- Boost-users - lists.boost.org

joining boost ranges in loop
by Ernest Zaslavsky 05 Feb '15

05 Feb '15

Actually the question is how to achieve it? The use of boost::any_range is possible but undesirable (performance considerations) Sample code: typedef boost::multi_index_container< int, bmi::indexed_by<bmi::hashed_unique<bmi::tag<struct someTag>, bmi::identity<int>>>> Datum; class bar { public: auto foo() const; private: Datum m_someDataContainer; }; SomeType bar::foo() const { SomeType retVal; for (auto i = 0u; i < 10; ++i) { retVal = boost::range::join(retVal, m_someDataContainer.equal_range(i)); } return retVal; } Here we can observe several problems. The return type cannot be easily deduced. The join would create joined_range<joined_range<joined_range and so on of nested types. The above can be achieved by using any_range like typedef boost::any_range<int, boost::random_access_traversal_tag, int&, std::ptrdiff_t> VersionRange; but above will create nested joined_range of any_range which, when the number of iterations is big enough, has tremendous impact on performance. So, what can be done here?

1 0

boost serialization + boost variant + boost blank
by Merrill Cornish 05 Feb '15

05 Feb '15

I'm running Boost 1.57.0 using serialization, variant, and boost::blank with that variant. I found that when compiled, serialization complains that blank does not contain a serialize method, which is true. I made a copy of blank.hpp, called it blank.h, and added template< class Archive > void serialize(Archive& ar, const unsigned int version) {} to the body of struct blank. That made the compiler happy, but is it a fix in general?

4 6

[statechart] fifo_scheduler<>::operator() in state constructor
by Andrew C 05 Feb '15

05 Feb '15

Hi, is it allowed to call fifo_scheduler<>::operator() from state constructor? If yes, then let's assume "scheduler" and "processor" are objects of "fifo_scheduler<>" and "fifo_scheduler<>::processor_handle" types respectively in the some state constructor code below: State(my_context ctx): my_base(ctx) { boost::intrusive_ptr<Event> pEvent(new Event()); // If both lines [A] and [B] are uncommented // then "unconsumed_event" method is called twice. // If just line [A] is uncommented then "unconsumed_event" method is called. // If just line [B] is uncommented then certain "react" method is called. // scheduler.queue_event(processor, pEvent); // [A] // this->post_event(Event()); // [B] scheduler(); } Note constructor calls "scheduler()" at the end. Other questions: Is this code correct and runs expectedly as described in comments? Why "unconsumed_event" is called if use "queue_event" instead of "post_event"? Thanks, Andrew.

1 0

Crashing Issue
by swati upadhyaya 04 Feb '15

04 Feb '15

In my code join() timed_joined() is crashing, I am using these API after interrupt(),Same piece of code is working for others but in my system its crashing, Is there some dependency problem I am using Boost version: 1.55.0 OS: Windows 8 IDE: Visual Studio 2010

2 1

[Variant] Assertion fails at variant/detail/visitation_impl.hpp:207
by Attila Rajmund Nohl 03 Feb '15

03 Feb '15

Hello! I'm using boost-1.55 on Linux. When I run a specific testcase, I get an abort at line 207 in the visitation_impl.hpp file. The comment says that "should never be here at runtime". Unfortunately I don't really understand under what circumstances could runtime end up here - could someone explain how? The variant has a recursive element, i.e. looks something like this: struct Resp; struct SR { } struct AR { std::vector<Resp> resp; } typedef variant< SR, recursive_wrapper<AR> > RMType; struct Resp { RMType rm; }

2 2

boost::ptr_container library question ?
by Danny K 02 Feb '15

02 Feb '15

The ptr_container supports an overload for std::auto_ptr<U>. I was curious if there was a way to add the same functionality for std::unique_ptr<U>. ie an overload that allows a std::unique_ptr to be added to a ptr_vector for example?

