christ, > Steven Watanabe > > _______________________________________________ > Boost-users mailing list > Boost-users(a)lists.boost.org > http://lists.boost.org/mailman/listinfo.cgi/boost-users >

2 1

[ASIO] Synchronous write on serial_port
by Daniele Barzotti 08 Apr '09

08 Apr '09

Hi, I'm developing a library that use ASIO Serial Port and I need to implement a synchronous write. The structure is: EuroATLib.cpp : library main file Serial.cpp : serial port class that wrap asio::serial_port ATDispatcher.cpp : a dispatcher object ATModem.cpp : the consumer object The problem is that my code writes characters on serial port very slowly (1 char every 10 seconds) and I can't understand why. Also, when the program ends, the io_service thread seems to be already alive: it continues to write to serial port for a while! The implementation is like this...can someone give me some feedback? PS: In the following code TBUFFER is std:string. In EuroATLib.cpp : ---------------------------------------------------------- // Main io_service static boost::asio::io_service io_service; static boost::scoped_ptr<boost::thread> io_service_thread; /*! \brief Factory function for ATModem objects creation */ EXTERN_C EUROATLIB_API ATModem* CALL CreateATModemObj() { static bool thread_started; if (!thread_started) { try { // run the IO service as a separate thread io_service_thread.reset(new boost::thread( boost::bind(&boost::asio::io_service::run, &io_service) )); thread_started = !thread_started; } catch (boost::thread_resource_error e) { // Failed to create the new thread return 0; } } // create the new object return new ATModem(io_service); }; ---------------------------------------------------------- In Serial.hpp : ---------------------------------------------------------- // the main IO service that runs this connection boost::asio::io_service& _io_service; // the serial port this instance is connected to boost::asio::serial_port _serialPort; ---------------------------------------------------------- In Serial.cpp : ---------------------------------------------------------- SerialPort::SerialPort(boost::asio::io_service &_io) : m_buffer_size(100), _io_service(_io), _serialPort(_io_service), _rx_timer(_io_service), ReadTimeOut(READ_TIMEOUT) { //init_dcs(); EosString = '\n'; _buf = 0; }; BOOL SerialPort::Write(TBUFFER &buffer) { _io_service.post(boost::bind(&SerialPort::do_write, this, buffer)); return true; }; void SerialPort::do_write(TBUFFER &buffer) { // size_t byte_written = boost::asio::write(_serialPort, boost::asio::buffer(buffer.c_str(), buffer.size())); size_t byte_written = _serialPort.write_some(boost::asio::buffer(buffer.c_str(), buffer.size())); }; ---------------------------------------------------------- In ATDispatcher.cpp simply have a SerialPort member and a function SendCommand that start a timer and call Serial.Write(cmd) In ATModem.cpp I have an ATDispatcher Object on which it calls disp_obj.SendCommand("AT+CMGS"): ---------------------------------------------------------- RET_VAL ATModem::Connect(std::string DevName) { if (DevName.length() > 0) SetComPort(DevName); if (!_serial_port.Open()) return -1; //Connect the OnDataReceived event to the ATDispatcher object _dispatcher->OnData(boost::bind(&EuroATLib::ATModem::OnDataReceived, this)); //Register Voice Service if ( _dispatcher->SendCommand("AT+CTSP=2,0,0") != ATDispatcher::RET_OK ) return DEV_AT_CMD_ERROR; //Register STATUS Service if ( _dispatcher->SendCommand("AT+CTSP=2,2,20") != ATDispatcher::RET_OK ) return DEV_AT_CMD_ERROR; this->SetStatus(STATE_READY); }; ---------------------------------------------------------- Finally in the main client program: // Create a ATModem object boost::scoped_ptr<ATModem> _devATModem( CreateATModemObj() ); _devATModem->Connect("\\\\.\\COM1");

2 5

Linking with dynamic boost thread library
by Igor 08 Apr '09

08 Apr '09

I build the thread library with the following cmd command: bjam.exe toolset=msvc --with-thread --prefix=C:\boost\ variant=debug,release link=shared install after which I get the following libs: boost_thread-vc80-mt.lib boost_thread-vc80-mt-1_38.dll boost_thread-vc80-mt-1_38.lib boost_thread-vc80-mt-gd.lib boost_thread-vc80-mt-gd-1_38.dll boost_thread-vc80-mt-gd-1_38.lib In my application, I placed #define BOOST_THREAD_DYN_LINK in all the cpp files where thread library is included. I get the following error: LINK : fatal error LNK1104: cannot open file 'libboost_thread-vc80-mt-gd-1_38.lib' How many mistakes have I done?

5 7

Re: [Boost-users] difference b/w regex lib files
by John Maddock 08 Apr '09

08 Apr '09

>I have a number of boost libs that I built (not being sure which one I > really needed). I understand the difference b/w them, with the exception > of > two. libboost*.lib are static link libraries, and boost*.lib are import libraries for the dll's. HTH, John.

2 1

Re: [Boost-users] boost build interface
by Vladimir Prus 08 Apr '09

08 Apr '09

Green, Jason M NSWCDL, W33 wrote: > Not that I am volunteering, but it isn't that hard to come up with a simple > dialog based mfc project that would set up the commands to properly run the > boost build. I did not claim that coming up with an MFC project is hard per se. However: 1. It will require *me* quite some time, for I was not doing any real windows development in years. 2. It might be best to use some more lightweight solution, but I don't know what are scripting options on modern Windows. - Volodya

3 2

[date_time] local time & date arithmetics
by Filip Konvička 08 Apr '09

08 Apr '09

Hi all, first, thanks for the date_time library. Although I see some issues with it, I still think it is highly needed and does a lot of good stuff for me. Recently I stumbled upon a problem with date arithmetic that involves local_date_time. The following code fragment local_date_time t( ptime(boost::gregorian::date(2008, 4, 7), time_duration(22, 0, 0)), boost::shared_ptr<posix_time_zone>(new posix_time_zone("CET+1CEST,M4.1.0,M10.1.0"))); cout << t.utc_time() << " UTC = " << t << endl; t -= boost::gregorian::days(6); cout << t.utc_time() << " UTC = " << t << endl; yields the following output 2008-Apr-07 22:00:00 UTC = 2008-Apr-08 00:00:00 CEST 2008-Apr-01 22:00:00 UTC = 2008-Apr-01 23:00:00 CET This indicates that the date operation just subtracted 6*24 hours from the internal representation of the local time. It missed the DST change within the time interval. I would rather expect the operation to do the following: 2008-Apr-07 22:00:00 UTC = 2008-Apr-08 00:00:00 CEST 2008-Apr-01 23:00:00 UTC = 2008-Apr-02 00:00:00 CET If the current behavior is intentional, it is not clear to me why have such operations as "+= days(...)" available in local_date_time at all, as they don't seem to obey the general local time semantics (at least as I understand it) - we could just use an UTC ptime for that, right? Another thing related to this is that many functions in local_date_time are inherited from the base class, which has a different semantics. From the user point of view, this seems to be quite dangerous - one has always to verify whether the function being called is overloaded in local_date_time (and so should have local time semantics) or whether it is defined in ptime and so works with the internal UTC representation. Why is the base class not protected? Thanks for any comments, Filip

1 0

Re: [Boost-users] difference b/w regex lib files
by Steven Watanabe 08 Apr '09

08 Apr '09

AMDG Green, Jason M NSWCDL, W33 wrote: > I have a number of boost libs that I built (not being sure which one I > really needed). I understand the difference b/w them, with the exception of > two. > > > > All of the libs are libboost_regex..lib except 2 which are just > boost_regex...lib > > What is the difference between the 2? > The boost_regex*.lib files are import libraries for boost_regex*.dll In Christ, Steven Watanabe

1 0

difference b/w regex lib files
by Green, Jason M NSWCDL, W33 08 Apr '09

08 Apr '09

I have a number of boost libs that I built (not being sure which one I really needed). I understand the difference b/w them, with the exception of two. All of the libs are libboost_regex..lib except 2 which are just boost_regex...lib What is the difference between the 2? Jason

1 0

[Boost-Users][interprocess]ERROR_ACCESS_DENIED when removing a message_queue
by Laurence Chambenois 07 Apr '09

07 Apr '09

Hi, I'm a beginner with boost, so please don't throw things at me ... I'm using boost message_queue in a sample program, on Windows, that just does : - create the message_queue - remove the message_queue It works fine, if my program does this : try { message_queue mq1(create_only, "toto", 10, 100); } catch (interprocess_exception &err) { return -1; } message_queue::remove("toto"); but if it does : message_queue *mq1=0; try { mq1 = new message_queue(create_only, "toto", 10, 100); } catch (interprocess_exception &err) { return -1; } message_queue::remove("toto"); then the remove return ERROR_ACCESS_DENIED (on tthe DeleteFile call). I don't understand why. Is somebody can tell me what is wrong with my code ? Thanks for your reply.

