- Boost-users - lists.boost.org

[distributions]: Inverse Gamma
by Thomas Mang 05 Aug '10

05 Aug '10

Hi, any plans of implementing the inverse gamma distribution as part of the distributions library ? What about multivariate (in particular the multivariate normal and t distributions and Wishart and Dirichlet ) ? thanks, Thomas

4 10

boost serialisation and enable_shared_from_this versus 1.43
by Avi Bahra 04 Aug '10

04 Aug '10

In 1.42, boost serialisation is broken when used with a class that is derived from enable_shared_from_this. This was supposed to fixed in the next release., but the online release notes for 1.43 make no mention of this. Does any one know if it made it into1.43 ? Best regards, Ta, Avi

2 1

Leandro Lima invited you to Dropbox
by Dropbox 04 Aug '10

04 Aug '10

Leandro Lima wants you to use Dropbox to sync and share files online and across computers. Get started here: http://www.dropbox.com/link/45.SygZEI0JRr.SygZEI0JRr/NjI3ODM0OTc5Nw - The Dropbox Team ____________________________________________________ To stop receiving invites from Dropbox, please go to http://www.dropbox.com/bl/207376696615/boost-users%40lists.boost.org

1 0

[GIL] Warnings when compiling items in channel_algorithms.hpp
by Nathan Crookston 04 Aug '10

04 Aug '10

All, While writing some image conversion code using scoped_channel_value I encountered several compiler warnings. The following minimal code demonstrates: Compiled with VC8, command line: "cl simple.cpp /Iboost /W3 /EHsc" ////// BEGIN CODE ///////////////////////////// #include "boost/gil/gil_all.hpp" namespace boost { namespace gil { const boost::uint16_t bits12_max_val = 0xFFF; struct bits12_min { static boost::uint16_t apply() { return 0; } }; struct bits12_max { static boost::uint16_t apply() { return bits12_max_val; } }; typedef boost::gil::scoped_channel_value<boost::uint16_t, bits12_min, bits12_max> bits12; namespace detail { template <> struct unsigned_integral_max_value<bits12> : public mpl::integral_c<uint32_t,bits12_max_val> {}; template <> struct unsigned_integral_num_bits<bits12> : public mpl::int_<12> {}; }//end detail GIL_DEFINE_BASE_TYPEDEFS(12, gray) const boost::uint16_t bits14_max_val = 0x3FFF; struct bits14_min { static boost::uint16_t apply() { return 0; } }; struct bits14_max { static boost::uint16_t apply() { return bits14_max_val; } }; typedef boost::gil::scoped_channel_value<boost::uint16_t, bits14_min, bits14_max> bits14; namespace detail { template <> struct unsigned_integral_max_value<bits14> : public mpl::integral_c<uint32_t,bits14_max_val> {}; template <> struct unsigned_integral_num_bits<bits14> : public mpl::int_<14> {}; }//end detail GIL_DEFINE_BASE_TYPEDEFS(14, gray) } } //end gil, boost using namespace boost::gil; int main() { gray14_image_t img14(2,2); color_converted_view<gray12_pixel_t>(view(img14))(1,0); gray32f_image_t img32f(2,2); color_converted_view<gray12_pixel_t>(view(img32f))(1,0); return 0; } ////// END CODE /////////////////////////////// It produces the following warnings, trimmed to only those I thought relevant: ///////////////// BEGIN Relevant Warnings ////////////////////////////////////// c:\cygwin\home\ncrookston\boost\boost\gil\channel_algorithm.hpp(257) : warning C 4244: 'argument' : conversion from 'const double' to 'boost::uint16_t', possible loss of data c:\cygwin\home\ncrookston\boost\boost\gil\channel_algorithm.hpp(252) : w hile compiling class template member function 'boost::gil::scoped_channel_value< BaseChannelValue,MinVal,MaxVal> boost::gil::detail::channel_converter_unsigned_i ntegral_nondivisible<SrcChannelV,DstChannelV,SrcLessThanDst,CannotFitInInteger>: :operator ()(SrcChannelV) const' with [ BaseChannelValue=boost::uint16_t, MinVal=boost::gil::bits12_min, MaxVal=boost::gil::bits12_max, SrcChannelV=boost::gil::scoped_channel_value<boost::uint16_t,boost:: gil::bits14_min,boost::gil::bits14_max>, DstChannelV=boost::gil::scoped_channel_value<boost::uint16_t,boost:: gil::bits12_min,boost::gil::bits12_max>, SrcLessThanDst=false, CannotFitInInteger=false ] <snip lots of instantiation backtrace> c:\cygwin\home\ncrookston\boost\boost\gil\channel_algorithm.hpp(268) : warning C 4244: 'argument' : conversion from 'float' to 'boost::uint16_t', possible loss o f data c:\cygwin\home\ncrookston\boost\boost\gil\channel_algorithm.hpp(268) : w hile compiling class template member function 'boost::gil::scoped_channel_value< BaseChannelValue,MinVal,MaxVal> boost::gil::channel_converter_unsigned<SrcChanne lV,DstChannelV>::operator ()(boost::gil::bits32f) const' with [ BaseChannelValue=boost::uint16_t, MinVal=boost::gil::bits12_min, MaxVal=boost::gil::bits12_max, SrcChannelV=boost::gil::scoped_channel_value<float,boost::gil::float _zero,boost::gil::float_one>, DstChannelV=boost::gil::scoped_channel_value<boost::uint16_t,boost:: gil::bits12_min,boost::gil::bits12_max> ] <snip lots more instantiation backtrace> /////////////////////// END Relevant Warnings ////////////////////////////////////////////////// Interestingly, g++ doesn't have an issue with the code. The following patch corrects the problem. I'd be happy to create a trac ticket, write some tests, etc. Please advise. ////////////// BEGIN Patch ////////////////////////////////////// Index: channel_algorithm.hpp =================================================================== --- channel_algorithm.hpp (revision 63088) +++ channel_algorithm.hpp (working copy) @@ -243,7 +243,18 @@ } }; +// Determines the base type of the channel. +template <typename Channel> +struct base_channel { typedef Channel type; }; +template <typename BaseChannelValue, typename MinVal, typename MaxVal> +struct base_channel<scoped_channel_value<BaseChannelValue,MinVal,MaxVal> > +: public base_channel<BaseChannelValue> {}; + +template <int NumBits> +struct base_channel<packed_channel_value<NumBits> > +{ typedef typename packed_channel_value<NumBits>::integer_t type; }; + // Both source and destination are unsigned integral channels, // the dst max value is less than (or equal to) the src max value, // and the src max value is not divisible by the dst max value @@ -253,8 +264,8 @@ typedef typename unsigned_integral_max_value<SrcChannelV>::value_type integer_t; static const double div = unsigned_integral_max_value<SrcChannelV>::value / double(unsigned_integral_max_value<DstChannelV>::value); - static const integer_t div2 = integer_t(div/2); - return DstChannelV((src + div2) / div); + static const integer_t div2 = static_cast<integer_t>(div/2); + return static_cast<typename base_channel<DstChannelV>::type>((src + div2) / div); } }; @@ -265,7 +276,8 @@ ///////////////////////////////////////////////////// template <typename DstChannelV> struct channel_converter_unsigned<bits32f,DstChannelV> : public std::unary_function<bits32f,DstChannelV> { - DstChannelV operator()(bits32f x) const { return DstChannelV(x*channel_traits<DstChannelV>::max_value()+0.5f); } + DstChannelV operator()(bits32f x) const { return + static_cast<typename detail::base_channel<DstChannelV>::type>(x*channel_traits<DstChannelV>::max_value()+0.5f); } }; template <typename SrcChannelV> struct channel_converter_unsigned<SrcChannelV,bits32f> : public std::unary_function<SrcChannelV,bits32f> { //////// END patch //////////////// Thanks, Nate

