October 2012 - Boost-users - lists.preview.boost.org

[mpi] Collectives on intercommunicator ?
by Alain O Miniussi 11 Oct '12

11 Oct '12

Am I correct in assuming that collective communications are not supported with Boost.MPI intercommunicators ? Thanks -- Alain Miniussi Responsable Tech. Centre de Calcul Haute Performance Observatoire de la Côte d'Azur - BP 4229 - 06304 Nice Cedex 4 tel: +33 4 92 00 30 09 https://crimson.oca.eu

2 2

Asio and ZeroMQ
by Matthias Vallentin 11 Oct '12

11 Oct '12

A while ago, Maciej Gajewski started an effort to use Asio to observe a ZeroMQ socket [1]. Based on his code, I put together a thin library that wraps the implementation details. Since other have requested a working example, here it is now at github: https://github.com/mavam/bmq Any feedback is highly appreciated. A basic usage example looks like this: #include <boost/asio/io_service.hpp> #include <bmq.h> int main() { boost::asio::io_service io; bmq::context ctx(1); bmq::component c(ctx, io); auto sink = c.add_sink(ZMQ_PAIR, "inproc://x"); c.subscribe(sink, [](bmq::message&& msg) { std::string str(static_cast<char*>(msg.data()), msg.size()); std::cout << str << std::endl; }); auto source = c.add_source(ZMQ_PAIR, "inproc://x"); bmq::message msg(const_cast<char*>("foo"), 3, nullptr); source->send(msg); // Execute message handler. io.poll(); return 0; } At this point, the library is really thin and only provides very rudimentary wrapping of ZeroMQ. In the future, I plan to extend it in order to facilitate applications that build on message-passing. The dataflow abstraction I envision is essentially a directed graph, where nodes represent components and edges communication links. 0mq provides the edges and the Asio handles asynchronous event processing at the component level. The motiviation is that this model lends itself well for scaling out distributed systems over large clusters. Matthias [1] http://permalink.gmane.org/gmane.network.zeromq.devel/9538

4 4

[1.52.0] Beta release candidates available
by Marshall Clow 10 Oct '12

10 Oct '12

Release candidate files for 1.52.0 beta 1 are available at http://boost.cowic.de/rc/ As always, the release managers would appreciate it if you download the candidate of your choice and give building it a try. Please report both success and failure, and anything else that is noteworthy. This helps ensure the candidates build OK before we push them out to SourceForge. The files (and associated md5s) are: 61a852c923c7e27d84fd2e6336d787f2 *boost_1_52_0_beta1_rc1.tar.gz 956fb476326ee634225c683612493fdd *boost_1_52_0_beta1_rc1.tar.bz2 703663eca3122152031c5555e26845c0 *boost_1_52_0_beta1_rc1.zip b277b6f36ead9df9a0754c868956f30c *boost_1_52_0_beta1_rc1.7z Thanks, --The release managers

4 10

[boost test] spelling error
by Ovanes Markarian 09 Oct '12

09 Oct '12

Can you please fix the spelling error in http://www.boost.org/doc/libs/1_51_0/libs/test/doc/html/tutorials/new-year-… BOOST_AUTO_EST_CASE( constructors_test ) Should be BOOST_AUTO_TEST_CASE This error is in both code snippets. Regards, Ovanes

2 1

Multiple Versions of Boost in the Same Executable
by Aaron_Wright＠selinc.com 09 Oct '12

09 Oct '12

I have a large executable that loads shared libraries as plugins. Some of those shared libraries dynamically link with an old boost version that's part of the system (/usr/lib) when they are loaded. I want to build my own plugin shared library, but I want to use a newer boost version. I tend to link boost in statically. My question is, will it cause any issues if I create a shared library that is statically linked with a newer version of boost, and have it loaded by the same executable that is dynamically linked with an old version of boost? I hoping this'll work, as I can't upgrade the system boost very easily. --- Aaron Wright

3 3

boost::bind issues
by Littlefield, Tyler 09 Oct '12

09 Oct '12

Hello all: I'm trying to boost::bind a couple of callbacks and was getting errors; could someone help me figure out what's going on? First, the errors: 19:22:36: Compiling staticObject.cpp. staticObject.cpp: In constructor â€˜StaticObject::StaticObject()â€™: staticObject.cpp:20:143: error: no matching function for call to â€˜OlcStringEntry::OlcStringEntry(std::string, int, boost::_bi::bind_t<std::basic_string<char>, boost::_mfi::cmf0<std::basic_string<char>, BaseObject>, boost::_bi::list1<boost::_bi::value<StaticObject*> > >, boost::_bi::bind_t<void, boost::_mfi::mf1<void, BaseObject, const std::basic_string<char>&>, boost::_bi::list2<boost::_bi::value<StaticObject*>, boost::arg<1> > >)â€™ staticObject.cpp:20:143: note: candidates are: In file included from staticObject.h:7:0, from staticObject.cpp:5: olc.h:48:3: note: OlcStringEntry::OlcStringEntry(const string&, FLAG, StringGetter&, StringSetter&) olc.h:48:3: note: no known conversion for argument 3 from â€˜boost::_bi::bind_t<std::basic_string<char>, boost::_mfi::cmf0<std::basic_string<char>, BaseObject>, boost::_bi::list1<boost::_bi::value<StaticObject*> > >â€™ to â€˜StringGetter& {aka boost::function<std::basic_string<char>()>&}â€™ olc.h:42:7: note: OlcStringEntry::OlcStringEntry(const OlcStringEntry&) olc.h:42:7: note: candidate expects 1 argument, 4 provided now, here's how I call it: AddOlc(new OlcStringEntry(std::string("name"), OF_NORMAL, boost::bind(&BaseObject::GetName, this), boost::bind(&BaseObject::SetName, this, _1))); baseobject's relevant methods look like: virtual std::string GetName() const; virtual void SetName(const std::string &s); and the relevant tidbits from olc.h: typedef boost::function<std::string()> StringGetter; typedef boost::function<void (const std::string&)> StringSetter; and OlcStringEntry's ctor looks like: OlcStringEntry(const std::string &name, FLAG flag, StringGetter& get, StringSetter& set); Any info here would be appreciated. Thanks, -- Take care, Ty http://tds-solutions.net The aspen project: a barebones light-weight mud engine: http://code.google.com/p/aspenmud He that will not reason is a bigot; he that cannot reason is a fool; he that dares not reason is a slave.

