September 2006 - Boost-users

[graph] breadth_first_search with a structure
by boost＠daniel-gohlke.de 12 Sep '06

12 Sep '06

Hello, I have a problem with the breadth_first_search. I read the documentation and examples an can't find a solution for my problem. In order breadth_first_search uses a bfs_discovery_visitor<unsigned int * > vis(&distanceSet[0]);. This distance vector storing all distances. So, now I need more then a vector. I want use a struct for storing the distances and later some other stuff e.g. probability of distances and cumulative probabilitys. I tried several code, but get every time errors from VC++ compiler. Can anybody tell me, how do I handle this vector<struct *>. The second question is, can I abort the further evaluation of distances, if any theshold is reached? The graph has about 57000 vertices an every vertice has about 6 edges. I am interested only for the vertices in a given range (e.g. 100). Now I get all distances and copy_if the vector of distances, if the given threshold isn't reached. This is fast enough for one vertice to all others, but the program calculates the distances for each of 57000 vertices. 20 minutes are too slow. Thanks for helping me.

2 1

Re: [Boost-users] Library Interface Design
by Oliver.Kowalke＠qimonda.com 12 Sep '06

12 Sep '06

Hello Scott, > I've always been a big fan of the "don't let users create > objects they can't really use" philosophy. I also prefer this style because it could be cumbersome to detect the source of exceptions thrown by half-initialized objects in complex applications. If half-initialized objects are allowed the library developer has to provide descriptive exceptions. with best regards, Oliver

1 0

array of strings in boost python
by ctrembla＠pcigeomatics.com 12 Sep '06

12 Sep '06

Hi, newbie in python and boost::python I have a method in c++ that returns a vector of strings, how can i expose this method in python, and how should i use it in python class A { public: std::vector<std::string> getName(); }; BOOST_PYTHON_MODULE(pysession) { class_<A>("A") .def("GetName" ,&A::getName) } now in python how could i call the function getName() ? here is my error in python TypeError: No to_python (by-value) converter found for C++ type: class std::vect or<class std::basic_string<char,struct std::char_traits<char>,class std::allocat or<char> >,class std::allocator<class std::basic_string<char,struct std::char_tr aits<char>,class std::allocator<char> > > > not valid thanks, Charles.

