- Boost-users - lists.boost.org

BGL: Trying to get my property bundle-using adjacency list to use a vector for its WeightMap.
by Geoff Hilton 14 Aug '08

14 Aug '08

Hi, I'm new to the boost and the BGL, and as the subject suggests, I'm trying to get my property bundle-using adjacency list to use a vector for its WeightMap. First off, the property bundles in question: template<size_t NumberOfBar> struct EdgeProperty { EdgeProperty(long double foo = 0):Foo(foo), Bar(NumberOfBar){} long double Foo; vector<long long> Bar; }; template<size_t NumberOfBar> struct VertexProperty { VertexProperty():Baz(NumberOfBar){} vector<pair<long long, long long> > Baz; }; The visitor: struct MyVisitor : public default_dijkstra_visitor { template <typename Vertex, typename Graph> void discover_vertex(Vertex u, const Graph& g) { cout << "discovered " << u << '\n'; } template <typename Edge, typename Graph> void tree_edge(Edge e, const Graph& g) { cout << "tree edge " << e << '\n'; } template <typename Edge, typename Graph> void examine_edge(Edge e, const Graph& g) { cout << "examining edge " << e << '\n'; } template <typename Vertex, typename Graph> void finish_vertex(Vertex u, const Graph& g) { cout << "finishing vertex " << u << endl; } }; The graph typedef: typedef adjacency_list< vecS, vecS, bidirectionalS, property<vertex_index_t, unsigned long long, VertexProperty<1> >, property<edge_index_t, unsigned long long, EdgeProperty<1> > > graph_t; Adding edges and vertices to the graph works without issue, as does assigning values to the properties, and navigating it, I was also able to get my test visitor to work.. using this code: int _tmain(int argc, _TCHAR* argv[]) { graph_t mygraph; //build graph //... //set up property accessors typedef EdgeProperty<1> EdgeProp; property_map<graph_t, long double EdgeProp::*>::type edge_foo_t = get(&EdgeProp::foo, mygraph); property_map<graph_t, vector<long long> EdgeProp::*>::type edge_bar_t = get(&EdgeProp::Bar, mygraph); property_map<graph_t, vertex_index_t>::type vertex_ind = get(vertex_index, mygraph); //dijkstra stuff size_t numberOfVertices = num_vertices(mygraph); std::vector<long double> distances(numberOfVertices, (numeric_limits<long double>::max)()); vector<vertex_descriptor> predecessors(numberOfVertices); MyVisitor vis; dijkstra_shortest_paths(mygraph, *vi, weight_map(edge_foo_t) .weight_map2(edge_bar_t) .distance_map( make_iterator_property_map(distances.begin(), vertex_ind) ) .predecessor_map( make_iterator_property_map(predecessors.begin(), vertex_ind) ) .visitor(vis) ); return 0; } However when I switch edge_bar_t and edge_foo_t within the weight_map/weight_map2 parentheses, I get this error (Visual Studio 2005) 'bool std::less<_Ty>::operator ()(const _Ty &,const _Ty &) const' : cannot convert parameter 1 from 'std::vector<_Ty>' to 'const D &' Of course that's normal because it's a vector. I figure I'm supposed to use make_iterator_property_map and somehow point it to &EdgeProp::Bar::begin() (or &EdgeProp::Bar::rbegin()) with a property_map<...> as I did with the others...but what I've tried so far hasn't worked. Any ideas? Thanks very much and best wishes, Geoff

1 1

Building boost 1.36
by Sean Farrow 13 Aug '08

13 Aug '08

Hi: I am trying to build boost 1.36, with the following command line: Bjam --toolset =msvc --build-type =complete. This gives me errors in bjam files. Any ideas what could be causing this error. Cheers Sean. __________ Information from ESET NOD32 Antivirus, version of virus signature database 3352 (20080813) __________ The message was checked by ESET NOD32 Antivirus. http://www.eset.com