2 8

matrix function template for triangular_matrix and matrix datatypes
by Max S. Kaznady 04 Aug '10

04 Aug '10

Suppose we are working in namespace boost::numeric::ublas and I have 2 functions 1) template <class T> save_matrix(matrix<T> const &m) 2) template <class T> save_matrix(triangular_matrix<T,lower> const &m) and the code in both overloaded save_matrix functions is identical. How can I write a templated function which accepts any type of a matrix? I tried this, but it didn't work: template <class T> save_matrix(T const &m) Thanks, Max

2 3

[interprocess] shared memory, condtions, deadlock
by Kai Benndorf 04 Aug '10

04 Aug '10

Hi all, I'm trying to use the shared memory functionality of the boost interprocess library for implementing a client and a server like process, where the server calculates something for the client. I use condition variables for the process synchronization. The implementation works initially, but I am getting deadlocks, now. I appended a pseudo code example of my implementation. One instance of the SharedData class is hold in shared memory. On deadlock, both processes are hanging in the wait() functions. What am I doing wrong? Many thanks for any hint. Regards Kai

1 0

[conversion] lexical_cast doesn't throw
by Gaetano Mendola 04 Aug '10

04 Aug '10

Hi all, I'm in the process to migrate our application from RedHat 5.5 (gcc 4.1.2) to Ubuntu 10.04 (gcc 4.4.3), one regression test is not passing due the fact that lexical_cast is not throwing, this is the line: unsigned int myNumber = boost::lexical_cast<unsigned int>("-123"); I'm using boost 1.40 on both servers. I have noticed that the same executable compiled on Ubuntu 10.04 while on Ubuntu doesn't throw, running it on RH 5.5 throws, so it seems a system library problem. Trying with a simple stringstream, it seems that with the following code: std::stringstream myStream; myStream << "-123"; unsigned int myUInt; myStream >> myUInt; doens't set the fail bit on Ubuntu 10.04, but it does on RedHat 5.5. Does anyone have an idea on what is going on Regards Gaetano Mendola