2 1

CSR graph and Dijkstra's to lower graph memory
by jer.k 08 Oct '12

08 Oct '12

Hello, This is my first post and I'm very new to BOOST, so please let me know if I have missed some of the mailing list etiquette :) When I run Dijkstra's on my graph, I quickly run out of memory and I'm trying to find ways to lower the memory needed to store the graph. My graphs has around ~3,600,000 nodes and ~580,000,000 edges (I will eventually need a graph at least 5x this number - more if possible). I only store a weight (double) at each edge (I don't need to store any information at each node) and find that while this only takes about 1.5 hrs to run, this takes up around ~95GB (I have access to a cluster so I can run these high memory jobs). I use an adjacency_list to store this graph. typedef adjacency_list < setS, vecS, undirectedS, no_property, property < edge_weight_t, double > > graph_t; My two questions are: 1) Is this a reasonable amount of memory to expect a graph of this size using an adjacency_list? Or am I doing some terribly wrong? This is much more memory needed than I expected. 2) From what I understand, the compressed sparse row (CSR) graph is a possible solution to lowering the memory requirements as I know the number of edges and the number of nodes prior to forming the graph. I've been going through the documentation and I followed this tread <http://boost.2283326.n4.nabble.com/Graph-On-Dijkstra-SSSP-performance-td444…> which had an example of using a CSR instead of an adjacency_list and running Dijkstra's. However I can't seem to get the last step (last reply by Jeremiah Willcock) of > "plus adding the weight map parameter explicitly > to the call to dijkstra_shortest_paths_no_color_map" to work successfully. I guess I'm not sure how to add the weigh map parameter to the Dijkstra's call. The following is the relevant code dealing with edge weights. ======= ... struct Edge_Cost { double weight;}; // Needed for bundled properties. Edge_Cost *edgeCosts = new Edge_Cost[2]; // Add 3 edges costs. edgeCosts[0].weight = (2.0); edgeCosts[1].weight = (5.2); edgeCosts[2].weight =(3.33); ... typedef compressed_sparse_row_graph<directedS, WebPage, // From the example docs on CSR Edge_Cost // The struct. > graph_t; graph_t g(boost::edges_are_unsorted, &the_edges[0], &the_edges[0] + numEdges, edgeCosts, // Add the weights here??? numNodes); ... boost::dijkstra_shortest_paths(g, // The graph. s, // VertexDescriptor - the source vertex. &p[0], &d[0], edgeCosts, // <--- WHAT DO I PUT HERE?? indexmap, std::less<int>(), closed_plus<int>(), (std::numeric_limits<int>::max)(), 0, default_dijkstra_visitor() ); ======= When I run this, I get several errors, the first is: error C2664: 'boost::get' : cannot convert parameter 2 from 'boost::detail::csr_edge_descriptor<Vertex,EdgeIndex>' to 'ptrdiff_t' c:\boost_1_49_0\boost_1_49_0\boost\graph\dijkstra_shortest_paths.hpp 162 I might be totally off here so please let me know :) Thanks! - Jeremy -- View this message in context: http://boost.2283326.n4.nabble.com/CSR-graph-and-Dijkstra-s-to-lower-graph-… Sent from the Boost - Users mailing list archive at Nabble.com.

4 9

Determining whether a key already exists in a labeled graph
by eric 08 Oct '12

08 Oct '12

Is there a way to tell if a labeled graph already contains a key? As I'm constructing my graph from database data, I would like to check whether a node's parent already exists, and if not create it. -- View this message in context: http://boost.2283326.n4.nabble.com/Determining-whether-a-key-already-exists… Sent from the Boost - Users mailing list archive at Nabble.com.

2 2

Linking problem - boost 1.51
by Daniel Santos 07 Oct '12

07 Oct '12

Dear list, I'm starting with boost (Boost.Asio) and I'm following the tutorial available at Boost.Asio web-page. I'm stuck at part 5 (Timer.5) because I'm getting the following linker error: *timer_5.cpp:(.text+0xb3): undefined reference to `boost::thread::~thread()' * *timer_5.cpp:(.text+0xe1): undefined reference to `boost::thread::~thread()' * *collect2: error: ld returned 1 exit status* I'm linking boost_system, boost_thread and pthread (I compiled the boost libraries and at some point I thought that I was including .hpp files from a different version of the libraries but I specified the paths of both .hpp and libraries and the problem remains). The system runs GNU/Linux 64bits (Fedora 17) with GCC 4.7.2 and boost 1.51.0. Any idea? Thanks i advance, -- Daniel

2 4

[Sandbox - endian] Why is endian still in sandbox?
by Nathaniel J Fries 07 Oct '12

07 Oct '12

A fairly novel (at least when put up against more intricate labours like asio, filesystem, and their like) project which appears to be finished and would be a very useful companion to existing libraries sits in the sandbox. Is there a reason for this?

3 3