October 2008 - Boost-users - lists.preview.boost.org

Re: [Boost-users] Apple iPod uses Boost
by Robert Marion 18 Oct '08

18 Oct '08

> From: Gennaro Prota > Subject: Re: [Boost-users] Apple iPod uses Boost > > Well, of course Boost isn't the only library using the Boost > Software License. And, are they distributing the sources? > Otherwise I don't see why reproducing the license. I haven't seen the Boost License being used outside of Boost, although certainly the license can be applied to any open source software. Since the Boost license is similar to the BSD style license, it is likely that an open source project would choose one of the more popular BSD or MIT style licenses. Also, the Boost license does not require that source be available for distribution. That is a characteristic of GPL, LGPL and other copyleft licenses. The Boost license is an attribution license. See http://www.boost.org/users/license.html. ________________________________ Robert Marion http://www.palamida.com

4 3

Must cast unsigned int for random generator seed
by Phillip Jones 18 Oct '08

18 Oct '08

I wanted to save the initial state of a random number generator and I ran into this issue - boost::mt19937 rng; unsigned int rseed = static_cast<unsigned int>(std::time(0)); //rng.seed( rseed ); // This causes an error rng.seed( static_cast<unsigned int>(rseed) ); // ok The error is "mersenne_twister.hpp:103: error: `gen' cannot be used as a function". Obviously not a big problem since I can just use the cast and keep going. But is there a reason for this behavior? It seems odd that one must cast an unsigned int as an unsigned int to make it work.

2 1

[Proto] Extracting types of sub-expression in transform
by Joel Falcou 18 Oct '08

18 Oct '08

Hello, i have a DSL where terminal are defined like : terminal< simd<proto::_> > What I want is that any binary operator that apply on those terminal should be available only if both lhs and rhs parts have the same underlying type. E.G : simd<float> + simd<float> is ok while (simd<char> + simd<char>)*simd<float> is not ok. How I can enforce this check ? Is having a transform that take an expression and returns the underlying type OR mpl::void_ if an error occurs the good solution ? Thanks in advance -- ___________________________________________ Joel Falcou - Assistant Professor PARALL Team - LRI - Universite Paris Sud XI Tel : (+33)1 69 15 66 35

3 6

Re: [Boost-users] return type of boost::iostreams::stream<T>::read()
by Struebing, Axel, le-tex 18 Oct '08

18 Oct '08

okay, it's within my grasp now. Primarily I get a basic_istream. To facilitate the 'while std::cin' idiom, there is a conversion of basic_istream to void* defined. That seems to hit me in my example. FYI: std::streamsize i = istr.read(buffer,10); is calling the read-function of my DeviceT template parameter, perfect. So I am probably going to use a read + gcount-combination. Thanks again Axel > Hallo Éric, > > although it is not an exact fit it will be of value in evaluating my > options. > > kind regards > Axel Strübing > > > > Reading the documentation about boost::iostreams library I could not > > > find the information which return type the boost::iostreams::stream's > > > class is giving me calling read(). > > > > > >From the documentation, a boost::iostreams::stream derives from an > > > std::basic_[i|o]stream, depending on the model of its underlying > > > Device. In your example, the device is a Source, so I guess that the > > > stream derives from std::basic_istream, whose read(char*, streamsize) > > > method returns a basic_istream, not a streamsize. > > > > The library offers the "streamsize read(stream, char*, streamsize)" > > function template. Using it in your example yould give something like > > (not tested): > > > > #include <boost/iostreams/stream.hpp> > > #include <boost/iostreams/device/file.hpp> > > #include <boost/iostreams/operations.hpp> > > > > int main(void){ > > namespace bio = boost::iostreams; > > bio::stream<bio::file_source> istr(std::string("in.pdf")); > > char buffer[10]; > > std::streamsize i = bio::read(istr,buffer,10); > > return 0; > > } > > > > > > HTH, > > > > Éric Malenfant > > --------------------------------------------- > > Two wrongs don't make a right but three rights make a left. > > --

1 0

nested template tests using boost.test BOOST_AUTO_TEST_CASE_TEMPLATE
by johanngoetz 18 Oct '08

18 Oct '08

It seems that one cannot do nested AUTO_TEST_CASE_TEMPLATE's with boost.test. Is there another way around this (besides unrolling manually)? I wish to do something like the following: i.e. testing every combination of two (or more?) lists of types. Example below. Thank you, Johann. define BOOST_TEST_MODULE std_map #define BOOST_TEST_DYN_LINK #include <map> #include <boost/test/unit_test.hpp> #include <boost/test/test_case_template.hpp> #include <boost/mpl/list.hpp> typedef boost::mpl::list<short, int, long, long long> test_int_types; typedef boost::mpl::list<float, double, long double> test_float_types; BOOST_AUTO_TEST_SUITE( std_map_suite ) BOOST_AUTO_TEST_CASE_TEMPLATE( int_template_case, Ti, test_int_types ) { BOOST_AUTO_TEST_CASE_TEMPLATE( float_template_case, Tf, test_float_types ) { std::map<Ti, Tf> m; m.insert(std::make_pair((Ti)0,(Tf)0)); BOOST_CHECK_EQUAL(m.size(), size_t(1)); } } BOOST_AUTO_TEST_SUITE_END() -- View this message in context: http://www.nabble.com/nested-template-tests-using-boost.test-BOOST_AUTO_TES… Sent from the Boost - Users mailing list archive at Nabble.com.

