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

[exception,serialization] transporting exceptions
by Brian Budge 05 Oct '12

05 Oct '12

Hi all - I like that boost::exception provides a mechanism to copy exceptions. This can be used to transport an exception across thread boundaries, which is nice. I'm running a program that can optionally start nodes using MPI to do back-end computations. I'd like to be able to "copy" the exception from one node to another. Are there any existing tools for something like a serialization of these kinds of exceptions? In non-MPI mode, I'd like to transport the exception across thread boundaries, pass it through some functions (causing actions to be taken along the way), and then finally to rethrow the exception. In MPI mode I want all of that PLUS, in place of one of the function calls, I'd like to transport the exception across MPI in some way. Are there any semi-automatic tools already in existence that will get me this (or close)? Or do I need to implement boost serialization for my entire exception hierarchy? If I do the latter, it seems I will need to operator new the exception, and will not easily be able to know when I can delete in a catch? Thanks, Brian

2 3

Trouble w/ fruchterman-reingold
by eric 05 Oct '12

05 Oct '12

I'm seeing a weird problem in fruchterman_reingold.hpp that after the displacements are calculated for every vertex, it gets down to the loop where it updates the positions (starting on line 327), and all the displacements are (0,0) by that point (they had been sensible values up in the calculation loop). Any idea what I could be doing to cause that? Here is my code: struct spring_attractive_force { template<typename Graph, typename T> T operator()(typename graph_traits<Graph>::edge_descriptor, T k, T d, const Graph&) const { return d; } }; struct magnetic_repulsive_force { template<typename Graph, typename T> T operator()(typename graph_traits<Graph>::vertex_descriptor, typename graph_traits<Graph>::vertex_descriptor, T k, T d, const Graph&) const { return 1 / (d * d); } }; typedef adjacency_list<listS, listS, undirectedS, // Vertex properties property<vertex_index_t, int, property<vertex_position_t, point> >, // Edge properties property<edge_weight_t, double> > Graph; enum {A, B, C, D, E, F, G, H}; simple_edge triangular_edges[14] = { {A, B}, {B, C}, {C, A}, {D, E}, {E, F}, {F, G}, {G, H}, {H, D}, {D, F}, {F, H}, {H, E}, {E, G}, {G, D} }; Graph g(&triangular_edges[0], &triangular_edges[13], 8); // Generate (x,y) coordinates for every node // - Randomly layout vertices of graph minstd_rand gen; rectangle_topology<> rect_top(gen, -25, -25, 25, 25); random_graph_layout(g, get(vertex_position, g), rect_top); // - Run Fruchterman Reingold algorithm on them typedef square_topology<> Topology; Topology topology(gen, 50.0); fruchterman_reingold_force_directed_layout(g, get(vertex_position, g), topology, attractive_force(spring_attractive_force()). repulsive_force(magnetic_repulsive_force()). force_pairs(all_force_pairs()). cooling(linear_cooling<double>(100))); -- View this message in context: http://boost.2283326.n4.nabble.com/Trouble-w-fruchterman-reingold-tp4636494… Sent from the Boost - Users mailing list archive at Nabble.com.