4 3

interprocess windows_shared_memory implementation
by Brent Arias 07 Apr '09

07 Apr '09

The documentation repeatedly states that "windows_shared_memory has no kernel persistence and is destroyed when all processes using that shared memory are gone." But I'm trying to find out what exactly is used for the implementation of winows_shared_memory? Is it implemented internally as a Windows named pipe? Is it based on using ReadProcessMemory? Anyone know? Brent Arias Software Architect Lumension Security(tm) Tel ~ (602) 692-8088 Fax ~ (480) 970-6323 brent.arias(a)lumension.com<mailto:brent.arias@lumension.com>

3 2

[Serialize] Serialize just the base class of polymorphic pointers
by Brandon Aubie 07 Apr '09

07 Apr '09

I have a base class BASE that stores data. It has a virtual function (empty in base class) that determines how this data is calculated, but there's no data stored in the derived classes. In my program, I'll load a pluging that is based on BASE and implements the function to calculate the data. Right now I have BOOST_EXPORT_CLASS in the plugin header files and I serialize boost::serialization::base_object<BASE>(*this) in the derived class. This lets me save the serialization to file just fine. When I load it back in though, my application isn't aware of the plugins (I'm just analyzing the data). But I don't need the plugins, since I just need the data. So when I try to load in my serialization, I get an "unregistered class" exception. Here's how it works /*************** CODE ******************/ class BASE { public: vector<double> data; virtual calculate() {}; private: friend class boost::serialization::access; template<class Archive> void serialize(Archive &ar, const unsigned int version) { ar & data; } } class Plugin: public BASE { public: virtual calculate() { /* do stuff */ } private: friend class boost::serialization::access; template<class Archive> void serialize(Archive &ar, const unsigned int version) { ar & boost::serialization::base_object<BASE>(*this); } } BOOST_CLASS_EXPORT(Plugin) /*************** CODE ******************/ Class Plugin is loaded dynamically at runtime via dlopen() on linux. So I have a BASE* that points to an instance of A. When I load in the serialized data though, I don't really care about A and would rather just have a BASE* that points to an instance of BASE. Since there's no data described in A. Can I do this somehow? Thanks

2 1

Re: [Boost-users] [mpl]... is there an mpl::string
by raindog＠macrohmasheen.com 07 Apr '09

07 Apr '09

L$$m$Ll$l$l$ ------Original Message------ From: Eric Niebler Sender: boost-users-bounces(a)lists.boost.org To: boost-users(a)lists.boost.org ReplyTo: boost-users(a)lists.boost.org Sent: Apr 7, 2009 10:21 AM Subject: Re: [Boost-users] [mpl]... is there an mpl::string Noah Roberts wrote: > Eric Niebler wrote: >>>>> Andy Stevenson wrote: >>>>>> >>>>>> I recall some discussion of there being an mpl::string >>>>>> template..... Can't see it in the mpl library. Is it elsewhere? >> >> It has been added to trunk as of revision 52208: >> >> https://svn.boost.org/trac/boost/changeset/52208 >> >> No doubt the regression tests will reveal portability problems. Once >> they have been worked out, we can move this to release. > > Why is this better than a metafunction c_str< Sequence >? > > typedef mpl::vector_c<char, 'h', 'e', 'l', 'l', 'o'> str; > > template < char const* sz > > struct x > {}; > > x< c_str<str>::value > test; > > Is there some reason that's not possible or is prone to problems avoided > by mpl::string? That is certainly a valid design. It's somewhat subjective, but IMO multi-character literals give a nicer compile-time string interface. Consider: // With mpl::vector_c mpl::vector_c<char, 'h','e','l','l','o',' ','w','o','r','l','d'> // With mpl::string mpl::string<'hell','o wo','rld'> Neither will win a beauty contest, but my preference is strongly for the latter. -- Eric Niebler BoostPro Computing http://www.boostpro.com _______________________________________________ Boost-users mailing list Boost-users(a)lists.boost.org http://lists.boost.org/mailman/listinfo.cgi/boost-users Sent from my Verizon Wireless BlackBerry

1 0

[program_options] Specify the same flag multiple times
by KSpam 07 Apr '09

07 Apr '09

I would like to be able to specify a flag multiple times on the command line. For instance, imagine a program that has two levels of verbosity. The first level of verbosity would be activated by specifying "-v" on the command line, and the second level of verbosity would be activated by specifying "-vv" or "-v -v" on the command line. Display verbose print statements: myProgram -v Display very verbose print statements: myProgram -vv In digging through the documentation, it is not evident to me how I can accomplish this with Boost.Program_Options. Any help would be appreciated. Thanks, Justin

1 0

building Boost with -fPIC on x86_64 linux box
by Francois Mauger 07 Apr '09

07 Apr '09

Hi boosters, He is my problem: I have a AMD 64 Linux box with boost installed on it, particularly with the Boost.Python component. Boost version is 1.34.1 (an old one but the same problem occurs with new 1.38.0 as explained below). also gcc version is 3.4.6. I use the standard: ./configure --with-xxxx; make; make install; sequence to build/install the libs. Unfortunately I cannot figure out if the -fPIC is activated while compiling the code... and I don't know how to check the PIC status from the resulting .so files (it is possible and does it make sense? I'm not an expert with this concept). I only know that if I missed the -fPIC flag on this system, I have linker errors. Next step, I have my own 'datatools' library with some classes using Boost stuff. I use plain makefiles for it and I garantee the -fPIC flags is activated to generated my 'libdatatools.so' shared library. Finally, I try to build a Boost.Python wrapper for my classes, trying to link against the 'libdatatools.so' file and 'libboost-xxxx.so' stuff. [Note this procedure works perfectly under 32 bits ubuntu system with gcc 4 and boost 1.34.1] Unfortunately I get this nasty well known error: >>> g++ datatools.cc ... -I.../sw/boost/install-1_38_0-Linux-x86_64-gcc34/include/boost-1_38 -I/usr/local/include/python2.5 -fno-strict-aliasing -DN DEBUG -g -O3 -Wall -I.../sw/datatools/pro/include -I.-shared -fPIC -O -DBOOST_PYTHON_DYNAMIC_LIB -L../Linux-x86_64/lib -ldatatools -lboost_iostreams-gcc34-mt -lboost_filesystem-gcc34-mt -fPIC -L.../sw/boost/install-1_38_0-Linux-x86_64-gcc34/lib -lboost_python-gcc34-mt -lboost_serialization-gcc34-mt -L/usr/local/lib/python2.5/config -lpthread -ldl -lutil -lpython2.5 -lm -o datatools.so /usr/bin/ld: /usr/local/lib/python2.5/config/libpython2.5.a(abstract.o): relocat ion R_X86_64_32 against `a local symbol' can not be used when making a shared object; recompile with -fPIC /usr/local/lib/python2.5/config/libpython2.5.a: could not read symbols: Bad value collect2: ld returned 1 exit status make: *** [datatools.so] Error 1 <<< So what is the problem? I used -fPIC for my classes and I used -fPIC during the Boost-Python wrapping procedure. Does it mean the Boost library itself has not been compiled with -fPIC despite the fact that it is recommended/mandatory with 64 bits architecture. Or does it mean there is something wrong with 'libpython2.5.a' which is provided by the sysadmin and out of my privilege scope? It seems this kind of issue has already been addressed before but scanning the list archives left me rather confused with no clear solution/hint. I tried to enforce the -fPIC flags adding the following line in user-config.jam with boost 1.38.0: >>> using gcc : : : <cxxflags>-fPIC ; <<< But it did not help... I'd really appreciate any hints about this issue. Thanks a lot. regards frc -- François Mauger Département de Physique - Université de Caen Basse-Normandie courriel/e-mail: mauger(a)lpccaen.in2p3.fr tél./phone: 02 31 45 25 12 / (+33) 2 31 45 25 12 fax: 02 31 45 25 49 / (+33) 2 31 45 25 49 Adresse/address: Laboratoire de Physique Corpusculaire de Caen (UMR 6534) ENSICAEN 6, Boulevard du Marechal Juin 14050 CAEN Cedex FRANCE

4 4

libboost_date_time: undefined reference to __sync_fetch_and_add_4
by Carsten Raas 07 Apr '09

07 Apr '09

Hello! I built a (complete) Boost 1.38 library with the Intel compilers 10.1.021, 11.0.073, and 11.0.081 for an EM64T (x86_64) Linux architecture. But I end up with the same problem for all builds: the shared version of libboost_date_time cannot be linked to this simple demo program: # cat conftest.cpp ----------------------------------------------------------------- #include <boost/date_time/date_generators.hpp> int main () { const char *ptr = boost::date_time::nth_as_str(0); return 0; } ----------------------------------------------------------------- # icpc -V Intel(R) C++ Intel(R) 64 Compiler Professional for applications running on Intel(R) 64, Version 11.0 Build 20090131 Package ID: l_cproc_p_11.0.081 ----------------------------------------------------------------- Static linking: # icpc -I /opt/boost/1.38/intel/include conftest.cpp \ /opt/boost/1.38/intel/lib/libboost_date_time.a => Works. ----------------------------------------------------------------- Shared linking: # icpc -I /opt/boost/1.38/intel/include conftest.cpp -L /opt/boost/1.38/intel/lib -lboost_date_time /opt/boost/1.38/intel/lib/libboost_date_time.so: undefined reference to `__sync_fetch_and_add_4' ----------------------------------------------------------------- Shared linking with gcc: # g++ -I /opt/boost/1.38/intel/include conftest.cpp \ -L /opt/boost/1.38/intel/lib -lboost_date_time /opt/boost/1.38/intel/lib/libboost_date_time.so: undefined reference to `__sync_fetch_and_add_4' collect2: ld returned 1 exit status ----------------------------------------------------------------- Shared linking with gcc to libraries compiled with gcc itself: # g++ -I /opt/boost/1.38/gcc/include conftest.cpp \ -L /opt/boost/1.38/gcc/lib -lboost_date_time => Works. ----------------------------------------------------------------- I have been googling around for a while, but all I found was related to GCC sync primitives (e.g. sync_lock_test_and_set, sync_fetch_and_add) and different architectures (armel, sparc, etc.). I also tried the workaround with #define BOOST_SP_USE_PTHREADS in boost/config/user.hpp, without success. I also checked 1.37 and today's SVN version, but they show the same problems. Any hints what I should look at next? Best regards, -Carsten