3 3

Re: [Boost-users] 3 times slower than c++
by Qinfeng(Javen) Shi 11 Sep '06

11 Sep '06

Dear all, We are writing a python toolbox and wrapping it from c++ is much faster than pure python. But my wrapped c++ code(called in python) is 3 times(not plus a constant time) slower than the orignial c++ one. I think the difference should be a constant not a factor. Because all the computation has been taken in *.so, the extra time cost is the time python load it at the beginning. Is that right? Any opinion or suggestion about why or how to speed it up? Best wishes! Javen On 9/11/06, boost-users-request(a)lists.boost.org < boost-users-request(a)lists.boost.org> wrote: > > Send Boost-users mailing list submissions to > boost-users(a)lists.boost.org > > To subscribe or unsubscribe via the World Wide Web, visit > http://lists.boost.org/mailman/listinfo.cgi/boost-users > or, via email, send a message with subject or body 'help' to > boost-users-request(a)lists.boost.org > > You can reach the person managing the list at > boost-users-owner(a)lists.boost.org > > When replying, please edit your Subject line so it is more specific > than "Re: Contents of Boost-users digest..." > > > Today's Topics: > > 1. Re: having a problem seeding random number generator. (Jim Ingram) > 2. Re: [string_algo] wildcard support? (Pavol Droba) > 3. Re: Absolute Quanfusion : was Re: [fusion] transform a vector > of values to refs (Andy Little) > 4. Properties (newbie) (Olivier Tournaire) > 5. 3 times slower than c++? (Qinfeng(Javen) Shi ) > 6. Re: 3 times slower than c++? (Andy Little) > 7. Re: Lock pointer implementation (Mattias Br?ndstr?m) > 8. Re: Lock pointer implementation (Mattias Brandstrom) > 9. Re: Link error with boost regex and STLport (Maik Beckmann) > > > ---------------------------------------------------------------------- > > Message: 1 > Date: Sun, 10 Sep 2006 15:38:41 +0000 (UTC) > From: Jim Ingram <ingramj(a)gmail.com> > Subject: Re: [Boost-users] having a problem seeding random number > generator. > To: boost-users(a)lists.boost.org > Message-ID: <loom.20060910T172921-895(a)post.gmane.org> > Content-Type: text/plain; charset=us-ascii > > Daniel James <daniel_james <at> fmail.co.uk> writes: > > > The problem is that 'variate_generator' stores a copy of the engine, not > > a reference. > > Wow. You'd think that between the docs, the header files, and the example > code, > I would have caught that. It does make a lot more sense to have the > generator > and distribution encapsulated inside a variate_generator instead of > maintaining > them seperately. > > Thanks for pointing that out. > > -- > Jim > > > > ------------------------------ > > Message: 2 > Date: Sun, 10 Sep 2006 20:29:09 +0200 > From: Pavol Droba <droba(a)topmail.sk> > Subject: Re: [Boost-users] [string_algo] wildcard support? > To: boost-users(a)lists.boost.org > Message-ID: <45045975.1090701(a)topmail.sk> > Content-Type: text/plain; charset=ISO-8859-1; format=flowed > > Hi, > > Andriy Tylychko (mail.ru) wrote: > > > > > > To community and especially to Pavol Droba as the author of string_algo: > > What do you think about popularity of wildcards as the tool for > end-users? > > Is it worth to add wildcards support to the library? To what library? Is > the > > boost-development mailing list a better place for this discussion? > > I can only express my personal opinion here and that is : the wildcards > should not be a directly supported in string_algo library. > > Reason is simple: the functionality provided by wildcards is superseeded > by regex in all respects. In addition, as you have explained in your > previous mail, it is easy to convert a wildcard to a regex. > > I can't say if the support could be incorporated into Boost.Regex. > This library already support several syntaxes, so wildcards could be > just another variant. But this is more question to John Maddock. > > You can also try to cleanup your conversion utility and offer it to > boost, to get a response from whole comunity. > > Regards, > Pavol > > > ------------------------------ > > Message: 3 > Date: Sun, 10 Sep 2006 19:33:32 +0100 > From: "Andy Little" <andy(a)servocomm.freeserve.co.uk> > Subject: Re: [Boost-users] Absolute Quanfusion : was Re: [fusion] > transform a vector of values to refs > To: boost-users(a)lists.boost.org > Message-ID: <ee1lpe$trq$1(a)sea.gmane.org> > Content-Type: text/plain; format=flowed; charset="utf-8"; > reply-type=original > > > "Joel de Guzman" <joel(a)boost-consulting.com> wrote in message > news:edv1e0$n73$1@sea.gmane.org... > > > > > Sounds good. Okidoki. > > Great. I will see if I can get on with it, rather than spend more time > mucking > about with fusion ;-) > > regards > Andy Little > > > > > ------------------------------ > > Message: 4 > Date: Sun, 10 Sep 2006 21:57:17 +0200 > From: Olivier Tournaire <olitour(a)yahoo.fr> > Subject: [Boost-users] Properties (newbie) > To: boost-users(a)lists.boost.org > Message-ID: <45046E1D.7040709(a)yahoo.fr> > Content-Type: text/plain; charset=ISO-8859-1; format=flowed > > Hi all, > > I am a new boost user (2 days old ...). I just want to know a few things. > > 1. I use an adjacency_list with my own properties to store my graph. I > just want to know if it is possible to manipulate directly the > properties' values. At the moment, I use get(...) to obtain the > property, make my change, and use put(...) to set the new value. > > 2. My graph cannot contain parallel edges, so I use a setS. However, the > documentation says that listS or vecS is more efficient in time and > space complexity. Is it possible to use std::vector or std::list and to > disallow parallele edges ? > > Thanks in advance > Olivier > > > ------------------------------ > > Message: 5 > Date: Mon, 11 Sep 2006 09:54:43 +1000 > From: "Qinfeng(Javen) Shi " <shiqinfeng(a)gmail.com> > Subject: [Boost-users] 3 times slower than c++? > To: boost-users(a)lists.boost.org > Message-ID: > <fe76f74b0609101654o12a61256w3b6f7c4ce24cd55a(a)mail.gmail.com> > Content-Type: text/plain; charset="iso-8859-1" > > Dear all, > > I wrapped a c++ code into python. It works but it is 3 times slower than > the > orignial c++ one. > Is that normal? Is there something I can do to speed it up? > > Thanks! > > -- > Qinfeng(Javen) Shi > > Research School of Information Sciences and Engineering > Australian National University > Locked Bag 8001 > Canberra ACT 2601 >

2 1

bcp not found ?
by Shintaro Fujiwara 11 Sep '06

11 Sep '06

Hi. I tried to extract some subset of boost by bcp command but couldn't make it. [root@mama lib]# bcp --boost="/usr/local/lib" boost/regex.hpp "/home/dir/projects/segatex2.15" -bash: bcp: command not found [root@mama lib]# rpm -qa|grep boost boost-jam-3.1.13-1 boost-devel-1.33.1-5 boost-doc-1.33.1-5 boost-1.33.1-5 What's wrong or what's missing? Thanks in advance.