2 13

link errors with boost serialization
by xavier 05 Oct '12

05 Oct '12

I have just built boost_1_51_1 with no problems on linux redhat 5.5 64-bit machine However, when i try to link a program that uses the g++ client_test.o -rdynamic -L/opt/boost_1_51_0/stage/lib -lboost_serialization -lboost_wserialization -o client_test it gives the following error: client_test.o: In function `boost::serialization::detail::extended_type_info_typeid_1<gps_position const>::extended_type_info_typeid_1()': client_test.cpp:(.text._ZN5boost13serialization6detail27extended_type_info_typeid_1IK12gps_positionEC1Ev[boost::serialization::detail::extended_type_info_typeid_1<gps_position const>::extended_type_info_typeid_1()]+0x12): undefined reference to `boost::serialization::detail::extended_type_info_typeid_0::extended_type_info_typeid_0()' client_test.cpp:(.text._ZN5boost13serialization6detail27extended_type_info_typeid_1IK12gps_positionEC1Ev[boost::serialization::detail::extended_type_info_typeid_1<gps_position const>::extended_type_info_typeid_1()]+0x27): undefined reference to `boost::serialization::extended_type_info::self_register()' client_test.cpp:(.text._ZN5boost13serialization6detail27extended_type_info_typeid_1IK12gps_positionEC1Ev[boost::serialization::detail::extended_type_info_typeid_1<gps_position const>::extended_type_info_typeid_1()]+0x3a): undefined reference to `boost::serialization::detail::extended_type_info_typeid_0::~extended_type_info_typeid_0()' client_test.o: In function `boost::serialization::detail::extended_type_info_typeid_1<gps_position const>::~extended_type_info_typeid_1()': client_test.cpp:(.text._ZN5boost13serialization6detail27extended_type_info_typeid_1IK12gps_positionED0Ev[boost::serialization::detail::extended_type_info_typeid_1<gps_position const>::~extended_type_info_typeid_1()]+0x1d): undefined reference to `boost::serialization::detail::extended_type_info_typeid_0::~extended_type_info_typeid_0()' client_test.o: In function `boost::serialization::detail::extended_type_info_typeid_1<gps_position const>::~extended_type_info_typeid_1()': client_test.cpp:(.text._ZN5boost13serialization6detail27extended_type_info_typeid_1IK12gps_positionED1Ev[boost::serialization::detail::extended_type_info_typeid_1<gps_position const>::~extended_type_info_typeid_1()]+0x1d): undefined reference to `boost::serialization::detail::extended_type_info_typeid_0::~extended_type_info_typeid_0()' client_test.o:(.rodata._ZTVN5boost13serialization6detail27extended_type_info_typeid_1IK12gps_positionEE[vtable for boost::serialization::detail::extended_type_info_typeid_1<gps_position const>]+0x10): undefined reference to `boost::serialization::detail::extended_type_info_typeid_0::less_than(boost::serialization::extended_type_info const&) const' client_test.o:(.rodata._ZTIN5boost13serialization6detail27extended_type_info_typeid_1IK12gps_positionEE[typeinfo for boost::serialization::detail::extended_type_info_typeid_1<gps_position const>]+0x10): undefined reference to `typeinfo for boost::serialization::detail::extended_type_info_typeid_0 What could be causing the problem ? -- View this message in context: http://boost.2283326.n4.nabble.com/link-errors-with-boost-serialization-tp4… Sent from the Boost - Users mailing list archive at Nabble.com.

1 0

Boost TraitsType Question
by Rawate, Vishal CIV NAVAIR-WD-5-2-7-1-1-0-E., 527110E 04 Oct '12

04 Oct '12

Hello, I am writing a function using templates in C++.I want to find the type of variable that is being to the function. I was wondering if I could use the Boost TypeTraits library or is there another library that I can use. I would highly appreciate any help. Thanking in advance. Vishal CodeSnipet: Template<typename T> Void setTag(T myVariable) { If(T == short) { Do something; } If(T == long) { Do something; } }

4 5

[Review Report] ODEInt accepted
by Steven Watanabe 04 Oct '12

04 Oct '12

AMDG The review of the ODEInt library submitted by Mario Mulansky and Karsten Ahnert ended last Friday. Many thanks to everyone who participated in this review. The library is ACCEPTED into Boost unconditionally. Summary: 7 ACCEPT 1 ABSTAIN Reviews: http://lists.boost.org/Archives/boost/2012/09/196733.php http://lists.boost.org/Archives/boost/2012/09/196882.php http://lists.boost.org/Archives/boost/2012/09/196883.php http://lists.boost.org/Archives/boost/2012/10/196966.php http://lists.boost.org/Archives/boost/2012/10/196965.php http://lists.boost.org/boost-users/2012/09/76089.php http://lists.boost.org/boost-users/2012/09/76058.php http://lists.boost.org/boost-users/2012/09/76069.php Other comments: http://lists.boost.org/Archives/boost/2012/09/196756.php The reviews were overwhelmingly positive. Most reviewers were satisfied with the library documentation and interfaces. Issues: These were points brought up by reviewers. The authors are free to accept or reject any of these at their discretion. - A number of minor documentation corrections. - Automatically deduce the correct algebra from the state type. - Use a default initial step size. - Allow breaking out of the algorithms In Christ, Steven Watanabe

2 1

Re: [Boost-users] [odeint] Can odeint integrate a 2-dimensional function?
by Karsten Ahnert 04 Oct '12