4 9

[threads] recursive shared_mutex
by joaquin＠tid.es 07 Apr '09

07 Apr '09

Hi, As the docs do not explicitly say otherwise, I guess I must assume that boost::shared_mutex is not recursive with respect to unique ownership, that is, the following behavior ensues: shared_mutex mutex; lock_guard<shared_mutex> lock1(mutex); lock_guard<shared_mutex> lock2(mutex); // DEADLOCK Is this so? In case boost::shared_mutex is indeed non-recursive, I've read the following assertion on Hinnant's paper N2406 (http://www.open-std.org/jtc1/sc22/wg21/docs/papers/2007/n2406.html ): "A recursive shared mutex could be built upon shared_mutex if desired, but this is not proposed." Does anyone has a reference on how to construct a recursive shared_mutex from a non-recursive one? Thank you, Joaquín M López Muñoz Telefónica, Investigación y Desarrollo

2 1

Boost, GCC, TR1 question
by Anthony Foglia 07 Apr '09

07 Apr '09

We've stumbled across an incompatiblity in mixing TR1 stuff from GCC 4.3 and Boost. The smallest test case is simply #include <tr1/array> #include <boost/tr1/tuple.hpp> This fails because GCC's array declares and specializes the GCC's tuple_size and tuple_element, while boost/tr1/tuple.hpp by default uses ::boost::fusion::tuple_size and tuple_element. It appears that defining BOOST_HAS_GCC_TR1 before including the boost libraries will solve the problem. Is there a better way? More importantly, this is code in part of library used by a number of other projects. Will we have to recompile the boost libraries (we're using the current Ubuntu package), and/or any of our projects that link against this code, or should linkage not be a problem? -- Anthony Foglia Princeton Consultants (609) 987-8787 x233

2 1

Boost.spirit: how to handle file parse errors?
by Joost Kraaijeveld 07 Apr '09

07 Apr '09

Hi, I am parsing a large file and experience errors. Is there an example available somewhere that shows how to print something like "error at file, line, mismatched string" somehow? TIA -- Groeten, Joost Kraaijeveld Askesis B.V. Molukkenstraat 14 6524NB Nijmegen tel: 024-3888063 / 06-51855277 fax: 024-3608416 web: www.askesis.nl

1 0

Selecting among different MPI versions
by Alain Leblanc 07 Apr '09

07 Apr '09

Hi, I'm trying to build boost 1.38 with MPI support. The problem is that my system has more than one version of MPI installed, and the boost build uses the wrong one. I don't know how boost decides which one to use. The following are the only default defined by gcc, and none of them contains mpi.h g++ -Wp,-v -E - ignoring nonexistent directory "/usr/lib/gcc/x86_64-redhat-linux/4.1.1/../../../../x86_64-redhat-linux/include" #include "..." search starts here: #include <...> search starts here: /usr/local/include /usr/lib/gcc/x86_64-redhat-linux/4.1.1/include /usr/include End of search list. When I try building boost, I get a bunch of messages of the form ...failed gcc.compile.c++ bin.v2/libs/mpi/build/gcc-4.1.1/release/threading-multi/graph_communicator.o... gcc.compile.c++ bin.v2/libs/mpi/build/gcc-4.1.1/release/threading-multi/group.o g++: /opt/intel/mpi-rt/3.1: No such file or directory g++: unrecognized option '-rpath' In file included from ./boost/mpi/config.hpp:20, from ./boost/mpi/exception.hpp:16, from ./boost/mpi/group.hpp:18, from libs/mpi/src/group.cpp:9: /usr/local/intel/impi/3.1/include/mpi.h:35:8: error: #error The wrong version of mpi.h file was included, check include path. and the version of MPI I want to use is in /usr/local/mpich/... So is there some option to tell boost which mpi.h to include. I tried setting the CPLUS_INCLUDE_PATH variable, but it didn't change anything. I have no control over the OS itself. Thanks, a

1 0

Intrusive list/ilist and rbtree marshalling support
by Brian Wood 07 Apr '09

07 Apr '09

I'm happy to announce that version 1.8 of the C++ Middleware Writer is available on line -- http://webEbenezer.net/cgi-bin/samb.cgi<http://webebenezer.net/cgi-bin/samb.cgi>. This version adds support for the following: * Exclusive marshalling support of boost::intrusive::list and boost::intrusive::rbtree. Neither Boost Intrusive nor Boost Serialization offer serialization support of the Intrusive containers. Rather than having two types called 'list,' we refer to boost::intrusive::list as an ilist. An example of this support is here -- http://webEbenezer.net/rbtree_marshalling.html<http://webebenezer.net/rbtree_marshalling.html>. * Several new options on the user interface including: + An "Error Handling" option that indicates whether the code should be written using return codes or exceptions. The exception support is new in this release. This option affects the signatures of the generated marshalling functions. + A "Code Structure" option that indicates whether the code should be written in header-only format or separated interface and implementation files. The separated interface and implementation support is new in this release. This option also affects the signatures of the generated marshalling functions. + A choice between unsigned short or unsigned int as the type for constants written by the C++ Middleware Writer. Constants that identify messages and user defined types are written by the software. Previously the constants were always unsigned ints. * A mechanism to turn off automatic code generation for a type. Adding a // comment within a class declaration with "hand_written_marshalling_code" in the text of the comment will disable default code generation for that type. For example, class bag { // hand_written_marshalling_code ... }; As the comment indicates, you will provide the implementation of the marshalling functions for this class. This option is not available with /* */ comments. Additionally, I would like to say that I find the Intrusive library to be sound and am using it internally in the C++ Middleware Writer. Thanks to the authors: Olaf Krzikalla and Ion Gaztañaga. Brian Wood Ebenezer Enterprises www.webEbenezer.net <http://www.webebenezer.net/> "Then Samuel took a rock and set it up between Mizpah and Shen. He named it Ebenezer [Rock of Help] and said, 'Until now the L-RD has helped us.'"

1 0

Matching constant input with input streams
by Eugene Wee 06 Apr '09

06 Apr '09

Hi, A fellow user on an online forum that I frequent asked about the use of C++ input streams to replace the use of scanf() and family where the format string included constant portions, e.g., the "a=" in "a=%d". The answer was to read the input as a string and then parse it. Another user chimed in to say that he preferred sscanf() for this reason, and that C++ input streams would be more useful if such constant portions could be easily matched. I wrote a manipulator that did just that as a proof of concept, but now I wonder if Boost already provides such a manipulator, or if say, Boost.Spirit with its file_iterator is deemed as the right solution. There certainly seems to be a trade-off between these two choices: a manipulator would be a simple add-on to existing standard library functionality, but Boost.Spirit is much more powerful. Thank you, Eugene Wee

1 0

[range_ex] range_ex conflicts with non_copyable
by 041021042 06 Apr '09

06 Apr '09

Hi， The following code works well with old version of range_ex (Dec 2008), but when I update range_ex to an newer version (after Feb 2009), the code cannot compile. code: #include <boost/range/algorithm.hpp> #include <boost/ptr_container/ptr_vector.hpp> struct A : boost::noncopyable { int a; A(int _a) : a(_a) {} bool operator == (const A& r) const { return a == r.a; } }; int main() { boost::ptr_vector<A> pva; A a(0); std::find(pva.begin(), pva.end(), a); // ok boost::find(pva, a); // cause an error } compiler: msvc2008 error message: e:\zxg\boost\range_ex\range.cpp(8) : error C2248: 'boost::noncopyable_::noncopyable::noncopyable' : cannot access private member declared in class 'boost::noncopyable_::noncopyable' d:\software\boost_1_37_0\boost\noncopyable.hpp(27) : see declaration of 'boost::noncopyable_::noncopyable::noncopyable' d:\software\boost_1_37_0\boost\noncopyable.hpp(22) : see declaration of 'boost::noncopyable_::noncopyable' This diagnostic occurred in the compiler generated function 'A::A(const A &)' Thanks, Roger Zhou

5 5

Reuse of boost config and other related headers
by Seweryn Habdank-Wojewódzki 06 Apr '09

