July 2011 - Boost - lists.boost.org

[assign v2] thoughts
by Gordon Woodhull 09 Jul '11

09 Jul '11

Hi Erwann, all, I had a thought about this library which I wasn't able to articulate during the review, and I don't think anyone else raised it. I think this library would be a useful utility if it had the right syntax, but I found issue especially with the use of operator| which is IMO not consistent with the way Boost.Range uses it. I would expect the pipe operator to just stream data and not affect the object on the LHS. I know everything is relative and in some EDSLs this might be appropriate, but it's a strain for me to think of pipe as anything other than unidirectional data flow. The syntax that you have now cal | do_csv_put<2>("jan", 31, "feb", 28, "mar", 31) looks kind of like Smalltalk or Objective C to me, like you are trying to send some arbitrary message to the container on the LHS. Did you consider keeping the overloaded operator+= from Assign v1, or another assignment operator that always modifies the LHS? cal += values<2>("jan", 31, "feb", 28, "mar", 31) is much clearer to me. And the assignment operators are expected to return a reference to the LHS. Yes, there is an ambiguity with other libraries which have meant += to mean "add to each element" but with the rare exception of containers containing elements of their own type, this should always be resolvable. If this is unacceptable, I think operator|=, meaning union-assign, also makes some sense. Or am I missing something? I always thought of Assign v1 as "the funny operator, library" and on at least one occasion didn't use it because I knew I'd have to explain what it's doing to my colleagues. So Assign v2 looks very powerful and useful to me, but in order to become a workhorse for everyone, it has to have the most blunt obvious syntax so that no one has to guess or explain anything. Cheers, Gordon

1 0

[graph] bug in ith_bandwidth?
by Joel Young 08 Jul '11

08 Jul '11

Is ith_bandwidth broken? It only measures the distance between indices when the source is greater than the sink. Should it look like this: template <typename Graph, typename VertexIndexMap> typename graph_traits<Graph>::vertices_size_type ith_bandwidth(typename graph_traits<Graph>::vertex_descriptor i, const Graph& g, VertexIndexMap index) { BOOST_USING_STD_MAX(); typedef typename graph_traits<Graph>::vertices_size_type size_type; size_type b = 0; typename graph_traits<Graph>::out_edge_iterator e, end; for (boost::tie(e, end) = out_edges(i, g); e != end; ++e) { int f_i = get(index, i); int f_j = get(index, target(*e, g)); using namespace std; // to call abs() unqualified if(f_i > f_j) b = max BOOST_PREVENT_MACRO_SUBSTITUTION (b, size_type(f_i - f_j)); else b = max BOOST_PREVENT_MACRO_SUBSTITUTION (b, size_type(f_j - f_i)); } return b; } Joel

2 1

Boost
by Joshua Boyce 08 Jul '11

08 Jul '11

Hey all, Ruben over at the MinGW-w64 project has started providing a Clang toolchain built upon MinGW-w64. http://lists.cs.uiuc.edu/pipermail/cfe-dev/2011-July/016005.html I just tried compiling Boost trunk r72754 with it, and figured I'd post the output here so work can commence bringing it in line with GCC via MinGW-w64. 90% of the errors seem to be a result of a 'call to deleted constructor' which seems to be caused by the shared_ptr library (or its usage)... Then the remaining 5% seem to be quite similar in nature, so it doesn't look like stuff will be too horrible to fix, but I don't know enough about the implementation of shared_ptr to fix it at the moment. One other thing to take into consideration will be the lack of a Clang Windows toolset. Currently when you set your toolset to Clang under MinGW, Boost treats it as Clang-Linux. I'm not sure what exactly that will affect... The script I used and the resulting output is available here: http://www.raptorfactor.com/code/Boost-Clang-MinGW-Output-20110708a.zip Thanks..

1 0

[fusion] random-access single_view
by Eric Niebler 07 Jul '11

07 Jul '11

Seems like a dumb thing, but it would actually help make my code simpler if single_view were random access. Patch attached. OK to commit (to trunk, of course)? The patch excludes the regenerated docs, which I would commit along with the rest. Thanks, -- Eric Niebler BoostPro Computing http://www.boostpro.com

2 1

More Boostcon videos available
by Marshall Clow 07 Jul '11

07 Jul '11

Monday: Micheal Wong - C++0x Progress and status http://blip.tv/boostcon/c-0x-progress-and-status-5349952 Ruediger Berlich - Distributed parametric optimization with the Geneva library http://blip.tv/boostcon/distributed-parametric-optimization-with-the-geneva… Tuesday: Joel Falcou - Automatic Hybrid MPI+OpenMP Code Generation http://blip.tv/boostcon/automatic-hybrid-mpi-openmp-code-generation-5349961 Sebastian Schaetz - Mastering the Cell Broadband Engine architecture through a Boost based parallel communication library http://blip.tv/boostcon/mastering-the-cell-broadband-engine-architecture-th… Beman Daves -- The proposed Boost B-tree library http://blip.tv/boostcon/the-proposed-boost-b-tree-library-5349968 Boris Kolpackov - Parsing C++ with GCC plugi http://blip.tv/boostcon/parsing-c-with-gcc-plugins-5349971 More next week.... -- Marshall Marshall Clow Idio Software <mailto:mclow.lists@gmail.com> A.D. 1517: Martin Luther nails his 95 Theses to the church door and is promptly moderated down to (-1, Flamebait). -- Yu Suzuki

