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

Boost release 1.40.0 is available
by Beman Dawes 27 Aug '09

27 Aug '09

Boost release 1.40.0 is available from SourceForge: http://sourceforge.net/projects/boost/files/ As always, the .7z and .zip files have Windows line endings, while the .bz2 and .gz files have POSIX line endings. Please download a file that has line endings appropriate for your platform. There are no new libraries in this release, but lots of libraries have been updated and had bugs fixed, including: * Accumulators * Asio * Circular Buffer * Filesystem * Foreach * Function * Fusion * Hash * Interprocess * Intrusive * MPL * Program Options * Proto * Python * Serialization * Unordered * Xpressive Enjoy! -- The Boost Release Managers

3 2

[serialization]
by Marco Canini 27 Aug '09

27 Aug '09

Hi all, I'm getting exhausted with trying to get a functioning test case. I'm attaching my code, which consits of two classses: the second class Route is having a const reference member to the first class Protocol. It seems that I am able to serialize the two classes correctly, however when I deserialize the reference member is set to NULL. What am I doing wrong? Further, I was trying to have save_construct_data(...const Route * t...) to access the private members of the class Route by making it friend of the class. However, doing so results in a compiler error. Any idea of how I am supposed to deal with that? Thank you very much! -- Marco Canini

1 0

Iterating over the seconds of a list of pairs
by Zachary Turner 27 Aug '09

27 Aug '09

I have an std::list<std::pair<double, float> > internal to a class. I'd like to expose begin() and end() functions from this class, that iterate over the second element of the list. So basically I want the user of the class to only see a sequence of floats. transform_iterator seems to fit the bill, but I think I'm seriously overcomplicating this. I've tried the following and it's not working. I feel like this pair_second / pair_first, and iterator should either already all be defined somewhere that I'm just not aware of, or be much shorter to write than this. template<class First, class Second> struct pair_second : public std::unary_function<std::pair<First, Second>, const Second&> { const Second& operator()(std::pair<First,Second>& pair) const { return pair.second; } }; template<class First, class Second> struct pair_first : public std::unary_function<std::pair<First, Second>, const First&> { const First& operator()(std::pair<First,Second>& pair) const { return pair.first; } }; template<class pair_type> struct first_iterator { typedef boost::transform_iterator< pair_first<typename pair_type::first_type, typename pair_type::second_type>, typename pair_type > type; }; template<class pair_type> struct second_iterator { typedef boost::transform_iterator< pair_second<typename pair_type::first_type, typename pair_type::second_type>, typename pair_type > type; }; typedef typename first_iterator<double_float_pair>::type float_iterator; float_iterator begin() { return float_iterator(items_.begin(), second_iterator<double_float_pair>()); } float_iterator end() { return float_iterator(items_.end(), second_iterator<double_float_pair>()); } Any advice would be appreciated. Thanks Zach

12 22

[date_time] cannot understand idea behind c_time::gmtime function
by Roman Shmelev 27 Aug '09

27 Aug '09

Hi! I've seen different behaviour under Windows and FreeBSD of the function boost::date_time::c_time::gmtime Looking inside it confused me: (Windows variant) inline static std::tm* gmtime(const std::time_t* t, std::tm* result) { // RESULT VALUE IS NOT USED. SO WHAT SENSE TO PASS ANYTHING HERE? result = std::gmtime(t); if (!result) boost::throw_exception(std::runtime_error("could not convert calendar time to UTC time")); return result; } Under FreeBSD everything is ok (gmtime_r is used). So, probably it is a bug? I mean, the idea of cross-platform library - is that functions will operate in similar way on different platforms. Also, please look at documentation http://www.boost.org/doc/libs/1_39_0/doc/html/boost/date_time/c_time.html it says c_time has 4 functions (as I understand, 2 functions were cloned :/ ) so it also seems to be a bug.

2 3

why send_to does not send all the packets..?
by Paolo Viotti 27 Aug '09

27 Aug '09

Hi, I'm using the boost asio library to create and manage a UDP socket. In my code I've a for loop in which I send some packets, as follows: for (it=refreshList.begin(); it !=refreshList.end(); it++){ [...] std::cout<<"___Refreshing: sent a KADEMLIA2_HELLO_REQ packet to "<< (*it)->getAddressString()<<":"<<(*it)->getUDPPort()<<std::endl; Packet* packet = new Packet(CList->getMyNodeID(),true); // KADEMLIA2_HELLO_REQ socket->send_to(boost::asio::buffer(packet->getPacket(), packet->getSize()), (*re)); } and actually it sends the packets, but just some of them: it stops sending after having sent about 30 packets out of 150, without throwing any exception or giving any error, while the loop continue and all the strings are regularly printed on std::cout. i'm really puzzled... is there any buffer to flush or any shutdown timer to take into account..? could it be due to another thread that is blocked on the receive_from function..? thanks, p_viotti