06 Apr '09

Dears, How could I reuse boost config and neighbour to prepare platform independent code for dll"s (.dll and .so)? Best regards, -- |\/\/| Seweryn Habdank-Wojewódzki \/\/ Professionalism in programming - www.accu.org ---------------------------------------------------- Hospicja w całej Polsce pomagają osobom umierającym i ich bliskim. Ty też możesz pomóc. Przekaż 1% podatku Fundacji Hospicyjnej (KRS 0000201002). http://klik.wp.pl/?adr=www.httz.pl&sid=685

1 0

Regarding the plod_iterator
by ashish deshpande 06 Apr '09

06 Apr '09

Hello all, I am using the plod_iterator() from the Boost Graph Library for generating a scale-free graph. Actually, I need the generated edges to be distinct. However, I'm getting repeated edges in the generated graph. Is there a way to make sure the edges of the generated scale-free graph are distinct. Thanks, Ashish

2 1

[filesystem] Making .zip files appear as directories
by Adam Nielsen 05 Apr '09

05 Apr '09

Hi all, I'm working on a library that manipulates file archives (although it won't work with .zip or .lha files, they're a perfect example of the types of files it will work with.) I'm wondering whether a good approach would be to extend the Boost Filesystem library so that it can open these files as if they were folders. That would allow the users of this library to work with the contents of the archive files in the same way they might work with "real" files in a directory on the local system. Can the Boost Filesystem library be extended in this manner? If so, does anyone have any pointers as to what classes would need to be implemented to achieve something like this? Would it still remain somewhat independent of the underlying filesystem code, so that it would still be cross-platform? Many thanks, Adam.

2 1

success of a multi-index container insertion
by Goran 04 Apr '09

04 Apr '09

Hello all! As I need to know if a insertion in my multi-index container succeded or not I've asked me and now you: How can I get a bool from inserting a new object? My problem: MyMultiIndex mmi; MyClass mc0; MyClass mc1 = mc0; mmi.insert(mc0); // mc0 inserted mmi.insert(mc1); // mc1 not inserted Thanks, Goran

2 1

asio/thread on vxWorks
by Dustin Spicuzza 04 Apr '09

04 Apr '09

Hey, We're using vxWorks 6.3, on a PPC target using GCC 3.4.x. I want to use boost asio and boost thread, but I'm running into various errors when compiling. Has anyone else been able to get these working under vxWorks correctly (or at least, compiling)? Thanks. Dustin PS: Yes, I know that in order to help with the errors, I should post them. I'm just looking to see if anyone else has been successful before I start patching the various non-compatibilities that I've seen so far... -- Innovation is just a problem away

1 0

add_const_ref type transformation metafct
by Hicham Mouline 04 Apr '09

04 Apr '09

Hello, I wish to have a metafct that adds "const &" to types that have it or don't have it... template< typename T > struct add_const_ref { typedef typename boost::remove_const<T>::type t1; typedef typename boost::add_const<t1>::type t2; typedef typename boost::add_reference<t2>::type type; }; Does this exist already in type traits or elsewhere? I couldn't find.... Can this be improved in terms of reducing the number of tmpl instantiations? Note : add_const_ref< const double& >::type is const double& for e.g. Rds,

3 2

issue using expat 2.0.1 with boost
by Sean Farrow 04 Apr '09

04 Apr '09

Hi: I am having an issue when trying to utilize the graphml library with the expat 2.0.1 library. The library fails to compaile saying it carn't find expat.h. My expat_include variable is set to: "c:\program files\expat\source\lib" and my expat_libpath is set to: "c:\program files\expat\bin\" Any help appreciated. Cheers Sean.

2 1

Re: [Boost-users] boost::transform_iterator, boost::array::iterator and category
by Hicham Mouline 04 Apr '09

04 Apr '09

>Try returning a reference const double& instead of double. >struct UnaryFunction { > typedef const double& result_type; > const double& operator()( const pair<double, double>& e ) const > { > return e.first; > } >}; >Random access iterators must return an lvalue. >In Christ, >Steven Watanabe Thank you, the transformed iterator's category now matches std::random_access_iterator_tag. So in general, if I am passed a iterator (regardless whether it has been transformed by transform_iterator<> or not), the most generic test for random access category in C++03 __is__ against std::random_access_iterator_tag, __not__ against boost::random_access_traversal_tag. The latter is some implementation detail for the Boost.Iterator library, right? Thanks again,

1 0

Thread safe intrusive ptr
by Jared Lee Richardson 04 Apr '09

04 Apr '09

I'm in the process of changing some of my shared_ptr's to intrusive_ptr's, and I'm trying to see how difficult it will be to make it thread-safe. I need intrusive_ptr's mostly due to size constraints(allocating an extremely large number of small objects), and also the ThisPtr issue. I tried to search and find any information about making them thread-safe, but what I found didn't explain the solution or the issues. What are the considerations for making intrusive_ptr thread safe? Is it as simple as a mutex-lock / atomic counter variable? Thanks, Jared

3 2

Re: [Boost-users] Help using math functions
by Mario Chacon 04 Apr '09

04 Apr '09

Great!... This physics engine it's amazing!! I'm doing my first examples and works excellent! Thank you so much for your recommendation!! salu2... masch... On Fri, Apr 3, 2009 at 3:04 PM, Hostile Fork <hostilefork(a)gmail.com> wrote: > I doubt boost::math is high level enough for what you want to do. > > You probably want a physics engine. > http://en.wikipedia.org/wiki/Physics_engine > Haven't used any so I don't know which is best in the C++ world. But > there's at least one popular free one: Box2D > > http://cplus.about.com/b/2007/11/19/box-2d-online-c-physics-engine-demo.htm > > On Apr 3, 2009, at 9:35 AM, Mario Chacon wrote: > > Hello: > First at all I do not know about math, but I need to do some animations. I > need to simulate a movement of a ball into a pipe. I have all the interface > working, but now I have to do the best way to give it life..:D. My fists > steps was giving X,Y coordinates but I do not like this way, so I need some > tips to make it more professional way. How can I use boost math functions to > make it? > > Thank you.. > masch... > ** > _______________________________________________ > Boost-users mailing list > Boost-users(a)lists.boost.org > http://lists.boost.org/mailman/listinfo.cgi/boost-users > > >

1 0

[Proto] lazy transform is not working
by Dave Jenkins 03 Apr '09

03 Apr '09

In the program below, I expected the proto::lazy transform to call the times2() function, but it doesn't. It does work correctly if you substitute times2<int> for proto::_state. Is this a bug or am I forgetting something? Thanks, Dave Jenkins #include <iostream> #include <boost/proto/proto.hpp> #include <boost/proto/transform.hpp> namespace proto = boost::proto; template<typename T, typename Callable = proto::callable> struct times2 { typedef T result_type; T operator()(T i) const { std::cout << "Called times2()\n"; return i * 2; } }; struct IntTimes2 : proto::when< proto::terminal<proto::_> , proto::lazy<proto::_state(proto::_value) > //, proto::lazy<times2<int>(proto::_value) > > {}; int main() { int dummy = 0; proto::terminal<int>::type i = {1}; IntTimes2()(i, times2<int>(), dummy); }

3 4

boost::transform_iterator, boost::array::iterator and category
by Hicham Mouline 03 Apr '09

03 Apr '09

Hello, I have boost::array< pair<double, double>, 16 >::iterator (it has iterator_traits<>::iterator_category random_access_iterator_tag) which I have transformed with typedef transform_iterator< UnaryFunction, ... > iteratorIntoFirst; struct UnaryFunction: public std::unary_function< pair<double, double>, double> { double operator()( const pair<double, double>& e ) const { return e.first; } } However on asserting the category is still random access, boost static assert fails: BOOST_STATIC_ASSERT(( boost::is_convertible< iteratorIntoFirst, std::random_access_iterator_tag >::value )); >From the docs http://www.boost.org/doc/libs/1_38_0/libs/iterator/doc/transform_iterator.ht ml If Iterator models Readable Lvalue Iterator and if Iterator models "Random Access Traversal Iterator", then iterator_category is convertible to random_access_iterator_tag. Any ideas, Regards,

2 1

[serialization] Multiply defined symbol: guid_initializer
by Robert Dailey 03 Apr '09

03 Apr '09

Hi, I'm currently trying to polymorphically serialize a class without using RTTI. I'm doing this currently in a file called StaticText.hpp: #ifndef RS_GUI_WIDGETS_STATICTEXT_HPP #define RS_GUI_WIDGETS_STATICTEXT_HPP #include <boost/scoped_ptr.hpp> #include <boost/serialization/access.hpp> #include <boost/serialization/base_object.hpp> #include <boost/serialization/string.hpp> #include <boost/serialization/extended_type_info_no_rtti.hpp> #include <boost/serialization/export.hpp> #include <rs/gui/widgets/Widget.hpp> #include <rs/gui/Text.hpp> namespace rs { class StaticText : public Widget { public: StaticText( std::string const& text ); void Update(); void Render( RenderSystem& renderer, Camera& camera ); char const* get_key() const; private: StaticText() {} template< class Archive > void serialize( Archive& archive, unsigned int version ) { archive & boost::serialization::base_object<Widget>( *this ); std::string text; archive & text; m_text.reset( new Text( text ) ); } boost::scoped_ptr<Text> m_text; friend class boost::serialization::access; }; } BOOST_CLASS_TYPE_INFO( rs::StaticText, extended_type_info_no_rtti<rs::StaticText> ) BOOST_CLASS_EXPORT( rs::StaticText ) #endif // RS_GUI_WIDGETS_STATICTEXT_HPP When I compile this into a LIB first, and then link the lib into an EXE, the linker says this (Visual Studio 2008): *gui.lib(StaticText.obj) : error LNK2005: "public: static struct boost::archive::detail::guid_initializer<class rs::StaticText> const & const boost::archive::detail::init_guid<class rs::StaticText>::guid_initializer" (?guid_initializer@?$init_guid@VStaticText@rs@@@detail@archive@boost@ @2ABU?$guid_initializer@VStaticText@rs@@@234@B) already defined in main.obj* * * If I move the 2 macro calls into the StaticText.cpp file, everything compiles/links just fine but I end up getting an exception thrown at runtime that says "unregistered void cast".