1 0

Re: [Boost-users] 3 times slower than c++
by Sohail Somani 11 Sep '06

11 Sep '06

> -----Original Message----- > From: boost-users-bounces(a)lists.boost.org > [mailto:boost-users-bounces@lists.boost.org] On Behalf Of > Qinfeng(Javen) Shi > Sent: Monday, September 11, 2006 3:46 AM > To: boost-users(a)lists.boost.org > Subject: Re: [Boost-users] 3 times slower than c++ > > Dear all, > > We are writing a python toolbox and wrapping it from c++ is > much faster than pure python. > But my wrapped c++ code(called in python) is 3 times(not plus > a constant time) slower than the orignial c++ one. > > I think the difference should be a constant not a factor. > Because all the computation has been taken in *.so, the extra > time cost is the time python load it at the beginning. > Is that right? > > Any opinion or suggestion about why or how to speed it up? Are you wrapping with boost.python?

1 0

3 times slower than c++?
by Qinfeng(Javen) Shi 11 Sep '06

11 Sep '06

Dear all, I wrapped a c++ code into python. It works but it is 3 times slower than the orignial c++ one. Is that normal? Is there something I can do to speed it up? Thanks! -- Qinfeng(Javen) Shi Research School of Information Sciences and Engineering Australian National University Locked Bag 8001 Canberra ACT 2601

4 4

Re: [Boost-users] array of strings in boost python
by ctrembla＠pcigeomatics.com 11 Sep '06

11 Sep '06

In article <87k64aad0w.fsf__22758.9953029112$1157972153$gmane $org(a)pereiro.luannocracy.com>, dave(a)boost-consulting.com says... > <ctrembla(a)pcigeomatics.com> writes: > > > Hi, > > > > newbie in python and boost::python > > Please post Boost.Python questions to the C++-sig: > http://www.boost.org/more/mailing_lists.htm#cplussig > > > I have a method in c++ that returns a vector of strings, how can i > > expose this method in python, and how should i use it in python > > Sorry to turn the question around on you, but how do you /want/ to be > able to use it in Python? That will determine how you should expose > it. > > You might look into > http://boost.org/libs/python/doc/v2/indexing.html#vector_indexing_suite_cla… > to see if it meets your needs. > > > hi, i have exposed my c++ class using the python.boost BOOST_PYTHON_MODULE(pysession) { class_<Session> >("Session") .def("Connect" ,&Session::Connect) .def("Disconnect",&Session::Disconnect) .def("SaveSession",& Session::SaveSession) .def("SaveSessionAs", &Session ::SaveSessionAs) .def("DeleteSession", & Session::DeleteSession) .def("GetListOfServer",& Session ::GetListOfServer) ; class_<std::vector<std::string> >("STD_String") .def(vector_indexing_suite<std::vector<std::string> >()) ; } i have compiled and it generated my pyd now in python , this is were i'm stuck, i was able to retreive the iterator, how can i iterate and how can i retreive the string within the vector? oString = self.oLogin.GetListOfServer() i=0 oList = range(oString.__len__()) oIter = oString.__iter__() for i in oList: oIter = oIter.next() error log oIter = oIter.next() TypeError: No Python class registered for C++ class class std::basic_string<char ,struct std::char_traits<char>,class std::allocator<char> > thanks, charles.

1 0

boost::thread_group signal for exception and exit
by Andrew Wingorodov 11 Sep '06

11 Sep '06

What interface exists for to send a signal of exit for boost::thread ? -- andr.ru

1 0

Link error with boost regex and STLport
by Gérald FAUVELLE 11 Sep '06

11 Sep '06

Hello, I've searched in both stlport and boost forums, but I couldn't find anything yet about my problem. I'm using boost 1.33, and I tried to compiled with STL port. Boost compiles fine (except boost python, but I'll have a look later) and when I link my program, I have the following errors : boost.lib(w32_regex_traits.obj) : error LNK2019: unresolved external symbol "__declspec(dllimport) public: __thiscall stlp_std::basic_string<unsigned short,class stlp_std::char_traits<unsigned short>,class stlp_std::allocator<unsigned short> >::basic_string<unsigned short,class stlp_std::char_traits<unsigned short>,class stlp_std::allocator<unsigned short> >(unsigned short const *,class stlp_std::allocator<unsigned short> const &)" Do you have an idea of what could be the problem, and how to solve it ? I tried with STLport 4.6.2, 5.0.2 and 5.1, and I have the same error. Thanks, Gérald

3 7