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

using boost::python v1
by Srivatsan Raghavan 25 Jul '02

25 Jul '02

hello all.. i've got some questions about using bpl, and i've searched the docs & the examples, and i can find no answer this example illustrates the issues i'm having <cpp> class B ; class C ; class A { public: A() ; virtual bool func1 ( ) ; virtual bool func2 ( ) = 0 ; virtual ~A() ; virtual int func3 ( std::list<C> ) ; protected : B m_Bobj ; std::string m_Aname ; } ; </cpp> now , i know how to expose the virtual and pure virtual functions to pyhon, i just write a derived class A_callback, and write some static default functions that forward the impl to the virtual function in A . my problem is in the func3() and the protected members do i need to write my own wrapper functions to python for mapping a list<C> ? or does bpl do this itself in some way? and how do i expose the protected members to python? if i do something like this : <cpp> boost::python::class_builder<A , A_callback> a_class ( this_module , "A" ) ; a_class.def ( &A::m_Bobj , "m_Bobj" ) ; </cpp> that line of code gets me this error (in msvc 6 : sp 5 ) : error C2248: 'm_Bobj': cannot access protected member declared in class 'Logging::Endpoint' <file-name> : see declaration of 'm_Bobj' how can i get this to work?

2 1

Is bellman_ford_shortest_paths supposed to work for undirected graphs?
by Louis Lavery 25 Jul '02

25 Jul '02

int main() { enum {A,B,Z}; char const name[] = "ABZ"; int const numVertex = static_cast<int>(Z); typedef std::pair<int,int> Edge; Edge edge_array[] = {Edge(B,A)}; int const numEdges = sizeof(edge_array) / sizeof(Edge); int const weight[numEdges] = {11}; typedef adjacency_list<vecS,vecS,undirectedS,no_property,property<edge_weight_t,int> > Graph; Graph g(edge_array, edge_array + numEdges, numVertex); Graph::edge_iterator ei, ei_end; property_map<Graph,edge_weight_t>::type weight_pmap = get(edge_weight, g); int i = 0; for(tie(ei, ei_end) = edges(g); ei != ei_end; ++ei, ++i) { weight_pmap[*ei] = weight[i]; } std::vector<int> parent(numVertex); for(i = 0; i < numVertex; ++i) { parent[i] = i; } std::vector<int> distance(numVertex, std::numeric_limits<int>::max()); distance[A] = 0; // Set source distance to zero bool const r = bellman_ford_shortest_paths(g, int (numVertex), weight_map(weight_pmap).distance_map(&distance[0]). predecessor_map(&parent[0])); if (r) { for(int i = 0; i < numVertex; ++i) { std::cout << name[i] << ": " << std::setw(3) << distance[i] << " " << name[parent[i]] << std::endl; } } else { std::cout << "negative cycle" << std::endl; } return EXIT_SUCCESS; } Expected Output:- A: 0 A B: 11 A Actual Output:- A: 0 A B: 2147483647 B Louis.

3 3

BGL: traversal algorithm for tree
by Björn Lindberg 25 Jul '02

25 Jul '02

I have implemented a rooted tree structure as a BGL graph. I need to be able to traverse the tree from root to leaves, while doing something at each vertex. I need to be able to do this as inorder and postorder traversal. My question is if it's possible to do this with any of the existing algorithms in BGL. My current thinking is to implement a function traversal( ... ), but if there already exists an alternative, I would prefer to use it. Björn