3 16

[ASIO] Windows native type for ASIO serial port?
by Jeff Diewald 03 Apr '09

03 Apr '09

I have a boost::asio::serial_port object. What is the specific Windows type returned by native()? I'm having trouble with my serial communication across the serial port, and want to get more diagnostic information by going to the native level. The serial port hangs occasionally on a synchronous read_some call, and I haven't been able to figure out why. Ideally, I'd like to build a timeout into the synchronous call. Any help is appreciated, Jeff Diewald JEOL USA

2 3

Re: [Boost-users] boost::transform_iterator, boost::array::iterator and category
by Hicham Mouline 03 Apr '09

03 Apr '09

>Hello >Following my previous post, >the transformed iterator has input iterator tag (not forward, not bidir, not random) >BOOST_STATIC_ASSERT(( > boost::is_convertible< > iteratorIntoFirst, > std::input_iterator_tag >>::value )); >But it should be random_access... Is there anything extra I need to put in the Unary fct of >transform_iterator? >Rds, There seems to be some boost::random_access_traversal_tag, and the static assert works with this tag??? Why isn't boost::random_access_traversal_tag equivalent to std::random_access_iterator_tag ? Rds,

1 0

Help using math functions
by Mario Chacon 03 Apr '09

03 Apr '09

Hello: First at all I do not know about math, but I need to do some animations. I need to simulate a movement of a ball into a pipe. I have all the interface working, but now I have to do the best way to give it life..:D. My fists steps was giving X,Y coordinates but I do not like this way, so I need some tips to make it more professional way. How can I use boost math functions to make it? Thank you.. masch... **

1 0

Re: [Boost-users] boost::transform_iterator, boost::array::iterator and category
by Hicham Mouline 03 Apr '09

03 Apr '09

Hello Following my previous post, the transformed iterator has input iterator tag (not forward, not bidir, not random) BOOST_STATIC_ASSERT(( boost::is_convertible< iteratorIntoFirst, std::input_iterator_tag >::value )); But it should be random_access... Is there anything extra I need to put in the Unary fct of transform_iterator? Rds,

1 0

[Multi-Index] crash on invariant_() violation
by Neil Hunt 03 Apr '09

03 Apr '09

> > This is definitely going to break the container's invariants, yes. Key > modification has to be done via modify() so as to give the container a > chance to reindex elements. So if you are quite sure this is happening > try to remove the offending behavior; after this my bet is that the > invariant assertion will go away. Otherwise please get back here. > > Just weighing in here with a related question. But first to say: I find multi-index to be invaluable. So thanks for a great library. In my code, the objects I am indexing against are altered elsewhere, and then my index owning code is notified. At this point, the index owning code calls: index.modify( index_iterator, NoOperationFunctor() ); That is, on the assumption that the modify function has no knowledge of what changes the functor are making, it must re-evaluate all indices. Therefore, it doesn't matter if the change is made inside the modify function as long as modify is called. (So, this statement from above does not appear strictly true: "Key modification has to be done via modify() so as to give the container a chance to reindex elements.") Because (in my case) the change is made elsewhere, I am calling modify with a functor that does nothing. In my particular case, I can guarantee that change made elsewhere won't affect the index that I use to find the iterator to pass to modify. This may seem a weird scenario, but it also seems weird that I have to call modify with a no-op functor. Is it possible/desirable to add a simple update_indices( itor ) function? Neil

2 1

question about using boost.function and boost.bind, V1.38
by Tan, Tom (Shanghai) 03 Apr '09

03 Apr '09

Can any one please explain to me why the former fails while the latter compilesThe definition: // usage in a ctor initialisation list ..., where m_button_monitor is of type ButtonMonitor, , m_button_monitor(boost::bind(boost::mem_fn(&CView::OnRun), this, ID_TOOLBUTTON_AUTO)) //definition 1 of ButtonMonitor that fails // the error message by MS VC9SP1: // error C2664: 'programmer_board::ButtonMonitor::ButtonMonitor(boost::function<Signatur e> &)' : //cannot convert parameter 1 from 'boost::_bi::bind_t<R,F,L>' to 'boost::function<Signature> &' // class ButtonMonitor { public: ButtonMonitor(boost::function<void()>& f) : m_f(f) { } void UpdateStatus(BUTTON_STATUS status) { m_f(); }; private: boost::function<void()>& m_f; }; //definition 2 of ButtonMonitor that works class ButtonMonitor { public: ButtonMonitor(const boost::function<void()>& f) : m_f(f) { } void UpdateStatus(BUTTON_STATUS status) { m_f(); }; private: const boost::function<void()>& m_f; }; Thanks,

3 2

[Q] eliminating false memory leak reports
by V S P 03 Apr '09

03 Apr '09

Hello, I have a class (below), and MS memory leak detector reports a leak on ptr.reset I have redefined 'new' as follows: #define MYDEBUG_NEW new( _NORMAL_BLOCK, __FILE__, __LINE__) In another example, also below I have an _impl structure that is allocated and assign to typedef boost::shared_ptr<cmoney_impl_dtl> t_money_storage; My understanding is that it is a 'false' detection is that correct? In otherwords, in my thinking -- as long as I have assigned an object to boost::shared_ptr -- I do not have remember anything about it anymore. (using visual studio 2008, 32 bit) Another question, since I have all of my heap objects using boost::shared_ptr is there any free tool that I can use to verify that those things do not leak (as some of the are actually DB connections so it is important for them to get completely destroyed). Thank you in advance ---------------------------- template <typename TThrdData> struct CService_ThreadData:private boost::noncopyable { CService_ThreadData (void){}; ~CService_ThreadData (void){}; ///creates a new thread data object if does not exists //Call this from every thread to establish thread's data //manager object static TThrdData& getdata(void) { /* YES, using static variable in this case is safe, because it is in the 'global' scope and therefore is intialized before Main starts Otherwise, static variables are not thread safe */ if (ptr.get() == 0) { ptr.reset(new TThrdData()); } return (TThrdData&)(*ptr); } private: static boost::thread_specific_ptr<TThrdData> ptr; };//end of class -------------------------- cmoney_t::cmoney_t (void) :v(new cmoney_impl_dtl) { } -- V S P toreason(a)fastmail.fm -- http://www.fastmail.fm - Access your email from home and the web

2 1

date_time question
by Amadeus W.M. 03 Apr '09

03 Apr '09

The program below reads a date from a string -> istringstream -> ptime. The string contains comma separated fields, the first of which is a date in the format %Y-%m-%d (without %H:%M:%S): 2009-04-03,12.34,56.78 If I define a facet with in the format %Y-%m-%d %H:%M:%S then the program outputs 2009-Apr-03 12:34:56 i.e. it interprets the numbers following the date as %H:%M:%S even though they are separated by commas and dots, rather than colons. Is this normal behaviour? Would it be possible to make it leave alone the fields following the date? That is, do not interpret 12.34,56.78 as %H:%M:%S. Thanks! // g++ -g -Wall -o dateIOSimgleFmt -I/usr/local/include/boost-1_38 dateIOSingleFmt.C #include <iostream> #include <sstream> #include <cstdlib> #include <string> #include <boost/date_time/posix_time/posix_time.hpp> using namespace std; using namespace boost::posix_time; int main(int argc, char * argv[]) { ptime pt(not_a_date_time); string datestr("2009-04-03,12.34,56.78"); const char * format="%Y-%m-%d %H:%M:%S"; istringstream INSTR(datestr); locale loc(INSTR.getloc(),new time_input_facet(format)); INSTR.imbue(loc); INSTR>>pt; cout << pt << endl; return 0; }

1 0

[asio] Setting UDP Source Port
by Scott McMurray 03 Apr '09

03 Apr '09

I'm writing an application that makes use of the source port in the UDP header. The only way I've found that sets it is by setting the endpoint on construction of the udp::socket. The problem with that, however, is that I have 3 different components that want to use that port, one listening and two sending. Understandably, I cannot bind the same port for all three. (Incidentally, while binding with the udp::socket constructor works, the bind call gives a bad file descriptor exception. I'm using 1.35, though, so that may have been fixed.) Is there some way to set the source port without binding? Since there's no session state, it seems like I should be able to send packets from any number of sockets. (I can't just share a socket, as they're not shared-objects thread-safe, according to the documentation.) Thanks, ~ Scott McMurray