2 1

[type_erasure] Problem with iterators
by Marco Guazzone 01 Feb '15

01 Feb '15

Hi, I'm new to the Type Erasure library. I need to write a base class with type-erased forward iterators. To do so I've taken inspiration from the print_sequence.cpp example I've found in the Type Erasure source tree. At the end of the email there is the whole code of my tentative, called "erasure.cpp". Unfortunately, I've problems (presumably) with const iterators which cause the following compile error: $ g++ -Wall -Wextra -pedantic -std=c++11 -I$HOME/sys/src/git/boost -o erasure erasure.cpp -lm *** [begin error] *** In file included from /home/sguazt/sys/src/git/boost/boost/type_erasure/iterator.hpp:20:0, from erasure.cpp:2: /home/sguazt/sys/src/git/boost/boost/type_erasure/operators.hpp: In instantiation of ‘static R boost::type_erasure::dereferenceable<R, T>::apply(const T&) [with R = double&; T = __gnu_cxx::__normal_iterator<const double*, std::vector<double> >]’: /home/sguazt/sys/src/git/boost/boost/type_erasure/detail/instantiate.hpp:91:9: required from ‘static void boost::type_erasure::detail::instantiate_concept7::apply(Concept*, Map*) [with Concept = boost::mpl::vector<boost::type_erasure::forward_iterator<>, boost::type_erasure::same_type<boost::type_erasure::deduced<boost::type_erasure::iterator_value_type<boost::type_erasure::_self> >, double>, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na>; Map = boost::mpl::map1<boost::mpl::pair<boost::type_erasure::_self, __gnu_cxx::__normal_iterator<const double*, std::vector<double> > > >]’ /home/sguazt/sys/src/git/boost/boost/type_erasure/any.hpp:225:13: required from ‘boost::type_erasure::any<Concept, T>::any(U&&) [with U = __gnu_cxx::__normal_iterator<const double*, std::vector<double> >&; Concept = boost::mpl::vector<boost::type_erasure::forward_iterator<>, boost::type_erasure::same_type<boost::type_erasure::deduced<boost::type_erasure::iterator_value_type<boost::type_erasure::_self> >, double>, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na, mpl_::na>; T = boost::type_erasure::_self]’ erasure.cpp:36:40: required from ‘std::vector<_RealType> base<T>::foo(IterT, IterT) [with IterT = __gnu_cxx::__normal_iterator<const double*, std::vector<double> >; T = double]’ erasure.cpp:77:73: required from here /home/sguazt/sys/src/git/boost/boost/type_erasure/operators.hpp:134:44: error: invalid initialization of reference of type ‘double&’ from expression of type ‘const double’ static R apply(const T& arg) { return *arg; } *** [/end error] *** The GCC version is 4.9.2. Here below is the content of "erasure.cpp". If I comment lines below the comment "THE FOLLOWING WON'T COMPILE" I get no error. So the problem is due to the use of cbegin/cend. *** [erasure.cpp] *** #include <boost/type_erasure/any.hpp> #include <boost/type_erasure/iterator.hpp> #include <boost/type_erasure/operators.hpp> #include <boost/type_erasure/same_type.hpp> #include <boost/mpl/vector.hpp> #include <boost/smart_ptr.hpp> #include <cmath> #include <cstddef> #include <iostream> #include <iterator> #include <vector> namespace mpl = boost::mpl; namespace te = boost::type_erasure; template <typename T> using fwd_iter_impl_t = te::any< mpl::vector< te::forward_iterator<>, te::same_type<te::forward_iterator<>::value_type,T>>>; template <typename T> class base { protected: typedef fwd_iter_impl_t<T> fwd_iter_type; public: virtual ~base() { } template <typename IterT> std::vector<T> foo(IterT first, IterT last) { return this->do_foo(first, last); } private: virtual std::vector<T> do_foo(fwd_iter_type first, fwd_iter_type last) = 0; }; // base template <typename T> class derived: public base<T> { private: typedef typename base<T>::fwd_iter_type fwd_iter_type; std::vector<T> do_foo(fwd_iter_type first, fwd_iter_type last) { const std::ptrdiff_t n = std::distance(first, last); std::vector<T> res(n); for (std::size_t i = 0; first != last; ++first, ++i) { res[i] = std::pow(*first,i)/n; } return res; } }; // derived int main() { std::vector<double> dv; dv.push_back(5); dv.push_back(8); dv.push_back(11); boost::scoped_ptr< base<double> > p_base(new derived<double>()); // begin/end const std::vector<double> foos = p_base->foo(dv.begin(), dv.end()); for (std::size_t i = 0; i < foos.size(); ++i) { std::cout << "foos[" << i << "]: " << foos[i] << std::endl; } //--- THE FOLLOWING WON'T COMPILE // cbegin/cend const std::vector<double> cfoos = p_base->foo(dv.cbegin(), dv.cend()); for (std::size_t i = 0; i < cfoos.size(); ++i) { std::cout << "cfoos[" << i << "]: " << cfoos[i] << std::endl; } } *** [/erasure.cpp] *** Thank you very much for your help. Best, -- Marco