2 14

iterator_adaptor Question
by Hickman, Greg 24 Jul '02

24 Jul '02

I have a class that holds a container of pointers to polymorphic Base classes. I'd like clients who iterate over those Bases using a mutating iterator type to be able to call non-const Base class member functions (but I do not want those clients to be able to replace the Base* directly). I'd also like clients who iterate using a const_iterator type to be able to call only const Base class member functions. In the following test code, I've attempted to use boost::iterator_adaptor to create a const iterator with these properties, but the const_iterator still allows calls to non-const operations. Can anyone help me see where I've gone wrong? #include <boost/iterator_adaptors.hpp> #include <boost/utility.hpp> #include <iostream> #include <list> class Base : private boost::noncopyable { public: virtual ~Base() {} virtual void foo() const = 0; }; class Derived : public Base { public: virtual void foo() const { std::cout << "Derived::foo" << std::endl; } }; struct Bases { typedef std::list<Base*> Bases_imp; typedef boost::iterator_adaptor< Bases_imp::iterator, boost::default_iterator_policies, Base*, Base* const&, Base* const* > iterator; typedef boost::iterator_adaptor< Bases_imp::const_iterator, boost::default_iterator_policies, Base const*, Base const* const&, Base const* const* > const_iterator; void foo() const { const_iterator i(bases_.begin()); const_iterator end(bases_.end()); for (; i != end; ++i) (*i)->foo(); // Oops - const_iterator allows call to non-const method } Bases_imp bases_; };

3 2

RE: Mail server problems (was: Boost Sockets)
by Jim.Hyslop 24 Jul '02

24 Jul '02

Jim.Hyslop (jim.hyslop(a)Leitch.com) wrote: > No, I don't think so, Phil - please tell us what you _really_ > think ;-) No, this is not a case of deja-vu - our Exchange server burped last night and re-sent a whole bunch of emails. Sorry 'bout that. -- Jim

1 0

boost shared_ptr - thread safe and non thread safe editions
by Ren� Dalby Carlsen 24 Jul '02

24 Jul '02

Hi. Is there any easy way to get a thread safe edition of the shared_ptr (BOOST_HAS_THREADS) and a "non" thread safe edition (BOOST_DISABLE_THREADS) inside the same project? The thread safe edition of the shared_ptr is a bit slower and I would like to able to use the non thread safe shared_ptr, when I known it only be used on the same thread. As a solution I thought I could do something like this, so I would not have to edit/copy any boost .hpp file : // // MyApp.cpp // #define BOOST_HAS_THREADS #include "boost/smart_ptr.hpp" namespace boost_nothread { #define BOOST_DISABLE_THREADS #include "boost/smart_ptr.hpp" } void f() { boost::shared_ptr <int> spIntTS; // thread safe smart pointer boost_nothread::shared_ptr <int> spInt; // non thread safe smart pointer } //////// The above sample does not work - what should I do instead? Regards, René

2 3

Python: class hierarchies; return pointer to base
by Jacek Generowicz 23 Jul '02

23 Jul '02

Hi, I'm trying to use Boost.Python to create an extension module based on some existing code. Therein I have a class hierarchy which I would like to expose in a way that would make it possible to subclass any of its classes in python. I run into a problem when I try to expose C++ functions which return pointers to the base class of the hierarchy. In the minimal example I am using for testing purposes I have: // An abstract base class class abstract { ... virtual void thing() const = 0; }; // A conctrete subclass implemented in C++ class concrete : public abstract { ... virtual void thing const { ... } }; // Callback to allow happy subclassing of abstract in python class abstract_callback : public abstract { ... }; // Callback to allow happy subclassing of concrete in python class concrete_callback : public abstract { ... }; BOOST_PYTHON_MODULE_INIT(abstract) { ... python::class_builder<abstract, abstract_callback> abs_class(this_module, "abstract"); ... python::class_builder<concrete, concrete_callback> con_class(this_module, "concrete"); con_class.declare_base(abs_class); ... } This all works fine, until I try to expose to python a function such as abstract* ret_p_abs( ... ); When compiling (gcc 2.95.2) the module, I get the error no matching function for call to `py_extension_class_converters (boost::python::type<abstract *>)' I infer from comments in the source code that this is to be expected if the class in question has not been wrapped as an extension class. This makes sense, as abstract has, indeed, not been wrapped ... but I don't think that it should be; abstract_callback was created for that purpose. How should I proceed ? Thanks,