2 1

Re: [Boost-users] [Multi-Index] crash on invariant_() violation
by Neil Hunt 03 Apr '09

03 Apr '09

> > ... it doesn't matter if the change is made inside the modify function > as long as modify is called. > > (So, this statement from above does not appear strictly true: > "Key modification has to be done via modify() so as to give the container a > chance to reindex elements.") > > Because (in my case) the change is made elsewhere, I am calling modify with > > a functor that does nothing. > I guess more generally, why is this paradigm implictly disallowed? index_itor->value++; index.update_indices( index_itor ); While I can see the benefit of tying the modification to the multi-index object via modify, there is no explicit reason that I can see why the above is invalid (as per my perfectly working code that uses this paradigm). And "forcing" the user to use a functor to modify can introduce annoying code overhead. Don't get me wrong, I think that using modify with a functor is very clear and helps prevent invalid indices. But, at the same time, it can be a pain to write a functor. (Here's to C++0x lambda's). But in my case, where the code that modifies the object is unaware of multi-indices on that object, it is not feasible to use modify() to make the change. Neil

1 0

[serialization] Private default constructors
by Robert Dailey 03 Apr '09

03 Apr '09

I have a class that can be constructed statically in code, or constructed by data (this is where boost::serialization comes in). For example: class foo { public: foo() {} foo( std::string text ) : m_text( text ) {} private: template< class Archive > void serialize( Archive& archive, unsigned int version ) { archive & m_text; } std::string m_text; friend class boost::serialization::access; }; There is one problem with this, though. It is possible for a user of this class to create a "foo" with the default constructor, thus enabling them to have "zombie" objects. The only thing I want using the default constructor is boost::serialization, not the user. Is there a way I can make the default constructor private and still have boost::serialization access it? Keep in mind that this must work polymorphically too.

1 1

[repost][filesystem][system] Linking errors with C++0X options on intel C++ 11.0
by jesseperla＠gmail.com 02 Apr '09

02 Apr '09

Sorry to repost, but I am at a loss and was hoping I have missed something. I am on Intel C++ 11.0 on windows. I built boost::filesystem and boost::system as static libraries with the following line in my userconfig.jam using intel-win : 11.0 : : <cxxflags>/Qstd:c++0x ; I am then compiling a project (that works with non-C++0x just fine) with /Qstd:c++0x in the command line. But when I try to link it with my calls to boost::filesystem header functions, I get the following link errors (it must be using auto linking) 1>etk_template.obj : error LNK2019: unresolved external symbol "class boost::system::error_category const & __cdecl boost::system::get_system_category(void)" (?get_system_category@system@boost@@YAABVerror_category@12@XZ) referenced in function _main 1>etk_template2.obj : error LNK2001: unresolved external symbol "class boost::system::error_category const & __cdecl boost::system::get_system_category(void)" (?get_system_category@system@boost@@YAABVerror_category@12@XZ) 1>etk_template.obj : error LNK2019: unresolved external symbol "class boost::system::error_category const & __cdecl boost::system::get_generic_category(void)" (?get_generic_category@system@boost@@YAABVerror_category@12@XZ) referenced in function _main 1>etk_template2.obj : error LNK2001: unresolved external symbol "class boost::system::error_category const & __cdecl boost::system::get_generic_category(void)" (?get_generic_category@system@boost@@YAABVerror_category@12@XZ) I am not changing any pre-compiler definitions for boost when I use /Qstd:c++0x. Should I be changing anything there? Am I out of luck? -Jesse

4 3

[BGL] A bug in the edge_range() function.
by Michael Olea 02 Apr '09

02 Apr '09

In boost 1.37.0 there is a "metaprogramming bug" in the edge_range() function: it fails to compile when the Directed Selector template argument to adjacency_list is set to directedS, but does work if the argument is either undirectedS, or bidirectionalS. Attached below is a short program illustrating the problem. The problem is that edge_range() instantiates a StoredEdge via a constructor taking 3 arguments, but when the Directed Selector is directedS StoredEdge is typedef'ed to stored_edge_property<vertex_descriptor, EdgeProperty>, which has no such constructor. One solution might be to create overloaded edge_range_dispatch() functions, and dispatch on Config::on_edge_storage. -- Michael

1 0

[serialization] Serializing "int" types...
by Robert Dailey 02 Apr '09

02 Apr '09

If I choose to serialize a variable with type "int", is this portable? For example, if I stream out this value on a 64-bit system, and then load it in on a 32-bit system, will this result in a buffer-overrun or anything else weird? I read the serialization docs but didn't see this topic in there. Sorry if I missed it. Help is appreciated. Thank you.

2 2

[asio] async_accept shares port?
by Roman Shmelev 02 Apr '09

02 Apr '09

HI! I've got another strange question, hope it is not very dumb. I'm launching two programs on my computer. Each tcp_acceptor in each program calls async_accept on one port and so is ready to accept connections. When I try to connect to the predefined port, I am connecting to the app that called async_accept later. As I thought, when trying to listen to port that is already accepting connections - another program will receive some error_code, but I do not see this in my case. Where is the mistake? Thanks!

1 0

[boost-users] lexical-cast throws exceptions
by Roman Shmelev 02 Apr '09

02 Apr '09

Please take a look at the simple piece of code. When it is compiled on MSVC 2008 (no sp) , Win Sp2 then we receive access violation on program exit Can you tell me the reason of such strange behaviour? (boost 1.38.0) #include "stdafx.h" #include <iostream> #include <boost/lexical_cast.hpp> //#include <boost/thread.hpp> using namespace std; class ctest { public: ~ctest () { boost::lexical_cast<string>(1); } }; ctest a; int main(int argc, char* argv[]) { boost::lexical_cast<string>(1); return 0; }

2 1

[Multi-Index] crash on invariant_() violation
by Dominique Devienne 02 Apr '09

02 Apr '09

We've recently introduced a boost multi_index_container into our code base. Now another seemingly unrelated change triggers an invariant violation in the bmi code during a unit test. I've tried to look at what kind of invariant is checked in the code, but got lost in its intricacies... Could I please get some pointers on what I should specifically look for, and maybe what invariant violations could be triggered from (or have been in the past)? Assertion failed: invariant_(), file c:\...\boost-1.38.0-debug\boost\multi_index_container.hpp, line 683 I don't have all the details of the crash yet, but the invariant is violated either during a call to empty() on the implicit index of the container (a unique_hashed index), or a call to erase() on that same index. Note that the indices use BOOST_MULTI_INDEX_CONST_MEM_FUN on a proxy object, which extracts the indexed value on the proxy-wrapped pointer. It's quite possible our code base makes a change to one of the indexed value without notifying the bmi. Is this something that would trigger an invariant violation? (it's on my todo list to avoid having something like this happening). Any insights in troubleshooting the above would be appreciated. Thanks, --DD

2 1

Re: [Boost-users] boost::array < pair<>... > iterator adaptor
by Hicham Mouline 02 Apr '09

02 Apr '09

>AMDG >Hicham Mouline wrote: >> I have a boost::array< pair<type1, type2>, maxsize > object, and I wish to >> iterate over the list of type1 elements, and the list of type2 elements... >> >> I gather I then need an iterator adaptor based on array's iterator and >> const_iterators... >> How may I write such a thing? >> >> I then can pass around iterators to type1 and type2 separately. >> >transform_iterator might be useful. I don't know off >the top of my head whether there is a more specialized >iterator adaptor in Boost somewhere. Great, this worked beautifully struct FirstElem : public std::unary_function< pair<>, double > { double operator()( const pair<>& p ) { return p.first; } }; typedef boost::transform_iterator< FirstElem, IterType > FirstElemIterType; FirstElemIterType feit ( container.begin() ); I can then pass feit around. Now, under normal assumptions, accessing first elements through feit should be as fast as direct access by manual loop.... inline calls should be optimized away? Thanks,

2 1

boost::array and operator< on private type (which is a pair) for using std algos
by Hicham Mouline 02 Apr '09

02 Apr '09

Hello, I'm not sure this is specific about boost::array... I apologize if it is not boost specific. I have this code http://codepad.org/h9uFoI1I This compiles under g++4.0 1. I find it strange that it does given that EntryType is a __private__ type inside C<T1,T2>. 2. It doesn't do what I expect: my operator< is defined in std namespace but is never called (i did this because with std::pair comes < operators frm the std namespace) I don't want to define operator< for all pairs, just for contexts (member functions of C) which use EntryType. Is this because I typedef EntryType? rds,

1 0

new idea for project
by divyanshu singh 02 Apr '09

02 Apr '09

HI all I am new to boost and i found its math function library really cool I like to play with numbers specially prime numbers I am thinking of developing a new library for boost that can generate various prime patterns adn will include other new algorithms related to prime numbers like aks algorithm to test whether a number is prime,Prime spiral algorithm,prime number generation algorithm This library can help people who love numbers and want to find new prime or number patterns first i would like to know that whether such library already exists in boost if not then hows the idea?