04 Oct '12

On 10/04/2012 03:28 PM, Rob Conde wrote: > Say I have an equation that is a function of x and y...can I use odeint to > integrate this function over a 2-d region (i.e. a square or a circle, etc.)? > If so, are there any examples of this? I don't think this is possible. This sound like integration of a known function. Odeint is devoted to solve ordinary differential equations. Karsten

2 1

[process] how to suggest bugs/fixes for boost.process?
by Brian Budge 04 Oct '12

04 Oct '12

Hi all - I've started using boost.process despite the fact that it is not an official boost library. I've founds some cases that don't compile or give warnings, and I'd like to log a bug with possible fixes somewhere. Given that the library is not officially in boost, how should I go about this? Thanks, Brian

3 2

Moving an std::string into a boost::thread in C++03
by Adi Shavit 04 Oct '12

04 Oct '12

Hi, I'm using boost 1.51 on multiple platforms and compilers without C++11. In my main thread, I have a very long, expensive to copy, std::string veryLongString, that I need to pass to a new thread for processing. After the new thread is created I have no more use for veryLongString, hence I'd like to *move *it into the boost::thread ctor. Obviously, if veryLongString was created as a shared_ptr<std::string> then I could just *copy* the shared_ptr into the thread ctor, but it wasn't, so I'd need to copy it anyway. How can I [boost::]move() veryLongString into the boost::thread ctor (probably using via boost::bind)? Is this possible? Thanks, Adi

3 3

Re: [Boost-users] [asio] Custom asynchronous service
by Marat Abrarov 04 Oct '12

04 Oct '12

Hi Acer, > I would like to adapt my class to asio for offering > async_do_something(param, callback) interfaces. http://sourceforge.net/projects/asio-samples/ especially (SVN): https://asio-samples.svn.sourceforge.net/svnroot/asio-samples/trunk/include… https://asio-samples.svn.sourceforge.net/svnroot/asio-samples/trunk/include… or https://asio-samples.svn.sourceforge.net/svnroot/asio-samples/trunk/include… https://asio-samples.svn.sourceforge.net/svnroot/asio-samples/trunk/include… > After reading examples presented in asio/examples/services/ and some asio > source code. > > I found the example does not do invoke callback stuff (e.g. > asio/examples/service/logger_service.cpp). > > In case of asio source code, there are classes such as > detail::wait_handler<> that wrap user callbacks for invoked later. It's not a wrapper for deferred invocation. Here's my implementation of handler parking (Asio custom memory allocation and custom invocation strategy are supported): https://asio-samples.svn.sourceforge.net/svnroot/asio-samples/trunk/include… Regards, Marat Abrarov.

1 0

[BGL] vecS for vertex container and vertex index adjustments
by Sergey Mitsyn 03 Oct '12

03 Oct '12

Hi all, I know that while using vecS for vertices one has no vertex iterator stability upon removing vertices. But I have a question if there exists something like vertex _index_ stability. The documentation says that indices for vertices are automatically adjusted so that they are still in contiguous range, but it is not specified how it is done. Is it okay to assume that if a vertex with an index I is removed, then vertices with indices < I have their indices unchanged, while vertices with indices > I have their indices decreased by one? The actual problem is removing all vertices meeting some condition. Is the following code ok? // traverse all vertices and remove those for which predicate // condition_to_remove_vertex returns true. // graph is adjacency_list<.., vecS, ...> vertex_iterator vi = vertices(graph).first, vi_end = vertices(graph).second; while (vi != vi_end) { if (condition_to_remove_vertex(vi, graph)) { vertices_size_type end_index = vertex_index[*vi_end] , vi_index = vertex_index[*vi]; remove_vertex_f(*vi, graph); // vi and vi_end are invalid --end_index; vi = vertices(graph).first + vi_index; vi_end = vertices(graph).first + end_index; // now vi and vi_end are valid again } else ++vi; } This way seems to be more effective than to start iteration again every time a vertex is removed. ------------ Sergey Mitsyn.