1 0

[1.47.0][fusion] fusion::copy is undocumented?
by Akira Takahashi 06 Jul '11

06 Jul '11

Hi. boost::fusion::copy algorithm merged release, but undocumented. Should not use this in 1.47.0? I think fusion::copy is very useful. Thanks. >>======================== Akira Takahashi mailto:faithandbrave@gmail.com blog:http://d.hatena.ne.jp/faith_and_brave/

2 3

[1.47.0] Permission to merge filesystem fix
by Beman Dawes 06 Jul '11

06 Jul '11

This changeset: http://svn.boost.org/trac/boost/changeset/72855 Has now cycled through many of the regression testers. Testing is still a little light on Windows, but I've tested locally on VC++ 8, 9, and 10, and late model mingw, too. This fixes a nasty POSIX case that was throwing before main() started if the environment LANG variable was whacked. OK to merge? --Beman

3 7

how to receive UDP in Windows using boost::asio@boost1.46.0
by Jinqiang Zhang 06 Jul '11

06 Jul '11

I have two classes UDPSender and UDPReceiver. Senarios: (1)It works when I run UDPSender in QNX and UDPReceiver in Linux. (2)It doesn't work when I run UDPSender in QNX and UDPReceiver in WindowsXP.(But I see UDP datagrams using Wireshark, meaning Sender is working as in Senario(1)) (3)It partially works in Windows. i) if I use Sender sending UDP messages to 127.0.0.1, Receiver can get messages ii) if I use Sender sending UDP messages to 192.168.1.137( local address in the network ), Receiver can NOT receive. My question is: *How can I get Receiver working on Windows?* ------------------------------------------------------------------------------------------------------------------------------------------ some code: UDPReceiver::UDPReceiver( boost::asio::io_service& service, unsigned short port ) : m_service( service ), m_socket( service, udp::endpoint(udp::v4(), port)), m_peerEndpoint(), m_buffer(), m_messageReceipient() { } void UDPReceiver::SetMessageReceipient( const boost::function<void(const char*, unsigned int)>& r ) { m_messageReceipient = r; } void UDPReceiver::Start() { m_socket.async_receive_from( boost::asio::buffer(m_buffer, m_buffer.size()), m_peerEndpoint, boost::bind( &UDPReceiver::HandleReceived, shared_from_this(), boost::asio::placeholders::error, boost::asio::placeholders::bytes_transferred)); } void UDPReceiver::Stop() { m_socket.close(); } void UDPReceiver::HandleReceived( const boost::system::error_code& error, size_t bytesReceived ) { assert( !m_messageReceipient.empty() ); if( error ) { LOG( Comms, Error, "UDPReceiver:" << error.message() ); } else{ m_messageReceipient( &m_buffer.front(), bytesReceived ); m_socket.async_receive_from( boost::asio::buffer(m_buffer, m_buffer.size()), m_peerEndpoint, boost::bind( &UDPReceiver::HandleReceived, shared_from_this(), boost::asio::placeholders::error, boost::asio::placeholders::bytes_transferred)); } }

2 1

Re: [boost] Logging library
by Ivan Le Lann 06 Jul '11

06 Jul '11

----- "Tigran Hayrapetyan" <tigran.human(a)gmail.com> a écrit : > >----- "Tigran Hayrapetyan" <tigran.human(a)gmail.com> a écrit : > > > >> Hello dear Boost community > >> > >>(...) > >> Existing logging libraries which I used, are writing log messages > >> linearly, > >> just one after another. > >> In contrast, as we know, almost any application's working structure > is > >> not > >> linear, but hierarchical instead, i.e. functions which call other > >> functions, > >> and so on. > >>(...) > > > >Have you tried Boost.Log and its attributes ? > > > >http://boost-log.sourceforge.net/libs/log/doc/html/log/detailed/attributes.… > >http://boost-log.sourceforge.net/libs/log/doc/html/log/detailed/utilities.h… > > > As I understand, that features of Boost.Log library are for managing > attributes of log messages, and not log messages (or their > placements) > themselves. > So the log messages are still placed linearly, and my proposal is to > place them hierarchically. NB: I'm not a Boost.Log expert at all. I was only curious. And my understanding is that how to handle attributes is left to the sink backend. I would try writing a new sink backend that handles a named_scope attribute. http://boost-log.sourceforge.net/libs/log/doc/html/log/extension.html#log.e… Ivan

1 0

(no subject)
by Bruce Adams 05 Jul '11

05 Jul '11

Hi, This probably ought to be a FAQ. But there is currently no database library in boost. There have been some contenders in the past. I had hopes for SOCI for instance (see e.g. http://lists.boost.org/Archives/boost/2004/10/74365.php) Are there any candidates on the horizon? If not, what unrelated projects would people here recommend (or recommend against). Just for context, I am looking for a "sword against anybody who got into belief that (db) programming in the language beginning with 'J' is easier. ;) " to quote the author of SOCI in the post I mentioned. However, I'm beginning to suspect maybe it is easier in Java, not through any fault in C++ itself but in the success of the Java community versus the lack of *recognised* success in the C++ community. By the way when I search the mailing lists from the link on http://www.boost.org/community/groups.html I (and many others) get: Google Sorry... We're sorry... ... but your computer or network may be sending automated queries. To protect our users, we can't process your request right now.See Google Help for more information. Regards, Bruce.

4 4