2 1

Fusion usage: `at_c' is not a member of `boost::fusion'
by Mauricio Gomes 02 Apr '09

02 Apr '09

Hi, boost::fusion::vector<int> v(1); int x = boost::fusion::at_c<0>(v); This code is giving me the error: `at_c' is not a member of `boost::fusion'| The first line compiles fine. I have all the includes that are showed in fusion::vector documentation page. Which include (or anything else) am I missing ? Thank you for your attention, -- Mauricio Gomes Pensar Digital 55-11-8319-9610 (Brazil - mobile)

2 1

[boost-users][bind, lambda] delayed assignment
by Igor R 02 Apr '09

02 Apr '09

Hello, I'd like to bild on-the-fly functor that would do something like: a_ = 1, b_ = 2; // a_, b_ - members But the assignment should be delayed until the functor is invoked. How can I do this with bind/lamda? Thanks!

2 1

[Asio] Version correspondence?
by Rush Manbert 01 Apr '09

01 Apr '09

I'm trying to figure out which standalone Asio library versions correspond to the Boost library versions for: 1.35.0 1.36.0 1.37.0 1.38.0 Is it possible to discover this from the header files? Is there some other way? I'm adding Asio-based classes to an existing library. I think I want to try and implement them using the version that was in Boost 1.35.0, but I would like to give library users the option to use the Boost Asio or the corresponding stand-alone Asio. I expect some users would appreciate that. That's why I need to know what standalone Asio version to get. - Rush

1 0

[Graph] Checking if a vertex_descriptor is valid
by Florian.PONROY＠fr.thalesgroup.com 01 Apr '09

01 Apr '09

Hi, I want to check if a vertex_descriptor is valid before using it. As I saw on this list, I perform the following test : Graph_t graph(10); vertexDescriptor srcVertex = boost::vertex(424242, graph); if (Graph_t::null_vertex() == boost::vertex(424242, graph)) { return; } Obviously, vertex indice 424242 does not exist, but test fails, and function does not return, leading the a crash of the following algorithm (BFS or Dijkstra). What did I do wrong? Thanks for your help. -- Florian PONROY Thales Land & Joint France Tel. : +33(0)1 41 304 363 Fax : +33(0)1 41 303 560 Email : florian.ponroy(a)fr.thalesgroup.com

3 2

[interprocess] access dynamically accessed shared region in another process
by Stephan Menzel 01 Apr '09

01 Apr '09

Hi there, I'm trying to do IPC via boost.interprocess and use a dynamic buffer. This is on windows, debug mt, boost 1.38. To do this, I first create a shared segment and then a dynamic area in it. like this: struct transfer_buffer { void *buf; size_t bufsize; boost::interprocess::interprocess_mutex mutex; size_t dmember; } typedef boost::interprocess::managed_windows_shared_memory shared_segment; ... using namespace boost::interprocess; shared_memory_object::remove(bufname); struct transfer_buffer *ret = 0; shared_segment_ptr seg;// I use a ptr to store the object away so it won't get lost seg.reset(new shared_segment(create_only, bufname, begin_size)); // and our maintanance struct in it ret = seg->construct<struct transfer_buffer>(unique_instance)(); ret->dmember = 42; const std::size_t Alignment = 128; ret->bufsize = 1982781; ret->buf = seg->allocate_aligned(ret->bufsize, Alignment); Then, process A writes into the dynamic buffer, which seems to work OK. In the other process however, I try to open it and read the data. Like this: seg.reset(new shared_segment(open_only, bufname)); ret = seg->find<struct transfer_buffer>(unique_instance).first; Now I can access everything fine in ret (the struct itself), including the mutex and data members. But I cannot access the dynamically allocated buffer. As soon as I access it (read it, memcpy, anything) the app crashes with what would appear to be a rather broken callstack. When I look at the transfer_buffer struct before the crash it looks as if the dynamic pointer points to some inaccessable area. Not what I wrote in it anyway. Do I have to do anything else (other than creating the segment and find() the unique transfer_buffer struct) in the other process to access the dynamic area? Or is this a bug? Any help appreciated... Stephan

3 5

boost::array < pair<>... > iterator adaptor
by Hicham Mouline 01 Apr '09

01 Apr '09

Hello, I have a boost::array< pair<type1, type2>, maxsize > object, and I wish to iterate over the list of type1 elements, and the list of type2 elements... I gather I then need an iterator adaptor based on array's iterator and const_iterators... How may I write such a thing? I then can pass around iterators to type1 and type2 separately. regards,

2 1

move() aware containers?
by Noah Roberts 01 Apr '09

01 Apr '09

I'm reading a pre-release of a book on C++0x threading. I'm using the boost::thread library since it's supposed to be pretty close. Now they're using std::vector<> though and I know it's not move ready. Is there a boost container I can mimic this with?

2 2

boost exception tutorial
by Robert Ramey 01 Apr '09

01 Apr '09

I'm investigating boost exception to see if it is helpful in a particular instance. I start by looking at the first Tutorial. I can't see how this is better than struc my_error : public std::exception { int m_ errno; my_error(int errno) : m_errno(errno) {} }; ... throw my_error(errno) try { ... } catch(my_error e){ cerr << e; } What am I missing here? This first example is two compilicated in that it requires that I delve into a bunch of other stuff (operator << for exception data?) and it doesn't provide any motivation. I think it would be more useful if it "spoon fed" the reader (me in this case) a little more. Something like: Traditional simple example like that above. explanation of why the example isn't "good enough" simple example reformulated using boost exception explanation of how this addresses the short comings of the original version. I think this would motivate more users to transition to the library. Robert Ramey

5 7

Re: [Boost-users] [statechart] state behaviour
by Andreas Huber 01 Apr '09

01 Apr '09

> So, if I post an event and then execute "template< class > DestinationState > result transit();" in custom reaction this > destination state will be given posted event. Am I right? Yep. Regards, -- Andreas Huber When replying by private email, please remove the words spam and trap from the address shown in the header.

1 0

Re: [Boost-users] statechart comparison
by Andreas Huber 01 Apr '09

01 Apr '09

Hi Brent > Is it still the case that boost statecharts require allocation and > destruction of a state object when entering and exiting that state? More precisely, the library issues "new State" upon state entry & and "delete State" upon state exit. So state construction & destruction *are* required but calls to global operator new and delete are not. The latter can be avoided by replacing operator new for your states or, on a more general level by passing a custom allocator to the state machine. Boost.Statechart was designed for reactive systems, where state machines are often serviced by dedicated threads. In such a scenario, the inter-thread communication usually introduces a much higher overhead than state construction & destruction. Regards, -- Andreas Huber When replying by private email, please remove the words spam and trap from the address shown in the header.

1 0

Re: [Boost-users] boost.statechart - a data driven state machine
by Andreas Huber 01 Apr '09

01 Apr '09

[snip] > I'd like this state machine to be completely data driven because it will > massively simplify testing and experimenting with different behaviours (ie. > I won't have to re-compile and produce new executables to try out a simple change). > > Instead of hard-coding all the states, events, transitions, etc., I would like > to read all those in from a config file (text, csv, xml, whatever). Boost.Statechart deliberately does not support dynamic configurability. The rationale can be found here: <http://www.boost.org/doc/libs/1_38_0/libs/statechart/doc/rationale.html#Dyn…> Regards, -- Andreas Huber When replying by private email, please remove the words spam and trap from the address shown in the header.

1 0

mpl::for_each state
by Skelly, Luke 01 Apr '09

01 Apr '09

Please ignore my last message. I somehow accidentally sent it while still editing! I am trying to create a std::map from an mpl::map. I've been trying to figure out how to use mpl::for_each to accomplish this. The problem I have is that the runtime function is passed to for_each by value and returns void. All the examples I've seen for mpl::for_each involve printing and not creating any new objects. Here is the attempt I've made: // creates std::map<size_t,size_t> from mpl::map<size_t,size_t> template<typename TypeMap> struct MapMaker { typedef std::map<size_t, size_t> map_type; map_type func_map; MapMaker() { boost::mpl::for_each<TypeMap>(*this); } template<typename U> void operator()(U x) { func_map[U::first::value]=U::second::value; } func_type operator[](size_t i) { return func_map[i]; } }; typedef boost::mpl::map< boost::mpl::pair<boost::mpl::size_t<1>, boost::mpl::size_t<2> >, boost::mpl::pair<boost::mpl::size_t<2>, boost::mpl::size_t<24> >, > myMap; int main() { MapMaker<myMap> m; size_t result = m[1]; } Of course this is a simplified version of what I'm really trying to do, which is create a map to function pointers. The problem is in the call to mpl::for_each in the MapMaker constructor. The std::map is created, but never returned. I've considered making func_map static, but that complicates things quite a bit, especially since MapMaker it is a class template. Luke Skelly

2 2

How to build signals2 with bjam?
by Stefan Funck 01 Apr '09

01 Apr '09

Hi all, I just checked out the current release from subversion and tried to build the signals2 lib with bjam --toolset=msvc --build-type=complete --with-signals2 stage but receive the message error: wrong library name 'signals2' in the --with-<library> option Building e.g. the thread or signals lib works fine. Does bjam have a list of "valid" libraries it checks against and signals2 is not yet contained in, or what am I doing wrong? Thanks for any help! Stefan

