February 2011 - Boost-users - lists.preview.boost.org

Interested in GSoC 2011
by Chaitali Shah 09 Feb '11

09 Feb '11

Hello members, Myself is Chaitali Shah, undergraduate student of Computer Engineering from Sardar Patel Institute of Technology,Mumbai. I am interested in participating in GSoC 2011 with Boost C++ libraries as mentoring organisation. As a computer engineering student I have good programming skills and am familiar with many new tools and technologies. I am new to the list and to the organization. I went through the projects of last year and found my interest over it. I saw that organization has already started with the preparations of GSoc 2011 [1] and I wish to work with the organization as a developer. I wanted to know if there are some task that i should work on so as to better understand the code base and architecture before the timeline. Thus, it will be helpful for me if you provide me or guide me with some information as pre-requisites. Thanks in advance. [1] http://lists.boost.org/Archives/boost/2010/10/172030.php -- Regards, Chaitali Shah Student at Sardar Patel Institute of Technology,Mumbai Gtalk: chaitalib.shah(a)gmail.com shahchaitali5(a)gmail.com

2 1

boost::python::import causes access violation
by Julian Bailey 08 Feb '11

08 Feb '11

Hi, I am running through the basic examples on the boost website for embedding boost::python. My code currently reads: #include <iostream> #include "./include/Python.h" #include "./boost/python.hpp" using namespace boost::python; int main(void) { Py_Initialize(); object main_module = import("__main__"); object main_namespace = main_module.attr("__dict__"); object ignored = exec("print Hello world",main_namespace); return 0; } The program links successfully and begins running, however an access violation occurs (NULL pointer) at the first line after "py_initialize();" when i call import("__main__"); Visual studio 2010s debugger says the exception occurs at: line 74 in object_attributes.hpp: inline object const_attribute_policies::get(object const& target, char const* key) { return python::getattr(target, key); } Python was built using Visual studio 2010 so i had access to the python debug library. Boost is version 1.45 and was built using : bjam --build-type=complete msvc stage --with-python I am linking against: boost_python-vc100-mt-gd-1_45.lib and boost_python-vc100-mt-gd-1_45.dll I have looked for soutions on the internet and have found nothing right now. Can anyone help me? note: i am building my application in MSVC 2010, this should work though shouldn't it? Dr Julian Bailey Electronic Systems & Devices, School of Electronics and Computer Science, University of Southampton, Southampton, SO17 1BJ, United Kingdom. Tel. +44 (0)2380 593545 email: jab(a)ecs.soton.ac.uk

2 1

SSL crash in boost asio 1.44
by Scott 08 Feb '11

08 Feb '11

I've been getting reports in the field about sporadic SSL crashes in our product (Linux version). After using a stress test for several hours, I got it to crash in asio 1.41. I then tried it in 1.44 and got the same crash on windows. It appears to be crashing in openssl_operator.hpp:158 The session_ object is definitely corrupt: session_ 0x000000022120ac30 {version=-17891602 type=-17891602 method=0xfeeefeeefeeefeee ...} version -17891602 int type -17891602 int + method 0xfeeefeeefeeefeee {version=??? ssl_new=??? + rbio 0xfeeefeeefeeefeee {method=??? callback=??? + wbio 0xfeeefeeefeeefeee {method=??? callback=??? + bbio 0xfeeefeeefeeefeee {method=??? callback=??? The callstack is this: NwConcentrator.exe!boost::asio::ssl::detail::openssl_operation< boost::asio::basic_stream_socket< boost::asio::ip::tcp,boost::asio::stream_socket_service< boost::asio::ip::tcp>> >::start() NwConcentrator.exe!boost::_mfi::mf0<int,boost::asio::ssl::detail:: openssl_operation< boost::asio::basic_stream_socket< boost::asio::ip::tcp,boost::asio::stream_socket_service< boost::asio::ip::tcp> ::operator() (boost::asio::ssl::detail::openssl_operation< boost::asio::basic_stream_socket< boost::asio::ip::tcp,boost::asio::stream_socket_service< boost::asio::ip::tcp>> > * p=0x00000002136cf440) Line 50 C++ Does 1.45 have a fix for this? I looked over the release notes but didn't see anything about SSL fixes. This crash is occurring on both Fedora/CentOS/Windows builds of boost asio 1.41 and 1.44. Thanks, Scott

