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

[iterators][range][rangeex] range concatenation
by Dmitry Vinogradov 12 Jan '10

12 Jan '10

Is it possible with Boost (or RangeEx) to concatenate ranges? Sample usage: template <class Range> void process_range(Range const &); std::vector<foo> v = ...; std::list<foo> l = ...; process_range(concat(v, l));

5 18

[Math/Statistical Distributions+Random] Random number generation
by Marco Guazzone 07 Jan '10

07 Jan '10

Hi all! I'm a "boost newbie" so sorry for too trivial questions. Just trying to use boost for some statistical experiments, involving statistical distribution fitting, quantile generation (for goodness-of-fit tests), random variate generation, and so on. . The "Math/Statistical Distribution" library is great and almost complete. But I'm just curious why the random number generation (provided by the "Random" lib) and this library have not been integrated. For instance, the boost::random::exponential_distribution and the boost::math::exponential_distribution are two different worlds. I think a function like this below would be very helpful: --- [snip] --- template <typename Generator, typename RealType, typename Policy> inline RealType rand(Generator& rng, const exponential_distribution<RealType, Policy>& dist) { return -1.0 / dist.lambda() * std::log(1.0-rng()); } --- [snip] --- or something similar (e.g., rand implemented as a class method) In this case only two objects are needed: * boost::random::mersenne_twister * boost::math::exponential_distribution Instead, actually I have to keep different objects. For instance, for the exponential distribution: * boost::random::exponential_distribution * boost::random::mersenne_twister * boost::random::variate_generator * boost::math::exponential_distribution Unluckely I can't use boost::math::exponential_distribution in place of boost::random::exponential_distribution since boost::random::variate_generator expects a specific interface for the distribution. Well, what is the rationale behind the choice of this separation? Do you think would be useful to provide this kind of integration? Thank you very much in advance!! Cheers -- Marco

4 13

Newbie question
by Michael Qi 07 Dec '09

07 Dec '09

Hi, I am trying to start to learn boost. Now, I am learning the getting started guide. And try to build a boost regex program in the guide. But Visual C++ 7.1 tell me error: LINK : fatal error LNK1104: Could not open file "D:\Program.obj" Could you help me? Thank you

2 1

timeconv.inl
by Dan Smithers 11 Sep '09

11 Sep '09

I am porting some inherited code to MSVS 2008 This is also bringing along an upgrade from boost 1.32 to a newer version (probably 1.37). I have found a thread wrapper class in the code that uses time_to directly. This is defined in lib/thread/src/timeconv.inl and doesn't look like its meant to be used directly. The original code base has timeconv.inl in the boost include directory. Was timeconv.inl moved at some point in the past? Is this as bad as I think it is? thanks dan

3 3

Re: [Boost-users] [Graph] Problems with adjacency_matrix
by Juan Antonio Farré Basurte 29 Aug '09

29 Aug '09

> > > Hello, > I have a problem with adjacency_matrix template. > It claims to implement VertexAndEdgeListGraphConcept, BidirectionalGraphConcept and MutablePropertyGraphConcept. > But when I try to compile, it complains about using the next functions with an adjacency_matrix: > > num_vertices(graph) > vertices(graph) > num_edges(graph) > edges(graph) > source(edge_descriptor, graph) > target(edge_descriptor, > graph) > > and also with > > get(&V::property, graph), where V is a vetex bundled-properties type get(&E::property, > graph), where E is an edge bundled-properties type > > (these get functions should return a property map for the corresponding bundled property > and they work correctly for adjacency_list, just like the rest). > > Is this a bug or an error in the documentation? > > Thanks a lot, > > Juan > Mmmm... I just tried to include adjacency_matrix.hpp and now I don't get errors about those functions anymore. It's strange to have to inclue this header in a template library that doesn't explicitly deal with any concrete implementations of graph concepts. In fact, I don't need to include adjacency_list.hpp for it to work with adjacency_list. I just need to include graph_concepts.hpp and graph_traits.hpp. Anyway, I'm still getting lots of errors in a code that perfectly compiles and works with adjacency lists. If anybody is interested, I can send the listing of compiler errors. Thanks again, Juan

9 20

Random library, discrete distribution
by Kevin Martin 18 Aug '09

18 Aug '09

I made a post sometime last year about requiring a distribution that could generate enums with specified probabilities. I find I am regularly requiring a discrete distribution like this - where I specify the probabilities. Am I the only one who requires it, or would it be useful to a wider audience? I'm just wondering whether to submit a feature request for it or not? Thanks, Kevin Martin This is what I'm using at the moment: template<class IntType = int> class discrete_distribution { public: typedef boost::uniform_int<>::input_type input_type; typedef IntType result_type; template<size_t N> discrete_distribution(const boost::array<int, N> &outputs, const boost::array<int, N> &weights) : outputs_(outputs.begin(), outputs.end()), weights_(weights.begin(), weights.end()) { for(int i=1;i!=weights_.size();++i) weights_[i] += weights_[i-1]; } inline void reset() {} template<class Engine> result_type operator()(Engine& eng) { boost::uniform_int<> ui(0, weights_.back()-1); boost::variate_generator<Engine&, boost::uniform_int<> > vg(eng, ui); int idx = std::lower_bound(weights_.begin(), weights_.end(), vg()) - weights_.begin(); return outputs_[idx]; } private: std::vector<int> outputs_; std::vector<int> weights_; };