3 2

boost::regex ++ sequence is invalid
by jc 01 Apr '09

01 Apr '09

Hi, This is a carry over from the wrong group(boost build). In short the "+ +" sequence is invalid in boost::regex and valid in PRCE via an addition made to Perl 5.10. So, I am porting regular expressions that are used in PRCE and having trouble getting around this limitation in boost::regex. snip1: const boost::regex regex_username_password( "(" "(?:" "[a-zA-Z0-9_.!~*'()&=+$,;?/-]++" "|" "%[0-9a-fA-F]{2}" ")+" ")" "(?:" ":" "(" "(?:" "[a-zA-Z0-9_.!~*'()&=+$,-]++" "|" "%[0-9a-fA-F]{2}" ")*" ")" ")?" "@" ); snip2: const boost::regex regex_url_params("\\s*;([^?>]++)"); Help! -s7

1 0

Compiling Boost 1.38.0 with xlc 8.0 (Linux 64)
by Matthieu Brucher 01 Apr '09

01 Apr '09

Hi, I'm trying to use Boost on a PowerPC with IBM's compiler. Is there a .jam somewhere to use? Matthieu -- Information System Engineer, Ph.D. Website: http://matthieu-brucher.developpez.com/ Blogs: http://matt.eifelle.com and http://blog.developpez.com/?blog=92 LinkedIn: http://www.linkedin.com/in/matthieubrucher

1 0

Socket iostream Question
by Anthrope 01 Apr '09

01 Apr '09

Hello all, I am developing a high-performance socket-based application using the boost asio library. The application implements a binary tcp-based protocol that a bi-directional exchange of fixed-size messages. To simplify the coding while not sacrificing flexibility or performance, I decided to use the TCP iostream classes. The approach seems to work well, and I like the fact that the streams are buffered and allow me control over when I flush them. There's one minor issue that I thought I would seek some guidance on: Diagnosing and reporting failures connecting to servers are problematic. Basically, if a connect fails, the only way to know about it is to check the fail bit of the iostream. It would be nice to know why a client is unable to connect. So, to that end, I thought I could construct the iostream object via the default constructor, grab its streambuf via the rdbuf() call and in turn grab a basic_socket object reference to the streambuf and then call connect() on it, i.e., tcp::iostream iStream ; boost::asio::basic_socket_streambuf<tcp>& buf(*(iStream.rdbuf())); boost::asio::basic_socket<tcp, stream_socket_service<tcp> >& mySock(buf) ; mySock.connect(endPoint, error) ; The call to connect() on mySock with the error_code argument enables me to the diagnose and report connection problems a bit better. Once the connection is established, we can resume using the iostream directly. I would like to hear from you as to whether this approach makes sense, and what the pitfalls might be. And, alternative approaches would be most welcome too! Thanks in advance, Prashanth Nandavanam -- View this message in context: http://www.nabble.com/Socket-iostream-Question-tp22805249p22805249.html Sent from the Boost - Users mailing list archive at Nabble.com.

2 1

Is boost license GPL compatible?
by Tommy Nordgren 31 Mar '09

31 Mar '09

That is, is it possible to distribute a library distributed under the boost license together with a software package distributed under the GPL. ----------------------------------- See the amazing new SF reel: Invasion of the man eating cucumbers from outer space. On congratulations for a fantastic parody, the producer replies : "What parody?" Tommy Nordgren tommy.nordgren(a)comhem.se

3 2

zip_iterator and sTL algorithms like random_shuffle
by Andreas Fabri 31 Mar '09

31 Mar '09

Hello, I have two vectors and want to random_shuffle them the same way using the zip_iterator. I thought the code below would do it, but when I execute it I see that both sequences get screwded up. What do I wrong? Best regards, andreas #include <vector> #include <boost/iterator/zip_iterator.hpp> #include <algorithm> #include <iostream> int main() { std::vector<int> I(8), J(8); for(int i = 0; i < I.size(); i++){ I[i] = J[i] = i; } std::random_shuffle(boost::make_zip_iterator(boost::make_tuple(I.begin(), J.begin())), boost::make_zip_iterator(boost::make_tuple(I.end(), J.end()))); for(int i = 0; i < I.size(); i++){ std::cout << I[i] << " " << J[i] << std::endl; } return 0; }

3 5

[Q] scoped_ptr for forward declared classes
by V S P 31 Mar '09

31 Mar '09

Hello I have a situation where I have struct cmoney_impl_dtl; //hidden impl is in Cpp file struct myclass { boost::scoped_ptr<hidden_impl> myimpl_ptr; } The compiler complains that the struct hidden_impl must be fully defined in the header file (which I cannot do) ------------- c:\usr\local\boost.msvc2008\boost_1_38_0\boost\checked_delete.hpp(32) : error C2027: use of undefined type 'cmoney_impl_dtl' c:\dbt_money.hpp(38) : see declaration of 'cmoney_impl_dtl' c:\usr\local\boost.msvc2008\boost_1_38_0\boost\scoped_ptr.hpp(80) : see reference to function template instantiation 'void boost::checked_delete<T>(T *)' being compiled ------------- is there a way around this? I was trying to use shared_ptr there, but MSVC reports memory leak on those objects... so I must be doing something wrong there , and therefore just wanted to try a simple ptr. thanks in advance, Vlad -- V S P toreason(a)fastmail.fm -- http://www.fastmail.fm - Same, same, but different...

2 4

warnings with boost::bind
by Hansi 31 Mar '09

31 Mar '09

Hello, I have a question. I use boost::bind with the following class: class B { int data; } class A { bool SetData(const B&){return true;}; } now when I make somewhere the following: A a; boost::bind<bool>(&A::SetData,&a); I get a warning 8>D:\boost\boost_1_34_1\boost/bind.hpp(1575) : warning C4180: qualifier applied to function type has no meaning; ignored 8> D:\boost\boost_1_34_1\boost/bind.hpp(1609) : see reference to class template instantiation 'boost::_bi::add_cref<Pm,I>' being compiled 8> with 8> [ 8> Pm=bool (__thiscall Microtec::Program::* )(const Microtec::TrackingInfo &), 8> I=1 8> ] 8> .\Program.cpp(2385) : see reference to class template instantiation 'boost::_bi::dm_result<Pm,A1>' being compiled 8> with 8> [ 8> Pm=bool (__thiscall Microtec::Program::* )(const Microtec::TrackingInfo &), 8> A1=bool 8> ] I think it is for the const reference. Is this a problem? Is there some way to avoid it (okay the way with pragma I know)? I use boost 1.34.1 and MSVC2005 Regards

2 5

Cyclic includes when using smart pointers.
by Jared Lee Richardson 31 Mar '09

31 Mar '09

Hey, I'm having trouble with cyclic includes in a project I'm working on. This is my first attempt at a project that heavily uses boost, and specifically, smart pointers. I have the following situation, simplified down. Header file 1: class Test1 { vector<Test2Ptr> mTest2Objects; }; typedef boost::shared_ptr<Test1> Test1Ptr; Header file 2: class Test2 { vector<Test1Ptr> mTest1Objects; }; typedef boost::shared_ptr<Test2> Test2Ptr; Now normally I'd get around this easily by doing a forward declaration, but I can't do that with a smart pointer definition. How can I work around this? So far I have found ways to work around the above(Since that's not precisely what I'm doing), but when I start inlining functions it will become very difficult to work around. Obviously something is wrong with my include structure or how I am going about this, but I don't know what... Any help? Is there a way I can forward-declare smart pointers? How are complicated cyclic includes usually resolved, if forward declares aren't enough(i.e., inline files)? Thanks, Jared

2 3

[concept check][mpl] (Conditional) Concept checking for functions
by Jesse Perla 31 Mar '09

31 Mar '09

One rough idea I had was that since I will always have boost::function declarations as the default type, I could try to see if the type F is compatible with the boost::function type? Could this be done in a simple metafunction similar to the following: template<class BoostF, class F> struct CheckFunctionCompatible { static void Check() { F f; BoostF fboost = f; //If this fails, then I am basically failing } }; But then if I instantiate, won't it also try to instantiate the constructor? template<class F> class { typedef CheckFunctionCompatible<boost::function<double (double)>, F> test_F_compatibility; }; Of course, this wouldn't work for detecting the signature of the function, but that is a 2nd order feature for me. Of course, this also assumes that F can be constructed with no parameters, but this also might be reasonable for me if there is no better solution. -Jesse

2 2

TupleIt and zip_iterator
by Andreas Fabri 31 Mar '09

31 Mar '09

Was Anthony Williams TupleIt, mentioned in http://www.cpptalk.net/7-vt9657.html?postdays=0&postorder=asc&start=90 and distributed in http://groups.yahoo.com/group/boost/files/tupleit.zip) the precursor of the boost::zip_iterator? In contrast to the boost::zip_iterator an TupleIt iterator range can be sorted. What are the advantages of the zip_iterator, besides being in boost. andreas

1 0