1 0

Boost::Spirit::Qi parsing into list
by Slide 08 Feb '11

08 Feb '11

I am looking at using Boost::Spirit for parsing a set of hardware register definitions. From the examples I have seen, it looks like one mainly creates a structure and then that is what is returned from the parsing of the input. I was wondering if anyone could point me towards some examples that might parse out a list of items and return them. The end goal is to parse in the register definitions and then generate C/C++ code from those register definitions to define a hardware access layer that can be used by my other libraries on my chip. So, I would parse the register definitions into a vector and then iterate over that vector to generate the necessary code. Thanks for any pointers. slide -- slide-o-blog http://slide-o-blog.blogspot.com/

2 1

Thread changes from Boost 1.38 to 1.45
by Ven Tadipatri 08 Feb '11

08 Feb '11

Hi, We upgraded our project from using Boost 1.38 to Boost 1.45 but now seem to be having some thread issues. The error doesn't happen for every run, but occasionally the wait method in condition_variable_any fails. At other times the program seems to run into an infinite loop, presumably waiting for a lock that's never released. I looked at the changes in http://www.boost.org/doc/libs/1_45_0/doc/html/thread/changes.html but don't see what could be causing the error to occur only after upgrading to Boost 1.45. Has anyone else experienced their code breaking when upgrading to the latest version of Boost? Thanks, Ven

5 10

Boost.MPI --- boost::mpi::group class
by Júlio Hoffimann 07 Feb '11

07 Feb '11

Hi, It's my first time using Boost and is amazing. I'm trying to understand why this simple example bellow don't works as expected: #include <iostream> #include <boost/mpi.hpp> using namespace std; namespace mpi = boost::mpi; int main ( int argc, char *argv[] ) { mpi::environment env( argc, argv ); mpi::communicator world; mpi::group subgroup( MPI_Group( world.group() ), true ); cout << "I'm process " << world.rank() << endl; cout << "But in subgroup i'm process " << subgroup.rank() << endl; cout << "And the subgroup size is... " << subgroup.size() << endl; return 0; } // ---------- end of function main ---------- *1. Why every process has rank 1 in subgroup? I trying to figure out what happens, but without success.* The mpi::boost::group class has no copy constructor explicitly defined. That is true for many other classes in Boost.MPI... *2. It's safe to use the copy constructor generated by the compiler as follows?* * * mpi::group subgroup = world.group().exclude( discardedRanks.begin(), discardedRanks.end() ); Thanks in advance for any help. Regards, Júlio.

1 2

math statistical: Kolmogorov-Smirnov distribution
by Matwey V. Kornilov 07 Feb '11

07 Feb '11

Hi, Isn't there KS-distribution in boost.math? I didn't find this but maybe I missed something. As far as I understand there is no way to perform KS-test using boost.math framework due to absence of needed statistic.

3 4

Re: [Boost-users] [boost] TTI library updated in the sandbox
by Dave Abrahams 07 Feb '11

07 Feb '11

Couple o' things: On Sun, Feb 6, 2011 at 12:39 PM, Edward Diener <eldiener(a)tropicsoft.com> wrote: > Features of the update include support for the older gcc 3.4.2 and 3.4.5, > finer-grained introspection for the basic macro metafunctions, documentation > examples, better and tighter design, and better use of Boost MPL methodology > ( metafunctions are now passed as lambda expressions and metafunction > globbing has been removed from the two metafunctions where it previously > existed ). You can view the changes in the History section of the > documentation. FWIW, passing lambda expressions instead of metafunction classes is a trade-off: it's more expressive, but will lengthen compile times. For the implementation of a boost library, it's probably better to err on the side of speed. > The TTI library is based on the type_traits_ext portion of the Concept > Traits Library, with improvements and additions, and also lifts > functionality, for the purposes of orthogonality, from Boost.MPL regarding > introspection of types and templates. Orthogonality in what sense? Duplicating functionality seems like the opposite of orthogonality. -- Dave Abrahams BoostPro Computing http://www.boostpro.com