3 14

C++ (boost) library for serial port communication
by Alexander Dietz 03 Oct '12

03 Oct '12

Hi all, I wonder if there is a boost-library that enables easy and robust communication to a serial port, just as e.g. the python library pyserial does. Maybe there is a boost-library which can be used for those purpose instead of implementing my own C-style classes and functions? Thanks a lot, Alex

5 7

A couple of odeint questions
by Tiago Gehring 02 Oct '12

02 Oct '12

Hello, I've been using the odeint library for some time now - and I really like it, thanks for the great work! I know that the odeint library is currently being reviewed but I'm actually writing here to ask a couple of questions about my usage scenario, hope that I'm in the right place. My usage scenario involves neural networks where I have a set of n equations for the N neurons themselves and another m equations for all the M synapses (connections) where usually M >> N and the two sets of equations depends on each other. What I'm trying to do is to put all these N + M equations into one large system and to integrate them at once while also keeping the state vectors separated. This is because of the different dimensionality of the two sets of equations and because my intention is to later use OpenCL to do the calculations (which as I understand works better if you use something like a vexcl::multivector<double, M> + another vexcl::multivector<double N> and I don't see how to pack all of this in one large state vector); So my first question is: can I somehow "combine" two sets of state vectors into one thing or integrate two systems in parallel (synchronously)? I tried the former (one combined state object) and even defined my own combined algebra (a std::pair of std::vectors - later vexcl::multivectors) but then stumbled into problems with my value type (which I defined as a std::array<double, 2>). It seems that I have to define all possible mathematical and comparison operations for my pairs (arrays). And my second question: my problem involves integrating the system until one of the neurons fires - at which point it's state variable (membrane potential) jumps to a fixed value, i.e., I have a sharp discontinuity in the state variable. My way of dealing with this right now was to implement a custom integration function (based on the adaptive integration one) which has a built in function that checks if the threshold was reached at every integration step. When this happens I break the integration, reset the state of the neuron in question to the fixed value and start integrating again. I wonder if it's possible to solve this in an easier way... Is it possible for example to change the state on the fly and can the system handle discontinuities? (i.e. can I integrate without stopping each time a neuron fires). Thanks a lot, Tiago

2 2

lisp like operetor collection on well kown C++ data structure
by Niitsuma Hirotaka 02 Oct '12

02 Oct '12

I wrote scm2cpp.hpp https://github.com/niitsuma/scm2cpp.hpp/blob/master/scm2cpp.hpp scm2cpp.hpp is lisp like operetor collection on well kown C++ data structure lisp function car cdr cons list-ref set-car! .. etc can use for std::vector std::list boost::ptr_list boost::fusion::list and std::pair is regard as cons example usage https://github.com/niitsuma/scm2cpp.hpp/blob/master/usage.cpp https://github.com/niitsuma/scm2cpp.hpp/blob/master/list-test.cpp Also eq? eqv? equal? quote symbol functions are implemented. plz see usege https://github.com/niitsuma/scm2cpp.hpp/blob/master/equal-test.cpp eq? function implimeted as direct addrest comparison template<typename T> bool is_eq(T & x, T & y) {return (&x)==(&y);} thus cons(T, std::list<T>) becomes boost::ptr_list<T> All lisp like operator keep same address for same object.

1 0

[Build] configure files copied into stage directory
by Alexander Khomyak 01 Oct '12

01 Oct '12

Hi. I build library binaries from source using build system. And now the system copies dll and lib-files into stage directory. But I also need to copy pdb-files. How to configure the build system to do so? Regards, Alexander Khomyak

1 0

[ICL] Specific interval_map
by A G 01 Oct '12

01 Oct '12

Hello, I am trying to see whether I can use interval_map for my project. I was initially amazed with interval_map and thought this would be it, but then I have some doubts: I need a interval tree like interval_map, but the aggregated values have the following equivalent property: equal to the length of the interval it is aggregated with. An example: [0:3] -> 3 [4:8] -> 4 Then if I add the interval [2:5] with aggregated value 3 (step 1) to this container, this should end up with: [0:1] -> 1 [2:3] -> 1 [4:5] -> 1 [6:8] -> 2 which I would later need to change into (step 2) : [0:1] -> 1 [2:5] -> 3 ([2:3] and [4:5] are merged because what matter is the last inserted interval: [2:3]). [6:8] -> 2 I am not sure I can do step 1 in just overloading the operator+= over my aggregated values. Could you please confirm ? If so, is there any part of the library I can reuse to do this task (like a function overloading, but which one) or do I have to rewrite an ad-hoc insert-like method myself ? Thank you for your hints. BR, AG.

2 2