3 11

Boost Serialization: undefined reference to `__assert'
by Kesseli Pascal 31 Jul '09

31 Jul '09

Hello everyone Today, I just happened to download the latest version of cygwin, including gcc 3.4.4. While a simple hello world works perfectly fine, trying to compile this code using a statically linked libserialization.a results in dozens of linker errors, all stating "undefined reference to `__assert'". #include <boost/assign/list_of.hpp> #include <boost/archive/xml_iarchive.hpp> #include <boost/archive/xml_oarchive.hpp> #include <boost/lambda/lambda.hpp> #include <boost/serialization/vector.hpp> #include <iostream> #include <fstream> using namespace boost::lambda; using namespace boost::assign; using namespace boost::archive; using namespace boost::serialization; using namespace std; int main() { { vector<string> v = list_of("123")("Boost")("Serialization"); ofstream ofs("test.xml"); xml_oarchive oxml(ofs); oxml << make_nvp("VectorTest", v); } { vector<string> v; ifstream ifs("test.xml"); xml_iarchive ixml(ifs); ixml >> make_nvp("VectorTest", v); for_each(v.begin(), v.end(), cout << _1 << "\n"); } return 0; } Is this a problem or known issue when using cygwin? I'm using eclipse in addition, by the way. The code works perfectly fine on other platforms, so I imagine it must be a problem when using the serialization library with cygwin. Thanks in advance for any help you can offer me here and best regards Pascal Kesseli

4 3

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

21 Jul '09

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

2 2

[Math Toolkit] Gamma Distribution
by Christian Henning 21 Jul '09

21 Jul '09

Hi there, the documention of the gamma distribution talks about two different forms. How can can generate the second one with shape=1 and scale=1? I know how to get the first form: boost::math::gamma_distribution<> dist( 1, 1 ); I don't see how to get the second one? Thanks, Christian

5 13

[Boost.Test] valgrind complains about invalid reads
by Peter Klotz 20 Jun '09

20 Jun '09

Boost 1.37.0, Linux, x86_64, gcc 4.1: Valgrind 3.3.1 complains when running this simple test: #define BOOST_TEST_MAIN #include <boost/test/unit_test.hpp> BOOST_AUTO_TEST_CASE( test1 ) { BOOST_CHECK( 1 == 1 ); } The valgrind messages: ==1836== Memcheck, a memory error detector. ==1836== Copyright (C) 2002-2007, and GNU GPL'd, by Julian Seward et al. ==1836== Using LibVEX rev 1854, a library for dynamic binary translation. ==1836== Copyright (C) 2004-2007, and GNU GPL'd, by OpenWorks LLP. ==1836== Using valgrind-3.3.1, a dynamic binary instrumentation framework. ==1836== Copyright (C) 2000-2007, and GNU GPL'd, by Julian Seward et al. ==1836== For more details, rerun with: -v ==1836== Running 1 test case... *** No errors detected ==1836== Invalid read of size 8 ==1836== at 0x41E41C: __tcf_1 (in RegressionTest) ==1836== by 0x3830632FA4: exit (in /lib64/libc-2.5.so) ==1836== by 0x383061D8BA: (below main) (in /lib64/libc-2.5.so) ==1836== Address 0x4c1ade0 is 40 bytes inside a block of size 48 free'd ==1836== at 0x4A0555C: operator delete(void*) (vg_replace_malloc.c:342) ==1836== by 0x42024C: std::_Rb_tree<unsigned long, std::pair<unsigned long const, boost::unit_test::test_unit*>, std::_Select1st<std::pair<unsigned long const, boost::unit_test::test_unit*> >, std::less<unsigned long>, std::allocator<std::pair<unsigned long const, boost::unit_test::test_unit*> > >::erase(unsigned long const&) (in RegressionTest) ==1836== by 0x417C11: boost::unit_test::test_unit::~test_unit() (in RegressionTest) ==1836== by 0x41E41B: __tcf_1 (in RegressionTest) ==1836== by 0x3830632FA4: exit (in /lib64/libc-2.5.so) ==1836== by 0x383061D8BA: (below main) (in /lib64/libc-2.5.so) ==1836== ==1836== ERROR SUMMARY: 1 errors from 1 contexts (suppressed: 5 from 1) ==1836== malloc/free: in use at exit: 0 bytes in 0 blocks. ==1836== malloc/free: 89 allocs, 89 frees, 38,935 bytes allocated. ==1836== For counts of detected errors, rerun with: -v ==1836== All heap blocks were freed -- no leaks are possible. I was not able to figure out what valgrind is really complaining about. The internal map of test units holds two entries. One corresponds to my BOOST_AUTO_TEST_CASE call (id=1), the other one seems to be used internally (id=65536). Both are properly erased in deregister_test_unit() which is called by ~test_unit(). Maybe someone with a little bit more insight into Boost.Test is willing to look at this. Regards, Peter.

5 11