3 4

[BOOST::XPRESSIVE] Trouble with functor
by Norbert Hauck 07 Feb '11

07 Feb '11

Hi again, and sorry for the repost - I had a fat finger accident and clicked the send button to early! If I managed to delete the old message prior to posting this one please ignore. To the point: Please find enclosed my code - df1.cpp - as an attachment. I tinkered around with it but since I'm new to the boost libraries as well as a hobby programmer at best I throw in the towel now. Maybe someone more experienced can help me out how to change my code so that it compiles without errors.The error message was: 1>c:\program files\boost\boost_1_45_0\boost\proto\context\default.hpp(441) : error C2664: 'int DateVerify_Impl::operator ()(const std::string &,wchar_t *) const' : cannot convert parameter 2 from 'wchar_t (*__w64 const )[11]' to 'wchar_t *' 1> Types pointed to are unrelated; conversion requires reinterpret_cast, C-style cast or function-style cast Are ALL parameters I pass to the operator() somehow const ? Pfff ..... maybe I should continue after my temperature went down again (flu) - it seems my brains are addled and I cannot figure out anything at the moment. Thanks a lot, Norbert -- NEU: FreePhone - kostenlos mobil telefonieren und surfen! Jetzt informieren: http://www.gmx.net/de/go/freephone

2 2

Re: [Boost-users] multi_array and variants
by Alle Meije Wink 07 Feb '11

07 Feb '11

Dear Kris, thanks for your reply. > > What I now have is a class bisArray: > > ... > > template<typename T> > > class bisArray: public bisObject { > > typedef boost::variant > > < boost::multi_array_ref<T, 1>, > > boost::multi_array_ref<T, 2>, > > boost::multi_array_ref<T, 3>, > > boost::multi_array_ref<T, 4>, > > boost::multi_array_ref<T, 5>, > > boost::multi_array_ref<T, 6>, > > boost::multi_array_ref<T, 7>, > > boost::multi_array_ref<T, 8>, > > boost::multi_array_ref<T, 9> > > > bisArray_t; > > bisArray_t _bisArray; > > ... > > > > and a class bisImage, which has a bisArray as well as a pointer to the > > data as read from file (1-dimensional structure): > > ... > > > > template<typename T> > > > > class bisImage: public bisObject { > > > > public: > > > > /** Constructor using existing array data > > Reformats array as bisImage > > */ > > bisImage ( T* _data, size_t _dimensions, std::vector<size_t> _sizes > > ): > > data(_data), dimensions(_dimensions) { > > > > This code (or You) comes from java or C#, doesn't it? ;-) In C++ it is > supposed to be this->[...] instead of this.[...] No, it's C (which doesn't use any of this -not properly at least). The whole this[...] business was not supposed to be there anyway I realised. The reason that I went to boost:: is that I was hoping to get one class for all data types (using T) and dimensionalities (using multi_array) > Second thing: the member storage is initialized with the default constructor > (which in this case does not exist => error ), and then assigned (below). > > I suggest a change to: > bisImage([...]) : [...], storage( _data, _dimensions, _size ) { [...] > and remove the assignment from the body of the constructor > > Note also: I have read in many places, that you should not use names > beginning with an underscore (_), because they are reserved for > compiler/library implementation. One of those places was Herb Sutter's gotw > series. Ah! The underscore names come from a library that I tried to imitate which has successfully incorporated some aspects of boost(it is called MIMAS). Have not reached the stage of 'nicifying' the code yet... I moved the storage thing to where you suggested (1 line up) and that does get rid of lots of errors. However what I am stuck with seems to be, in essence, still the same thing: >>> /home/amwink/programs/cpp/bis/src/bisniftiimage.cpp||In function ‘void bis::bisReadNifti(std::string)’:| >>> /home/amwink/programs/cpp/bis/src/bisniftiimage.cpp|102|warning: unused variable ‘testimage’| >>> /usr/include/boost/variant/variant.hpp||In constructor ‘boost::variant<T0, T1, T2, T3, T4, T5, T6, T7, T8, T9, T10, T11, T12, T13, T14, T15, T16, T17, T18, T19>::variant() [with T0_ = boost::multi_array_ref<unsigned char, 1ul>, T1 = boost::multi_array_ref<unsigned char, 2ul>, T2 = boost::multi_array_ref<unsigned char, 3ul>, T3 = boost::multi_array_ref<unsigned char, 4ul>, T4 = boost::multi_array_ref<unsigned char, 5ul>, T5 = boost::multi_array_ref<unsigned char, 6ul>, T6 = boost::multi_array_ref<unsigned char, 7ul>, T7 = boost::multi_array_ref<unsigned char, 8ul>, T8 = boost::multi_array_ref<unsigned char, 9ul>, T9 = boost::detail::variant::void_, T10 = boost::detail::variant::void_, T11 = boost::detail::variant::void_, T12 = boost::detail::variant::void_, T13 = boost::detail::variant::void_, T14 = boost::detail::variant::void_, T15 = boost::detail::variant::void_, T16 = boost::detail::variant::void_, T17 = boost::detail::variant::void_, T18 = boost::detail::variant::void_, T19 = boost::detail::variant::void_]’:| >>> /home/amwink/programs/cpp/bis/src/bisarray.hpp|52|instantiated from ‘bis::bisArray<T>::bisArray(T*, size_t, std::vector<long unsigned int, std::allocator<long unsigned int> >) [with T = unsigned char]’| >>> /home/amwink/programs/cpp/bis/src/bisimage.hpp|33|instantiated from ‘bis::bisImage<T>::bisImage(T*, size_t, std::vector<long unsigned int, std::allocator<long unsigned int> >) [with T = unsigned char]’| >>> /home/amwink/programs/cpp/bis/src/bisniftiimage.cpp|126|instantiated from here| >>> /usr/include/boost/variant/variant.hpp|1198|error: no matching function for call to ‘boost::multi_array_ref<unsigned char, 1ul>::multi_array_ref()’| >>> /usr/include/boost/multi_array/multi_array_ref.hpp|623|note: candidates are: boost::multi_array_ref<T, NumDims>::multi_array_ref(T*, const typename boost::const_multi_array_ref<T, NumDims, T*>::storage_order_type&, const typename boost::const_multi_array_ref<T, NumDims, T*>::index*, const typename boost::const_multi_array_ref<T, NumDims, T*>::size_type*) [with T = unsigned char, long unsigned int NumDims = 1ul]| >>> /usr/include/boost/multi_array/multi_array_ref.hpp|469|note: boost::multi_array_ref<T, NumDims>::multi_array_ref(T*, const boost::detail::multi_array::extent_gen<NumDims>&, const boost::general_storage_order<NumDims>&) [with T = unsigned char, long unsigned int NumDims = 1ul]| >>> /usr/include/boost/multi_array/multi_array_ref.hpp|463|note: boost::multi_array_ref<T, NumDims>::multi_array_ref(T*, const boost::detail::multi_array::extent_gen<NumDims>&) [with T = unsigned char, long unsigned int NumDims = 1ul]| >>> /usr/include/boost/multi_array/multi_array_ref.hpp|417|note: boost::multi_array_ref<unsigned char, 1ul>::multi_array_ref(const boost::multi_array_ref<unsigned char, 1ul>&)| >>> ||=== Build finished: 1 errors, 1 warnings ===| So boost.variant now sugegsts that multi_array_ref() is called, a constructor with no arguments! I don't use that, and I don't even think it exists (because it produces a multi_array view of an *existing* C-style array! The only way I knew to initialise a bisArray was with using a switch statement -proper C: switch dims{ case 1: _bisarray=multi_array_ref<T,1> (data, sizes); break; case 2: _bisarray=multi_array_ref<T,2> (data, sizes); break; ... } That is the only way multi_array_ref is called! Any idea why a problem with the *default* constructor would turn up? Thanks for your suggestions! My error list already fits on one screen :) Alle Meije

3 3