2 2

bjam output
by Simon Pickles 13 Aug '08

13 Aug '08

Hi, I have finally got my head round boost::python and I'm loving it. I am on Ubuntu Hardy, stackless python 2.5, gcc 4.2.3. Two questions: 1) When using bjam to build libraries which expose C++ classes and functions to python, how can I use bjam to output the built libraries into a different folder to /bin/gcc-4.2.3/debug? 2) is this actually a debug library? if so, how do I create release versions? Many thanks Simon

3 4

Unattended regression tests are blocked by Windows Dialog
by Olav Kindt 13 Aug '08

13 Aug '08

Hi all, I have searched through some of the previous postings on this list, but have not been able to find the solution - so I'll try here. I have a test case that (due to an error of course) tries to run code like this: class MyObject { public: virtual void f() {}; }; void MyFailingTestCase() { MyObject * a = NULL; a->f(); } When running these tests in release mode - the whole process is stopped by a windows dialog letting be know that there was a system error (Access Violation) and asks the user to click OK/Debug. I have also tried to pass the --catch_system_errors=yes, but that did not help I am running 1_33_1 version of the library. (libboost_unit_test_framework-vc80-mt-1_33_1.lib) Does anybody know a way around this problem? Olav

5 5

Re: [Boost-users] Unattended regression tests are blocked byWindowsDialog
by Olav Kindt 13 Aug '08

13 Aug '08

>>> I have a test case that (due to an error of course) tries to run code >>> like this: >>> >[snip] >>> >>> void MyFailingTestCase() >>> { >>> MyObject * a = NULL; >>> >>> a->f(); >>> } >>> >>> When running these tests in release mode - the whole process is >>> stopped by a windows dialog letting be know that there was a system >>> error (Access Violation) and asks the user to click OK/Debug. >>> >>> I have also tried to pass the --catch_system_errors=yes, but that did >>> not help >>> >>> I am running 1_33_1 version of the library. >>> (libboost_unit_test_framework-vc80-mt-1_33_1.lib) >>> >>> Does anybody know a way around this problem? >> >> The only reasonable solution is to fix the test. According >> to standard C++ the expression "a->f()" using an a with >> a null pointer value causes undefined behaviour. >> Why don't you use a valid argument? Possible examples are: > >I believe the above is just an example. Under Windows, the dereferencing of >the null pointer leads to an access violation which can be handled via SEH >(Structured Exception Handling). MSVC provides a way to translate strutured >exceptions to C++ exceptions, which Boost.Test is supposed to use. >Regards, >Johan If I understand you (and the documentation) correctly - then the Boost.Test framework should have caught this (Structured Exception), and just reported the test as failed. Does this mean that the framework is broken (I'm using 1.33) Does anyone else have any experience with this - will an upgrade to 1.35 help? Olav

1 0

[boost-users][thread] upgrading lock
by Igor R 13 Aug '08

13 Aug '08

Hello, This's rather basic question, but I couldn't find any example in the docs. typedef boost::shared_lock<boost::shared_mutex> shared_lock; shared_mutex mutex_; /// { shared_lock lock(mutex_); // reading locked data... // here I want to midify the data, how can I upgrade the lock to be exclusive? } Thank you in advance.

2 1

general question on sentry in streams
by Ovanes Markarian 13 Aug '08

13 Aug '08

Hello all, this is probably a wrong place to write this post, but I hope to receive your thoughts anyway. Writing to moderated c++ group might open my email address to spammers, so I decided to write my post here. One of my collegues told me once that C++ streams are slower then Java. We did some trials (on a Win32 platform with Visual C++ 7.1) and he was able to prove this. Today when I read the standard on output steams I came accross the point, that most all stream functions must construct the nested sentry type. I suddenly remembered the article written by Nathan Myers on locales available at: http://www.cantrip.org/locale.html There he states, that in sentry a locking might be made in a multi-threaded environment. I looked in the implemenation of my STL distribution (which by default bundled with VC++ 7.1) and here we go. sentry locks in ctor and unlocks in dtor. So writing a character by character or integer by integer will produce the whole bunch of lock/unlock sequences, since standard states (in §27.6.2.5.4) that every call to operator << has to create sentry object instance and that explains to me why the std streams (at least on Windows are much slower as on Java). Would be great to hear your thoughts... Regards, Ovanes