2 2

Using boost.signals with boost.bind
by Sergey Kishchenko 18 Oct '08

18 Oct '08

Hi, I have one problem with boost.bind and boost.signals I can't understand. Source code is the best description, therefore #include <iostream> #include <boost/signals.hpp> #include <boost/shared_ptr.hpp> #include <boost/bind.hpp> class Func : public boost::noncopyable { public: void operator()(int i) { std::cout<<"Func::operator("<< i <<")\n"; } }; typedef boost::signal<void ()> signal_type; Func f; signal_type::slot_type binded_func = boost::bind<void>(boost::ref(f),42); int main() { boost::shared_ptr<signal_type> apSig(new signal_type()); apSig->connect(binded_func); (*apSig)(); apSig.reset(new signal_type()); boost::signals::connection con = apSig->connect(binded_func); std::cout<<"Connected: "<<con.connected()<<" Blocked: "<<con.blocked()<<"\n"; } Program output: Func::operator(42) Connected: 0 Blocked: 1 Ie, if i create new signal and then try to connect to the same slot, it fails. If I use not boost::bind object but only boost::function as slot, all works ok. I can not understand, why? I used Boost 1.34.1, gcc 4.1.2 Regards, Sergey Kishchenko

6 6

Re: [Boost-users] boost::mpl::set and boost::mpl::sort -- does not compile -- any help appreciated
by peter_foelsche＠agilent.com 17 Oct '08

17 Oct '08

I found the answer meanwhile. Peter #include <boost/mpl/set.hpp> #include <boost/mpl/sort.hpp> #include <boost/mpl/inserter.hpp> #include <boost/mpl/insert.hpp> #include <boost/mpl/placeholders.hpp> #include <boost/mpl/front.hpp> #include <boost/mpl/deref.hpp> #include <boost/mpl/for_each.hpp> #include <iostream> typedef boost::mpl::set< boost::mpl::int_<1>, boost::mpl::int_<5>, boost::mpl::int_<3>, boost::mpl::int_<0> > CSet; typedef boost::mpl::inserter< boost::mpl::set<>, boost::mpl::insert<boost::mpl::_1, boost::mpl::_2> > CInserter; typedef boost::mpl::sort< CSet, boost::mpl::less< boost::mpl::_1, boost::mpl::_2 >, CInserter >::type COrderedSet;

1 0

boost::mpl::set and boost::mpl::sort -- does not compile -- any help appreciated
by peter_foelsche＠agilent.com 17 Oct '08

17 Oct '08

Somehow I need an ordered set. How to do this? Any help appreciated. Peter #include <boost/mpl/set.hpp> #include <boost/mpl/sort.hpp> typedef boost::mpl::sort< boost::mpl::set< boost::mpl::int_<1>, boost::mpl::int_<5>, boost::mpl::int_<3> > >::type CFirstSet; int main(int argc, char **argv) { CFirstSet s; } "The first point (using an init() function in preference to a constructor) is bogus. Using constructors and exception handling is a more general and systematic way of dealing with resource acquisition and initialization errors. This style is a relic of pre-exception C++." -- Stroustrup

1 0

[Thread] Bug in 1.35.0 and 1.36.0 timed wait code.
by Ross Wightman 17 Oct '08

17 Oct '08

I posted this shortly after the 1.36.0 release but I neglected to put the [thread] tag in the subject. Since I noticed 1.37 is already being finalized I figured I'd post again. There is a bug in the get_due_time(...) function in lib/src/win32/thread.cpp line 343 when boost::date_time is setup to have nanosecond resolution. The integer division always results in a due_time of 0 so the timed waits never wait. I fixed my local copy by changing the order of operations but I haven't investigated to see if this could cause an overflow in situations where nanosecond time resolution has not been configured. In general I question whether it was a good idea to have the threading library depend on date_time? There is a lot of functionality and code in date_time that the thread library doesn't need. The C++0x standard suggests a much simpler system_time class. While there is overlap between such a system_time class and the date_time class, the smaller compile time / code size costs and lower chance of bugs when only using the threading library seems worth it. Regards, Ross

1 0

Difficulty compiling - ASIO/BIND
by David 17 Oct '08

17 Oct '08

Hi, I'm trying to use asio, and I run into a compiler error when using async_connect. I don't think this is an asio issue, perhaps a compiler or bind issue. Errors and code snippet below. I'm perplexed by the error, as apparently the compiler can't if 'this' should be passed as a pointer, or by reference. I also tried alternate syntax using bind and mem_fn as mentioned in the bind docs, but that gave exactly the same error. What's causing this error, and what can I do to fix it? using gcc 4.2.3: .../boost_1_36_0/boost/bind.hpp:292: error: no match for call to ‘(boost::_mfi::mf1<void, Connection, boost::system::error_code&>) (Connection*&, boost::asio::error::basic_errors&)’ .../boost_1_36_0/boost/bind/mem_fn_template.hpp:160: note: candidates are: R boost::_mfi::mf1<R, T, A1>::operator()(T*, A1) const [with R = void, T = Connection, A1 = boost::system::error_code&] .../boost_1_36_0/boost/bind/mem_fn_template.hpp:179: note: R boost::_mfi::mf1<R, T, A1>::operator()(T&, A1) const [with R = void, T = Connection, A1 = boost::system::error_code&] //code class Connection { public: //... void onConnected(boost::system::error_code&); //... }; _socket.lowest_layer().async_connect(*_epit, boost::bind(&Connection::onConnected, this, placeholders::error));

2 1