3 9

[Asio] Handlers of asycn_send_to are called in which order?
by Attila Rajmund Nohl 31 Jan '15

31 Jan '15

Hello! My code sends UDP packets using async_send_to. If all packets are successfully sent, are their handlers called in the same order as the async_send_to calls happened? I mean if my code looks something like this: socket.async_send_to(buffer1, destination, handler1); socket.async_send_to(buffer2, destination, handler2); Is it guaranteed that handler2 is always called after handler1?

3 2

[iostreams] Devices and WOULD_BLOCK
by David Hawkins 31 Jan '15

31 Jan '15

Hi all, Boost newcomer here ... I have an application where I need to encode/decode a serialized data stream. I have used Boost.Iostreams filtering_stream filter to implement encoding and decoding function of each layer. The device interface is via a USB device, eg., think USB-to-Serial device, but without the classic Virtual COM Port or ttyUSB0 layer available. For example, the FTDI FT232H USB-to-MPSSE interface is a suitable cable http://www.ftdichip.com/Products/Cables/USBMPSSE.htm The cable can be used to implement either JTAG or SPI mode access (the mode selection introduces different filters into the filtering_stream stack). I have the filters working, and am working on getting a device working. Rather than deal with the details of the final hardware, I figured I'd simplify the design by creating a client/server design via sockets; the client code matches what I would use with real hardware, while the server code emulates the hardware. I started with this example for a socket device http://stackoverflow.com/questions/12023166/is-there-a-boostiostreams-bidir… The "problem" with the protocol I need to decode (which I have no choice to change) is that the data stream possibly contains escaped characters, so there is no way to know the read data length at the socket layer - its up to the filtering_stream layers to request data from the device layer until a complete packet is decoded. This all sounds good in theory, but in practice the filter layers attempt to read in blocks, and the read size is often larger than the read data that the device layer will supply. This lead to higher-level layers blocking in read(). I figured (perhaps incorrectly) that I could deal with this using non-blocking socket reads. After reading Boost.Iostreams non-blocking support in Section 3.6 http://www.boost.org/doc/libs/1_57_0/libs/iostreams/doc/index.html http://www.boost.org/doc/libs/1_57_0/libs/iostreams/doc/guide/asynchronous.… I modified the socket device example in the link above to; * put the socket in non-blocking mode before constructing the device * change the device 'read' procedure to return WOULD_BLOCK rather than throw an exception This does not work, and its not due to the filter layers. There is an issue with the device layer. Here's the issue ... given a design with a filtering_stream created only with a socket_device and no filters, if I trace the code in a debugger (Boost 1.57.0 source, Visual Studio 2012 under Win7), the socket_device read method call return sequence is boost/iostreams/read.hpp - read_device_impl read template at line 187 - read at line 52 boost/iostreams/detail/adapter/concept_adapter.hpp - device_wrapper_impl read at line 169 - read at line 77 boost/iostreams/detail/streambuf/indirect_streambuf.hpp - line 258 i.e., this source file https://github.com/boostorg/iostreams/blob/master/include/boost/iostreams/d… and this particular block of code // Read from source. std::streamsize chars = obj().read(buf.data() + pback_size_, buf.size() - pback_size_, next_); if (chars == -1) { this->set_true_eof(true); chars = 0; } setg(eback(), gptr(), buf.data() + pback_size_ + chars); The code tests for EOF (-1), but not WOULD_BLOCK (-2), so after this point, since chars is -2, things go bad. So I guess my question now is; Have I just bumped into the as-yet-unsupported part of Boost.Iostreams support for asynchronous I/O? Cheers, Dave PS. I can post example code if anyone wants to trace the code for themselves, I just figured I'd post the question to start with.

4 15

Joining boost::asio::strands
by Cesar Mello 30 Jan '15

30 Jan '15

Hi, After spawning several paralell jobs through calls to boost::asio::spawn inside a loop, is there an easy way to wait for all of them to finish? I would like to join them to aggregate the results. Please can you give me any clue? Thank you a lot! Best regards Cesar

1 0