1 0

1.36.0 && issue tracking
by Tim St. Clair 13 Aug '08

13 Aug '08

Folks - I was wondering what method is used to submit bugs which we find, other then spamming the list. I found quite a few in 1.35.0 and I'm hoping 1.36.0 fixes a lot of those issues. However if I find something, how/where can I submit that info. Is there a bugzilla server somewhere which people actually use? -- Regards, Timothy St. Clair [timothysc(a)gmail.com]

2 1

Re: [Boost-users] boost::math::tools and boost::distributions
by Jesse Perla 13 Aug '08

13 Aug '08

> > >Docs here: > > > http://www.boost.org/doc/libs/1_35_0/libs/math/doc/sf_and_dist/html/math_to… > > > >Note that these are considered "experimental" in the sense that their > >interfaces haven't been reviewed for formal inclusion in Boost. They are > >all rigorously tested though as they're fundamental to the implementation > of > >the rest of the library. So... please do use them, and if you find any > >interface SNAFU's that can be improved let us know! BTW the most likely > >reason for the interfaces to change is to facilitate improvements to their > >implementation and especially their efficiency. > > > >HTH, John. (sorry John, I sent my last email before getting the digest) Perfect. Since a snapshot of the interfaces exists for 1.35, then a port over to newer versions of boost is reasonable. Hopefully we can turn this into a top level project that does the baseline optimization/root finding people need before shopping for really specific tools/algorithms. -------- A couple of other questions on this beautiful distributions library - which I see as a great starting point for a lot of operations treating them as measures: * Is there any planned or existing support in the library for multi-dimensional distributions? In particular, bivariate norms with a covariance matrix passed in? Will the library design and notation support these sorts of extensions without too many problems? * Is there any way in the current library to pass in a user created discrete valued probability measure and have it as a "distribution" here? If not, I think I may write it, but as I am not a very good library programmer I would love to exploit any other work out there. I would need this to work for at least 1 and 2 dimension measures. * If we were to use a function to bundle a discrete valued function, is there a nice class/pattern out there that already combines the x-values and y-values into a single class? (I also want this later for implementing different interpolation algorithms returning functors after passing in a discrete function in the constructor). * Then I would write a specialization of this (2nd and higher dimension) for a markov measure. * Then, I want to write a general operator for unconditional expectation passing in a function object, and over a generic distribution as a measure from this library (though I will use it for the discrete measures to begin with). Specializing for different types of distributions, this could use exact analytical calculations, quadrature, or use monte-carlo methods. * Last, I want to write a conditional expectation operator that takes in a function object mapping reals^2 to reals, a distribution (would have to be 2 dimensional), the conditional (which would be a real for now, but could be generic), and what dimension the condition applies to. * Eventually, we may be able to tie together this operator with boost::lambda and boost::bind and we could up with a very elegant way to write out functions that involve conditional expectations. It may not have the performance of hand-tweaked fortran loops, but it should be a good enough place to start and infinitely preferable to matlab. The main use case I have for this is to implement easy/flexible notation for conditional distribution functions when writing out bellman equations for dynamic programming in economics. Most cases of this involve two dimensional markov measures (often approximated by Tauchen's algorithm from ARMA processes) If anyone is interested in this topic, has ideas, wants to work jointly, or plans to tell me that I am insane, please email me. My main fear is that I am a library user with a pretty good ability to think in generics, but with absolutely no chance of ever understanding what all of the crazy non-algorithmic code in these libraries does or writing it myself. -Jesse

1 0

Building boost with xml support (graph) and icu support (regex)
by Sean Farrow 13 Aug '08

13 Aug '08

Hi: How can I build boost with xml support in boost::graph, and icu support in Regex. What version of other libraries do I need? Cheers Sean. __________ Information from ESET NOD32 Antivirus, version of virus signature database 3352 (20080813) __________ The message was checked by ESET NOD32 Antivirus. http://www.eset.com

1 0