1 0

[interprocess] shared_memory_object design
by Christian Henning 27 Aug '09

27 Aug '09

Hi there, a colleague of mine is questioning the design decisions in the shared_memory_object class. The assignment operator and copy constructor are private. Why? Also the constructor takes a type instead of a value as the first parameter. In his eyes it makes it very inflexible since the creation of or attaching to shared_memory area has to known at compile time. A few pointers on why such a design was chosen would be very helpful. Regards, Christian

1 0

[assign] Inconvenience with list_of
by dariomt＠gmail.com 27 Aug '09

27 Aug '09

Hi list, I found a small inconvenience while trying to use boost::assign::list_of with the following code #include <vector> #include <boost/assign/list_of.hpp> struct A { // no implicit conversion from int explicit A(int i) {} // constructor from two ints A(int i, int j) {} }; void works_with_two_ints() { std::vector<A> v = boost::assign::list_of<A>(1,2)(3,4)(5,6); } #if 0 void does_not_work_with_one_int() { std::vector<A> v = boost::assign::list_of<A>(1)(2)(3); } #endif Looking into the code, it seems as if assign_detail::generic_list<A> had two slightly different versions of operator() generated with some preprocessor magic // 1) accepts a fully constructed A generic_list& operator()( const A& u ) { this->push_back( u ); return *this; } // 2) accepts two generic paramenters forwarding them to A's two parameter constructor template <typename U, typename V> generic_list& operator()(const U& u, const V& v) { this->push_back( A(u,v) ); return *this; } I'm missing another operator() that takes one generic parameter and forwards it to A's one parameter constructor, for the case where it is not implicitly convertible. // 3) accepts a generic paramenter forwarding it to A's one parameter constructor template <typename U> generic_list& operator()(const U& u) { this->push_back( A(u) ); return *this; } That saves me from typing A's everywhere std::vector<A> v = boost::assign::list_of(A(1))(A(2))(A(3)); Does this make sense? Thanks!

2 1

[ASIO] work/development flow example
by smbrown＠teledyne.com 27 Aug '09

27 Aug '09

Afternoon all, I'm using the ASIO library to develop a simple application and have a favour to ask. Up until now I have been making many mistakes as I expected, but what I didn't expect was having so much trouble debugging the problems. Perhaps this is normal for applications with threads and callbacks. Any insight you could share about your development work flow and use of debuggers would be gratefully received. I'm using MinGW and VS2K8 on XP and with either debugger trying to find the context of the devassert/exception is hard. Simon P. S. This is a repost, I think our mail server ate the original as I saw no sign of it on the list. Apologies if I'm wrong.

1 0

[boost-users][function_types] Function synthesis and property tags
by yuryk 27 Aug '09

27 Aug '09

I'm trying to synthesize a function type G from another function type F with some changes on return and arg types. What I need to remain constant, are all these additional properties that are returned by components<F>::tag. I want them all as a bunch, don't really fancy querying them all one by one, they sorta look contradictive or at least redundant. The simple approach of passing the same tag to the function_type<Seq, Tag> does not make a change - it leads to the same result as if null_tag were passed. How do you generate a function type with the same properties as the given one? Sample code: //-------------------------------------------------------------------------------------------------- typedef int (__stdcall * pfn1)(char); typedef int ( * pfn2)(char); typedef remove_pointer<pfn1>::type fn1; typedef ft::components<fn1> comps1; // generate type hopefully like pfn1 typedef ft::function_pointer<comps1, //create tag from mask and bits ft::detail::property_tag<comps1::bits::value, comps1::mask::value> >::type pfn_gen1; BOOST_MPL_ASSERT((is_same<pfn_gen1, pfn2>)); // this one passes, should fail, though BOOST_MPL_ASSERT((is_same<pfn_gen1, pfn1>)); // I wish, but fails //-------------------------------------------------------------------------------------------------- -- View this message in context: http://www.nabble.com/-boost-users--function_types--Function-synthesis-and-… Sent from the Boost - Users mailing list archive at Nabble.com.

1 0

[iostreams] how to flush mapped_file
by Stefan Strasser 27 Aug '09

27 Aug '09

Hi, I'm new to the iostreams library, how do you flush/sync a iostreams::mapped_file? iostreams::flush(mapped_file) doesn't seem to work, although it works with a normal file device. are mapped_file's not flushable? the documentation of concept Flushable strangly says that it is an implementation detail, if a device is flushable. it should result in a call to msync() on posix systems and must not be an implementation detail, but the flush must be guaranteed. is this not possible with iostreams? thank you

1 0