1 0

Thread object pointers.
by Albert Schueller 04 Aug '10

04 Aug '10

Tue Aug 3 14:10:04 PDT 2010 Hi all, I'm trying to get up to speed on libboost's thread class using the tutorial information here: http://www.drdobbs.com/cpp/184401518 I'm curious why the two programs below seem to behave differently. In LISTING 1, three threads are declared as thrd1, thrd2, and thrd3 explicitly and each loops 10 times with random length sleeps. LISTING 2 is identical, except for the fact the three thread variables are allocated to boost::thread pointers. When I run LISTING 1 it appears that the threads are acting independently as I would hope. When I run LISTING 2, it seems that the threads are not independent, but rather thrd1 does its job, followed by thrd2, followed by thrd3. Am I not locking properly? Any help would be appreciated. Albert /////////////////////////////////////////////////////////////////////////// //LISTING 1 #include <boost/thread/thread.hpp> #include <boost/thread/mutex.hpp> #include <iostream> #include <unistd.h> #include <cstdlib> boost::mutex io_mutex; struct count { count(int id) : id(id) { } void operator()() { int upause; for (int i = 0; i < 10; ++i) { upause = rand()%1000000; usleep(upause); boost::mutex::scoped_lock lock(io_mutex); std::cout << id << ": " << i << " paused " << upause << std::endl; } } int id; }; int main(int argc, char* argv[]) { boost::thread thrd1(count(1)); boost::thread thrd2(count(2)); boost::thread thrd3(count(3)); thrd1.join(); thrd2.join(); thrd3.join(); return 0; } /////////////////////////////////////////////////////////////////////////// //LISTING 2 #include <boost/thread/thread.hpp> #include <boost/thread/mutex.hpp> #include <iostream> #include <unistd.h> #include <cstdlib> boost::mutex io_mutex; struct count { count(int id) : id(id) { } void operator()() { int upause; for (int i = 0; i < 10; ++i) { upause = rand()%1000000; usleep(upause); boost::mutex::scoped_lock lock(io_mutex); std::cout << id << ": " << i << "paused (" << upause << ")" << std::endl; } } int id; }; int main(int argc, char* argv[]) { int i; int num_thrds = 3; boost::thread *thrd[4]; for(i=0;i<num_thrds;i++) { thrd[i] = new boost::thread(count(i+1)); thrd[i]->join(); } // deallocate thread objects for(i=0;i<num_thrds;i++) { delete thrd[i]; } return 0; }

5 7

Open a filename with a wstring or wchar_t *
by Timothy Madden 03 Aug '10

03 Aug '10

Hello Surprisingly enough, C++ file-based streams can be opened with a char * string (for the filename) only, while modern computer systems have Unicode filenames. I see boost::iostreams::basic_file also gets constructed from a char * only. My project name has Korean characters, and I work on a Latin1 Windows system, and on my system the narrow-characters set simply does not contain Korean characters. Is there a (good) way to open a file with a wstring in boost ? Thank you, Timothy Madden

6 14

Problem with boost::bind, boost::shared_ptr and std::find_if algorithm
by przemyslaw.sliwa＠uk.bnpparibas.com 03 Aug '10

03 Aug '10

All, I have the following data sructure: struct WeightedTimeHolder { double time; double dt; double weight; double dailyVol; double varDelta_t; double cutToCutdailyVol; double extraVar; double varPlusEventVar; double aggrVar; double implVol; int cut; } and typedef multimap<int, boost::shared_ptr<WeightedTimeHolder> > DataMap. Now given value of cut in this data structure I would like to find the next position in DataMap which corresposnds to this gven value, I do it in the following manner: void nextPositionForCut(DataMap::iterator& pos, CutOffType cut) const { if((*pos).second->cut!=cut) while((*(++pos)).second->cut!=cut && pos!=m_pData->end()) {} // convert to boost::bind for better use! //pos = find_if(pos, m_pData->end(), boost::bind(&DataMap::value_type::second::value_type::cut, _1) == cut); } Since this looks really ugly I would like to ask you how this can be converto to find_if algorithm with a little help of boost::bind. I tried but could not succees. Thanks for help, Przemyslaw ___________________________________________________________ This e-mail may contain confidential and/or privileged information. If you are not the intended recipient (or have received this e-mail in error) please notify the sender immediately and delete this e-mail. Any unauthorised copying, disclosure or distribution of the material in this e-mail is prohibited. Please refer to http://www.bnpparibas.co.uk/en/information/legal_information.asp?Code=ECAS-… for additional disclosures.

2 1