1 0

Regexp linking problems with VC++6.0 + STLPort
by Noel Llopis 22 Jul '02

22 Jul '02

I've read through all the other previous messages dealing with problems with Regexp and VC++, but I wasn't able to find a good answer for my problem. I compiled Regexp++ without any problems. I added the correct directory path for the libraries, so that happens automatically (in my case it links with boost_regex_vc6-stlport_mssd.lib (multithreaded, static run, static link, debug). However, as soon as I try to use any regexp function or class from my code, it comes back with a link error. For example, trying to use the RegExp class, I get: error LNK2001: unresolved external symbol "public: __thiscall boost::RegEx::RegEx(class _STL::basic_string<char,struct std::char_traits <char>,class _STL::allocator<char> > const &,bool)" (?? 0RegEx@boost@@QAE@ABV?$basic_string@DU?$char_trai ts@D@std@@V?$allocator@D@_STL@@@_STL@@_N@Z) bin/XXXXXX.exe : fatal error LNK1120: 1 unresolved externals Can anybody guess as to what might be wrong? I'm using VC++ 6.0 with the latest upgrades and STLPort 4.5.3. Thanks. --Noel llopis(a)comcast.net

2 3

RE: [Boost-Users] Boost Config problem - MSVC 7, boost 1.28.0
by Tom Matelich 22 Jul '02

22 Jul '02

Cool, thanks. I really hate modifying third party libraries. > -----Original Message----- > From: John Maddock [mailto:john_maddock@compuserve.com] > Sent: Saturday, July 20, 2002 5:03 AM > To: Boost-Users(a)yahoogroups.com > Subject: Re: [Boost-Users] Boost Config problem - MSVC 7, boost 1.28.0 > > > > I was running into problems with min and max being macros > when attempting > to > > port my code to .NET. When I went to look at visualc.hpp, > I saw that > > NOMINMAX was only defined when _MSC_VER <= 1200. My break > seems to have > > happened where a boost file was included, then a windows > header came in, > so > > the macros were enabled. > > > > I know we are not supposed to rely on Boost.Config in our > code, but it > sure > > is nice :). > > I've modified the config system so that NOMINMAX is defined > in the win32 > platform configuration rather than for VC6 specifically, > likewise the undefs > for min and max are in there now as well, hopefully this won't break > anything ... :-) > > John Maddock > http://ourworld.compuserve.com/homepages/john_maddock/index.htm > > > > ------------------------ Yahoo! Groups Sponsor > ---------------------~--> > Access your PC just like Web Mail > http://us.click.yahoo.com/r5uw2C/zncEAA/Ey.GAA/EbFolB/TM > -------------------------------------------------------------- > -------~-> > > Info: <http://www.boost.org> > Wiki: <http://www.crystalclearsoftware.com/cgi-bin/boost_wiki/wiki.pl> > Unsubscribe: <mailto:boost-users-unsubscribe@yahoogroups.com> > > > Your use of Yahoo! Groups is subject to > http://docs.yahoo.com/info/terms/ > > ----------------------------------------------------------------------- DISCLAIMER: Information contained in this message and/or attachment(s) may contain confidential information of Zetec, Inc. If you have received this transmission in error, please notify the sender by return email. -----------------------------------------------------------------------

1 0

HELP, boost::thread in freebsd, run example/monitor core dumped
by wugui 22 Jul '02

22 Jul '02

system: FreeBSD 4.6 gcc: version 2.95.3 prog: example/monitor cause: recursive_mutex ? do_test<boost::recursive_mutex>(); link with: libboost_threadd.a libboost_threadds.a assertion "res == 0" failed: file "libs/thread/build/../src/condition.cpp", line 347 link with: libboost_thread.a libboost_threads.a OK, BUT ! I'M NOT SURE THAT !

1 1